summaryrefslogtreecommitdiffstats
path: root/arch/x86/platform
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/platform')
-rw-r--r--arch/x86/platform/Makefile16
-rw-r--r--arch/x86/platform/atom/Makefile1
-rw-r--r--arch/x86/platform/atom/punit_atom_debug.c181
-rw-r--r--arch/x86/platform/ce4100/Makefile1
-rw-r--r--arch/x86/platform/ce4100/ce4100.c160
-rw-r--r--arch/x86/platform/ce4100/falconfalls.dts433
-rw-r--r--arch/x86/platform/efi/Makefile7
-rw-r--r--arch/x86/platform/efi/early_printk.c236
-rw-r--r--arch/x86/platform/efi/efi.c1046
-rw-r--r--arch/x86/platform/efi/efi_32.c97
-rw-r--r--arch/x86/platform/efi/efi_64.c1039
-rw-r--r--arch/x86/platform/efi/efi_stub_32.S124
-rw-r--r--arch/x86/platform/efi/efi_stub_64.S58
-rw-r--r--arch/x86/platform/efi/efi_thunk_64.S153
-rw-r--r--arch/x86/platform/efi/quirks.c655
-rw-r--r--arch/x86/platform/geode/Makefile3
-rw-r--r--arch/x86/platform/geode/alix.c200
-rw-r--r--arch/x86/platform/geode/geos.c122
-rw-r--r--arch/x86/platform/geode/net5501.c148
-rw-r--r--arch/x86/platform/goldfish/Makefile1
-rw-r--r--arch/x86/platform/goldfish/goldfish.c63
-rw-r--r--arch/x86/platform/intel-mid/Makefile6
-rw-r--r--arch/x86/platform/intel-mid/device_libs/Makefile33
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c95
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_bma023.c20
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_bt.c108
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_emc1403.c43
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_gpio_keys.c85
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_lis331.c41
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_max7315.c81
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mpu3050.c36
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mrfld_pinctrl.c43
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mrfld_power_btn.c82
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mrfld_rtc.c48
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c47
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mrfld_spidev.c54
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_mrfld_wdt.c86
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic.c87
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic.h19
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic_audio.c46
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic_battery.c36
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic_gpio.c47
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic_ocd.c48
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic_power_btn.c35
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_msic_thermal.c36
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_pcal9555a.c99
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_tc35876x.c36
-rw-r--r--arch/x86/platform/intel-mid/device_libs/platform_tca6416.c57
-rw-r--r--arch/x86/platform/intel-mid/intel-mid.c216
-rw-r--r--arch/x86/platform/intel-mid/intel_mid_vrtc.c177
-rw-r--r--arch/x86/platform/intel-mid/pwr.c488
-rw-r--r--arch/x86/platform/intel-mid/sfi.c547
-rw-r--r--arch/x86/platform/intel-quark/Makefile2
-rw-r--r--arch/x86/platform/intel-quark/imr.c600
-rw-r--r--arch/x86/platform/intel-quark/imr_selftest.c127
-rw-r--r--arch/x86/platform/intel/Makefile1
-rw-r--r--arch/x86/platform/intel/iosf_mbi.c391
-rw-r--r--arch/x86/platform/iris/Makefile1
-rw-r--r--arch/x86/platform/iris/iris.c136
-rw-r--r--arch/x86/platform/olpc/Makefile6
-rw-r--r--arch/x86/platform/olpc/olpc-xo1-pm.c200
-rw-r--r--arch/x86/platform/olpc/olpc-xo1-rtc.c84
-rw-r--r--arch/x86/platform/olpc/olpc-xo1-sci.c642
-rw-r--r--arch/x86/platform/olpc/olpc-xo15-sci.c235
-rw-r--r--arch/x86/platform/olpc/olpc.c408
-rw-r--r--arch/x86/platform/olpc/olpc_dt.c304
-rw-r--r--arch/x86/platform/olpc/olpc_ofw.c120
-rw-r--r--arch/x86/platform/olpc/xo1-wakeup.S125
-rw-r--r--arch/x86/platform/scx200/Makefile2
-rw-r--r--arch/x86/platform/scx200/scx200_32.c129
-rw-r--r--arch/x86/platform/sfi/Makefile1
-rw-r--r--arch/x86/platform/sfi/sfi.c114
-rw-r--r--arch/x86/platform/ts5500/Makefile1
-rw-r--r--arch/x86/platform/ts5500/ts5500.c347
-rw-r--r--arch/x86/platform/uv/Makefile1
-rw-r--r--arch/x86/platform/uv/bios_uv.c240
-rw-r--r--arch/x86/platform/uv/tlb_uv.c2284
-rw-r--r--arch/x86/platform/uv/uv_irq.c217
-rw-r--r--arch/x86/platform/uv/uv_nmi.c1066
-rw-r--r--arch/x86/platform/uv/uv_sysfs.c76
-rw-r--r--arch/x86/platform/uv/uv_time.c416
81 files changed, 15902 insertions, 0 deletions
diff --git a/arch/x86/platform/Makefile b/arch/x86/platform/Makefile
new file mode 100644
index 000000000..d0e835470
--- /dev/null
+++ b/arch/x86/platform/Makefile
@@ -0,0 +1,16 @@
+# SPDX-License-Identifier: GPL-2.0
+# Platform specific code goes here
+obj-y += atom/
+obj-y += ce4100/
+obj-y += efi/
+obj-y += geode/
+obj-y += goldfish/
+obj-y += iris/
+obj-y += intel/
+obj-y += intel-mid/
+obj-y += intel-quark/
+obj-y += olpc/
+obj-y += scx200/
+obj-y += sfi/
+obj-y += ts5500/
+obj-y += uv/
diff --git a/arch/x86/platform/atom/Makefile b/arch/x86/platform/atom/Makefile
new file mode 100644
index 000000000..57be88fa3
--- /dev/null
+++ b/arch/x86/platform/atom/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_PUNIT_ATOM_DEBUG) += punit_atom_debug.o
diff --git a/arch/x86/platform/atom/punit_atom_debug.c b/arch/x86/platform/atom/punit_atom_debug.c
new file mode 100644
index 000000000..41dae0f0d
--- /dev/null
+++ b/arch/x86/platform/atom/punit_atom_debug.c
@@ -0,0 +1,181 @@
+/*
+ * Intel SOC Punit device state debug driver
+ * Punit controls power management for North Complex devices (Graphics
+ * blocks, Image Signal Processing, video processing, display, DSP etc.)
+ *
+ * Copyright (c) 2015, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/device.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/io.h>
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include <asm/iosf_mbi.h>
+
+/* Subsystem config/status Video processor */
+#define VED_SS_PM0 0x32
+/* Subsystem config/status ISP (Image Signal Processor) */
+#define ISP_SS_PM0 0x39
+/* Subsystem config/status Input/output controller */
+#define MIO_SS_PM 0x3B
+/* Shift bits for getting status for video, isp and i/o */
+#define SSS_SHIFT 24
+
+/* Power gate status reg */
+#define PWRGT_STATUS 0x61
+/* Shift bits for getting status for graphics rendering */
+#define RENDER_POS 0
+/* Shift bits for getting status for media control */
+#define MEDIA_POS 2
+/* Shift bits for getting status for Valley View/Baytrail display */
+#define VLV_DISPLAY_POS 6
+
+/* Subsystem config/status display for Cherry Trail SOC */
+#define CHT_DSP_SSS 0x36
+/* Shift bits for getting status for display */
+#define CHT_DSP_SSS_POS 16
+
+struct punit_device {
+ char *name;
+ int reg;
+ int sss_pos;
+};
+
+static const struct punit_device punit_device_tng[] = {
+ { "DISPLAY", CHT_DSP_SSS, SSS_SHIFT },
+ { "VED", VED_SS_PM0, SSS_SHIFT },
+ { "ISP", ISP_SS_PM0, SSS_SHIFT },
+ { "MIO", MIO_SS_PM, SSS_SHIFT },
+ { NULL }
+};
+
+static const struct punit_device punit_device_byt[] = {
+ { "GFX RENDER", PWRGT_STATUS, RENDER_POS },
+ { "GFX MEDIA", PWRGT_STATUS, MEDIA_POS },
+ { "DISPLAY", PWRGT_STATUS, VLV_DISPLAY_POS },
+ { "VED", VED_SS_PM0, SSS_SHIFT },
+ { "ISP", ISP_SS_PM0, SSS_SHIFT },
+ { "MIO", MIO_SS_PM, SSS_SHIFT },
+ { NULL }
+};
+
+static const struct punit_device punit_device_cht[] = {
+ { "GFX RENDER", PWRGT_STATUS, RENDER_POS },
+ { "GFX MEDIA", PWRGT_STATUS, MEDIA_POS },
+ { "DISPLAY", CHT_DSP_SSS, CHT_DSP_SSS_POS },
+ { "VED", VED_SS_PM0, SSS_SHIFT },
+ { "ISP", ISP_SS_PM0, SSS_SHIFT },
+ { "MIO", MIO_SS_PM, SSS_SHIFT },
+ { NULL }
+};
+
+static const char * const dstates[] = {"D0", "D0i1", "D0i2", "D0i3"};
+
+static int punit_dev_state_show(struct seq_file *seq_file, void *unused)
+{
+ u32 punit_pwr_status;
+ struct punit_device *punit_devp = seq_file->private;
+ int index;
+ int status;
+
+ seq_puts(seq_file, "\n\nPUNIT NORTH COMPLEX DEVICES :\n");
+ while (punit_devp->name) {
+ status = iosf_mbi_read(BT_MBI_UNIT_PMC, MBI_REG_READ,
+ punit_devp->reg, &punit_pwr_status);
+ if (status) {
+ seq_printf(seq_file, "%9s : Read Failed\n",
+ punit_devp->name);
+ } else {
+ index = (punit_pwr_status >> punit_devp->sss_pos) & 3;
+ seq_printf(seq_file, "%9s : %s\n", punit_devp->name,
+ dstates[index]);
+ }
+ punit_devp++;
+ }
+
+ return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(punit_dev_state);
+
+static struct dentry *punit_dbg_file;
+
+static int punit_dbgfs_register(struct punit_device *punit_device)
+{
+ static struct dentry *dev_state;
+
+ punit_dbg_file = debugfs_create_dir("punit_atom", NULL);
+ if (!punit_dbg_file)
+ return -ENXIO;
+
+ dev_state = debugfs_create_file("dev_power_state", 0444,
+ punit_dbg_file, punit_device,
+ &punit_dev_state_fops);
+ if (!dev_state) {
+ pr_err("punit_dev_state register failed\n");
+ debugfs_remove(punit_dbg_file);
+ return -ENXIO;
+ }
+
+ return 0;
+}
+
+static void punit_dbgfs_unregister(void)
+{
+ debugfs_remove_recursive(punit_dbg_file);
+}
+
+#define ICPU(model, drv_data) \
+ { X86_VENDOR_INTEL, 6, model, X86_FEATURE_MWAIT,\
+ (kernel_ulong_t)&drv_data }
+
+static const struct x86_cpu_id intel_punit_cpu_ids[] = {
+ ICPU(INTEL_FAM6_ATOM_SILVERMONT, punit_device_byt),
+ ICPU(INTEL_FAM6_ATOM_SILVERMONT_MID, punit_device_tng),
+ ICPU(INTEL_FAM6_ATOM_AIRMONT, punit_device_cht),
+ {}
+};
+
+MODULE_DEVICE_TABLE(x86cpu, intel_punit_cpu_ids);
+
+static int __init punit_atom_debug_init(void)
+{
+ const struct x86_cpu_id *id;
+ int ret;
+
+ id = x86_match_cpu(intel_punit_cpu_ids);
+ if (!id)
+ return -ENODEV;
+
+ ret = punit_dbgfs_register((struct punit_device *)id->driver_data);
+ if (ret < 0)
+ return ret;
+
+ return 0;
+}
+
+static void __exit punit_atom_debug_exit(void)
+{
+ punit_dbgfs_unregister();
+}
+
+module_init(punit_atom_debug_init);
+module_exit(punit_atom_debug_exit);
+
+MODULE_AUTHOR("Kumar P, Mahesh <mahesh.kumar.p@intel.com>");
+MODULE_AUTHOR("Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>");
+MODULE_DESCRIPTION("Driver for Punit devices states debugging");
+MODULE_LICENSE("GPL v2");
diff --git a/arch/x86/platform/ce4100/Makefile b/arch/x86/platform/ce4100/Makefile
new file mode 100644
index 000000000..91fc92971
--- /dev/null
+++ b/arch/x86/platform/ce4100/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_X86_INTEL_CE) += ce4100.o
diff --git a/arch/x86/platform/ce4100/ce4100.c b/arch/x86/platform/ce4100/ce4100.c
new file mode 100644
index 000000000..ce4b06733
--- /dev/null
+++ b/arch/x86/platform/ce4100/ce4100.c
@@ -0,0 +1,160 @@
+/*
+ * Intel CE4100 platform specific setup code
+ *
+ * (C) Copyright 2010 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/irq.h>
+#include <linux/reboot.h>
+#include <linux/serial_reg.h>
+#include <linux/serial_8250.h>
+
+#include <asm/ce4100.h>
+#include <asm/prom.h>
+#include <asm/setup.h>
+#include <asm/i8259.h>
+#include <asm/io.h>
+#include <asm/io_apic.h>
+#include <asm/emergency-restart.h>
+
+/*
+ * The CE4100 platform has an internal 8051 Microcontroller which is
+ * responsible for signaling to the external Power Management Unit the
+ * intention to reset, reboot or power off the system. This 8051 device has
+ * its command register mapped at I/O port 0xcf9 and the value 0x4 is used
+ * to power off the system.
+ */
+static void ce4100_power_off(void)
+{
+ outb(0x4, 0xcf9);
+}
+
+#ifdef CONFIG_SERIAL_8250
+
+static unsigned int mem_serial_in(struct uart_port *p, int offset)
+{
+ offset = offset << p->regshift;
+ return readl(p->membase + offset);
+}
+
+/*
+ * The UART Tx interrupts are not set under some conditions and therefore serial
+ * transmission hangs. This is a silicon issue and has not been root caused. The
+ * workaround for this silicon issue checks UART_LSR_THRE bit and UART_LSR_TEMT
+ * bit of LSR register in interrupt handler to see whether at least one of these
+ * two bits is set, if so then process the transmit request. If this workaround
+ * is not applied, then the serial transmission may hang. This workaround is for
+ * errata number 9 in Errata - B step.
+*/
+
+static unsigned int ce4100_mem_serial_in(struct uart_port *p, int offset)
+{
+ unsigned int ret, ier, lsr;
+
+ if (offset == UART_IIR) {
+ offset = offset << p->regshift;
+ ret = readl(p->membase + offset);
+ if (ret & UART_IIR_NO_INT) {
+ /* see if the TX interrupt should have really set */
+ ier = mem_serial_in(p, UART_IER);
+ /* see if the UART's XMIT interrupt is enabled */
+ if (ier & UART_IER_THRI) {
+ lsr = mem_serial_in(p, UART_LSR);
+ /* now check to see if the UART should be
+ generating an interrupt (but isn't) */
+ if (lsr & (UART_LSR_THRE | UART_LSR_TEMT))
+ ret &= ~UART_IIR_NO_INT;
+ }
+ }
+ } else
+ ret = mem_serial_in(p, offset);
+ return ret;
+}
+
+static void ce4100_mem_serial_out(struct uart_port *p, int offset, int value)
+{
+ offset = offset << p->regshift;
+ writel(value, p->membase + offset);
+}
+
+static void ce4100_serial_fixup(int port, struct uart_port *up,
+ u32 *capabilites)
+{
+#ifdef CONFIG_EARLY_PRINTK
+ /*
+ * Over ride the legacy port configuration that comes from
+ * asm/serial.h. Using the ioport driver then switching to the
+ * PCI memmaped driver hangs the IOAPIC
+ */
+ if (up->iotype != UPIO_MEM32) {
+ up->uartclk = 14745600;
+ up->mapbase = 0xdffe0200;
+ set_fixmap_nocache(FIX_EARLYCON_MEM_BASE,
+ up->mapbase & PAGE_MASK);
+ up->membase =
+ (void __iomem *)__fix_to_virt(FIX_EARLYCON_MEM_BASE);
+ up->membase += up->mapbase & ~PAGE_MASK;
+ up->mapbase += port * 0x100;
+ up->membase += port * 0x100;
+ up->iotype = UPIO_MEM32;
+ up->regshift = 2;
+ up->irq = 4;
+ }
+#endif
+ up->iobase = 0;
+ up->serial_in = ce4100_mem_serial_in;
+ up->serial_out = ce4100_mem_serial_out;
+
+ *capabilites |= (1 << 12);
+}
+
+static __init void sdv_serial_fixup(void)
+{
+ serial8250_set_isa_configurator(ce4100_serial_fixup);
+}
+
+#else
+static inline void sdv_serial_fixup(void) {};
+#endif
+
+static void __init sdv_arch_setup(void)
+{
+ sdv_serial_fixup();
+}
+
+static void sdv_pci_init(void)
+{
+ x86_of_pci_init();
+}
+
+/*
+ * CE4100 specific x86_init function overrides and early setup
+ * calls.
+ */
+void __init x86_ce4100_early_setup(void)
+{
+ x86_init.oem.arch_setup = sdv_arch_setup;
+ x86_init.resources.probe_roms = x86_init_noop;
+ x86_init.mpparse.get_smp_config = x86_init_uint_noop;
+ x86_init.mpparse.find_smp_config = x86_init_noop;
+ x86_init.mpparse.setup_ioapic_ids = setup_ioapic_ids_from_mpc_nocheck;
+ x86_init.pci.init = ce4100_pci_init;
+ x86_init.pci.init_irq = sdv_pci_init;
+
+ /*
+ * By default, the reboot method is ACPI which is supported by the
+ * CE4100 bootloader CEFDK using FADT.ResetReg Address and ResetValue
+ * the bootloader will however issue a system power off instead of
+ * reboot. By using BOOT_KBD we ensure proper system reboot as
+ * expected.
+ */
+ reboot_type = BOOT_KBD;
+
+ pm_power_off = ce4100_power_off;
+}
diff --git a/arch/x86/platform/ce4100/falconfalls.dts b/arch/x86/platform/ce4100/falconfalls.dts
new file mode 100644
index 000000000..ce874f872
--- /dev/null
+++ b/arch/x86/platform/ce4100/falconfalls.dts
@@ -0,0 +1,433 @@
+/*
+ * CE4100 on Falcon Falls
+ *
+ * (c) Copyright 2010 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; version 2 of the License.
+ */
+/dts-v1/;
+/ {
+ model = "intel,falconfalls";
+ compatible = "intel,falconfalls";
+ #address-cells = <1>;
+ #size-cells = <1>;
+
+ cpus {
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ cpu@0 {
+ device_type = "cpu";
+ compatible = "intel,ce4100";
+ reg = <0>;
+ lapic = <&lapic0>;
+ };
+ };
+
+ soc@0 {
+ #address-cells = <1>;
+ #size-cells = <1>;
+ compatible = "intel,ce4100-cp";
+ ranges;
+
+ ioapic1: interrupt-controller@fec00000 {
+ #interrupt-cells = <2>;
+ compatible = "intel,ce4100-ioapic";
+ interrupt-controller;
+ reg = <0xfec00000 0x1000>;
+ };
+
+ timer@fed00000 {
+ compatible = "intel,ce4100-hpet";
+ reg = <0xfed00000 0x200>;
+ };
+
+ lapic0: interrupt-controller@fee00000 {
+ compatible = "intel,ce4100-lapic";
+ reg = <0xfee00000 0x1000>;
+ };
+
+ pci@3fc {
+ #address-cells = <3>;
+ #size-cells = <2>;
+ compatible = "intel,ce4100-pci", "pci";
+ device_type = "pci";
+ bus-range = <0 0>;
+ ranges = <0x2000000 0 0xbffff000 0xbffff000 0 0x1000
+ 0x2000000 0 0xdffe0000 0xdffe0000 0 0x1000
+ 0x0000000 0 0x0 0x0 0 0x100>;
+
+ /* Secondary IO-APIC */
+ ioapic2: interrupt-controller@0,1 {
+ #interrupt-cells = <2>;
+ compatible = "intel,ce4100-ioapic";
+ interrupt-controller;
+ reg = <0x100 0x0 0x0 0x0 0x0>;
+ assigned-addresses = <0x02000000 0x0 0xbffff000 0x0 0x1000>;
+ };
+
+ pci@1,0 {
+ #address-cells = <3>;
+ #size-cells = <2>;
+ compatible = "intel,ce4100-pci", "pci";
+ device_type = "pci";
+ bus-range = <1 1>;
+ reg = <0x0800 0x0 0x0 0x0 0x0>;
+ ranges = <0x2000000 0 0xdffe0000 0x2000000 0 0xdffe0000 0 0x1000>;
+
+ interrupt-parent = <&ioapic2>;
+
+ display@2,0 {
+ compatible = "pci8086,2e5b.2",
+ "pci8086,2e5b",
+ "pciclass038000",
+ "pciclass0380";
+
+ reg = <0x11000 0x0 0x0 0x0 0x0>;
+ interrupts = <0 1>;
+ };
+
+ multimedia@3,0 {
+ compatible = "pci8086,2e5c.2",
+ "pci8086,2e5c",
+ "pciclass048000",
+ "pciclass0480";
+
+ reg = <0x11800 0x0 0x0 0x0 0x0>;
+ interrupts = <2 1>;
+ };
+
+ multimedia@4,0 {
+ compatible = "pci8086,2e5d.2",
+ "pci8086,2e5d",
+ "pciclass048000",
+ "pciclass0480";
+
+ reg = <0x12000 0x0 0x0 0x0 0x0>;
+ interrupts = <4 1>;
+ };
+
+ multimedia@4,1 {
+ compatible = "pci8086,2e5e.2",
+ "pci8086,2e5e",
+ "pciclass048000",
+ "pciclass0480";
+
+ reg = <0x12100 0x0 0x0 0x0 0x0>;
+ interrupts = <5 1>;
+ };
+
+ sound@6,0 {
+ compatible = "pci8086,2e5f.2",
+ "pci8086,2e5f",
+ "pciclass040100",
+ "pciclass0401";
+
+ reg = <0x13000 0x0 0x0 0x0 0x0>;
+ interrupts = <6 1>;
+ };
+
+ sound@6,1 {
+ compatible = "pci8086,2e5f.2",
+ "pci8086,2e5f",
+ "pciclass040100",
+ "pciclass0401";
+
+ reg = <0x13100 0x0 0x0 0x0 0x0>;
+ interrupts = <7 1>;
+ };
+
+ sound@6,2 {
+ compatible = "pci8086,2e60.2",
+ "pci8086,2e60",
+ "pciclass040100",
+ "pciclass0401";
+
+ reg = <0x13200 0x0 0x0 0x0 0x0>;
+ interrupts = <8 1>;
+ };
+
+ display@8,0 {
+ compatible = "pci8086,2e61.2",
+ "pci8086,2e61",
+ "pciclass038000",
+ "pciclass0380";
+
+ reg = <0x14000 0x0 0x0 0x0 0x0>;
+ interrupts = <9 1>;
+ };
+
+ display@8,1 {
+ compatible = "pci8086,2e62.2",
+ "pci8086,2e62",
+ "pciclass038000",
+ "pciclass0380";
+
+ reg = <0x14100 0x0 0x0 0x0 0x0>;
+ interrupts = <10 1>;
+ };
+
+ multimedia@8,2 {
+ compatible = "pci8086,2e63.2",
+ "pci8086,2e63",
+ "pciclass048000",
+ "pciclass0480";
+
+ reg = <0x14200 0x0 0x0 0x0 0x0>;
+ interrupts = <11 1>;
+ };
+
+ entertainment-encryption@9,0 {
+ compatible = "pci8086,2e64.2",
+ "pci8086,2e64",
+ "pciclass101000",
+ "pciclass1010";
+
+ reg = <0x14800 0x0 0x0 0x0 0x0>;
+ interrupts = <12 1>;
+ };
+
+ localbus@a,0 {
+ compatible = "pci8086,2e65.2",
+ "pci8086,2e65",
+ "pciclassff0000",
+ "pciclassff00";
+
+ reg = <0x15000 0x0 0x0 0x0 0x0>;
+ };
+
+ serial@b,0 {
+ compatible = "pci8086,2e66.2",
+ "pci8086,2e66",
+ "pciclass070003",
+ "pciclass0700";
+
+ reg = <0x15800 0x0 0x0 0x0 0x0>;
+ interrupts = <14 1>;
+ };
+
+ pcigpio: gpio@b,1 {
+ #gpio-cells = <2>;
+ #interrupt-cells = <2>;
+ compatible = "pci8086,2e67.2",
+ "pci8086,2e67",
+ "pciclassff0000",
+ "pciclassff00";
+
+ reg = <0x15900 0x0 0x0 0x0 0x0>;
+ interrupts = <15 1>;
+ interrupt-controller;
+ gpio-controller;
+ intel,muxctl = <0>;
+ };
+
+ i2c-controller@b,2 {
+ #address-cells = <2>;
+ #size-cells = <1>;
+ compatible = "pci8086,2e68.2",
+ "pci8086,2e68",
+ "pciclass,ff0000",
+ "pciclass,ff00";
+
+ reg = <0x15a00 0x0 0x0 0x0 0x0>;
+ interrupts = <16 1>;
+ ranges = <0 0 0x02000000 0 0xdffe0500 0x100
+ 1 0 0x02000000 0 0xdffe0600 0x100
+ 2 0 0x02000000 0 0xdffe0700 0x100>;
+
+ i2c@0 {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ compatible = "intel,ce4100-i2c-controller";
+ reg = <0 0 0x100>;
+ };
+
+ i2c@1 {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ compatible = "intel,ce4100-i2c-controller";
+ reg = <1 0 0x100>;
+
+ gpio@26 {
+ #gpio-cells = <2>;
+ compatible = "ti,pcf8575";
+ reg = <0x26>;
+ gpio-controller;
+ };
+ };
+
+ i2c@2 {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ compatible = "intel,ce4100-i2c-controller";
+ reg = <2 0 0x100>;
+
+ gpio@26 {
+ #gpio-cells = <2>;
+ compatible = "ti,pcf8575";
+ reg = <0x26>;
+ gpio-controller;
+ };
+ };
+ };
+
+ smard-card@b,3 {
+ compatible = "pci8086,2e69.2",
+ "pci8086,2e69",
+ "pciclass070500",
+ "pciclass0705";
+
+ reg = <0x15b00 0x0 0x0 0x0 0x0>;
+ interrupts = <15 1>;
+ };
+
+ spi-controller@b,4 {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ compatible =
+ "pci8086,2e6a.2",
+ "pci8086,2e6a",
+ "pciclass,ff0000",
+ "pciclass,ff00";
+
+ reg = <0x15c00 0x0 0x0 0x0 0x0>;
+ interrupts = <15 1>;
+
+ dac@0 {
+ compatible = "ti,pcm1755";
+ reg = <0>;
+ spi-max-frequency = <115200>;
+ };
+
+ dac@1 {
+ compatible = "ti,pcm1609a";
+ reg = <1>;
+ spi-max-frequency = <115200>;
+ };
+
+ eeprom@2 {
+ compatible = "atmel,at93c46";
+ reg = <2>;
+ spi-max-frequency = <115200>;
+ };
+ };
+
+ multimedia@b,7 {
+ compatible = "pci8086,2e6d.2",
+ "pci8086,2e6d",
+ "pciclassff0000",
+ "pciclassff00";
+
+ reg = <0x15f00 0x0 0x0 0x0 0x0>;
+ };
+
+ ethernet@c,0 {
+ compatible = "pci8086,2e6e.2",
+ "pci8086,2e6e",
+ "pciclass020000",
+ "pciclass0200";
+
+ reg = <0x16000 0x0 0x0 0x0 0x0>;
+ interrupts = <21 1>;
+ };
+
+ clock@c,1 {
+ compatible = "pci8086,2e6f.2",
+ "pci8086,2e6f",
+ "pciclassff0000",
+ "pciclassff00";
+
+ reg = <0x16100 0x0 0x0 0x0 0x0>;
+ interrupts = <3 1>;
+ };
+
+ usb@d,0 {
+ compatible = "pci8086,2e70.2",
+ "pci8086,2e70",
+ "pciclass0c0320",
+ "pciclass0c03";
+
+ reg = <0x16800 0x0 0x0 0x0 0x0>;
+ interrupts = <22 1>;
+ };
+
+ usb@d,1 {
+ compatible = "pci8086,2e70.2",
+ "pci8086,2e70",
+ "pciclass0c0320",
+ "pciclass0c03";
+
+ reg = <0x16900 0x0 0x0 0x0 0x0>;
+ interrupts = <22 1>;
+ };
+
+ sata@e,0 {
+ compatible = "pci8086,2e71.0",
+ "pci8086,2e71",
+ "pciclass010601",
+ "pciclass0106";
+
+ reg = <0x17000 0x0 0x0 0x0 0x0>;
+ interrupts = <23 1>;
+ };
+
+ flash@f,0 {
+ compatible = "pci8086,701.1",
+ "pci8086,701",
+ "pciclass050100",
+ "pciclass0501";
+
+ reg = <0x17800 0x0 0x0 0x0 0x0>;
+ interrupts = <13 1>;
+ };
+
+ entertainment-encryption@10,0 {
+ compatible = "pci8086,702.1",
+ "pci8086,702",
+ "pciclass101000",
+ "pciclass1010";
+
+ reg = <0x18000 0x0 0x0 0x0 0x0>;
+ };
+
+ co-processor@11,0 {
+ compatible = "pci8086,703.1",
+ "pci8086,703",
+ "pciclass0b4000",
+ "pciclass0b40";
+
+ reg = <0x18800 0x0 0x0 0x0 0x0>;
+ interrupts = <1 1>;
+ };
+
+ multimedia@12,0 {
+ compatible = "pci8086,704.0",
+ "pci8086,704",
+ "pciclass048000",
+ "pciclass0480";
+
+ reg = <0x19000 0x0 0x0 0x0 0x0>;
+ };
+ };
+
+ isa@1f,0 {
+ #address-cells = <2>;
+ #size-cells = <1>;
+ compatible = "isa";
+ reg = <0xf800 0x0 0x0 0x0 0x0>;
+ ranges = <1 0 0 0 0 0x100>;
+
+ rtc@70 {
+ compatible = "intel,ce4100-rtc", "motorola,mc146818";
+ interrupts = <8 3>;
+ interrupt-parent = <&ioapic1>;
+ ctrl-reg = <2>;
+ freq-reg = <0x26>;
+ reg = <1 0x70 2>;
+ };
+ };
+ };
+ };
+};
diff --git a/arch/x86/platform/efi/Makefile b/arch/x86/platform/efi/Makefile
new file mode 100644
index 000000000..e4dc3862d
--- /dev/null
+++ b/arch/x86/platform/efi/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0
+OBJECT_FILES_NON_STANDARD_efi_thunk_$(BITS).o := y
+OBJECT_FILES_NON_STANDARD_efi_stub_$(BITS).o := y
+
+obj-$(CONFIG_EFI) += quirks.o efi.o efi_$(BITS).o efi_stub_$(BITS).o
+obj-$(CONFIG_EARLY_PRINTK_EFI) += early_printk.o
+obj-$(CONFIG_EFI_MIXED) += efi_thunk_$(BITS).o
diff --git a/arch/x86/platform/efi/early_printk.c b/arch/x86/platform/efi/early_printk.c
new file mode 100644
index 000000000..c3e6be110
--- /dev/null
+++ b/arch/x86/platform/efi/early_printk.c
@@ -0,0 +1,236 @@
+/*
+ * Copyright (C) 2013 Intel Corporation; author Matt Fleming
+ *
+ * This file is part of the Linux kernel, and is made available under
+ * the terms of the GNU General Public License version 2.
+ */
+
+#include <linux/console.h>
+#include <linux/efi.h>
+#include <linux/font.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <asm/setup.h>
+
+static const struct font_desc *font;
+static u32 efi_x, efi_y;
+static void *efi_fb;
+static bool early_efi_keep;
+
+/*
+ * efi earlyprintk need use early_ioremap to map the framebuffer.
+ * But early_ioremap is not usable for earlyprintk=efi,keep, ioremap should
+ * be used instead. ioremap will be available after paging_init() which is
+ * earlier than initcall callbacks. Thus adding this early initcall function
+ * early_efi_map_fb to map the whole efi framebuffer.
+ */
+static __init int early_efi_map_fb(void)
+{
+ unsigned long base, size;
+
+ if (!early_efi_keep)
+ return 0;
+
+ base = boot_params.screen_info.lfb_base;
+ size = boot_params.screen_info.lfb_size;
+ efi_fb = ioremap(base, size);
+
+ return efi_fb ? 0 : -ENOMEM;
+}
+early_initcall(early_efi_map_fb);
+
+/*
+ * early_efi_map maps efi framebuffer region [start, start + len -1]
+ * In case earlyprintk=efi,keep we have the whole framebuffer mapped already
+ * so just return the offset efi_fb + start.
+ */
+static __ref void *early_efi_map(unsigned long start, unsigned long len)
+{
+ unsigned long base;
+
+ base = boot_params.screen_info.lfb_base;
+
+ if (efi_fb)
+ return (efi_fb + start);
+ else
+ return early_ioremap(base + start, len);
+}
+
+static __ref void early_efi_unmap(void *addr, unsigned long len)
+{
+ if (!efi_fb)
+ early_iounmap(addr, len);
+}
+
+static void early_efi_clear_scanline(unsigned int y)
+{
+ unsigned long *dst;
+ u16 len;
+
+ len = boot_params.screen_info.lfb_linelength;
+ dst = early_efi_map(y*len, len);
+ if (!dst)
+ return;
+
+ memset(dst, 0, len);
+ early_efi_unmap(dst, len);
+}
+
+static void early_efi_scroll_up(void)
+{
+ unsigned long *dst, *src;
+ u16 len;
+ u32 i, height;
+
+ len = boot_params.screen_info.lfb_linelength;
+ height = boot_params.screen_info.lfb_height;
+
+ for (i = 0; i < height - font->height; i++) {
+ dst = early_efi_map(i*len, len);
+ if (!dst)
+ return;
+
+ src = early_efi_map((i + font->height) * len, len);
+ if (!src) {
+ early_efi_unmap(dst, len);
+ return;
+ }
+
+ memmove(dst, src, len);
+
+ early_efi_unmap(src, len);
+ early_efi_unmap(dst, len);
+ }
+}
+
+static void early_efi_write_char(u32 *dst, unsigned char c, unsigned int h)
+{
+ const u32 color_black = 0x00000000;
+ const u32 color_white = 0x00ffffff;
+ const u8 *src;
+ u8 s8;
+ int m;
+
+ src = font->data + c * font->height;
+ s8 = *(src + h);
+
+ for (m = 0; m < 8; m++) {
+ if ((s8 >> (7 - m)) & 1)
+ *dst = color_white;
+ else
+ *dst = color_black;
+ dst++;
+ }
+}
+
+static void
+early_efi_write(struct console *con, const char *str, unsigned int num)
+{
+ struct screen_info *si;
+ unsigned int len;
+ const char *s;
+ void *dst;
+
+ si = &boot_params.screen_info;
+ len = si->lfb_linelength;
+
+ while (num) {
+ unsigned int linemax;
+ unsigned int h, count = 0;
+
+ for (s = str; *s && *s != '\n'; s++) {
+ if (count == num)
+ break;
+ count++;
+ }
+
+ linemax = (si->lfb_width - efi_x) / font->width;
+ if (count > linemax)
+ count = linemax;
+
+ for (h = 0; h < font->height; h++) {
+ unsigned int n, x;
+
+ dst = early_efi_map((efi_y + h) * len, len);
+ if (!dst)
+ return;
+
+ s = str;
+ n = count;
+ x = efi_x;
+
+ while (n-- > 0) {
+ early_efi_write_char(dst + x*4, *s, h);
+ x += font->width;
+ s++;
+ }
+
+ early_efi_unmap(dst, len);
+ }
+
+ num -= count;
+ efi_x += count * font->width;
+ str += count;
+
+ if (num > 0 && *s == '\n') {
+ efi_x = 0;
+ efi_y += font->height;
+ str++;
+ num--;
+ }
+
+ if (efi_x + font->width > si->lfb_width) {
+ efi_x = 0;
+ efi_y += font->height;
+ }
+
+ if (efi_y + font->height > si->lfb_height) {
+ u32 i;
+
+ efi_y -= font->height;
+ early_efi_scroll_up();
+
+ for (i = 0; i < font->height; i++)
+ early_efi_clear_scanline(efi_y + i);
+ }
+ }
+}
+
+static __init int early_efi_setup(struct console *con, char *options)
+{
+ struct screen_info *si;
+ u16 xres, yres;
+ u32 i;
+
+ si = &boot_params.screen_info;
+ xres = si->lfb_width;
+ yres = si->lfb_height;
+
+ /*
+ * early_efi_write_char() implicitly assumes a framebuffer with
+ * 32-bits per pixel.
+ */
+ if (si->lfb_depth != 32)
+ return -ENODEV;
+
+ font = get_default_font(xres, yres, -1, -1);
+ if (!font)
+ return -ENODEV;
+
+ efi_y = rounddown(yres, font->height) - font->height;
+ for (i = 0; i < (yres - efi_y) / font->height; i++)
+ early_efi_scroll_up();
+
+ /* early_console_register will unset CON_BOOT in case ,keep */
+ if (!(con->flags & CON_BOOT))
+ early_efi_keep = true;
+ return 0;
+}
+
+struct console early_efi_console = {
+ .name = "earlyefi",
+ .write = early_efi_write,
+ .setup = early_efi_setup,
+ .flags = CON_PRINTBUFFER,
+ .index = -1,
+};
diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
new file mode 100644
index 000000000..e7f19dec1
--- /dev/null
+++ b/arch/x86/platform/efi/efi.c
@@ -0,0 +1,1046 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Common EFI (Extensible Firmware Interface) support functions
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 1999 VA Linux Systems
+ * Copyright (C) 1999 Walt Drummond <drummond@valinux.com>
+ * Copyright (C) 1999-2002 Hewlett-Packard Co.
+ * David Mosberger-Tang <davidm@hpl.hp.com>
+ * Stephane Eranian <eranian@hpl.hp.com>
+ * Copyright (C) 2005-2008 Intel Co.
+ * Fenghua Yu <fenghua.yu@intel.com>
+ * Bibo Mao <bibo.mao@intel.com>
+ * Chandramouli Narayanan <mouli@linux.intel.com>
+ * Huang Ying <ying.huang@intel.com>
+ * Copyright (C) 2013 SuSE Labs
+ * Borislav Petkov <bp@suse.de> - runtime services VA mapping
+ *
+ * Copied from efi_32.c to eliminate the duplicated code between EFI
+ * 32/64 support code. --ying 2007-10-26
+ *
+ * All EFI Runtime Services are not implemented yet as EFI only
+ * supports physical mode addressing on SoftSDV. This is to be fixed
+ * in a future version. --drummond 1999-07-20
+ *
+ * Implemented EFI runtime services and virtual mode calls. --davidm
+ *
+ * Goutham Rao: <goutham.rao@intel.com>
+ * Skip non-WB memory and ignore empty memory ranges.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/efi.h>
+#include <linux/efi-bgrt.h>
+#include <linux/export.h>
+#include <linux/bootmem.h>
+#include <linux/slab.h>
+#include <linux/memblock.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/time.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+#include <linux/bcd.h>
+
+#include <asm/setup.h>
+#include <asm/efi.h>
+#include <asm/e820/api.h>
+#include <asm/time.h>
+#include <asm/set_memory.h>
+#include <asm/tlbflush.h>
+#include <asm/x86_init.h>
+#include <asm/uv/uv.h>
+
+static struct efi efi_phys __initdata;
+static efi_system_table_t efi_systab __initdata;
+
+static efi_config_table_type_t arch_tables[] __initdata = {
+#ifdef CONFIG_X86_UV
+ {UV_SYSTEM_TABLE_GUID, "UVsystab", &efi.uv_systab},
+#endif
+ {NULL_GUID, NULL, NULL},
+};
+
+u64 efi_setup; /* efi setup_data physical address */
+
+static int add_efi_memmap __initdata;
+static int __init setup_add_efi_memmap(char *arg)
+{
+ add_efi_memmap = 1;
+ return 0;
+}
+early_param("add_efi_memmap", setup_add_efi_memmap);
+
+static efi_status_t __init phys_efi_set_virtual_address_map(
+ unsigned long memory_map_size,
+ unsigned long descriptor_size,
+ u32 descriptor_version,
+ efi_memory_desc_t *virtual_map)
+{
+ efi_status_t status;
+ unsigned long flags;
+ pgd_t *save_pgd;
+
+ save_pgd = efi_call_phys_prolog();
+
+ /* Disable interrupts around EFI calls: */
+ local_irq_save(flags);
+ status = efi_call_phys(efi_phys.set_virtual_address_map,
+ memory_map_size, descriptor_size,
+ descriptor_version, virtual_map);
+ local_irq_restore(flags);
+
+ efi_call_phys_epilog(save_pgd);
+
+ return status;
+}
+
+void __init efi_find_mirror(void)
+{
+ efi_memory_desc_t *md;
+ u64 mirror_size = 0, total_size = 0;
+
+ for_each_efi_memory_desc(md) {
+ unsigned long long start = md->phys_addr;
+ unsigned long long size = md->num_pages << EFI_PAGE_SHIFT;
+
+ total_size += size;
+ if (md->attribute & EFI_MEMORY_MORE_RELIABLE) {
+ memblock_mark_mirror(start, size);
+ mirror_size += size;
+ }
+ }
+ if (mirror_size)
+ pr_info("Memory: %lldM/%lldM mirrored memory\n",
+ mirror_size>>20, total_size>>20);
+}
+
+/*
+ * Tell the kernel about the EFI memory map. This might include
+ * more than the max 128 entries that can fit in the e820 legacy
+ * (zeropage) memory map.
+ */
+
+static void __init do_add_efi_memmap(void)
+{
+ efi_memory_desc_t *md;
+
+ for_each_efi_memory_desc(md) {
+ unsigned long long start = md->phys_addr;
+ unsigned long long size = md->num_pages << EFI_PAGE_SHIFT;
+ int e820_type;
+
+ switch (md->type) {
+ case EFI_LOADER_CODE:
+ case EFI_LOADER_DATA:
+ case EFI_BOOT_SERVICES_CODE:
+ case EFI_BOOT_SERVICES_DATA:
+ case EFI_CONVENTIONAL_MEMORY:
+ if (md->attribute & EFI_MEMORY_WB)
+ e820_type = E820_TYPE_RAM;
+ else
+ e820_type = E820_TYPE_RESERVED;
+ break;
+ case EFI_ACPI_RECLAIM_MEMORY:
+ e820_type = E820_TYPE_ACPI;
+ break;
+ case EFI_ACPI_MEMORY_NVS:
+ e820_type = E820_TYPE_NVS;
+ break;
+ case EFI_UNUSABLE_MEMORY:
+ e820_type = E820_TYPE_UNUSABLE;
+ break;
+ case EFI_PERSISTENT_MEMORY:
+ e820_type = E820_TYPE_PMEM;
+ break;
+ default:
+ /*
+ * EFI_RESERVED_TYPE EFI_RUNTIME_SERVICES_CODE
+ * EFI_RUNTIME_SERVICES_DATA EFI_MEMORY_MAPPED_IO
+ * EFI_MEMORY_MAPPED_IO_PORT_SPACE EFI_PAL_CODE
+ */
+ e820_type = E820_TYPE_RESERVED;
+ break;
+ }
+ e820__range_add(start, size, e820_type);
+ }
+ e820__update_table(e820_table);
+}
+
+int __init efi_memblock_x86_reserve_range(void)
+{
+ struct efi_info *e = &boot_params.efi_info;
+ struct efi_memory_map_data data;
+ phys_addr_t pmap;
+ int rv;
+
+ if (efi_enabled(EFI_PARAVIRT))
+ return 0;
+
+#ifdef CONFIG_X86_32
+ /* Can't handle data above 4GB at this time */
+ if (e->efi_memmap_hi) {
+ pr_err("Memory map is above 4GB, disabling EFI.\n");
+ return -EINVAL;
+ }
+ pmap = e->efi_memmap;
+#else
+ pmap = (e->efi_memmap | ((__u64)e->efi_memmap_hi << 32));
+#endif
+ data.phys_map = pmap;
+ data.size = e->efi_memmap_size;
+ data.desc_size = e->efi_memdesc_size;
+ data.desc_version = e->efi_memdesc_version;
+
+ rv = efi_memmap_init_early(&data);
+ if (rv)
+ return rv;
+
+ if (add_efi_memmap)
+ do_add_efi_memmap();
+
+ WARN(efi.memmap.desc_version != 1,
+ "Unexpected EFI_MEMORY_DESCRIPTOR version %ld",
+ efi.memmap.desc_version);
+
+ memblock_reserve(pmap, efi.memmap.nr_map * efi.memmap.desc_size);
+
+ return 0;
+}
+
+#define OVERFLOW_ADDR_SHIFT (64 - EFI_PAGE_SHIFT)
+#define OVERFLOW_ADDR_MASK (U64_MAX << OVERFLOW_ADDR_SHIFT)
+#define U64_HIGH_BIT (~(U64_MAX >> 1))
+
+static bool __init efi_memmap_entry_valid(const efi_memory_desc_t *md, int i)
+{
+ u64 end = (md->num_pages << EFI_PAGE_SHIFT) + md->phys_addr - 1;
+ u64 end_hi = 0;
+ char buf[64];
+
+ if (md->num_pages == 0) {
+ end = 0;
+ } else if (md->num_pages > EFI_PAGES_MAX ||
+ EFI_PAGES_MAX - md->num_pages <
+ (md->phys_addr >> EFI_PAGE_SHIFT)) {
+ end_hi = (md->num_pages & OVERFLOW_ADDR_MASK)
+ >> OVERFLOW_ADDR_SHIFT;
+
+ if ((md->phys_addr & U64_HIGH_BIT) && !(end & U64_HIGH_BIT))
+ end_hi += 1;
+ } else {
+ return true;
+ }
+
+ pr_warn_once(FW_BUG "Invalid EFI memory map entries:\n");
+
+ if (end_hi) {
+ pr_warn("mem%02u: %s range=[0x%016llx-0x%llx%016llx] (invalid)\n",
+ i, efi_md_typeattr_format(buf, sizeof(buf), md),
+ md->phys_addr, end_hi, end);
+ } else {
+ pr_warn("mem%02u: %s range=[0x%016llx-0x%016llx] (invalid)\n",
+ i, efi_md_typeattr_format(buf, sizeof(buf), md),
+ md->phys_addr, end);
+ }
+ return false;
+}
+
+static void __init efi_clean_memmap(void)
+{
+ efi_memory_desc_t *out = efi.memmap.map;
+ const efi_memory_desc_t *in = out;
+ const efi_memory_desc_t *end = efi.memmap.map_end;
+ int i, n_removal;
+
+ for (i = n_removal = 0; in < end; i++) {
+ if (efi_memmap_entry_valid(in, i)) {
+ if (out != in)
+ memcpy(out, in, efi.memmap.desc_size);
+ out = (void *)out + efi.memmap.desc_size;
+ } else {
+ n_removal++;
+ }
+ in = (void *)in + efi.memmap.desc_size;
+ }
+
+ if (n_removal > 0) {
+ u64 size = efi.memmap.nr_map - n_removal;
+
+ pr_warn("Removing %d invalid memory map entries.\n", n_removal);
+ efi_memmap_install(efi.memmap.phys_map, size);
+ }
+}
+
+void __init efi_print_memmap(void)
+{
+ efi_memory_desc_t *md;
+ int i = 0;
+
+ for_each_efi_memory_desc(md) {
+ char buf[64];
+
+ pr_info("mem%02u: %s range=[0x%016llx-0x%016llx] (%lluMB)\n",
+ i++, efi_md_typeattr_format(buf, sizeof(buf), md),
+ md->phys_addr,
+ md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT) - 1,
+ (md->num_pages >> (20 - EFI_PAGE_SHIFT)));
+ }
+}
+
+static int __init efi_systab_init(void *phys)
+{
+ if (efi_enabled(EFI_64BIT)) {
+ efi_system_table_64_t *systab64;
+ struct efi_setup_data *data = NULL;
+ u64 tmp = 0;
+
+ if (efi_setup) {
+ data = early_memremap(efi_setup, sizeof(*data));
+ if (!data)
+ return -ENOMEM;
+ }
+ systab64 = early_memremap((unsigned long)phys,
+ sizeof(*systab64));
+ if (systab64 == NULL) {
+ pr_err("Couldn't map the system table!\n");
+ if (data)
+ early_memunmap(data, sizeof(*data));
+ return -ENOMEM;
+ }
+
+ efi_systab.hdr = systab64->hdr;
+ efi_systab.fw_vendor = data ? (unsigned long)data->fw_vendor :
+ systab64->fw_vendor;
+ tmp |= data ? data->fw_vendor : systab64->fw_vendor;
+ efi_systab.fw_revision = systab64->fw_revision;
+ efi_systab.con_in_handle = systab64->con_in_handle;
+ tmp |= systab64->con_in_handle;
+ efi_systab.con_in = systab64->con_in;
+ tmp |= systab64->con_in;
+ efi_systab.con_out_handle = systab64->con_out_handle;
+ tmp |= systab64->con_out_handle;
+ efi_systab.con_out = systab64->con_out;
+ tmp |= systab64->con_out;
+ efi_systab.stderr_handle = systab64->stderr_handle;
+ tmp |= systab64->stderr_handle;
+ efi_systab.stderr = systab64->stderr;
+ tmp |= systab64->stderr;
+ efi_systab.runtime = data ?
+ (void *)(unsigned long)data->runtime :
+ (void *)(unsigned long)systab64->runtime;
+ tmp |= data ? data->runtime : systab64->runtime;
+ efi_systab.boottime = (void *)(unsigned long)systab64->boottime;
+ tmp |= systab64->boottime;
+ efi_systab.nr_tables = systab64->nr_tables;
+ efi_systab.tables = data ? (unsigned long)data->tables :
+ systab64->tables;
+ tmp |= data ? data->tables : systab64->tables;
+
+ early_memunmap(systab64, sizeof(*systab64));
+ if (data)
+ early_memunmap(data, sizeof(*data));
+#ifdef CONFIG_X86_32
+ if (tmp >> 32) {
+ pr_err("EFI data located above 4GB, disabling EFI.\n");
+ return -EINVAL;
+ }
+#endif
+ } else {
+ efi_system_table_32_t *systab32;
+
+ systab32 = early_memremap((unsigned long)phys,
+ sizeof(*systab32));
+ if (systab32 == NULL) {
+ pr_err("Couldn't map the system table!\n");
+ return -ENOMEM;
+ }
+
+ efi_systab.hdr = systab32->hdr;
+ efi_systab.fw_vendor = systab32->fw_vendor;
+ efi_systab.fw_revision = systab32->fw_revision;
+ efi_systab.con_in_handle = systab32->con_in_handle;
+ efi_systab.con_in = systab32->con_in;
+ efi_systab.con_out_handle = systab32->con_out_handle;
+ efi_systab.con_out = systab32->con_out;
+ efi_systab.stderr_handle = systab32->stderr_handle;
+ efi_systab.stderr = systab32->stderr;
+ efi_systab.runtime = (void *)(unsigned long)systab32->runtime;
+ efi_systab.boottime = (void *)(unsigned long)systab32->boottime;
+ efi_systab.nr_tables = systab32->nr_tables;
+ efi_systab.tables = systab32->tables;
+
+ early_memunmap(systab32, sizeof(*systab32));
+ }
+
+ efi.systab = &efi_systab;
+
+ /*
+ * Verify the EFI Table
+ */
+ if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE) {
+ pr_err("System table signature incorrect!\n");
+ return -EINVAL;
+ }
+ if ((efi.systab->hdr.revision >> 16) == 0)
+ pr_err("Warning: System table version %d.%02d, expected 1.00 or greater!\n",
+ efi.systab->hdr.revision >> 16,
+ efi.systab->hdr.revision & 0xffff);
+
+ return 0;
+}
+
+static int __init efi_runtime_init32(void)
+{
+ efi_runtime_services_32_t *runtime;
+
+ runtime = early_memremap((unsigned long)efi.systab->runtime,
+ sizeof(efi_runtime_services_32_t));
+ if (!runtime) {
+ pr_err("Could not map the runtime service table!\n");
+ return -ENOMEM;
+ }
+
+ /*
+ * We will only need *early* access to the SetVirtualAddressMap
+ * EFI runtime service. All other runtime services will be called
+ * via the virtual mapping.
+ */
+ efi_phys.set_virtual_address_map =
+ (efi_set_virtual_address_map_t *)
+ (unsigned long)runtime->set_virtual_address_map;
+ early_memunmap(runtime, sizeof(efi_runtime_services_32_t));
+
+ return 0;
+}
+
+static int __init efi_runtime_init64(void)
+{
+ efi_runtime_services_64_t *runtime;
+
+ runtime = early_memremap((unsigned long)efi.systab->runtime,
+ sizeof(efi_runtime_services_64_t));
+ if (!runtime) {
+ pr_err("Could not map the runtime service table!\n");
+ return -ENOMEM;
+ }
+
+ /*
+ * We will only need *early* access to the SetVirtualAddressMap
+ * EFI runtime service. All other runtime services will be called
+ * via the virtual mapping.
+ */
+ efi_phys.set_virtual_address_map =
+ (efi_set_virtual_address_map_t *)
+ (unsigned long)runtime->set_virtual_address_map;
+ early_memunmap(runtime, sizeof(efi_runtime_services_64_t));
+
+ return 0;
+}
+
+static int __init efi_runtime_init(void)
+{
+ int rv;
+
+ /*
+ * Check out the runtime services table. We need to map
+ * the runtime services table so that we can grab the physical
+ * address of several of the EFI runtime functions, needed to
+ * set the firmware into virtual mode.
+ *
+ * When EFI_PARAVIRT is in force then we could not map runtime
+ * service memory region because we do not have direct access to it.
+ * However, runtime services are available through proxy functions
+ * (e.g. in case of Xen dom0 EFI implementation they call special
+ * hypercall which executes relevant EFI functions) and that is why
+ * they are always enabled.
+ */
+
+ if (!efi_enabled(EFI_PARAVIRT)) {
+ if (efi_enabled(EFI_64BIT))
+ rv = efi_runtime_init64();
+ else
+ rv = efi_runtime_init32();
+
+ if (rv)
+ return rv;
+ }
+
+ set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+
+ return 0;
+}
+
+void __init efi_init(void)
+{
+ efi_char16_t *c16;
+ char vendor[100] = "unknown";
+ int i = 0;
+
+#ifdef CONFIG_X86_32
+ if (boot_params.efi_info.efi_systab_hi ||
+ boot_params.efi_info.efi_memmap_hi) {
+ pr_info("Table located above 4GB, disabling EFI.\n");
+ return;
+ }
+ efi_phys.systab = (efi_system_table_t *)boot_params.efi_info.efi_systab;
+#else
+ efi_phys.systab = (efi_system_table_t *)
+ (boot_params.efi_info.efi_systab |
+ ((__u64)boot_params.efi_info.efi_systab_hi<<32));
+#endif
+
+ if (efi_systab_init(efi_phys.systab))
+ return;
+
+ efi.config_table = (unsigned long)efi.systab->tables;
+ efi.fw_vendor = (unsigned long)efi.systab->fw_vendor;
+ efi.runtime = (unsigned long)efi.systab->runtime;
+
+ /*
+ * Show what we know for posterity
+ */
+ c16 = early_memremap_ro(efi.systab->fw_vendor,
+ sizeof(vendor) * sizeof(efi_char16_t));
+ if (c16) {
+ for (i = 0; i < sizeof(vendor) - 1 && c16[i]; ++i)
+ vendor[i] = c16[i];
+ vendor[i] = '\0';
+ early_memunmap(c16, sizeof(vendor) * sizeof(efi_char16_t));
+ } else {
+ pr_err("Could not map the firmware vendor!\n");
+ }
+
+ pr_info("EFI v%u.%.02u by %s\n",
+ efi.systab->hdr.revision >> 16,
+ efi.systab->hdr.revision & 0xffff, vendor);
+
+ if (efi_reuse_config(efi.systab->tables, efi.systab->nr_tables))
+ return;
+
+ if (efi_config_init(arch_tables))
+ return;
+
+ /*
+ * Note: We currently don't support runtime services on an EFI
+ * that doesn't match the kernel 32/64-bit mode.
+ */
+
+ if (!efi_runtime_supported())
+ pr_info("No EFI runtime due to 32/64-bit mismatch with kernel\n");
+ else {
+ if (efi_runtime_disabled() || efi_runtime_init()) {
+ efi_memmap_unmap();
+ return;
+ }
+ }
+
+ efi_clean_memmap();
+
+ if (efi_enabled(EFI_DBG))
+ efi_print_memmap();
+}
+
+void __init efi_set_executable(efi_memory_desc_t *md, bool executable)
+{
+ u64 addr, npages;
+
+ addr = md->virt_addr;
+ npages = md->num_pages;
+
+ memrange_efi_to_native(&addr, &npages);
+
+ if (executable)
+ set_memory_x(addr, npages);
+ else
+ set_memory_nx(addr, npages);
+}
+
+void __init runtime_code_page_mkexec(void)
+{
+ efi_memory_desc_t *md;
+
+ /* Make EFI runtime service code area executable */
+ for_each_efi_memory_desc(md) {
+ if (md->type != EFI_RUNTIME_SERVICES_CODE)
+ continue;
+
+ efi_set_executable(md, true);
+ }
+}
+
+void __init efi_memory_uc(u64 addr, unsigned long size)
+{
+ unsigned long page_shift = 1UL << EFI_PAGE_SHIFT;
+ u64 npages;
+
+ npages = round_up(size, page_shift) / page_shift;
+ memrange_efi_to_native(&addr, &npages);
+ set_memory_uc(addr, npages);
+}
+
+void __init old_map_region(efi_memory_desc_t *md)
+{
+ u64 start_pfn, end_pfn, end;
+ unsigned long size;
+ void *va;
+
+ start_pfn = PFN_DOWN(md->phys_addr);
+ size = md->num_pages << PAGE_SHIFT;
+ end = md->phys_addr + size;
+ end_pfn = PFN_UP(end);
+
+ if (pfn_range_is_mapped(start_pfn, end_pfn)) {
+ va = __va(md->phys_addr);
+
+ if (!(md->attribute & EFI_MEMORY_WB))
+ efi_memory_uc((u64)(unsigned long)va, size);
+ } else
+ va = efi_ioremap(md->phys_addr, size,
+ md->type, md->attribute);
+
+ md->virt_addr = (u64) (unsigned long) va;
+ if (!va)
+ pr_err("ioremap of 0x%llX failed!\n",
+ (unsigned long long)md->phys_addr);
+}
+
+/* Merge contiguous regions of the same type and attribute */
+static void __init efi_merge_regions(void)
+{
+ efi_memory_desc_t *md, *prev_md = NULL;
+
+ for_each_efi_memory_desc(md) {
+ u64 prev_size;
+
+ if (!prev_md) {
+ prev_md = md;
+ continue;
+ }
+
+ if (prev_md->type != md->type ||
+ prev_md->attribute != md->attribute) {
+ prev_md = md;
+ continue;
+ }
+
+ prev_size = prev_md->num_pages << EFI_PAGE_SHIFT;
+
+ if (md->phys_addr == (prev_md->phys_addr + prev_size)) {
+ prev_md->num_pages += md->num_pages;
+ md->type = EFI_RESERVED_TYPE;
+ md->attribute = 0;
+ continue;
+ }
+ prev_md = md;
+ }
+}
+
+static void __init get_systab_virt_addr(efi_memory_desc_t *md)
+{
+ unsigned long size;
+ u64 end, systab;
+
+ size = md->num_pages << EFI_PAGE_SHIFT;
+ end = md->phys_addr + size;
+ systab = (u64)(unsigned long)efi_phys.systab;
+ if (md->phys_addr <= systab && systab < end) {
+ systab += md->virt_addr - md->phys_addr;
+ efi.systab = (efi_system_table_t *)(unsigned long)systab;
+ }
+}
+
+static void *realloc_pages(void *old_memmap, int old_shift)
+{
+ void *ret;
+
+ ret = (void *)__get_free_pages(GFP_KERNEL, old_shift + 1);
+ if (!ret)
+ goto out;
+
+ /*
+ * A first-time allocation doesn't have anything to copy.
+ */
+ if (!old_memmap)
+ return ret;
+
+ memcpy(ret, old_memmap, PAGE_SIZE << old_shift);
+
+out:
+ free_pages((unsigned long)old_memmap, old_shift);
+ return ret;
+}
+
+/*
+ * Iterate the EFI memory map in reverse order because the regions
+ * will be mapped top-down. The end result is the same as if we had
+ * mapped things forward, but doesn't require us to change the
+ * existing implementation of efi_map_region().
+ */
+static inline void *efi_map_next_entry_reverse(void *entry)
+{
+ /* Initial call */
+ if (!entry)
+ return efi.memmap.map_end - efi.memmap.desc_size;
+
+ entry -= efi.memmap.desc_size;
+ if (entry < efi.memmap.map)
+ return NULL;
+
+ return entry;
+}
+
+/*
+ * efi_map_next_entry - Return the next EFI memory map descriptor
+ * @entry: Previous EFI memory map descriptor
+ *
+ * This is a helper function to iterate over the EFI memory map, which
+ * we do in different orders depending on the current configuration.
+ *
+ * To begin traversing the memory map @entry must be %NULL.
+ *
+ * Returns %NULL when we reach the end of the memory map.
+ */
+static void *efi_map_next_entry(void *entry)
+{
+ if (!efi_enabled(EFI_OLD_MEMMAP) && efi_enabled(EFI_64BIT)) {
+ /*
+ * Starting in UEFI v2.5 the EFI_PROPERTIES_TABLE
+ * config table feature requires us to map all entries
+ * in the same order as they appear in the EFI memory
+ * map. That is to say, entry N must have a lower
+ * virtual address than entry N+1. This is because the
+ * firmware toolchain leaves relative references in
+ * the code/data sections, which are split and become
+ * separate EFI memory regions. Mapping things
+ * out-of-order leads to the firmware accessing
+ * unmapped addresses.
+ *
+ * Since we need to map things this way whether or not
+ * the kernel actually makes use of
+ * EFI_PROPERTIES_TABLE, let's just switch to this
+ * scheme by default for 64-bit.
+ */
+ return efi_map_next_entry_reverse(entry);
+ }
+
+ /* Initial call */
+ if (!entry)
+ return efi.memmap.map;
+
+ entry += efi.memmap.desc_size;
+ if (entry >= efi.memmap.map_end)
+ return NULL;
+
+ return entry;
+}
+
+static bool should_map_region(efi_memory_desc_t *md)
+{
+ /*
+ * Runtime regions always require runtime mappings (obviously).
+ */
+ if (md->attribute & EFI_MEMORY_RUNTIME)
+ return true;
+
+ /*
+ * 32-bit EFI doesn't suffer from the bug that requires us to
+ * reserve boot services regions, and mixed mode support
+ * doesn't exist for 32-bit kernels.
+ */
+ if (IS_ENABLED(CONFIG_X86_32))
+ return false;
+
+ /*
+ * Map all of RAM so that we can access arguments in the 1:1
+ * mapping when making EFI runtime calls.
+ */
+ if (IS_ENABLED(CONFIG_EFI_MIXED) && !efi_is_native()) {
+ if (md->type == EFI_CONVENTIONAL_MEMORY ||
+ md->type == EFI_LOADER_DATA ||
+ md->type == EFI_LOADER_CODE)
+ return true;
+ }
+
+ /*
+ * Map boot services regions as a workaround for buggy
+ * firmware that accesses them even when they shouldn't.
+ *
+ * See efi_{reserve,free}_boot_services().
+ */
+ if (md->type == EFI_BOOT_SERVICES_CODE ||
+ md->type == EFI_BOOT_SERVICES_DATA)
+ return true;
+
+ return false;
+}
+
+/*
+ * Map the efi memory ranges of the runtime services and update new_mmap with
+ * virtual addresses.
+ */
+static void * __init efi_map_regions(int *count, int *pg_shift)
+{
+ void *p, *new_memmap = NULL;
+ unsigned long left = 0;
+ unsigned long desc_size;
+ efi_memory_desc_t *md;
+
+ desc_size = efi.memmap.desc_size;
+
+ p = NULL;
+ while ((p = efi_map_next_entry(p))) {
+ md = p;
+
+ if (!should_map_region(md))
+ continue;
+
+ efi_map_region(md);
+ get_systab_virt_addr(md);
+
+ if (left < desc_size) {
+ new_memmap = realloc_pages(new_memmap, *pg_shift);
+ if (!new_memmap)
+ return NULL;
+
+ left += PAGE_SIZE << *pg_shift;
+ (*pg_shift)++;
+ }
+
+ memcpy(new_memmap + (*count * desc_size), md, desc_size);
+
+ left -= desc_size;
+ (*count)++;
+ }
+
+ return new_memmap;
+}
+
+static void __init kexec_enter_virtual_mode(void)
+{
+#ifdef CONFIG_KEXEC_CORE
+ efi_memory_desc_t *md;
+ unsigned int num_pages;
+
+ efi.systab = NULL;
+
+ /*
+ * We don't do virtual mode, since we don't do runtime services, on
+ * non-native EFI. With efi=old_map, we don't do runtime services in
+ * kexec kernel because in the initial boot something else might
+ * have been mapped at these virtual addresses.
+ */
+ if (!efi_is_native() || efi_enabled(EFI_OLD_MEMMAP)) {
+ efi_memmap_unmap();
+ clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+ return;
+ }
+
+ if (efi_alloc_page_tables()) {
+ pr_err("Failed to allocate EFI page tables\n");
+ clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+ return;
+ }
+
+ /*
+ * Map efi regions which were passed via setup_data. The virt_addr is a
+ * fixed addr which was used in first kernel of a kexec boot.
+ */
+ for_each_efi_memory_desc(md) {
+ efi_map_region_fixed(md); /* FIXME: add error handling */
+ get_systab_virt_addr(md);
+ }
+
+ /*
+ * Unregister the early EFI memmap from efi_init() and install
+ * the new EFI memory map.
+ */
+ efi_memmap_unmap();
+
+ if (efi_memmap_init_late(efi.memmap.phys_map,
+ efi.memmap.desc_size * efi.memmap.nr_map)) {
+ pr_err("Failed to remap late EFI memory map\n");
+ clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+ return;
+ }
+
+ BUG_ON(!efi.systab);
+
+ num_pages = ALIGN(efi.memmap.nr_map * efi.memmap.desc_size, PAGE_SIZE);
+ num_pages >>= PAGE_SHIFT;
+
+ if (efi_setup_page_tables(efi.memmap.phys_map, num_pages)) {
+ clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+ return;
+ }
+
+ efi_sync_low_kernel_mappings();
+
+ /*
+ * Now that EFI is in virtual mode, update the function
+ * pointers in the runtime service table to the new virtual addresses.
+ *
+ * Call EFI services through wrapper functions.
+ */
+ efi.runtime_version = efi_systab.hdr.revision;
+
+ efi_native_runtime_setup();
+
+ efi.set_virtual_address_map = NULL;
+
+ if (efi_enabled(EFI_OLD_MEMMAP) && (__supported_pte_mask & _PAGE_NX))
+ runtime_code_page_mkexec();
+#endif
+}
+
+/*
+ * This function will switch the EFI runtime services to virtual mode.
+ * Essentially, we look through the EFI memmap and map every region that
+ * has the runtime attribute bit set in its memory descriptor into the
+ * efi_pgd page table.
+ *
+ * The old method which used to update that memory descriptor with the
+ * virtual address obtained from ioremap() is still supported when the
+ * kernel is booted with efi=old_map on its command line. Same old
+ * method enabled the runtime services to be called without having to
+ * thunk back into physical mode for every invocation.
+ *
+ * The new method does a pagetable switch in a preemption-safe manner
+ * so that we're in a different address space when calling a runtime
+ * function. For function arguments passing we do copy the PUDs of the
+ * kernel page table into efi_pgd prior to each call.
+ *
+ * Specially for kexec boot, efi runtime maps in previous kernel should
+ * be passed in via setup_data. In that case runtime ranges will be mapped
+ * to the same virtual addresses as the first kernel, see
+ * kexec_enter_virtual_mode().
+ */
+static void __init __efi_enter_virtual_mode(void)
+{
+ int count = 0, pg_shift = 0;
+ void *new_memmap = NULL;
+ efi_status_t status;
+ unsigned long pa;
+
+ efi.systab = NULL;
+
+ if (efi_alloc_page_tables()) {
+ pr_err("Failed to allocate EFI page tables\n");
+ goto err;
+ }
+
+ efi_merge_regions();
+ new_memmap = efi_map_regions(&count, &pg_shift);
+ if (!new_memmap) {
+ pr_err("Error reallocating memory, EFI runtime non-functional!\n");
+ goto err;
+ }
+
+ pa = __pa(new_memmap);
+
+ /*
+ * Unregister the early EFI memmap from efi_init() and install
+ * the new EFI memory map that we are about to pass to the
+ * firmware via SetVirtualAddressMap().
+ */
+ efi_memmap_unmap();
+
+ if (efi_memmap_init_late(pa, efi.memmap.desc_size * count)) {
+ pr_err("Failed to remap late EFI memory map\n");
+ goto err;
+ }
+
+ if (efi_enabled(EFI_DBG)) {
+ pr_info("EFI runtime memory map:\n");
+ efi_print_memmap();
+ }
+
+ if (WARN_ON(!efi.systab))
+ goto err;
+
+ if (efi_setup_page_tables(pa, 1 << pg_shift))
+ goto err;
+
+ efi_sync_low_kernel_mappings();
+
+ if (efi_is_native()) {
+ status = phys_efi_set_virtual_address_map(
+ efi.memmap.desc_size * count,
+ efi.memmap.desc_size,
+ efi.memmap.desc_version,
+ (efi_memory_desc_t *)pa);
+ } else {
+ status = efi_thunk_set_virtual_address_map(
+ efi_phys.set_virtual_address_map,
+ efi.memmap.desc_size * count,
+ efi.memmap.desc_size,
+ efi.memmap.desc_version,
+ (efi_memory_desc_t *)pa);
+ }
+
+ if (status != EFI_SUCCESS) {
+ pr_err("Unable to switch EFI into virtual mode (status=%lx)!\n",
+ status);
+ goto err;
+ }
+
+ /*
+ * Now that EFI is in virtual mode, update the function
+ * pointers in the runtime service table to the new virtual addresses.
+ *
+ * Call EFI services through wrapper functions.
+ */
+ efi.runtime_version = efi_systab.hdr.revision;
+
+ if (efi_is_native())
+ efi_native_runtime_setup();
+ else
+ efi_thunk_runtime_setup();
+
+ efi.set_virtual_address_map = NULL;
+
+ /*
+ * Apply more restrictive page table mapping attributes now that
+ * SVAM() has been called and the firmware has performed all
+ * necessary relocation fixups for the new virtual addresses.
+ */
+ efi_runtime_update_mappings();
+
+ /* clean DUMMY object */
+ efi_delete_dummy_variable();
+ return;
+
+err:
+ clear_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+}
+
+void __init efi_enter_virtual_mode(void)
+{
+ if (efi_enabled(EFI_PARAVIRT))
+ return;
+
+ if (efi_setup)
+ kexec_enter_virtual_mode();
+ else
+ __efi_enter_virtual_mode();
+
+ efi_dump_pagetable();
+}
+
+static int __init arch_parse_efi_cmdline(char *str)
+{
+ if (!str) {
+ pr_warn("need at least one option\n");
+ return -EINVAL;
+ }
+
+ if (parse_option_str(str, "old_map"))
+ set_bit(EFI_OLD_MEMMAP, &efi.flags);
+
+ return 0;
+}
+early_param("efi", arch_parse_efi_cmdline);
diff --git a/arch/x86/platform/efi/efi_32.c b/arch/x86/platform/efi/efi_32.c
new file mode 100644
index 000000000..995965712
--- /dev/null
+++ b/arch/x86/platform/efi/efi_32.c
@@ -0,0 +1,97 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Extensible Firmware Interface
+ *
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 1999 VA Linux Systems
+ * Copyright (C) 1999 Walt Drummond <drummond@valinux.com>
+ * Copyright (C) 1999-2002 Hewlett-Packard Co.
+ * David Mosberger-Tang <davidm@hpl.hp.com>
+ * Stephane Eranian <eranian@hpl.hp.com>
+ *
+ * All EFI Runtime Services are not implemented yet as EFI only
+ * supports physical mode addressing on SoftSDV. This is to be fixed
+ * in a future version. --drummond 1999-07-20
+ *
+ * Implemented EFI runtime services and virtual mode calls. --davidm
+ *
+ * Goutham Rao: <goutham.rao@intel.com>
+ * Skip non-WB memory and ignore empty memory ranges.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/ioport.h>
+#include <linux/efi.h>
+
+#include <asm/io.h>
+#include <asm/desc.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/tlbflush.h>
+#include <asm/efi.h>
+
+/*
+ * To make EFI call EFI runtime service in physical addressing mode we need
+ * prolog/epilog before/after the invocation to claim the EFI runtime service
+ * handler exclusively and to duplicate a memory mapping in low memory space,
+ * say 0 - 3G.
+ */
+
+int __init efi_alloc_page_tables(void)
+{
+ return 0;
+}
+
+void efi_sync_low_kernel_mappings(void) {}
+
+void __init efi_dump_pagetable(void)
+{
+#ifdef CONFIG_EFI_PGT_DUMP
+ ptdump_walk_pgd_level(NULL, swapper_pg_dir);
+#endif
+}
+
+int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
+{
+ return 0;
+}
+
+void __init efi_map_region(efi_memory_desc_t *md)
+{
+ old_map_region(md);
+}
+
+void __init efi_map_region_fixed(efi_memory_desc_t *md) {}
+void __init parse_efi_setup(u64 phys_addr, u32 data_len) {}
+
+pgd_t * __init efi_call_phys_prolog(void)
+{
+ struct desc_ptr gdt_descr;
+ pgd_t *save_pgd;
+
+ /* Current pgd is swapper_pg_dir, we'll restore it later: */
+ save_pgd = swapper_pg_dir;
+ load_cr3(initial_page_table);
+ __flush_tlb_all();
+
+ gdt_descr.address = get_cpu_gdt_paddr(0);
+ gdt_descr.size = GDT_SIZE - 1;
+ load_gdt(&gdt_descr);
+
+ return save_pgd;
+}
+
+void __init efi_call_phys_epilog(pgd_t *save_pgd)
+{
+ load_fixmap_gdt(0);
+ load_cr3(save_pgd);
+ __flush_tlb_all();
+}
+
+void __init efi_runtime_update_mappings(void)
+{
+ if (__supported_pte_mask & _PAGE_NX)
+ runtime_code_page_mkexec();
+}
diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c
new file mode 100644
index 000000000..77d05b560
--- /dev/null
+++ b/arch/x86/platform/efi/efi_64.c
@@ -0,0 +1,1039 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * x86_64 specific EFI support functions
+ * Based on Extensible Firmware Interface Specification version 1.0
+ *
+ * Copyright (C) 2005-2008 Intel Co.
+ * Fenghua Yu <fenghua.yu@intel.com>
+ * Bibo Mao <bibo.mao@intel.com>
+ * Chandramouli Narayanan <mouli@linux.intel.com>
+ * Huang Ying <ying.huang@intel.com>
+ *
+ * Code to convert EFI to E820 map has been implemented in elilo bootloader
+ * based on a EFI patch by Edgar Hucek. Based on the E820 map, the page table
+ * is setup appropriately for EFI runtime code.
+ * - mouli 06/14/2007.
+ *
+ */
+
+#define pr_fmt(fmt) "efi: " fmt
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+#include <linux/spinlock.h>
+#include <linux/bootmem.h>
+#include <linux/ioport.h>
+#include <linux/mc146818rtc.h>
+#include <linux/efi.h>
+#include <linux/export.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
+#include <linux/reboot.h>
+#include <linux/slab.h>
+#include <linux/ucs2_string.h>
+#include <linux/mem_encrypt.h>
+#include <linux/sched/task.h>
+
+#include <asm/setup.h>
+#include <asm/page.h>
+#include <asm/e820/api.h>
+#include <asm/pgtable.h>
+#include <asm/tlbflush.h>
+#include <asm/proto.h>
+#include <asm/efi.h>
+#include <asm/cacheflush.h>
+#include <asm/fixmap.h>
+#include <asm/realmode.h>
+#include <asm/time.h>
+#include <asm/pgalloc.h>
+
+/*
+ * We allocate runtime services regions top-down, starting from -4G, i.e.
+ * 0xffff_ffff_0000_0000 and limit EFI VA mapping space to 64G.
+ */
+static u64 efi_va = EFI_VA_START;
+
+struct efi_scratch efi_scratch;
+
+static void __init early_code_mapping_set_exec(int executable)
+{
+ efi_memory_desc_t *md;
+
+ if (!(__supported_pte_mask & _PAGE_NX))
+ return;
+
+ /* Make EFI service code area executable */
+ for_each_efi_memory_desc(md) {
+ if (md->type == EFI_RUNTIME_SERVICES_CODE ||
+ md->type == EFI_BOOT_SERVICES_CODE)
+ efi_set_executable(md, executable);
+ }
+}
+
+pgd_t * __init efi_call_phys_prolog(void)
+{
+ unsigned long vaddr, addr_pgd, addr_p4d, addr_pud;
+ pgd_t *save_pgd, *pgd_k, *pgd_efi;
+ p4d_t *p4d, *p4d_k, *p4d_efi;
+ pud_t *pud;
+
+ int pgd;
+ int n_pgds, i, j;
+
+ if (!efi_enabled(EFI_OLD_MEMMAP)) {
+ efi_switch_mm(&efi_mm);
+ return NULL;
+ }
+
+ early_code_mapping_set_exec(1);
+
+ n_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT), PGDIR_SIZE);
+ save_pgd = kmalloc_array(n_pgds, sizeof(*save_pgd), GFP_KERNEL);
+
+ /*
+ * Build 1:1 identity mapping for efi=old_map usage. Note that
+ * PAGE_OFFSET is PGDIR_SIZE aligned when KASLR is disabled, while
+ * it is PUD_SIZE ALIGNED with KASLR enabled. So for a given physical
+ * address X, the pud_index(X) != pud_index(__va(X)), we can only copy
+ * PUD entry of __va(X) to fill in pud entry of X to build 1:1 mapping.
+ * This means here we can only reuse the PMD tables of the direct mapping.
+ */
+ for (pgd = 0; pgd < n_pgds; pgd++) {
+ addr_pgd = (unsigned long)(pgd * PGDIR_SIZE);
+ vaddr = (unsigned long)__va(pgd * PGDIR_SIZE);
+ pgd_efi = pgd_offset_k(addr_pgd);
+ save_pgd[pgd] = *pgd_efi;
+
+ p4d = p4d_alloc(&init_mm, pgd_efi, addr_pgd);
+ if (!p4d) {
+ pr_err("Failed to allocate p4d table!\n");
+ goto out;
+ }
+
+ for (i = 0; i < PTRS_PER_P4D; i++) {
+ addr_p4d = addr_pgd + i * P4D_SIZE;
+ p4d_efi = p4d + p4d_index(addr_p4d);
+
+ pud = pud_alloc(&init_mm, p4d_efi, addr_p4d);
+ if (!pud) {
+ pr_err("Failed to allocate pud table!\n");
+ goto out;
+ }
+
+ for (j = 0; j < PTRS_PER_PUD; j++) {
+ addr_pud = addr_p4d + j * PUD_SIZE;
+
+ if (addr_pud > (max_pfn << PAGE_SHIFT))
+ break;
+
+ vaddr = (unsigned long)__va(addr_pud);
+
+ pgd_k = pgd_offset_k(vaddr);
+ p4d_k = p4d_offset(pgd_k, vaddr);
+ pud[j] = *pud_offset(p4d_k, vaddr);
+ }
+ }
+ pgd_offset_k(pgd * PGDIR_SIZE)->pgd &= ~_PAGE_NX;
+ }
+
+out:
+ __flush_tlb_all();
+
+ return save_pgd;
+}
+
+void __init efi_call_phys_epilog(pgd_t *save_pgd)
+{
+ /*
+ * After the lock is released, the original page table is restored.
+ */
+ int pgd_idx, i;
+ int nr_pgds;
+ pgd_t *pgd;
+ p4d_t *p4d;
+ pud_t *pud;
+
+ if (!efi_enabled(EFI_OLD_MEMMAP)) {
+ efi_switch_mm(efi_scratch.prev_mm);
+ return;
+ }
+
+ nr_pgds = DIV_ROUND_UP((max_pfn << PAGE_SHIFT) , PGDIR_SIZE);
+
+ for (pgd_idx = 0; pgd_idx < nr_pgds; pgd_idx++) {
+ pgd = pgd_offset_k(pgd_idx * PGDIR_SIZE);
+ set_pgd(pgd_offset_k(pgd_idx * PGDIR_SIZE), save_pgd[pgd_idx]);
+
+ if (!pgd_present(*pgd))
+ continue;
+
+ for (i = 0; i < PTRS_PER_P4D; i++) {
+ p4d = p4d_offset(pgd,
+ pgd_idx * PGDIR_SIZE + i * P4D_SIZE);
+
+ if (!p4d_present(*p4d))
+ continue;
+
+ pud = (pud_t *)p4d_page_vaddr(*p4d);
+ pud_free(&init_mm, pud);
+ }
+
+ p4d = (p4d_t *)pgd_page_vaddr(*pgd);
+ p4d_free(&init_mm, p4d);
+ }
+
+ kfree(save_pgd);
+
+ __flush_tlb_all();
+ early_code_mapping_set_exec(0);
+}
+
+EXPORT_SYMBOL_GPL(efi_mm);
+
+/*
+ * We need our own copy of the higher levels of the page tables
+ * because we want to avoid inserting EFI region mappings (EFI_VA_END
+ * to EFI_VA_START) into the standard kernel page tables. Everything
+ * else can be shared, see efi_sync_low_kernel_mappings().
+ *
+ * We don't want the pgd on the pgd_list and cannot use pgd_alloc() for the
+ * allocation.
+ */
+int __init efi_alloc_page_tables(void)
+{
+ pgd_t *pgd, *efi_pgd;
+ p4d_t *p4d;
+ pud_t *pud;
+ gfp_t gfp_mask;
+
+ if (efi_enabled(EFI_OLD_MEMMAP))
+ return 0;
+
+ gfp_mask = GFP_KERNEL | __GFP_ZERO;
+ efi_pgd = (pgd_t *)__get_free_pages(gfp_mask, PGD_ALLOCATION_ORDER);
+ if (!efi_pgd)
+ goto fail;
+
+ pgd = efi_pgd + pgd_index(EFI_VA_END);
+ p4d = p4d_alloc(&init_mm, pgd, EFI_VA_END);
+ if (!p4d)
+ goto free_pgd;
+
+ pud = pud_alloc(&init_mm, p4d, EFI_VA_END);
+ if (!pud)
+ goto free_p4d;
+
+ efi_mm.pgd = efi_pgd;
+ mm_init_cpumask(&efi_mm);
+ init_new_context(NULL, &efi_mm);
+
+ return 0;
+
+free_p4d:
+ if (pgtable_l5_enabled())
+ free_page((unsigned long)pgd_page_vaddr(*pgd));
+free_pgd:
+ free_pages((unsigned long)efi_pgd, PGD_ALLOCATION_ORDER);
+fail:
+ return -ENOMEM;
+}
+
+/*
+ * Add low kernel mappings for passing arguments to EFI functions.
+ */
+void efi_sync_low_kernel_mappings(void)
+{
+ unsigned num_entries;
+ pgd_t *pgd_k, *pgd_efi;
+ p4d_t *p4d_k, *p4d_efi;
+ pud_t *pud_k, *pud_efi;
+ pgd_t *efi_pgd = efi_mm.pgd;
+
+ if (efi_enabled(EFI_OLD_MEMMAP))
+ return;
+
+ /*
+ * We can share all PGD entries apart from the one entry that
+ * covers the EFI runtime mapping space.
+ *
+ * Make sure the EFI runtime region mappings are guaranteed to
+ * only span a single PGD entry and that the entry also maps
+ * other important kernel regions.
+ */
+ MAYBE_BUILD_BUG_ON(pgd_index(EFI_VA_END) != pgd_index(MODULES_END));
+ MAYBE_BUILD_BUG_ON((EFI_VA_START & PGDIR_MASK) !=
+ (EFI_VA_END & PGDIR_MASK));
+
+ pgd_efi = efi_pgd + pgd_index(PAGE_OFFSET);
+ pgd_k = pgd_offset_k(PAGE_OFFSET);
+
+ num_entries = pgd_index(EFI_VA_END) - pgd_index(PAGE_OFFSET);
+ memcpy(pgd_efi, pgd_k, sizeof(pgd_t) * num_entries);
+
+ /*
+ * As with PGDs, we share all P4D entries apart from the one entry
+ * that covers the EFI runtime mapping space.
+ */
+ BUILD_BUG_ON(p4d_index(EFI_VA_END) != p4d_index(MODULES_END));
+ BUILD_BUG_ON((EFI_VA_START & P4D_MASK) != (EFI_VA_END & P4D_MASK));
+
+ pgd_efi = efi_pgd + pgd_index(EFI_VA_END);
+ pgd_k = pgd_offset_k(EFI_VA_END);
+ p4d_efi = p4d_offset(pgd_efi, 0);
+ p4d_k = p4d_offset(pgd_k, 0);
+
+ num_entries = p4d_index(EFI_VA_END);
+ memcpy(p4d_efi, p4d_k, sizeof(p4d_t) * num_entries);
+
+ /*
+ * We share all the PUD entries apart from those that map the
+ * EFI regions. Copy around them.
+ */
+ BUILD_BUG_ON((EFI_VA_START & ~PUD_MASK) != 0);
+ BUILD_BUG_ON((EFI_VA_END & ~PUD_MASK) != 0);
+
+ p4d_efi = p4d_offset(pgd_efi, EFI_VA_END);
+ p4d_k = p4d_offset(pgd_k, EFI_VA_END);
+ pud_efi = pud_offset(p4d_efi, 0);
+ pud_k = pud_offset(p4d_k, 0);
+
+ num_entries = pud_index(EFI_VA_END);
+ memcpy(pud_efi, pud_k, sizeof(pud_t) * num_entries);
+
+ pud_efi = pud_offset(p4d_efi, EFI_VA_START);
+ pud_k = pud_offset(p4d_k, EFI_VA_START);
+
+ num_entries = PTRS_PER_PUD - pud_index(EFI_VA_START);
+ memcpy(pud_efi, pud_k, sizeof(pud_t) * num_entries);
+}
+
+/*
+ * Wrapper for slow_virt_to_phys() that handles NULL addresses.
+ */
+static inline phys_addr_t
+virt_to_phys_or_null_size(void *va, unsigned long size)
+{
+ phys_addr_t pa;
+
+ if (!va)
+ return 0;
+
+ if (virt_addr_valid(va))
+ return virt_to_phys(va);
+
+ pa = slow_virt_to_phys(va);
+
+ /* check if the object crosses a page boundary */
+ if (WARN_ON((pa ^ (pa + size - 1)) & PAGE_MASK))
+ return 0;
+
+ return pa;
+}
+
+#define virt_to_phys_or_null(addr) \
+ virt_to_phys_or_null_size((addr), sizeof(*(addr)))
+
+int __init efi_setup_page_tables(unsigned long pa_memmap, unsigned num_pages)
+{
+ unsigned long pfn, text, pf;
+ struct page *page;
+ unsigned npages;
+ pgd_t *pgd = efi_mm.pgd;
+
+ if (efi_enabled(EFI_OLD_MEMMAP))
+ return 0;
+
+ /*
+ * It can happen that the physical address of new_memmap lands in memory
+ * which is not mapped in the EFI page table. Therefore we need to go
+ * and ident-map those pages containing the map before calling
+ * phys_efi_set_virtual_address_map().
+ */
+ pfn = pa_memmap >> PAGE_SHIFT;
+ pf = _PAGE_NX | _PAGE_RW | _PAGE_ENC;
+ if (kernel_map_pages_in_pgd(pgd, pfn, pa_memmap, num_pages, pf)) {
+ pr_err("Error ident-mapping new memmap (0x%lx)!\n", pa_memmap);
+ return 1;
+ }
+
+ /*
+ * Certain firmware versions are way too sentimential and still believe
+ * they are exclusive and unquestionable owners of the first physical page,
+ * even though they explicitly mark it as EFI_CONVENTIONAL_MEMORY
+ * (but then write-access it later during SetVirtualAddressMap()).
+ *
+ * Create a 1:1 mapping for this page, to avoid triple faults during early
+ * boot with such firmware. We are free to hand this page to the BIOS,
+ * as trim_bios_range() will reserve the first page and isolate it away
+ * from memory allocators anyway.
+ */
+ pf = _PAGE_RW;
+ if (sev_active())
+ pf |= _PAGE_ENC;
+
+ if (kernel_map_pages_in_pgd(pgd, 0x0, 0x0, 1, pf)) {
+ pr_err("Failed to create 1:1 mapping for the first page!\n");
+ return 1;
+ }
+
+ /*
+ * When making calls to the firmware everything needs to be 1:1
+ * mapped and addressable with 32-bit pointers. Map the kernel
+ * text and allocate a new stack because we can't rely on the
+ * stack pointer being < 4GB.
+ */
+ if (!IS_ENABLED(CONFIG_EFI_MIXED) || efi_is_native())
+ return 0;
+
+ page = alloc_page(GFP_KERNEL|__GFP_DMA32);
+ if (!page) {
+ pr_err("Unable to allocate EFI runtime stack < 4GB\n");
+ return 1;
+ }
+
+ efi_scratch.phys_stack = page_to_phys(page + 1); /* stack grows down */
+
+ npages = (_etext - _text) >> PAGE_SHIFT;
+ text = __pa(_text);
+ pfn = text >> PAGE_SHIFT;
+
+ pf = _PAGE_RW | _PAGE_ENC;
+ if (kernel_map_pages_in_pgd(pgd, pfn, text, npages, pf)) {
+ pr_err("Failed to map kernel text 1:1\n");
+ return 1;
+ }
+
+ return 0;
+}
+
+static void __init __map_region(efi_memory_desc_t *md, u64 va)
+{
+ unsigned long flags = _PAGE_RW;
+ unsigned long pfn;
+ pgd_t *pgd = efi_mm.pgd;
+
+ if (!(md->attribute & EFI_MEMORY_WB))
+ flags |= _PAGE_PCD;
+
+ if (sev_active() && md->type != EFI_MEMORY_MAPPED_IO)
+ flags |= _PAGE_ENC;
+
+ pfn = md->phys_addr >> PAGE_SHIFT;
+ if (kernel_map_pages_in_pgd(pgd, pfn, va, md->num_pages, flags))
+ pr_warn("Error mapping PA 0x%llx -> VA 0x%llx!\n",
+ md->phys_addr, va);
+}
+
+void __init efi_map_region(efi_memory_desc_t *md)
+{
+ unsigned long size = md->num_pages << PAGE_SHIFT;
+ u64 pa = md->phys_addr;
+
+ if (efi_enabled(EFI_OLD_MEMMAP))
+ return old_map_region(md);
+
+ /*
+ * Make sure the 1:1 mappings are present as a catch-all for b0rked
+ * firmware which doesn't update all internal pointers after switching
+ * to virtual mode and would otherwise crap on us.
+ */
+ __map_region(md, md->phys_addr);
+
+ /*
+ * Enforce the 1:1 mapping as the default virtual address when
+ * booting in EFI mixed mode, because even though we may be
+ * running a 64-bit kernel, the firmware may only be 32-bit.
+ */
+ if (!efi_is_native () && IS_ENABLED(CONFIG_EFI_MIXED)) {
+ md->virt_addr = md->phys_addr;
+ return;
+ }
+
+ efi_va -= size;
+
+ /* Is PA 2M-aligned? */
+ if (!(pa & (PMD_SIZE - 1))) {
+ efi_va &= PMD_MASK;
+ } else {
+ u64 pa_offset = pa & (PMD_SIZE - 1);
+ u64 prev_va = efi_va;
+
+ /* get us the same offset within this 2M page */
+ efi_va = (efi_va & PMD_MASK) + pa_offset;
+
+ if (efi_va > prev_va)
+ efi_va -= PMD_SIZE;
+ }
+
+ if (efi_va < EFI_VA_END) {
+ pr_warn(FW_WARN "VA address range overflow!\n");
+ return;
+ }
+
+ /* Do the VA map */
+ __map_region(md, efi_va);
+ md->virt_addr = efi_va;
+}
+
+/*
+ * kexec kernel will use efi_map_region_fixed to map efi runtime memory ranges.
+ * md->virt_addr is the original virtual address which had been mapped in kexec
+ * 1st kernel.
+ */
+void __init efi_map_region_fixed(efi_memory_desc_t *md)
+{
+ __map_region(md, md->phys_addr);
+ __map_region(md, md->virt_addr);
+}
+
+void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size,
+ u32 type, u64 attribute)
+{
+ unsigned long last_map_pfn;
+
+ if (type == EFI_MEMORY_MAPPED_IO)
+ return ioremap(phys_addr, size);
+
+ last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size);
+ if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) {
+ unsigned long top = last_map_pfn << PAGE_SHIFT;
+ efi_ioremap(top, size - (top - phys_addr), type, attribute);
+ }
+
+ if (!(attribute & EFI_MEMORY_WB))
+ efi_memory_uc((u64)(unsigned long)__va(phys_addr), size);
+
+ return (void __iomem *)__va(phys_addr);
+}
+
+void __init parse_efi_setup(u64 phys_addr, u32 data_len)
+{
+ efi_setup = phys_addr + sizeof(struct setup_data);
+}
+
+static int __init efi_update_mappings(efi_memory_desc_t *md, unsigned long pf)
+{
+ unsigned long pfn;
+ pgd_t *pgd = efi_mm.pgd;
+ int err1, err2;
+
+ /* Update the 1:1 mapping */
+ pfn = md->phys_addr >> PAGE_SHIFT;
+ err1 = kernel_map_pages_in_pgd(pgd, pfn, md->phys_addr, md->num_pages, pf);
+ if (err1) {
+ pr_err("Error while updating 1:1 mapping PA 0x%llx -> VA 0x%llx!\n",
+ md->phys_addr, md->virt_addr);
+ }
+
+ err2 = kernel_map_pages_in_pgd(pgd, pfn, md->virt_addr, md->num_pages, pf);
+ if (err2) {
+ pr_err("Error while updating VA mapping PA 0x%llx -> VA 0x%llx!\n",
+ md->phys_addr, md->virt_addr);
+ }
+
+ return err1 || err2;
+}
+
+static int __init efi_update_mem_attr(struct mm_struct *mm, efi_memory_desc_t *md)
+{
+ unsigned long pf = 0;
+
+ if (md->attribute & EFI_MEMORY_XP)
+ pf |= _PAGE_NX;
+
+ if (!(md->attribute & EFI_MEMORY_RO))
+ pf |= _PAGE_RW;
+
+ if (sev_active())
+ pf |= _PAGE_ENC;
+
+ return efi_update_mappings(md, pf);
+}
+
+void __init efi_runtime_update_mappings(void)
+{
+ efi_memory_desc_t *md;
+
+ if (efi_enabled(EFI_OLD_MEMMAP)) {
+ if (__supported_pte_mask & _PAGE_NX)
+ runtime_code_page_mkexec();
+ return;
+ }
+
+ /*
+ * Use the EFI Memory Attribute Table for mapping permissions if it
+ * exists, since it is intended to supersede EFI_PROPERTIES_TABLE.
+ */
+ if (efi_enabled(EFI_MEM_ATTR)) {
+ efi_memattr_apply_permissions(NULL, efi_update_mem_attr);
+ return;
+ }
+
+ /*
+ * EFI_MEMORY_ATTRIBUTES_TABLE is intended to replace
+ * EFI_PROPERTIES_TABLE. So, use EFI_PROPERTIES_TABLE to update
+ * permissions only if EFI_MEMORY_ATTRIBUTES_TABLE is not
+ * published by the firmware. Even if we find a buggy implementation of
+ * EFI_MEMORY_ATTRIBUTES_TABLE, don't fall back to
+ * EFI_PROPERTIES_TABLE, because of the same reason.
+ */
+
+ if (!efi_enabled(EFI_NX_PE_DATA))
+ return;
+
+ for_each_efi_memory_desc(md) {
+ unsigned long pf = 0;
+
+ if (!(md->attribute & EFI_MEMORY_RUNTIME))
+ continue;
+
+ if (!(md->attribute & EFI_MEMORY_WB))
+ pf |= _PAGE_PCD;
+
+ if ((md->attribute & EFI_MEMORY_XP) ||
+ (md->type == EFI_RUNTIME_SERVICES_DATA))
+ pf |= _PAGE_NX;
+
+ if (!(md->attribute & EFI_MEMORY_RO) &&
+ (md->type != EFI_RUNTIME_SERVICES_CODE))
+ pf |= _PAGE_RW;
+
+ if (sev_active())
+ pf |= _PAGE_ENC;
+
+ efi_update_mappings(md, pf);
+ }
+}
+
+void __init efi_dump_pagetable(void)
+{
+#ifdef CONFIG_EFI_PGT_DUMP
+ if (efi_enabled(EFI_OLD_MEMMAP))
+ ptdump_walk_pgd_level(NULL, swapper_pg_dir);
+ else
+ ptdump_walk_pgd_level(NULL, efi_mm.pgd);
+#endif
+}
+
+/*
+ * Makes the calling thread switch to/from efi_mm context. Can be used
+ * for SetVirtualAddressMap() i.e. current->active_mm == init_mm as well
+ * as during efi runtime calls i.e current->active_mm == current_mm.
+ * We are not mm_dropping()/mm_grabbing() any mm, because we are not
+ * losing/creating any references.
+ */
+void efi_switch_mm(struct mm_struct *mm)
+{
+ task_lock(current);
+ efi_scratch.prev_mm = current->active_mm;
+ current->active_mm = mm;
+ switch_mm(efi_scratch.prev_mm, mm, NULL);
+ task_unlock(current);
+}
+
+#ifdef CONFIG_EFI_MIXED
+extern efi_status_t efi64_thunk(u32, ...);
+
+static DEFINE_SPINLOCK(efi_runtime_lock);
+
+#define runtime_service32(func) \
+({ \
+ u32 table = (u32)(unsigned long)efi.systab; \
+ u32 *rt, *___f; \
+ \
+ rt = (u32 *)(table + offsetof(efi_system_table_32_t, runtime)); \
+ ___f = (u32 *)(*rt + offsetof(efi_runtime_services_32_t, func)); \
+ *___f; \
+})
+
+/*
+ * Switch to the EFI page tables early so that we can access the 1:1
+ * runtime services mappings which are not mapped in any other page
+ * tables. This function must be called before runtime_service32().
+ *
+ * Also, disable interrupts because the IDT points to 64-bit handlers,
+ * which aren't going to function correctly when we switch to 32-bit.
+ */
+#define efi_thunk(f, ...) \
+({ \
+ efi_status_t __s; \
+ u32 __func; \
+ \
+ arch_efi_call_virt_setup(); \
+ \
+ __func = runtime_service32(f); \
+ __s = efi64_thunk(__func, __VA_ARGS__); \
+ \
+ arch_efi_call_virt_teardown(); \
+ \
+ __s; \
+})
+
+efi_status_t efi_thunk_set_virtual_address_map(
+ void *phys_set_virtual_address_map,
+ unsigned long memory_map_size,
+ unsigned long descriptor_size,
+ u32 descriptor_version,
+ efi_memory_desc_t *virtual_map)
+{
+ efi_status_t status;
+ unsigned long flags;
+ u32 func;
+
+ efi_sync_low_kernel_mappings();
+ local_irq_save(flags);
+
+ efi_switch_mm(&efi_mm);
+
+ func = (u32)(unsigned long)phys_set_virtual_address_map;
+ status = efi64_thunk(func, memory_map_size, descriptor_size,
+ descriptor_version, virtual_map);
+
+ efi_switch_mm(efi_scratch.prev_mm);
+ local_irq_restore(flags);
+
+ return status;
+}
+
+static efi_status_t efi_thunk_get_time(efi_time_t *tm, efi_time_cap_t *tc)
+{
+ efi_status_t status;
+ u32 phys_tm, phys_tc;
+ unsigned long flags;
+
+ spin_lock(&rtc_lock);
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_tm = virt_to_phys_or_null(tm);
+ phys_tc = virt_to_phys_or_null(tc);
+
+ status = efi_thunk(get_time, phys_tm, phys_tc);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+ spin_unlock(&rtc_lock);
+
+ return status;
+}
+
+static efi_status_t efi_thunk_set_time(efi_time_t *tm)
+{
+ efi_status_t status;
+ u32 phys_tm;
+ unsigned long flags;
+
+ spin_lock(&rtc_lock);
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_tm = virt_to_phys_or_null(tm);
+
+ status = efi_thunk(set_time, phys_tm);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+ spin_unlock(&rtc_lock);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_get_wakeup_time(efi_bool_t *enabled, efi_bool_t *pending,
+ efi_time_t *tm)
+{
+ efi_status_t status;
+ u32 phys_enabled, phys_pending, phys_tm;
+ unsigned long flags;
+
+ spin_lock(&rtc_lock);
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_enabled = virt_to_phys_or_null(enabled);
+ phys_pending = virt_to_phys_or_null(pending);
+ phys_tm = virt_to_phys_or_null(tm);
+
+ status = efi_thunk(get_wakeup_time, phys_enabled,
+ phys_pending, phys_tm);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+ spin_unlock(&rtc_lock);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_set_wakeup_time(efi_bool_t enabled, efi_time_t *tm)
+{
+ efi_status_t status;
+ u32 phys_tm;
+ unsigned long flags;
+
+ spin_lock(&rtc_lock);
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_tm = virt_to_phys_or_null(tm);
+
+ status = efi_thunk(set_wakeup_time, enabled, phys_tm);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+ spin_unlock(&rtc_lock);
+
+ return status;
+}
+
+static unsigned long efi_name_size(efi_char16_t *name)
+{
+ return ucs2_strsize(name, EFI_VAR_NAME_LEN) + 1;
+}
+
+static efi_status_t
+efi_thunk_get_variable(efi_char16_t *name, efi_guid_t *vendor,
+ u32 *attr, unsigned long *data_size, void *data)
+{
+ u8 buf[24] __aligned(8);
+ efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+ efi_status_t status;
+ u32 phys_name, phys_vendor, phys_attr;
+ u32 phys_data_size, phys_data;
+ unsigned long flags;
+
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ *vnd = *vendor;
+
+ phys_data_size = virt_to_phys_or_null(data_size);
+ phys_vendor = virt_to_phys_or_null(vnd);
+ phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
+ phys_attr = virt_to_phys_or_null(attr);
+ phys_data = virt_to_phys_or_null_size(data, *data_size);
+
+ if (!phys_name || (data && !phys_data))
+ status = EFI_INVALID_PARAMETER;
+ else
+ status = efi_thunk(get_variable, phys_name, phys_vendor,
+ phys_attr, phys_data_size, phys_data);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_set_variable(efi_char16_t *name, efi_guid_t *vendor,
+ u32 attr, unsigned long data_size, void *data)
+{
+ u8 buf[24] __aligned(8);
+ efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+ u32 phys_name, phys_vendor, phys_data;
+ efi_status_t status;
+ unsigned long flags;
+
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ *vnd = *vendor;
+
+ phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
+ phys_vendor = virt_to_phys_or_null(vnd);
+ phys_data = virt_to_phys_or_null_size(data, data_size);
+
+ if (!phys_name || (data && !phys_data))
+ status = EFI_INVALID_PARAMETER;
+ else
+ status = efi_thunk(set_variable, phys_name, phys_vendor,
+ attr, data_size, phys_data);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_set_variable_nonblocking(efi_char16_t *name, efi_guid_t *vendor,
+ u32 attr, unsigned long data_size,
+ void *data)
+{
+ u8 buf[24] __aligned(8);
+ efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+ u32 phys_name, phys_vendor, phys_data;
+ efi_status_t status;
+ unsigned long flags;
+
+ if (!spin_trylock_irqsave(&efi_runtime_lock, flags))
+ return EFI_NOT_READY;
+
+ *vnd = *vendor;
+
+ phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
+ phys_vendor = virt_to_phys_or_null(vnd);
+ phys_data = virt_to_phys_or_null_size(data, data_size);
+
+ if (!phys_name || (data && !phys_data))
+ status = EFI_INVALID_PARAMETER;
+ else
+ status = efi_thunk(set_variable, phys_name, phys_vendor,
+ attr, data_size, phys_data);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_get_next_variable(unsigned long *name_size,
+ efi_char16_t *name,
+ efi_guid_t *vendor)
+{
+ u8 buf[24] __aligned(8);
+ efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
+ efi_status_t status;
+ u32 phys_name_size, phys_name, phys_vendor;
+ unsigned long flags;
+
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ *vnd = *vendor;
+
+ phys_name_size = virt_to_phys_or_null(name_size);
+ phys_vendor = virt_to_phys_or_null(vnd);
+ phys_name = virt_to_phys_or_null_size(name, *name_size);
+
+ if (!phys_name)
+ status = EFI_INVALID_PARAMETER;
+ else
+ status = efi_thunk(get_next_variable, phys_name_size,
+ phys_name, phys_vendor);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ *vendor = *vnd;
+ return status;
+}
+
+static efi_status_t
+efi_thunk_get_next_high_mono_count(u32 *count)
+{
+ efi_status_t status;
+ u32 phys_count;
+ unsigned long flags;
+
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_count = virt_to_phys_or_null(count);
+ status = efi_thunk(get_next_high_mono_count, phys_count);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ return status;
+}
+
+static void
+efi_thunk_reset_system(int reset_type, efi_status_t status,
+ unsigned long data_size, efi_char16_t *data)
+{
+ u32 phys_data;
+ unsigned long flags;
+
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_data = virt_to_phys_or_null_size(data, data_size);
+
+ efi_thunk(reset_system, reset_type, status, data_size, phys_data);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+}
+
+static efi_status_t
+efi_thunk_update_capsule(efi_capsule_header_t **capsules,
+ unsigned long count, unsigned long sg_list)
+{
+ /*
+ * To properly support this function we would need to repackage
+ * 'capsules' because the firmware doesn't understand 64-bit
+ * pointers.
+ */
+ return EFI_UNSUPPORTED;
+}
+
+static efi_status_t
+efi_thunk_query_variable_info(u32 attr, u64 *storage_space,
+ u64 *remaining_space,
+ u64 *max_variable_size)
+{
+ efi_status_t status;
+ u32 phys_storage, phys_remaining, phys_max;
+ unsigned long flags;
+
+ if (efi.runtime_version < EFI_2_00_SYSTEM_TABLE_REVISION)
+ return EFI_UNSUPPORTED;
+
+ spin_lock_irqsave(&efi_runtime_lock, flags);
+
+ phys_storage = virt_to_phys_or_null(storage_space);
+ phys_remaining = virt_to_phys_or_null(remaining_space);
+ phys_max = virt_to_phys_or_null(max_variable_size);
+
+ status = efi_thunk(query_variable_info, attr, phys_storage,
+ phys_remaining, phys_max);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_query_variable_info_nonblocking(u32 attr, u64 *storage_space,
+ u64 *remaining_space,
+ u64 *max_variable_size)
+{
+ efi_status_t status;
+ u32 phys_storage, phys_remaining, phys_max;
+ unsigned long flags;
+
+ if (efi.runtime_version < EFI_2_00_SYSTEM_TABLE_REVISION)
+ return EFI_UNSUPPORTED;
+
+ if (!spin_trylock_irqsave(&efi_runtime_lock, flags))
+ return EFI_NOT_READY;
+
+ phys_storage = virt_to_phys_or_null(storage_space);
+ phys_remaining = virt_to_phys_or_null(remaining_space);
+ phys_max = virt_to_phys_or_null(max_variable_size);
+
+ status = efi_thunk(query_variable_info, attr, phys_storage,
+ phys_remaining, phys_max);
+
+ spin_unlock_irqrestore(&efi_runtime_lock, flags);
+
+ return status;
+}
+
+static efi_status_t
+efi_thunk_query_capsule_caps(efi_capsule_header_t **capsules,
+ unsigned long count, u64 *max_size,
+ int *reset_type)
+{
+ /*
+ * To properly support this function we would need to repackage
+ * 'capsules' because the firmware doesn't understand 64-bit
+ * pointers.
+ */
+ return EFI_UNSUPPORTED;
+}
+
+void efi_thunk_runtime_setup(void)
+{
+ efi.get_time = efi_thunk_get_time;
+ efi.set_time = efi_thunk_set_time;
+ efi.get_wakeup_time = efi_thunk_get_wakeup_time;
+ efi.set_wakeup_time = efi_thunk_set_wakeup_time;
+ efi.get_variable = efi_thunk_get_variable;
+ efi.get_next_variable = efi_thunk_get_next_variable;
+ efi.set_variable = efi_thunk_set_variable;
+ efi.set_variable_nonblocking = efi_thunk_set_variable_nonblocking;
+ efi.get_next_high_mono_count = efi_thunk_get_next_high_mono_count;
+ efi.reset_system = efi_thunk_reset_system;
+ efi.query_variable_info = efi_thunk_query_variable_info;
+ efi.query_variable_info_nonblocking = efi_thunk_query_variable_info_nonblocking;
+ efi.update_capsule = efi_thunk_update_capsule;
+ efi.query_capsule_caps = efi_thunk_query_capsule_caps;
+}
+#endif /* CONFIG_EFI_MIXED */
diff --git a/arch/x86/platform/efi/efi_stub_32.S b/arch/x86/platform/efi/efi_stub_32.S
new file mode 100644
index 000000000..ab2e91e76
--- /dev/null
+++ b/arch/x86/platform/efi/efi_stub_32.S
@@ -0,0 +1,124 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * EFI call stub for IA32.
+ *
+ * This stub allows us to make EFI calls in physical mode with interrupts
+ * turned off.
+ */
+
+#include <linux/linkage.h>
+#include <asm/page_types.h>
+
+/*
+ * efi_call_phys(void *, ...) is a function with variable parameters.
+ * All the callers of this function assure that all the parameters are 4-bytes.
+ */
+
+/*
+ * In gcc calling convention, EBX, ESP, EBP, ESI and EDI are all callee save.
+ * So we'd better save all of them at the beginning of this function and restore
+ * at the end no matter how many we use, because we can not assure EFI runtime
+ * service functions will comply with gcc calling convention, too.
+ */
+
+.text
+ENTRY(efi_call_phys)
+ /*
+ * 0. The function can only be called in Linux kernel. So CS has been
+ * set to 0x0010, DS and SS have been set to 0x0018. In EFI, I found
+ * the values of these registers are the same. And, the corresponding
+ * GDT entries are identical. So I will do nothing about segment reg
+ * and GDT, but change GDT base register in prolog and epilog.
+ */
+
+ /*
+ * 1. Now I am running with EIP = <physical address> + PAGE_OFFSET.
+ * But to make it smoothly switch from virtual mode to flat mode.
+ * The mapping of lower virtual memory has been created in prolog and
+ * epilog.
+ */
+ movl $1f, %edx
+ subl $__PAGE_OFFSET, %edx
+ jmp *%edx
+1:
+
+ /*
+ * 2. Now on the top of stack is the return
+ * address in the caller of efi_call_phys(), then parameter 1,
+ * parameter 2, ..., param n. To make things easy, we save the return
+ * address of efi_call_phys in a global variable.
+ */
+ popl %edx
+ movl %edx, saved_return_addr
+ /* get the function pointer into ECX*/
+ popl %ecx
+ movl %ecx, efi_rt_function_ptr
+ movl $2f, %edx
+ subl $__PAGE_OFFSET, %edx
+ pushl %edx
+
+ /*
+ * 3. Clear PG bit in %CR0.
+ */
+ movl %cr0, %edx
+ andl $0x7fffffff, %edx
+ movl %edx, %cr0
+ jmp 1f
+1:
+
+ /*
+ * 4. Adjust stack pointer.
+ */
+ subl $__PAGE_OFFSET, %esp
+
+ /*
+ * 5. Call the physical function.
+ */
+ jmp *%ecx
+
+2:
+ /*
+ * 6. After EFI runtime service returns, control will return to
+ * following instruction. We'd better readjust stack pointer first.
+ */
+ addl $__PAGE_OFFSET, %esp
+
+ /*
+ * 7. Restore PG bit
+ */
+ movl %cr0, %edx
+ orl $0x80000000, %edx
+ movl %edx, %cr0
+ jmp 1f
+1:
+ /*
+ * 8. Now restore the virtual mode from flat mode by
+ * adding EIP with PAGE_OFFSET.
+ */
+ movl $1f, %edx
+ jmp *%edx
+1:
+
+ /*
+ * 9. Balance the stack. And because EAX contain the return value,
+ * we'd better not clobber it.
+ */
+ leal efi_rt_function_ptr, %edx
+ movl (%edx), %ecx
+ pushl %ecx
+
+ /*
+ * 10. Push the saved return address onto the stack and return.
+ */
+ leal saved_return_addr, %edx
+ movl (%edx), %ecx
+ pushl %ecx
+ ret
+ENDPROC(efi_call_phys)
+.previous
+
+.data
+saved_return_addr:
+ .long 0
+efi_rt_function_ptr:
+ .long 0
diff --git a/arch/x86/platform/efi/efi_stub_64.S b/arch/x86/platform/efi/efi_stub_64.S
new file mode 100644
index 000000000..74628ec78
--- /dev/null
+++ b/arch/x86/platform/efi/efi_stub_64.S
@@ -0,0 +1,58 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Function calling ABI conversion from Linux to EFI for x86_64
+ *
+ * Copyright (C) 2007 Intel Corp
+ * Bibo Mao <bibo.mao@intel.com>
+ * Huang Ying <ying.huang@intel.com>
+ */
+
+#include <linux/linkage.h>
+#include <asm/segment.h>
+#include <asm/msr.h>
+#include <asm/processor-flags.h>
+#include <asm/page_types.h>
+
+#define SAVE_XMM \
+ mov %rsp, %rax; \
+ subq $0x70, %rsp; \
+ and $~0xf, %rsp; \
+ mov %rax, (%rsp); \
+ mov %cr0, %rax; \
+ clts; \
+ mov %rax, 0x8(%rsp); \
+ movaps %xmm0, 0x60(%rsp); \
+ movaps %xmm1, 0x50(%rsp); \
+ movaps %xmm2, 0x40(%rsp); \
+ movaps %xmm3, 0x30(%rsp); \
+ movaps %xmm4, 0x20(%rsp); \
+ movaps %xmm5, 0x10(%rsp)
+
+#define RESTORE_XMM \
+ movaps 0x60(%rsp), %xmm0; \
+ movaps 0x50(%rsp), %xmm1; \
+ movaps 0x40(%rsp), %xmm2; \
+ movaps 0x30(%rsp), %xmm3; \
+ movaps 0x20(%rsp), %xmm4; \
+ movaps 0x10(%rsp), %xmm5; \
+ mov 0x8(%rsp), %rsi; \
+ mov %rsi, %cr0; \
+ mov (%rsp), %rsp
+
+ENTRY(efi_call)
+ pushq %rbp
+ movq %rsp, %rbp
+ SAVE_XMM
+ mov 16(%rbp), %rax
+ subq $48, %rsp
+ mov %r9, 32(%rsp)
+ mov %rax, 40(%rsp)
+ mov %r8, %r9
+ mov %rcx, %r8
+ mov %rsi, %rcx
+ call *%rdi
+ addq $48, %rsp
+ RESTORE_XMM
+ popq %rbp
+ ret
+ENDPROC(efi_call)
diff --git a/arch/x86/platform/efi/efi_thunk_64.S b/arch/x86/platform/efi/efi_thunk_64.S
new file mode 100644
index 000000000..46c58b087
--- /dev/null
+++ b/arch/x86/platform/efi/efi_thunk_64.S
@@ -0,0 +1,153 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (C) 2014 Intel Corporation; author Matt Fleming
+ *
+ * Support for invoking 32-bit EFI runtime services from a 64-bit
+ * kernel.
+ *
+ * The below thunking functions are only used after ExitBootServices()
+ * has been called. This simplifies things considerably as compared with
+ * the early EFI thunking because we can leave all the kernel state
+ * intact (GDT, IDT, etc) and simply invoke the the 32-bit EFI runtime
+ * services from __KERNEL32_CS. This means we can continue to service
+ * interrupts across an EFI mixed mode call.
+ *
+ * We do however, need to handle the fact that we're running in a full
+ * 64-bit virtual address space. Things like the stack and instruction
+ * addresses need to be accessible by the 32-bit firmware, so we rely on
+ * using the identity mappings in the EFI page table to access the stack
+ * and kernel text (see efi_setup_page_tables()).
+ */
+
+#include <linux/linkage.h>
+#include <asm/page_types.h>
+#include <asm/segment.h>
+
+ .text
+ .code64
+ENTRY(efi64_thunk)
+ push %rbp
+ push %rbx
+
+ /*
+ * Switch to 1:1 mapped 32-bit stack pointer.
+ */
+ movq %rsp, efi_saved_sp(%rip)
+ movq efi_scratch(%rip), %rsp
+
+ /*
+ * Calculate the physical address of the kernel text.
+ */
+ movq $__START_KERNEL_map, %rax
+ subq phys_base(%rip), %rax
+
+ /*
+ * Push some physical addresses onto the stack. This is easier
+ * to do now in a code64 section while the assembler can address
+ * 64-bit values. Note that all the addresses on the stack are
+ * 32-bit.
+ */
+ subq $16, %rsp
+ leaq efi_exit32(%rip), %rbx
+ subq %rax, %rbx
+ movl %ebx, 8(%rsp)
+
+ leaq __efi64_thunk(%rip), %rbx
+ subq %rax, %rbx
+ call *%rbx
+
+ movq efi_saved_sp(%rip), %rsp
+ pop %rbx
+ pop %rbp
+ retq
+ENDPROC(efi64_thunk)
+
+/*
+ * We run this function from the 1:1 mapping.
+ *
+ * This function must be invoked with a 1:1 mapped stack.
+ */
+ENTRY(__efi64_thunk)
+ movl %ds, %eax
+ push %rax
+ movl %es, %eax
+ push %rax
+ movl %ss, %eax
+ push %rax
+
+ subq $32, %rsp
+ movl %esi, 0x0(%rsp)
+ movl %edx, 0x4(%rsp)
+ movl %ecx, 0x8(%rsp)
+ movq %r8, %rsi
+ movl %esi, 0xc(%rsp)
+ movq %r9, %rsi
+ movl %esi, 0x10(%rsp)
+
+ leaq 1f(%rip), %rbx
+ movq %rbx, func_rt_ptr(%rip)
+
+ /* Switch to 32-bit descriptor */
+ pushq $__KERNEL32_CS
+ leaq efi_enter32(%rip), %rax
+ pushq %rax
+ lretq
+
+1: addq $32, %rsp
+
+ pop %rbx
+ movl %ebx, %ss
+ pop %rbx
+ movl %ebx, %es
+ pop %rbx
+ movl %ebx, %ds
+
+ /*
+ * Convert 32-bit status code into 64-bit.
+ */
+ test %rax, %rax
+ jz 1f
+ movl %eax, %ecx
+ andl $0x0fffffff, %ecx
+ andl $0xf0000000, %eax
+ shl $32, %rax
+ or %rcx, %rax
+1:
+ ret
+ENDPROC(__efi64_thunk)
+
+ENTRY(efi_exit32)
+ movq func_rt_ptr(%rip), %rax
+ push %rax
+ mov %rdi, %rax
+ ret
+ENDPROC(efi_exit32)
+
+ .code32
+/*
+ * EFI service pointer must be in %edi.
+ *
+ * The stack should represent the 32-bit calling convention.
+ */
+ENTRY(efi_enter32)
+ movl $__KERNEL_DS, %eax
+ movl %eax, %ds
+ movl %eax, %es
+ movl %eax, %ss
+
+ call *%edi
+
+ /* We must preserve return value */
+ movl %eax, %edi
+
+ movl 72(%esp), %eax
+ pushl $__KERNEL_CS
+ pushl %eax
+
+ lret
+ENDPROC(efi_enter32)
+
+ .data
+ .balign 8
+func_rt_ptr: .quad 0
+efi_saved_sp: .quad 0
diff --git a/arch/x86/platform/efi/quirks.c b/arch/x86/platform/efi/quirks.c
new file mode 100644
index 000000000..006eb09e9
--- /dev/null
+++ b/arch/x86/platform/efi/quirks.c
@@ -0,0 +1,655 @@
+#define pr_fmt(fmt) "efi: " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/time.h>
+#include <linux/types.h>
+#include <linux/efi.h>
+#include <linux/slab.h>
+#include <linux/memblock.h>
+#include <linux/bootmem.h>
+#include <linux/acpi.h>
+#include <linux/dmi.h>
+
+#include <asm/e820/api.h>
+#include <asm/efi.h>
+#include <asm/uv/uv.h>
+#include <asm/cpu_device_id.h>
+
+#define EFI_MIN_RESERVE 5120
+
+#define EFI_DUMMY_GUID \
+ EFI_GUID(0x4424ac57, 0xbe4b, 0x47dd, 0x9e, 0x97, 0xed, 0x50, 0xf0, 0x9f, 0x92, 0xa9)
+
+#define QUARK_CSH_SIGNATURE 0x5f435348 /* _CSH */
+#define QUARK_SECURITY_HEADER_SIZE 0x400
+
+/*
+ * Header prepended to the standard EFI capsule on Quark systems the are based
+ * on Intel firmware BSP.
+ * @csh_signature: Unique identifier to sanity check signed module
+ * presence ("_CSH").
+ * @version: Current version of CSH used. Should be one for Quark A0.
+ * @modulesize: Size of the entire module including the module header
+ * and payload.
+ * @security_version_number_index: Index of SVN to use for validation of signed
+ * module.
+ * @security_version_number: Used to prevent against roll back of modules.
+ * @rsvd_module_id: Currently unused for Clanton (Quark).
+ * @rsvd_module_vendor: Vendor Identifier. For Intel products value is
+ * 0x00008086.
+ * @rsvd_date: BCD representation of build date as yyyymmdd, where
+ * yyyy=4 digit year, mm=1-12, dd=1-31.
+ * @headersize: Total length of the header including including any
+ * padding optionally added by the signing tool.
+ * @hash_algo: What Hash is used in the module signing.
+ * @cryp_algo: What Crypto is used in the module signing.
+ * @keysize: Total length of the key data including including any
+ * padding optionally added by the signing tool.
+ * @signaturesize: Total length of the signature including including any
+ * padding optionally added by the signing tool.
+ * @rsvd_next_header: 32-bit pointer to the next Secure Boot Module in the
+ * chain, if there is a next header.
+ * @rsvd: Reserved, padding structure to required size.
+ *
+ * See also QuartSecurityHeader_t in
+ * Quark_EDKII_v1.2.1.1/QuarkPlatformPkg/Include/QuarkBootRom.h
+ * from https://downloadcenter.intel.com/download/23197/Intel-Quark-SoC-X1000-Board-Support-Package-BSP
+ */
+struct quark_security_header {
+ u32 csh_signature;
+ u32 version;
+ u32 modulesize;
+ u32 security_version_number_index;
+ u32 security_version_number;
+ u32 rsvd_module_id;
+ u32 rsvd_module_vendor;
+ u32 rsvd_date;
+ u32 headersize;
+ u32 hash_algo;
+ u32 cryp_algo;
+ u32 keysize;
+ u32 signaturesize;
+ u32 rsvd_next_header;
+ u32 rsvd[2];
+};
+
+static const efi_char16_t efi_dummy_name[] = L"DUMMY";
+
+static bool efi_no_storage_paranoia;
+
+/*
+ * Some firmware implementations refuse to boot if there's insufficient
+ * space in the variable store. The implementation of garbage collection
+ * in some FW versions causes stale (deleted) variables to take up space
+ * longer than intended and space is only freed once the store becomes
+ * almost completely full.
+ *
+ * Enabling this option disables the space checks in
+ * efi_query_variable_store() and forces garbage collection.
+ *
+ * Only enable this option if deleting EFI variables does not free up
+ * space in your variable store, e.g. if despite deleting variables
+ * you're unable to create new ones.
+ */
+static int __init setup_storage_paranoia(char *arg)
+{
+ efi_no_storage_paranoia = true;
+ return 0;
+}
+early_param("efi_no_storage_paranoia", setup_storage_paranoia);
+
+/*
+ * Deleting the dummy variable which kicks off garbage collection
+*/
+void efi_delete_dummy_variable(void)
+{
+ efi.set_variable_nonblocking((efi_char16_t *)efi_dummy_name,
+ &EFI_DUMMY_GUID,
+ EFI_VARIABLE_NON_VOLATILE |
+ EFI_VARIABLE_BOOTSERVICE_ACCESS |
+ EFI_VARIABLE_RUNTIME_ACCESS, 0, NULL);
+}
+
+/*
+ * In the nonblocking case we do not attempt to perform garbage
+ * collection if we do not have enough free space. Rather, we do the
+ * bare minimum check and give up immediately if the available space
+ * is below EFI_MIN_RESERVE.
+ *
+ * This function is intended to be small and simple because it is
+ * invoked from crash handler paths.
+ */
+static efi_status_t
+query_variable_store_nonblocking(u32 attributes, unsigned long size)
+{
+ efi_status_t status;
+ u64 storage_size, remaining_size, max_size;
+
+ status = efi.query_variable_info_nonblocking(attributes, &storage_size,
+ &remaining_size,
+ &max_size);
+ if (status != EFI_SUCCESS)
+ return status;
+
+ if (remaining_size - size < EFI_MIN_RESERVE)
+ return EFI_OUT_OF_RESOURCES;
+
+ return EFI_SUCCESS;
+}
+
+/*
+ * Some firmware implementations refuse to boot if there's insufficient space
+ * in the variable store. Ensure that we never use more than a safe limit.
+ *
+ * Return EFI_SUCCESS if it is safe to write 'size' bytes to the variable
+ * store.
+ */
+efi_status_t efi_query_variable_store(u32 attributes, unsigned long size,
+ bool nonblocking)
+{
+ efi_status_t status;
+ u64 storage_size, remaining_size, max_size;
+
+ if (!(attributes & EFI_VARIABLE_NON_VOLATILE))
+ return 0;
+
+ if (nonblocking)
+ return query_variable_store_nonblocking(attributes, size);
+
+ status = efi.query_variable_info(attributes, &storage_size,
+ &remaining_size, &max_size);
+ if (status != EFI_SUCCESS)
+ return status;
+
+ /*
+ * We account for that by refusing the write if permitting it would
+ * reduce the available space to under 5KB. This figure was provided by
+ * Samsung, so should be safe.
+ */
+ if ((remaining_size - size < EFI_MIN_RESERVE) &&
+ !efi_no_storage_paranoia) {
+
+ /*
+ * Triggering garbage collection may require that the firmware
+ * generate a real EFI_OUT_OF_RESOURCES error. We can force
+ * that by attempting to use more space than is available.
+ */
+ unsigned long dummy_size = remaining_size + 1024;
+ void *dummy = kzalloc(dummy_size, GFP_KERNEL);
+
+ if (!dummy)
+ return EFI_OUT_OF_RESOURCES;
+
+ status = efi.set_variable((efi_char16_t *)efi_dummy_name,
+ &EFI_DUMMY_GUID,
+ EFI_VARIABLE_NON_VOLATILE |
+ EFI_VARIABLE_BOOTSERVICE_ACCESS |
+ EFI_VARIABLE_RUNTIME_ACCESS,
+ dummy_size, dummy);
+
+ if (status == EFI_SUCCESS) {
+ /*
+ * This should have failed, so if it didn't make sure
+ * that we delete it...
+ */
+ efi_delete_dummy_variable();
+ }
+
+ kfree(dummy);
+
+ /*
+ * The runtime code may now have triggered a garbage collection
+ * run, so check the variable info again
+ */
+ status = efi.query_variable_info(attributes, &storage_size,
+ &remaining_size, &max_size);
+
+ if (status != EFI_SUCCESS)
+ return status;
+
+ /*
+ * There still isn't enough room, so return an error
+ */
+ if (remaining_size - size < EFI_MIN_RESERVE)
+ return EFI_OUT_OF_RESOURCES;
+ }
+
+ return EFI_SUCCESS;
+}
+EXPORT_SYMBOL_GPL(efi_query_variable_store);
+
+/*
+ * The UEFI specification makes it clear that the operating system is
+ * free to do whatever it wants with boot services code after
+ * ExitBootServices() has been called. Ignoring this recommendation a
+ * significant bunch of EFI implementations continue calling into boot
+ * services code (SetVirtualAddressMap). In order to work around such
+ * buggy implementations we reserve boot services region during EFI
+ * init and make sure it stays executable. Then, after
+ * SetVirtualAddressMap(), it is discarded.
+ *
+ * However, some boot services regions contain data that is required
+ * by drivers, so we need to track which memory ranges can never be
+ * freed. This is done by tagging those regions with the
+ * EFI_MEMORY_RUNTIME attribute.
+ *
+ * Any driver that wants to mark a region as reserved must use
+ * efi_mem_reserve() which will insert a new EFI memory descriptor
+ * into efi.memmap (splitting existing regions if necessary) and tag
+ * it with EFI_MEMORY_RUNTIME.
+ */
+void __init efi_arch_mem_reserve(phys_addr_t addr, u64 size)
+{
+ phys_addr_t new_phys, new_size;
+ struct efi_mem_range mr;
+ efi_memory_desc_t md;
+ int num_entries;
+ void *new;
+
+ if (efi_mem_desc_lookup(addr, &md) ||
+ md.type != EFI_BOOT_SERVICES_DATA) {
+ pr_err("Failed to lookup EFI memory descriptor for %pa\n", &addr);
+ return;
+ }
+
+ if (addr + size > md.phys_addr + (md.num_pages << EFI_PAGE_SHIFT)) {
+ pr_err("Region spans EFI memory descriptors, %pa\n", &addr);
+ return;
+ }
+
+ size += addr % EFI_PAGE_SIZE;
+ size = round_up(size, EFI_PAGE_SIZE);
+ addr = round_down(addr, EFI_PAGE_SIZE);
+
+ mr.range.start = addr;
+ mr.range.end = addr + size - 1;
+ mr.attribute = md.attribute | EFI_MEMORY_RUNTIME;
+
+ num_entries = efi_memmap_split_count(&md, &mr.range);
+ num_entries += efi.memmap.nr_map;
+
+ new_size = efi.memmap.desc_size * num_entries;
+
+ new_phys = efi_memmap_alloc(num_entries);
+ if (!new_phys) {
+ pr_err("Could not allocate boot services memmap\n");
+ return;
+ }
+
+ new = early_memremap_prot(new_phys, new_size,
+ pgprot_val(pgprot_encrypted(FIXMAP_PAGE_NORMAL)));
+ if (!new) {
+ pr_err("Failed to map new boot services memmap\n");
+ return;
+ }
+
+ efi_memmap_insert(&efi.memmap, new, &mr);
+ early_memunmap(new, new_size);
+
+ efi_memmap_install(new_phys, num_entries);
+ e820__range_update(addr, size, E820_TYPE_RAM, E820_TYPE_RESERVED);
+ e820__update_table(e820_table);
+}
+
+/*
+ * Helper function for efi_reserve_boot_services() to figure out if we
+ * can free regions in efi_free_boot_services().
+ *
+ * Use this function to ensure we do not free regions owned by somebody
+ * else. We must only reserve (and then free) regions:
+ *
+ * - Not within any part of the kernel
+ * - Not the BIOS reserved area (E820_TYPE_RESERVED, E820_TYPE_NVS, etc)
+ */
+static bool can_free_region(u64 start, u64 size)
+{
+ if (start + size > __pa_symbol(_text) && start <= __pa_symbol(_end))
+ return false;
+
+ if (!e820__mapped_all(start, start+size, E820_TYPE_RAM))
+ return false;
+
+ return true;
+}
+
+void __init efi_reserve_boot_services(void)
+{
+ efi_memory_desc_t *md;
+
+ for_each_efi_memory_desc(md) {
+ u64 start = md->phys_addr;
+ u64 size = md->num_pages << EFI_PAGE_SHIFT;
+ bool already_reserved;
+
+ if (md->type != EFI_BOOT_SERVICES_CODE &&
+ md->type != EFI_BOOT_SERVICES_DATA)
+ continue;
+
+ already_reserved = memblock_is_region_reserved(start, size);
+
+ /*
+ * Because the following memblock_reserve() is paired
+ * with free_bootmem_late() for this region in
+ * efi_free_boot_services(), we must be extremely
+ * careful not to reserve, and subsequently free,
+ * critical regions of memory (like the kernel image) or
+ * those regions that somebody else has already
+ * reserved.
+ *
+ * A good example of a critical region that must not be
+ * freed is page zero (first 4Kb of memory), which may
+ * contain boot services code/data but is marked
+ * E820_TYPE_RESERVED by trim_bios_range().
+ */
+ if (!already_reserved) {
+ memblock_reserve(start, size);
+
+ /*
+ * If we are the first to reserve the region, no
+ * one else cares about it. We own it and can
+ * free it later.
+ */
+ if (can_free_region(start, size))
+ continue;
+ }
+
+ /*
+ * We don't own the region. We must not free it.
+ *
+ * Setting this bit for a boot services region really
+ * doesn't make sense as far as the firmware is
+ * concerned, but it does provide us with a way to tag
+ * those regions that must not be paired with
+ * free_bootmem_late().
+ */
+ md->attribute |= EFI_MEMORY_RUNTIME;
+ }
+}
+
+void __init efi_free_boot_services(void)
+{
+ phys_addr_t new_phys, new_size;
+ efi_memory_desc_t *md;
+ int num_entries = 0;
+ void *new, *new_md;
+
+ for_each_efi_memory_desc(md) {
+ unsigned long long start = md->phys_addr;
+ unsigned long long size = md->num_pages << EFI_PAGE_SHIFT;
+ size_t rm_size;
+
+ if (md->type != EFI_BOOT_SERVICES_CODE &&
+ md->type != EFI_BOOT_SERVICES_DATA) {
+ num_entries++;
+ continue;
+ }
+
+ /* Do not free, someone else owns it: */
+ if (md->attribute & EFI_MEMORY_RUNTIME) {
+ num_entries++;
+ continue;
+ }
+
+ /*
+ * Nasty quirk: if all sub-1MB memory is used for boot
+ * services, we can get here without having allocated the
+ * real mode trampoline. It's too late to hand boot services
+ * memory back to the memblock allocator, so instead
+ * try to manually allocate the trampoline if needed.
+ *
+ * I've seen this on a Dell XPS 13 9350 with firmware
+ * 1.4.4 with SGX enabled booting Linux via Fedora 24's
+ * grub2-efi on a hard disk. (And no, I don't know why
+ * this happened, but Linux should still try to boot rather
+ * panicing early.)
+ */
+ rm_size = real_mode_size_needed();
+ if (rm_size && (start + rm_size) < (1<<20) && size >= rm_size) {
+ set_real_mode_mem(start, rm_size);
+ start += rm_size;
+ size -= rm_size;
+ }
+
+ free_bootmem_late(start, size);
+ }
+
+ if (!num_entries)
+ return;
+
+ new_size = efi.memmap.desc_size * num_entries;
+ new_phys = efi_memmap_alloc(num_entries);
+ if (!new_phys) {
+ pr_err("Failed to allocate new EFI memmap\n");
+ return;
+ }
+
+ new = memremap(new_phys, new_size, MEMREMAP_WB);
+ if (!new) {
+ pr_err("Failed to map new EFI memmap\n");
+ return;
+ }
+
+ /*
+ * Build a new EFI memmap that excludes any boot services
+ * regions that are not tagged EFI_MEMORY_RUNTIME, since those
+ * regions have now been freed.
+ */
+ new_md = new;
+ for_each_efi_memory_desc(md) {
+ if (!(md->attribute & EFI_MEMORY_RUNTIME) &&
+ (md->type == EFI_BOOT_SERVICES_CODE ||
+ md->type == EFI_BOOT_SERVICES_DATA))
+ continue;
+
+ memcpy(new_md, md, efi.memmap.desc_size);
+ new_md += efi.memmap.desc_size;
+ }
+
+ memunmap(new);
+
+ if (efi_memmap_install(new_phys, num_entries)) {
+ pr_err("Could not install new EFI memmap\n");
+ return;
+ }
+}
+
+/*
+ * A number of config table entries get remapped to virtual addresses
+ * after entering EFI virtual mode. However, the kexec kernel requires
+ * their physical addresses therefore we pass them via setup_data and
+ * correct those entries to their respective physical addresses here.
+ *
+ * Currently only handles smbios which is necessary for some firmware
+ * implementation.
+ */
+int __init efi_reuse_config(u64 tables, int nr_tables)
+{
+ int i, sz, ret = 0;
+ void *p, *tablep;
+ struct efi_setup_data *data;
+
+ if (!efi_setup)
+ return 0;
+
+ if (!efi_enabled(EFI_64BIT))
+ return 0;
+
+ data = early_memremap(efi_setup, sizeof(*data));
+ if (!data) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ if (!data->smbios)
+ goto out_memremap;
+
+ sz = sizeof(efi_config_table_64_t);
+
+ p = tablep = early_memremap(tables, nr_tables * sz);
+ if (!p) {
+ pr_err("Could not map Configuration table!\n");
+ ret = -ENOMEM;
+ goto out_memremap;
+ }
+
+ for (i = 0; i < efi.systab->nr_tables; i++) {
+ efi_guid_t guid;
+
+ guid = ((efi_config_table_64_t *)p)->guid;
+
+ if (!efi_guidcmp(guid, SMBIOS_TABLE_GUID))
+ ((efi_config_table_64_t *)p)->table = data->smbios;
+ p += sz;
+ }
+ early_memunmap(tablep, nr_tables * sz);
+
+out_memremap:
+ early_memunmap(data, sizeof(*data));
+out:
+ return ret;
+}
+
+static const struct dmi_system_id sgi_uv1_dmi[] = {
+ { NULL, "SGI UV1",
+ { DMI_MATCH(DMI_PRODUCT_NAME, "Stoutland Platform"),
+ DMI_MATCH(DMI_PRODUCT_VERSION, "1.0"),
+ DMI_MATCH(DMI_BIOS_VENDOR, "SGI.COM"),
+ }
+ },
+ { } /* NULL entry stops DMI scanning */
+};
+
+void __init efi_apply_memmap_quirks(void)
+{
+ /*
+ * Once setup is done earlier, unmap the EFI memory map on mismatched
+ * firmware/kernel architectures since there is no support for runtime
+ * services.
+ */
+ if (!efi_runtime_supported()) {
+ pr_info("Setup done, disabling due to 32/64-bit mismatch\n");
+ efi_memmap_unmap();
+ }
+
+ /* UV2+ BIOS has a fix for this issue. UV1 still needs the quirk. */
+ if (dmi_check_system(sgi_uv1_dmi))
+ set_bit(EFI_OLD_MEMMAP, &efi.flags);
+}
+
+/*
+ * For most modern platforms the preferred method of powering off is via
+ * ACPI. However, there are some that are known to require the use of
+ * EFI runtime services and for which ACPI does not work at all.
+ *
+ * Using EFI is a last resort, to be used only if no other option
+ * exists.
+ */
+bool efi_reboot_required(void)
+{
+ if (!acpi_gbl_reduced_hardware)
+ return false;
+
+ efi_reboot_quirk_mode = EFI_RESET_WARM;
+ return true;
+}
+
+bool efi_poweroff_required(void)
+{
+ return acpi_gbl_reduced_hardware || acpi_no_s5;
+}
+
+#ifdef CONFIG_EFI_CAPSULE_QUIRK_QUARK_CSH
+
+static int qrk_capsule_setup_info(struct capsule_info *cap_info, void **pkbuff,
+ size_t hdr_bytes)
+{
+ struct quark_security_header *csh = *pkbuff;
+
+ /* Only process data block that is larger than the security header */
+ if (hdr_bytes < sizeof(struct quark_security_header))
+ return 0;
+
+ if (csh->csh_signature != QUARK_CSH_SIGNATURE ||
+ csh->headersize != QUARK_SECURITY_HEADER_SIZE)
+ return 1;
+
+ /* Only process data block if EFI header is included */
+ if (hdr_bytes < QUARK_SECURITY_HEADER_SIZE +
+ sizeof(efi_capsule_header_t))
+ return 0;
+
+ pr_debug("Quark security header detected\n");
+
+ if (csh->rsvd_next_header != 0) {
+ pr_err("multiple Quark security headers not supported\n");
+ return -EINVAL;
+ }
+
+ *pkbuff += csh->headersize;
+ cap_info->total_size = csh->headersize;
+
+ /*
+ * Update the first page pointer to skip over the CSH header.
+ */
+ cap_info->phys[0] += csh->headersize;
+
+ /*
+ * cap_info->capsule should point at a virtual mapping of the entire
+ * capsule, starting at the capsule header. Our image has the Quark
+ * security header prepended, so we cannot rely on the default vmap()
+ * mapping created by the generic capsule code.
+ * Given that the Quark firmware does not appear to care about the
+ * virtual mapping, let's just point cap_info->capsule at our copy
+ * of the capsule header.
+ */
+ cap_info->capsule = &cap_info->header;
+
+ return 1;
+}
+
+#define ICPU(family, model, quirk_handler) \
+ { X86_VENDOR_INTEL, family, model, X86_FEATURE_ANY, \
+ (unsigned long)&quirk_handler }
+
+static const struct x86_cpu_id efi_capsule_quirk_ids[] = {
+ ICPU(5, 9, qrk_capsule_setup_info), /* Intel Quark X1000 */
+ { }
+};
+
+int efi_capsule_setup_info(struct capsule_info *cap_info, void *kbuff,
+ size_t hdr_bytes)
+{
+ int (*quirk_handler)(struct capsule_info *, void **, size_t);
+ const struct x86_cpu_id *id;
+ int ret;
+
+ if (hdr_bytes < sizeof(efi_capsule_header_t))
+ return 0;
+
+ cap_info->total_size = 0;
+
+ id = x86_match_cpu(efi_capsule_quirk_ids);
+ if (id) {
+ /*
+ * The quirk handler is supposed to return
+ * - a value > 0 if the setup should continue, after advancing
+ * kbuff as needed
+ * - 0 if not enough hdr_bytes are available yet
+ * - a negative error code otherwise
+ */
+ quirk_handler = (typeof(quirk_handler))id->driver_data;
+ ret = quirk_handler(cap_info, &kbuff, hdr_bytes);
+ if (ret <= 0)
+ return ret;
+ }
+
+ memcpy(&cap_info->header, kbuff, sizeof(cap_info->header));
+
+ cap_info->total_size += cap_info->header.imagesize;
+
+ return __efi_capsule_setup_info(cap_info);
+}
+
+#endif
diff --git a/arch/x86/platform/geode/Makefile b/arch/x86/platform/geode/Makefile
new file mode 100644
index 000000000..5b51194f4
--- /dev/null
+++ b/arch/x86/platform/geode/Makefile
@@ -0,0 +1,3 @@
+obj-$(CONFIG_ALIX) += alix.o
+obj-$(CONFIG_NET5501) += net5501.o
+obj-$(CONFIG_GEOS) += geos.o
diff --git a/arch/x86/platform/geode/alix.c b/arch/x86/platform/geode/alix.c
new file mode 100644
index 000000000..1865c196f
--- /dev/null
+++ b/arch/x86/platform/geode/alix.c
@@ -0,0 +1,200 @@
+/*
+ * System Specific setup for PCEngines ALIX.
+ * At the moment this means setup of GPIO control of LEDs
+ * on Alix.2/3/6 boards.
+ *
+ *
+ * Copyright (C) 2008 Constantin Baranov <const@mimas.ru>
+ * Copyright (C) 2011 Ed Wildgoose <kernel@wildgooses.com>
+ * and Philip Prindeville <philipp@redfish-solutions.com>
+ *
+ * TODO: There are large similarities with leds-net5501.c
+ * by Alessandro Zummo <a.zummo@towertech.it>
+ * In the future leds-net5501.c should be migrated over to platform
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/moduleparam.h>
+#include <linux/leds.h>
+#include <linux/platform_device.h>
+#include <linux/gpio.h>
+#include <linux/input.h>
+#include <linux/gpio_keys.h>
+#include <linux/dmi.h>
+
+#include <asm/geode.h>
+
+#define BIOS_SIGNATURE_TINYBIOS 0xf0000
+#define BIOS_SIGNATURE_COREBOOT 0x500
+#define BIOS_REGION_SIZE 0x10000
+
+/*
+ * This driver is not modular, but to keep back compatibility
+ * with existing use cases, continuing with module_param is
+ * the easiest way forward.
+ */
+static bool force = 0;
+module_param(force, bool, 0444);
+/* FIXME: Award bios is not automatically detected as Alix platform */
+MODULE_PARM_DESC(force, "Force detection as ALIX.2/ALIX.3 platform");
+
+static struct gpio_keys_button alix_gpio_buttons[] = {
+ {
+ .code = KEY_RESTART,
+ .gpio = 24,
+ .active_low = 1,
+ .desc = "Reset button",
+ .type = EV_KEY,
+ .wakeup = 0,
+ .debounce_interval = 100,
+ .can_disable = 0,
+ }
+};
+static struct gpio_keys_platform_data alix_buttons_data = {
+ .buttons = alix_gpio_buttons,
+ .nbuttons = ARRAY_SIZE(alix_gpio_buttons),
+ .poll_interval = 20,
+};
+
+static struct platform_device alix_buttons_dev = {
+ .name = "gpio-keys-polled",
+ .id = 1,
+ .dev = {
+ .platform_data = &alix_buttons_data,
+ }
+};
+
+static struct gpio_led alix_leds[] = {
+ {
+ .name = "alix:1",
+ .gpio = 6,
+ .default_trigger = "default-on",
+ .active_low = 1,
+ },
+ {
+ .name = "alix:2",
+ .gpio = 25,
+ .default_trigger = "default-off",
+ .active_low = 1,
+ },
+ {
+ .name = "alix:3",
+ .gpio = 27,
+ .default_trigger = "default-off",
+ .active_low = 1,
+ },
+};
+
+static struct gpio_led_platform_data alix_leds_data = {
+ .num_leds = ARRAY_SIZE(alix_leds),
+ .leds = alix_leds,
+};
+
+static struct platform_device alix_leds_dev = {
+ .name = "leds-gpio",
+ .id = -1,
+ .dev.platform_data = &alix_leds_data,
+};
+
+static struct platform_device *alix_devs[] __initdata = {
+ &alix_buttons_dev,
+ &alix_leds_dev,
+};
+
+static void __init register_alix(void)
+{
+ /* Setup LED control through leds-gpio driver */
+ platform_add_devices(alix_devs, ARRAY_SIZE(alix_devs));
+}
+
+static bool __init alix_present(unsigned long bios_phys,
+ const char *alix_sig,
+ size_t alix_sig_len)
+{
+ const size_t bios_len = BIOS_REGION_SIZE;
+ const char *bios_virt;
+ const char *scan_end;
+ const char *p;
+ char name[64];
+
+ if (force) {
+ printk(KERN_NOTICE "%s: forced to skip BIOS test, "
+ "assume system is ALIX.2/ALIX.3\n",
+ KBUILD_MODNAME);
+ return true;
+ }
+
+ bios_virt = phys_to_virt(bios_phys);
+ scan_end = bios_virt + bios_len - (alix_sig_len + 2);
+ for (p = bios_virt; p < scan_end; p++) {
+ const char *tail;
+ char *a;
+
+ if (memcmp(p, alix_sig, alix_sig_len) != 0)
+ continue;
+
+ memcpy(name, p, sizeof(name));
+
+ /* remove the first \0 character from string */
+ a = strchr(name, '\0');
+ if (a)
+ *a = ' ';
+
+ /* cut the string at a newline */
+ a = strchr(name, '\r');
+ if (a)
+ *a = '\0';
+
+ tail = p + alix_sig_len;
+ if ((tail[0] == '2' || tail[0] == '3' || tail[0] == '6')) {
+ printk(KERN_INFO
+ "%s: system is recognized as \"%s\"\n",
+ KBUILD_MODNAME, name);
+ return true;
+ }
+ }
+
+ return false;
+}
+
+static bool __init alix_present_dmi(void)
+{
+ const char *vendor, *product;
+
+ vendor = dmi_get_system_info(DMI_SYS_VENDOR);
+ if (!vendor || strcmp(vendor, "PC Engines"))
+ return false;
+
+ product = dmi_get_system_info(DMI_PRODUCT_NAME);
+ if (!product || (strcmp(product, "ALIX.2D") && strcmp(product, "ALIX.6")))
+ return false;
+
+ printk(KERN_INFO "%s: system is recognized as \"%s %s\"\n",
+ KBUILD_MODNAME, vendor, product);
+
+ return true;
+}
+
+static int __init alix_init(void)
+{
+ const char tinybios_sig[] = "PC Engines ALIX.";
+ const char coreboot_sig[] = "PC Engines\0ALIX.";
+
+ if (!is_geode())
+ return 0;
+
+ if (alix_present(BIOS_SIGNATURE_TINYBIOS, tinybios_sig, sizeof(tinybios_sig) - 1) ||
+ alix_present(BIOS_SIGNATURE_COREBOOT, coreboot_sig, sizeof(coreboot_sig) - 1) ||
+ alix_present_dmi())
+ register_alix();
+
+ return 0;
+}
+device_initcall(alix_init);
diff --git a/arch/x86/platform/geode/geos.c b/arch/x86/platform/geode/geos.c
new file mode 100644
index 000000000..4fcdb9131
--- /dev/null
+++ b/arch/x86/platform/geode/geos.c
@@ -0,0 +1,122 @@
+/*
+ * System Specific setup for Traverse Technologies GEOS.
+ * At the moment this means setup of GPIO control of LEDs.
+ *
+ * Copyright (C) 2008 Constantin Baranov <const@mimas.ru>
+ * Copyright (C) 2011 Ed Wildgoose <kernel@wildgooses.com>
+ * and Philip Prindeville <philipp@redfish-solutions.com>
+ *
+ * TODO: There are large similarities with leds-net5501.c
+ * by Alessandro Zummo <a.zummo@towertech.it>
+ * In the future leds-net5501.c should be migrated over to platform
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/leds.h>
+#include <linux/platform_device.h>
+#include <linux/gpio.h>
+#include <linux/input.h>
+#include <linux/gpio_keys.h>
+#include <linux/dmi.h>
+
+#include <asm/geode.h>
+
+static struct gpio_keys_button geos_gpio_buttons[] = {
+ {
+ .code = KEY_RESTART,
+ .gpio = 3,
+ .active_low = 1,
+ .desc = "Reset button",
+ .type = EV_KEY,
+ .wakeup = 0,
+ .debounce_interval = 100,
+ .can_disable = 0,
+ }
+};
+static struct gpio_keys_platform_data geos_buttons_data = {
+ .buttons = geos_gpio_buttons,
+ .nbuttons = ARRAY_SIZE(geos_gpio_buttons),
+ .poll_interval = 20,
+};
+
+static struct platform_device geos_buttons_dev = {
+ .name = "gpio-keys-polled",
+ .id = 1,
+ .dev = {
+ .platform_data = &geos_buttons_data,
+ }
+};
+
+static struct gpio_led geos_leds[] = {
+ {
+ .name = "geos:1",
+ .gpio = 6,
+ .default_trigger = "default-on",
+ .active_low = 1,
+ },
+ {
+ .name = "geos:2",
+ .gpio = 25,
+ .default_trigger = "default-off",
+ .active_low = 1,
+ },
+ {
+ .name = "geos:3",
+ .gpio = 27,
+ .default_trigger = "default-off",
+ .active_low = 1,
+ },
+};
+
+static struct gpio_led_platform_data geos_leds_data = {
+ .num_leds = ARRAY_SIZE(geos_leds),
+ .leds = geos_leds,
+};
+
+static struct platform_device geos_leds_dev = {
+ .name = "leds-gpio",
+ .id = -1,
+ .dev.platform_data = &geos_leds_data,
+};
+
+static struct platform_device *geos_devs[] __initdata = {
+ &geos_buttons_dev,
+ &geos_leds_dev,
+};
+
+static void __init register_geos(void)
+{
+ /* Setup LED control through leds-gpio driver */
+ platform_add_devices(geos_devs, ARRAY_SIZE(geos_devs));
+}
+
+static int __init geos_init(void)
+{
+ const char *vendor, *product;
+
+ if (!is_geode())
+ return 0;
+
+ vendor = dmi_get_system_info(DMI_SYS_VENDOR);
+ if (!vendor || strcmp(vendor, "Traverse Technologies"))
+ return 0;
+
+ product = dmi_get_system_info(DMI_PRODUCT_NAME);
+ if (!product || strcmp(product, "Geos"))
+ return 0;
+
+ printk(KERN_INFO "%s: system is recognized as \"%s %s\"\n",
+ KBUILD_MODNAME, vendor, product);
+
+ register_geos();
+
+ return 0;
+}
+device_initcall(geos_init);
diff --git a/arch/x86/platform/geode/net5501.c b/arch/x86/platform/geode/net5501.c
new file mode 100644
index 000000000..a2f6b982a
--- /dev/null
+++ b/arch/x86/platform/geode/net5501.c
@@ -0,0 +1,148 @@
+/*
+ * System Specific setup for Soekris net5501
+ * At the moment this means setup of GPIO control of LEDs and buttons
+ * on net5501 boards.
+ *
+ *
+ * Copyright (C) 2008-2009 Tower Technologies
+ * Written by Alessandro Zummo <a.zummo@towertech.it>
+ *
+ * Copyright (C) 2008 Constantin Baranov <const@mimas.ru>
+ * Copyright (C) 2011 Ed Wildgoose <kernel@wildgooses.com>
+ * and Philip Prindeville <philipp@redfish-solutions.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2
+ * as published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/leds.h>
+#include <linux/platform_device.h>
+#include <linux/gpio.h>
+#include <linux/input.h>
+#include <linux/gpio_keys.h>
+
+#include <asm/geode.h>
+
+#define BIOS_REGION_BASE 0xffff0000
+#define BIOS_REGION_SIZE 0x00010000
+
+static struct gpio_keys_button net5501_gpio_buttons[] = {
+ {
+ .code = KEY_RESTART,
+ .gpio = 24,
+ .active_low = 1,
+ .desc = "Reset button",
+ .type = EV_KEY,
+ .wakeup = 0,
+ .debounce_interval = 100,
+ .can_disable = 0,
+ }
+};
+static struct gpio_keys_platform_data net5501_buttons_data = {
+ .buttons = net5501_gpio_buttons,
+ .nbuttons = ARRAY_SIZE(net5501_gpio_buttons),
+ .poll_interval = 20,
+};
+
+static struct platform_device net5501_buttons_dev = {
+ .name = "gpio-keys-polled",
+ .id = 1,
+ .dev = {
+ .platform_data = &net5501_buttons_data,
+ }
+};
+
+static struct gpio_led net5501_leds[] = {
+ {
+ .name = "net5501:1",
+ .gpio = 6,
+ .default_trigger = "default-on",
+ .active_low = 0,
+ },
+};
+
+static struct gpio_led_platform_data net5501_leds_data = {
+ .num_leds = ARRAY_SIZE(net5501_leds),
+ .leds = net5501_leds,
+};
+
+static struct platform_device net5501_leds_dev = {
+ .name = "leds-gpio",
+ .id = -1,
+ .dev.platform_data = &net5501_leds_data,
+};
+
+static struct platform_device *net5501_devs[] __initdata = {
+ &net5501_buttons_dev,
+ &net5501_leds_dev,
+};
+
+static void __init register_net5501(void)
+{
+ /* Setup LED control through leds-gpio driver */
+ platform_add_devices(net5501_devs, ARRAY_SIZE(net5501_devs));
+}
+
+struct net5501_board {
+ u16 offset;
+ u16 len;
+ char *sig;
+};
+
+static struct net5501_board __initdata boards[] = {
+ { 0xb7b, 7, "net5501" }, /* net5501 v1.33/1.33c */
+ { 0xb1f, 7, "net5501" }, /* net5501 v1.32i */
+};
+
+static bool __init net5501_present(void)
+{
+ int i;
+ unsigned char *rombase, *bios;
+ bool found = false;
+
+ rombase = ioremap(BIOS_REGION_BASE, BIOS_REGION_SIZE - 1);
+ if (!rombase) {
+ printk(KERN_ERR "%s: failed to get rombase\n", KBUILD_MODNAME);
+ return found;
+ }
+
+ bios = rombase + 0x20; /* null terminated */
+
+ if (memcmp(bios, "comBIOS", 7))
+ goto unmap;
+
+ for (i = 0; i < ARRAY_SIZE(boards); i++) {
+ unsigned char *model = rombase + boards[i].offset;
+
+ if (!memcmp(model, boards[i].sig, boards[i].len)) {
+ printk(KERN_INFO "%s: system is recognized as \"%s\"\n",
+ KBUILD_MODNAME, model);
+
+ found = true;
+ break;
+ }
+ }
+
+unmap:
+ iounmap(rombase);
+ return found;
+}
+
+static int __init net5501_init(void)
+{
+ if (!is_geode())
+ return 0;
+
+ if (!net5501_present())
+ return 0;
+
+ register_net5501();
+
+ return 0;
+}
+device_initcall(net5501_init);
diff --git a/arch/x86/platform/goldfish/Makefile b/arch/x86/platform/goldfish/Makefile
new file mode 100644
index 000000000..f030b532f
--- /dev/null
+++ b/arch/x86/platform/goldfish/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_GOLDFISH) += goldfish.o
diff --git a/arch/x86/platform/goldfish/goldfish.c b/arch/x86/platform/goldfish/goldfish.c
new file mode 100644
index 000000000..0d17c0aaf
--- /dev/null
+++ b/arch/x86/platform/goldfish/goldfish.c
@@ -0,0 +1,63 @@
+/*
+ * Copyright (C) 2007 Google, Inc.
+ * Copyright (C) 2011 Intel, Inc.
+ * Copyright (C) 2013 Intel, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+
+/*
+ * Where in virtual device memory the IO devices (timers, system controllers
+ * and so on)
+ */
+
+#define GOLDFISH_PDEV_BUS_BASE (0xff001000)
+#define GOLDFISH_PDEV_BUS_END (0xff7fffff)
+#define GOLDFISH_PDEV_BUS_IRQ (4)
+
+#define GOLDFISH_TTY_BASE (0x2000)
+
+static struct resource goldfish_pdev_bus_resources[] = {
+ {
+ .start = GOLDFISH_PDEV_BUS_BASE,
+ .end = GOLDFISH_PDEV_BUS_END,
+ .flags = IORESOURCE_MEM,
+ },
+ {
+ .start = GOLDFISH_PDEV_BUS_IRQ,
+ .end = GOLDFISH_PDEV_BUS_IRQ,
+ .flags = IORESOURCE_IRQ,
+ }
+};
+
+static bool goldfish_enable __initdata;
+
+static int __init goldfish_setup(char *str)
+{
+ goldfish_enable = true;
+ return 0;
+}
+__setup("goldfish", goldfish_setup);
+
+static int __init goldfish_init(void)
+{
+ if (!goldfish_enable)
+ return -ENODEV;
+
+ platform_device_register_simple("goldfish_pdev_bus", -1,
+ goldfish_pdev_bus_resources, 2);
+ return 0;
+}
+device_initcall(goldfish_init);
diff --git a/arch/x86/platform/intel-mid/Makefile b/arch/x86/platform/intel-mid/Makefile
new file mode 100644
index 000000000..5cf886c86
--- /dev/null
+++ b/arch/x86/platform/intel-mid/Makefile
@@ -0,0 +1,6 @@
+obj-$(CONFIG_X86_INTEL_MID) += intel-mid.o intel_mid_vrtc.o pwr.o
+
+# SFI specific code
+ifdef CONFIG_X86_INTEL_MID
+obj-$(CONFIG_SFI) += sfi.o device_libs/
+endif
diff --git a/arch/x86/platform/intel-mid/device_libs/Makefile b/arch/x86/platform/intel-mid/device_libs/Makefile
new file mode 100644
index 000000000..480fed21c
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/Makefile
@@ -0,0 +1,33 @@
+# SPDX-License-Identifier: GPL-2.0
+# Family-Level Interface Shim (FLIS)
+obj-$(subst m,y,$(CONFIG_PINCTRL_MERRIFIELD)) += platform_mrfld_pinctrl.o
+# SDHCI Devices
+obj-$(subst m,y,$(CONFIG_MMC_SDHCI_PCI)) += platform_mrfld_sd.o
+# WiFi + BT
+obj-$(subst m,y,$(CONFIG_BRCMFMAC_SDIO)) += platform_bcm43xx.o
+obj-$(subst m,y,$(CONFIG_BT_HCIUART_BCM)) += platform_bt.o
+# IPC Devices
+obj-$(subst m,y,$(CONFIG_MFD_INTEL_MSIC)) += platform_msic.o
+obj-$(subst m,y,$(CONFIG_SND_MFLD_MACHINE)) += platform_msic_audio.o
+obj-$(subst m,y,$(CONFIG_GPIO_MSIC)) += platform_msic_gpio.o
+obj-$(subst m,y,$(CONFIG_MFD_INTEL_MSIC)) += platform_msic_ocd.o
+obj-$(subst m,y,$(CONFIG_MFD_INTEL_MSIC)) += platform_msic_battery.o
+obj-$(subst m,y,$(CONFIG_INTEL_MID_POWER_BUTTON)) += platform_msic_power_btn.o
+obj-$(subst m,y,$(CONFIG_INTEL_MFLD_THERMAL)) += platform_msic_thermal.o
+# SPI Devices
+obj-$(subst m,y,$(CONFIG_SPI_SPIDEV)) += platform_mrfld_spidev.o
+# I2C Devices
+obj-$(subst m,y,$(CONFIG_SENSORS_EMC1403)) += platform_emc1403.o
+obj-$(subst m,y,$(CONFIG_SENSORS_LIS3LV02D)) += platform_lis331.o
+obj-$(subst m,y,$(CONFIG_MPU3050_I2C)) += platform_mpu3050.o
+obj-$(subst m,y,$(CONFIG_INPUT_BMA150)) += platform_bma023.o
+obj-$(subst m,y,$(CONFIG_DRM_MEDFIELD)) += platform_tc35876x.o
+# I2C GPIO Expanders
+obj-$(subst m,y,$(CONFIG_GPIO_PCA953X)) += platform_max7315.o
+obj-$(subst m,y,$(CONFIG_GPIO_PCA953X)) += platform_pcal9555a.o
+obj-$(subst m,y,$(CONFIG_GPIO_PCA953X)) += platform_tca6416.o
+# MISC Devices
+obj-$(subst m,y,$(CONFIG_KEYBOARD_GPIO)) += platform_gpio_keys.o
+obj-$(subst m,y,$(CONFIG_INTEL_MID_POWER_BUTTON)) += platform_mrfld_power_btn.o
+obj-$(subst m,y,$(CONFIG_RTC_DRV_CMOS)) += platform_mrfld_rtc.o
+obj-$(subst m,y,$(CONFIG_INTEL_MID_WATCHDOG)) += platform_mrfld_wdt.o
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c b/arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c
new file mode 100644
index 000000000..4392c15ed
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_bcm43xx.c
@@ -0,0 +1,95 @@
+/*
+ * platform_bcm43xx.c: bcm43xx platform data initilization file
+ *
+ * (C) Copyright 2016 Intel Corporation
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/gpio.h>
+#include <linux/platform_device.h>
+#include <linux/regulator/machine.h>
+#include <linux/regulator/fixed.h>
+#include <linux/sfi.h>
+
+#include <asm/intel-mid.h>
+
+#define WLAN_SFI_GPIO_IRQ_NAME "WLAN-interrupt"
+#define WLAN_SFI_GPIO_ENABLE_NAME "WLAN-enable"
+
+#define WLAN_DEV_NAME "0000:00:01.3"
+
+static struct regulator_consumer_supply bcm43xx_vmmc_supply = {
+ .dev_name = WLAN_DEV_NAME,
+ .supply = "vmmc",
+};
+
+static struct regulator_init_data bcm43xx_vmmc_data = {
+ .constraints = {
+ .valid_ops_mask = REGULATOR_CHANGE_STATUS,
+ },
+ .num_consumer_supplies = 1,
+ .consumer_supplies = &bcm43xx_vmmc_supply,
+};
+
+static struct fixed_voltage_config bcm43xx_vmmc = {
+ .supply_name = "bcm43xx-vmmc-regulator",
+ /*
+ * Announce 2.0V here to be compatible with SDIO specification. The
+ * real voltage and signaling are still 1.8V.
+ */
+ .microvolts = 2000000, /* 1.8V */
+ .gpio = -EINVAL,
+ .startup_delay = 250 * 1000, /* 250ms */
+ .enable_high = 1, /* active high */
+ .enabled_at_boot = 0, /* disabled at boot */
+ .init_data = &bcm43xx_vmmc_data,
+};
+
+static struct platform_device bcm43xx_vmmc_regulator = {
+ .name = "reg-fixed-voltage",
+ .id = PLATFORM_DEVID_AUTO,
+ .dev = {
+ .platform_data = &bcm43xx_vmmc,
+ },
+};
+
+static int __init bcm43xx_regulator_register(void)
+{
+ int ret;
+
+ bcm43xx_vmmc.gpio = get_gpio_by_name(WLAN_SFI_GPIO_ENABLE_NAME);
+ ret = platform_device_register(&bcm43xx_vmmc_regulator);
+ if (ret) {
+ pr_err("%s: vmmc regulator register failed\n", __func__);
+ return ret;
+ }
+
+ return 0;
+}
+
+static void __init *bcm43xx_platform_data(void *info)
+{
+ int ret;
+
+ ret = bcm43xx_regulator_register();
+ if (ret)
+ return NULL;
+
+ pr_info("Using generic wifi platform data\n");
+
+ /* For now it's empty */
+ return NULL;
+}
+
+static const struct devs_id bcm43xx_clk_vmmc_dev_id __initconst = {
+ .name = "bcm43xx_clk_vmmc",
+ .type = SFI_DEV_TYPE_SD,
+ .get_platform_data = &bcm43xx_platform_data,
+};
+
+sfi_device(bcm43xx_clk_vmmc_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_bma023.c b/arch/x86/platform/intel-mid/device_libs/platform_bma023.c
new file mode 100644
index 000000000..c26cf393d
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_bma023.c
@@ -0,0 +1,20 @@
+/*
+ * platform_bma023.c: bma023 platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <asm/intel-mid.h>
+
+static const struct devs_id bma023_dev_id __initconst = {
+ .name = "bma023",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+};
+
+sfi_device(bma023_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_bt.c b/arch/x86/platform/intel-mid/device_libs/platform_bt.c
new file mode 100644
index 000000000..31dce7813
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_bt.c
@@ -0,0 +1,108 @@
+/*
+ * Bluetooth platform data initialization file
+ *
+ * (C) Copyright 2017 Intel Corporation
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/gpio/machine.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+
+#include <asm/cpu_device_id.h>
+#include <asm/intel-family.h>
+#include <asm/intel-mid.h>
+
+struct bt_sfi_data {
+ struct device *dev;
+ const char *name;
+ int (*setup)(struct bt_sfi_data *ddata);
+};
+
+static struct gpiod_lookup_table tng_bt_sfi_gpio_table = {
+ .dev_id = "hci_bcm",
+ .table = {
+ GPIO_LOOKUP("0000:00:0c.0", -1, "device-wakeup", GPIO_ACTIVE_HIGH),
+ GPIO_LOOKUP("0000:00:0c.0", -1, "shutdown", GPIO_ACTIVE_HIGH),
+ GPIO_LOOKUP("0000:00:0c.0", -1, "host-wakeup", GPIO_ACTIVE_HIGH),
+ { },
+ },
+};
+
+#define TNG_BT_SFI_GPIO_DEVICE_WAKEUP "bt_wakeup"
+#define TNG_BT_SFI_GPIO_SHUTDOWN "BT-reset"
+#define TNG_BT_SFI_GPIO_HOST_WAKEUP "bt_uart_enable"
+
+static int __init tng_bt_sfi_setup(struct bt_sfi_data *ddata)
+{
+ struct gpiod_lookup_table *table = &tng_bt_sfi_gpio_table;
+ struct gpiod_lookup *lookup = table->table;
+ struct pci_dev *pdev;
+
+ /* Connected to /dev/ttyS0 */
+ pdev = pci_get_domain_bus_and_slot(0, 0, PCI_DEVFN(4, 1));
+ if (!pdev)
+ return -ENODEV;
+
+ ddata->dev = &pdev->dev;
+ ddata->name = table->dev_id;
+
+ lookup[0].chip_hwnum = get_gpio_by_name(TNG_BT_SFI_GPIO_DEVICE_WAKEUP);
+ lookup[1].chip_hwnum = get_gpio_by_name(TNG_BT_SFI_GPIO_SHUTDOWN);
+ lookup[2].chip_hwnum = get_gpio_by_name(TNG_BT_SFI_GPIO_HOST_WAKEUP);
+
+ gpiod_add_lookup_table(table);
+ return 0;
+}
+
+static struct bt_sfi_data tng_bt_sfi_data __initdata = {
+ .setup = tng_bt_sfi_setup,
+};
+
+#define ICPU(model, ddata) \
+ { X86_VENDOR_INTEL, 6, model, X86_FEATURE_ANY, (kernel_ulong_t)&ddata }
+
+static const struct x86_cpu_id bt_sfi_cpu_ids[] = {
+ ICPU(INTEL_FAM6_ATOM_SILVERMONT_MID, tng_bt_sfi_data),
+ {}
+};
+
+static int __init bt_sfi_init(void)
+{
+ struct platform_device_info info;
+ struct platform_device *pdev;
+ const struct x86_cpu_id *id;
+ struct bt_sfi_data *ddata;
+ int ret;
+
+ id = x86_match_cpu(bt_sfi_cpu_ids);
+ if (!id)
+ return -ENODEV;
+
+ ddata = (struct bt_sfi_data *)id->driver_data;
+ if (!ddata)
+ return -ENODEV;
+
+ ret = ddata->setup(ddata);
+ if (ret)
+ return ret;
+
+ memset(&info, 0, sizeof(info));
+ info.fwnode = ddata->dev->fwnode;
+ info.parent = ddata->dev;
+ info.name = ddata->name,
+ info.id = PLATFORM_DEVID_NONE,
+
+ pdev = platform_device_register_full(&info);
+ if (IS_ERR(pdev))
+ return PTR_ERR(pdev);
+
+ dev_info(ddata->dev, "Registered Bluetooth device: %s\n", ddata->name);
+ return 0;
+}
+device_initcall(bt_sfi_init);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_emc1403.c b/arch/x86/platform/intel-mid/device_libs/platform_emc1403.c
new file mode 100644
index 000000000..c259fb6c8
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_emc1403.c
@@ -0,0 +1,43 @@
+/*
+ * platform_emc1403.c: emc1403 platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/gpio.h>
+#include <linux/i2c.h>
+#include <asm/intel-mid.h>
+
+static void __init *emc1403_platform_data(void *info)
+{
+ static short intr2nd_pdata;
+ struct i2c_board_info *i2c_info = info;
+ int intr = get_gpio_by_name("thermal_int");
+ int intr2nd = get_gpio_by_name("thermal_alert");
+
+ if (intr < 0)
+ return NULL;
+ if (intr2nd < 0)
+ return NULL;
+
+ i2c_info->irq = intr + INTEL_MID_IRQ_OFFSET;
+ intr2nd_pdata = intr2nd + INTEL_MID_IRQ_OFFSET;
+
+ return &intr2nd_pdata;
+}
+
+static const struct devs_id emc1403_dev_id __initconst = {
+ .name = "emc1403",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &emc1403_platform_data,
+};
+
+sfi_device(emc1403_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_gpio_keys.c b/arch/x86/platform/intel-mid/device_libs/platform_gpio_keys.c
new file mode 100644
index 000000000..e639e3116
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_gpio_keys.c
@@ -0,0 +1,85 @@
+/*
+ * platform_gpio_keys.c: gpio_keys platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/input.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/gpio.h>
+#include <linux/gpio_keys.h>
+#include <linux/platform_device.h>
+#include <asm/intel-mid.h>
+
+#define DEVICE_NAME "gpio-keys"
+
+/*
+ * we will search these buttons in SFI GPIO table (by name)
+ * and register them dynamically. Please add all possible
+ * buttons here, we will shrink them if no GPIO found.
+ */
+static struct gpio_keys_button gpio_button[] = {
+ {KEY_POWER, -1, 1, "power_btn", EV_KEY, 0, 3000},
+ {KEY_PROG1, -1, 1, "prog_btn1", EV_KEY, 0, 20},
+ {KEY_PROG2, -1, 1, "prog_btn2", EV_KEY, 0, 20},
+ {SW_LID, -1, 1, "lid_switch", EV_SW, 0, 20},
+ {KEY_VOLUMEUP, -1, 1, "vol_up", EV_KEY, 0, 20},
+ {KEY_VOLUMEDOWN, -1, 1, "vol_down", EV_KEY, 0, 20},
+ {KEY_MUTE, -1, 1, "mute_enable", EV_KEY, 0, 20},
+ {KEY_VOLUMEUP, -1, 1, "volume_up", EV_KEY, 0, 20},
+ {KEY_VOLUMEDOWN, -1, 1, "volume_down", EV_KEY, 0, 20},
+ {KEY_CAMERA, -1, 1, "camera_full", EV_KEY, 0, 20},
+ {KEY_CAMERA_FOCUS, -1, 1, "camera_half", EV_KEY, 0, 20},
+ {SW_KEYPAD_SLIDE, -1, 1, "MagSw1", EV_SW, 0, 20},
+ {SW_KEYPAD_SLIDE, -1, 1, "MagSw2", EV_SW, 0, 20},
+};
+
+static struct gpio_keys_platform_data gpio_keys = {
+ .buttons = gpio_button,
+ .rep = 1,
+ .nbuttons = -1, /* will fill it after search */
+};
+
+static struct platform_device pb_device = {
+ .name = DEVICE_NAME,
+ .id = -1,
+ .dev = {
+ .platform_data = &gpio_keys,
+ },
+};
+
+/*
+ * Shrink the non-existent buttons, register the gpio button
+ * device if there is some
+ */
+static int __init pb_keys_init(void)
+{
+ struct gpio_keys_button *gb = gpio_button;
+ int i, good = 0;
+
+ for (i = 0; i < ARRAY_SIZE(gpio_button); i++) {
+ gb[i].gpio = get_gpio_by_name(gb[i].desc);
+ pr_debug("info[%2d]: name = %s, gpio = %d\n", i, gb[i].desc,
+ gb[i].gpio);
+ if (gb[i].gpio < 0)
+ continue;
+
+ if (i != good)
+ gb[good] = gb[i];
+ good++;
+ }
+
+ if (good) {
+ gpio_keys.nbuttons = good;
+ return platform_device_register(&pb_device);
+ }
+ return 0;
+}
+late_initcall(pb_keys_init);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_lis331.c b/arch/x86/platform/intel-mid/device_libs/platform_lis331.c
new file mode 100644
index 000000000..a35cf912d
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_lis331.c
@@ -0,0 +1,41 @@
+/*
+ * platform_lis331.c: lis331 platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/i2c.h>
+#include <linux/gpio.h>
+#include <asm/intel-mid.h>
+
+static void __init *lis331dl_platform_data(void *info)
+{
+ static short intr2nd_pdata;
+ struct i2c_board_info *i2c_info = info;
+ int intr = get_gpio_by_name("accel_int");
+ int intr2nd = get_gpio_by_name("accel_2");
+
+ if (intr < 0)
+ return NULL;
+ if (intr2nd < 0)
+ return NULL;
+
+ i2c_info->irq = intr + INTEL_MID_IRQ_OFFSET;
+ intr2nd_pdata = intr2nd + INTEL_MID_IRQ_OFFSET;
+
+ return &intr2nd_pdata;
+}
+
+static const struct devs_id lis331dl_dev_id __initconst = {
+ .name = "i2c_accel",
+ .type = SFI_DEV_TYPE_I2C,
+ .get_platform_data = &lis331dl_platform_data,
+};
+
+sfi_device(lis331dl_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_max7315.c b/arch/x86/platform/intel-mid/device_libs/platform_max7315.c
new file mode 100644
index 000000000..58337b2bc
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_max7315.c
@@ -0,0 +1,81 @@
+/*
+ * platform_max7315.c: max7315 platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/gpio.h>
+#include <linux/i2c.h>
+#include <linux/platform_data/pca953x.h>
+#include <asm/intel-mid.h>
+
+#define MAX7315_NUM 2
+
+static void __init *max7315_platform_data(void *info)
+{
+ static struct pca953x_platform_data max7315_pdata[MAX7315_NUM];
+ static int nr;
+ struct pca953x_platform_data *max7315 = &max7315_pdata[nr];
+ struct i2c_board_info *i2c_info = info;
+ int gpio_base, intr;
+ char base_pin_name[SFI_NAME_LEN + 1];
+ char intr_pin_name[SFI_NAME_LEN + 1];
+
+ if (nr == MAX7315_NUM) {
+ pr_err("too many max7315s, we only support %d\n",
+ MAX7315_NUM);
+ return NULL;
+ }
+ /* we have several max7315 on the board, we only need load several
+ * instances of the same pca953x driver to cover them
+ */
+ strcpy(i2c_info->type, "max7315");
+ if (nr++) {
+ snprintf(base_pin_name, sizeof(base_pin_name),
+ "max7315_%d_base", nr);
+ snprintf(intr_pin_name, sizeof(intr_pin_name),
+ "max7315_%d_int", nr);
+ } else {
+ strcpy(base_pin_name, "max7315_base");
+ strcpy(intr_pin_name, "max7315_int");
+ }
+
+ gpio_base = get_gpio_by_name(base_pin_name);
+ intr = get_gpio_by_name(intr_pin_name);
+
+ if (gpio_base < 0)
+ return NULL;
+ max7315->gpio_base = gpio_base;
+ if (intr != -1) {
+ i2c_info->irq = intr + INTEL_MID_IRQ_OFFSET;
+ max7315->irq_base = gpio_base + INTEL_MID_IRQ_OFFSET;
+ } else {
+ i2c_info->irq = -1;
+ max7315->irq_base = -1;
+ }
+ return max7315;
+}
+
+static const struct devs_id max7315_dev_id __initconst = {
+ .name = "i2c_max7315",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &max7315_platform_data,
+};
+
+static const struct devs_id max7315_2_dev_id __initconst = {
+ .name = "i2c_max7315_2",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &max7315_platform_data,
+};
+
+sfi_device(max7315_dev_id);
+sfi_device(max7315_2_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mpu3050.c b/arch/x86/platform/intel-mid/device_libs/platform_mpu3050.c
new file mode 100644
index 000000000..ee22864bb
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mpu3050.c
@@ -0,0 +1,36 @@
+/*
+ * platform_mpu3050.c: mpu3050 platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/gpio.h>
+#include <linux/i2c.h>
+#include <asm/intel-mid.h>
+
+static void *mpu3050_platform_data(void *info)
+{
+ struct i2c_board_info *i2c_info = info;
+ int intr = get_gpio_by_name("mpu3050_int");
+
+ if (intr < 0)
+ return NULL;
+
+ i2c_info->irq = intr + INTEL_MID_IRQ_OFFSET;
+ return NULL;
+}
+
+static const struct devs_id mpu3050_dev_id __initconst = {
+ .name = "mpu3050",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &mpu3050_platform_data,
+};
+
+sfi_device(mpu3050_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_pinctrl.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_pinctrl.c
new file mode 100644
index 000000000..4de8a664e
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_pinctrl.c
@@ -0,0 +1,43 @@
+/*
+ * Intel Merrifield FLIS platform device initialization file
+ *
+ * Copyright (C) 2016, Intel Corporation
+ *
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/platform_device.h>
+
+#include <asm/intel-mid.h>
+
+#define FLIS_BASE_ADDR 0xff0c0000
+#define FLIS_LENGTH 0x8000
+
+static struct resource mrfld_pinctrl_mmio_resource = {
+ .start = FLIS_BASE_ADDR,
+ .end = FLIS_BASE_ADDR + FLIS_LENGTH - 1,
+ .flags = IORESOURCE_MEM,
+};
+
+static struct platform_device mrfld_pinctrl_device = {
+ .name = "pinctrl-merrifield",
+ .id = PLATFORM_DEVID_NONE,
+ .resource = &mrfld_pinctrl_mmio_resource,
+ .num_resources = 1,
+};
+
+static int __init mrfld_pinctrl_init(void)
+{
+ if (intel_mid_identify_cpu() == INTEL_MID_CPU_CHIP_TANGIER)
+ return platform_device_register(&mrfld_pinctrl_device);
+
+ return -ENODEV;
+}
+arch_initcall(mrfld_pinctrl_init);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_power_btn.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_power_btn.c
new file mode 100644
index 000000000..a6c3705a2
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_power_btn.c
@@ -0,0 +1,82 @@
+/*
+ * Intel Merrifield power button support
+ *
+ * (C) Copyright 2017 Intel Corporation
+ *
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/ioport.h>
+#include <linux/platform_device.h>
+#include <linux/sfi.h>
+
+#include <asm/intel-mid.h>
+#include <asm/intel_scu_ipc.h>
+
+static struct resource mrfld_power_btn_resources[] = {
+ {
+ .flags = IORESOURCE_IRQ,
+ },
+};
+
+static struct platform_device mrfld_power_btn_dev = {
+ .name = "msic_power_btn",
+ .id = PLATFORM_DEVID_NONE,
+ .num_resources = ARRAY_SIZE(mrfld_power_btn_resources),
+ .resource = mrfld_power_btn_resources,
+};
+
+static int mrfld_power_btn_scu_status_change(struct notifier_block *nb,
+ unsigned long code, void *data)
+{
+ if (code == SCU_DOWN) {
+ platform_device_unregister(&mrfld_power_btn_dev);
+ return 0;
+ }
+
+ return platform_device_register(&mrfld_power_btn_dev);
+}
+
+static struct notifier_block mrfld_power_btn_scu_notifier = {
+ .notifier_call = mrfld_power_btn_scu_status_change,
+};
+
+static int __init register_mrfld_power_btn(void)
+{
+ if (intel_mid_identify_cpu() != INTEL_MID_CPU_CHIP_TANGIER)
+ return -ENODEV;
+
+ /*
+ * We need to be sure that the SCU IPC is ready before
+ * PMIC power button device can be registered:
+ */
+ intel_scu_notifier_add(&mrfld_power_btn_scu_notifier);
+
+ return 0;
+}
+arch_initcall(register_mrfld_power_btn);
+
+static void __init *mrfld_power_btn_platform_data(void *info)
+{
+ struct resource *res = mrfld_power_btn_resources;
+ struct sfi_device_table_entry *pentry = info;
+
+ res->start = res->end = pentry->irq;
+ return NULL;
+}
+
+static const struct devs_id mrfld_power_btn_dev_id __initconst = {
+ .name = "bcove_power_btn",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &mrfld_power_btn_platform_data,
+};
+
+sfi_device(mrfld_power_btn_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_rtc.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_rtc.c
new file mode 100644
index 000000000..3135416df
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_rtc.c
@@ -0,0 +1,48 @@
+/*
+ * Intel Merrifield legacy RTC initialization file
+ *
+ * (C) Copyright 2017 Intel Corporation
+ *
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+
+#include <asm/hw_irq.h>
+#include <asm/intel-mid.h>
+#include <asm/io_apic.h>
+#include <asm/time.h>
+#include <asm/x86_init.h>
+
+static int __init mrfld_legacy_rtc_alloc_irq(void)
+{
+ struct irq_alloc_info info;
+ int ret;
+
+ if (!x86_platform.legacy.rtc)
+ return -ENODEV;
+
+ ioapic_set_alloc_attr(&info, NUMA_NO_NODE, 1, 0);
+ ret = mp_map_gsi_to_irq(RTC_IRQ, IOAPIC_MAP_ALLOC, &info);
+ if (ret < 0) {
+ pr_info("Failed to allocate RTC interrupt. Disabling RTC\n");
+ x86_platform.legacy.rtc = 0;
+ return ret;
+ }
+
+ return 0;
+}
+
+static int __init mrfld_legacy_rtc_init(void)
+{
+ if (intel_mid_identify_cpu() != INTEL_MID_CPU_CHIP_TANGIER)
+ return -ENODEV;
+
+ return mrfld_legacy_rtc_alloc_irq();
+}
+arch_initcall(mrfld_legacy_rtc_init);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c
new file mode 100644
index 000000000..00c4a034a
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_sd.c
@@ -0,0 +1,47 @@
+/*
+ * SDHCI platform data initilisation file
+ *
+ * (C) Copyright 2016 Intel Corporation
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/pci.h>
+
+#include <linux/mmc/sdhci-pci-data.h>
+
+#include <asm/intel-mid.h>
+
+#define INTEL_MRFLD_SD 2
+#define INTEL_MRFLD_SD_CD_GPIO 77
+
+static struct sdhci_pci_data mrfld_sdhci_pci_data = {
+ .rst_n_gpio = -EINVAL,
+ .cd_gpio = INTEL_MRFLD_SD_CD_GPIO,
+};
+
+static struct sdhci_pci_data *
+mrfld_sdhci_pci_get_data(struct pci_dev *pdev, int slotno)
+{
+ unsigned int func = PCI_FUNC(pdev->devfn);
+
+ if (func == INTEL_MRFLD_SD)
+ return &mrfld_sdhci_pci_data;
+
+ return NULL;
+}
+
+static int __init mrfld_sd_init(void)
+{
+ if (intel_mid_identify_cpu() != INTEL_MID_CPU_CHIP_TANGIER)
+ return -ENODEV;
+
+ sdhci_pci_get_data = mrfld_sdhci_pci_get_data;
+ return 0;
+}
+arch_initcall(mrfld_sd_init);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_spidev.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_spidev.c
new file mode 100644
index 000000000..27186ad65
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_spidev.c
@@ -0,0 +1,54 @@
+/*
+ * spidev platform data initilization file
+ *
+ * (C) Copyright 2014, 2016 Intel Corporation
+ * Authors: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ * Dan O'Donovan <dan@emutex.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/sfi.h>
+#include <linux/spi/pxa2xx_spi.h>
+#include <linux/spi/spi.h>
+
+#include <asm/intel-mid.h>
+
+#define MRFLD_SPI_DEFAULT_DMA_BURST 8
+#define MRFLD_SPI_DEFAULT_TIMEOUT 500
+
+/* GPIO pin for spidev chipselect */
+#define MRFLD_SPIDEV_GPIO_CS 111
+
+static struct pxa2xx_spi_chip spidev_spi_chip = {
+ .dma_burst_size = MRFLD_SPI_DEFAULT_DMA_BURST,
+ .timeout = MRFLD_SPI_DEFAULT_TIMEOUT,
+ .gpio_cs = MRFLD_SPIDEV_GPIO_CS,
+};
+
+static void __init *spidev_platform_data(void *info)
+{
+ struct spi_board_info *spi_info = info;
+
+ if (intel_mid_identify_cpu() != INTEL_MID_CPU_CHIP_TANGIER)
+ return ERR_PTR(-ENODEV);
+
+ spi_info->mode = SPI_MODE_0;
+ spi_info->controller_data = &spidev_spi_chip;
+
+ return NULL;
+}
+
+static const struct devs_id spidev_dev_id __initconst = {
+ .name = "spidev",
+ .type = SFI_DEV_TYPE_SPI,
+ .delay = 0,
+ .get_platform_data = &spidev_platform_data,
+};
+
+sfi_device(spidev_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_mrfld_wdt.c b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_wdt.c
new file mode 100644
index 000000000..2acd6be13
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_mrfld_wdt.c
@@ -0,0 +1,86 @@
+/*
+ * Intel Merrifield watchdog platform device library file
+ *
+ * (C) Copyright 2014 Intel Corporation
+ * Author: David Cohen <david.a.cohen@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/platform_data/intel-mid_wdt.h>
+
+#include <asm/intel-mid.h>
+#include <asm/intel_scu_ipc.h>
+#include <asm/io_apic.h>
+#include <asm/hw_irq.h>
+
+#define TANGIER_EXT_TIMER0_MSI 12
+
+static struct platform_device wdt_dev = {
+ .name = "intel_mid_wdt",
+ .id = -1,
+};
+
+static int tangier_probe(struct platform_device *pdev)
+{
+ struct irq_alloc_info info;
+ struct intel_mid_wdt_pdata *pdata = pdev->dev.platform_data;
+ int gsi = TANGIER_EXT_TIMER0_MSI;
+ int irq;
+
+ if (!pdata)
+ return -EINVAL;
+
+ /* IOAPIC builds identity mapping between GSI and IRQ on MID */
+ ioapic_set_alloc_attr(&info, cpu_to_node(0), 1, 0);
+ irq = mp_map_gsi_to_irq(gsi, IOAPIC_MAP_ALLOC, &info);
+ if (irq < 0) {
+ dev_warn(&pdev->dev, "cannot find interrupt %d in ioapic\n", gsi);
+ return irq;
+ }
+
+ pdata->irq = irq;
+ return 0;
+}
+
+static struct intel_mid_wdt_pdata tangier_pdata = {
+ .probe = tangier_probe,
+};
+
+static int wdt_scu_status_change(struct notifier_block *nb,
+ unsigned long code, void *data)
+{
+ if (code == SCU_DOWN) {
+ platform_device_unregister(&wdt_dev);
+ return 0;
+ }
+
+ return platform_device_register(&wdt_dev);
+}
+
+static struct notifier_block wdt_scu_notifier = {
+ .notifier_call = wdt_scu_status_change,
+};
+
+static int __init register_mid_wdt(void)
+{
+ if (intel_mid_identify_cpu() != INTEL_MID_CPU_CHIP_TANGIER)
+ return -ENODEV;
+
+ wdt_dev.dev.platform_data = &tangier_pdata;
+
+ /*
+ * We need to be sure that the SCU IPC is ready before watchdog device
+ * can be registered:
+ */
+ intel_scu_notifier_add(&wdt_scu_notifier);
+
+ return 0;
+}
+arch_initcall(register_mid_wdt);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic.c b/arch/x86/platform/intel-mid/device_libs/platform_msic.c
new file mode 100644
index 000000000..e421106c1
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic.c
@@ -0,0 +1,87 @@
+/*
+ * platform_msic.c: MSIC platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/init.h>
+#include <linux/sfi.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel_scu_ipc.h>
+#include <asm/intel-mid.h>
+#include "platform_msic.h"
+
+struct intel_msic_platform_data msic_pdata;
+
+static struct resource msic_resources[] = {
+ {
+ .start = INTEL_MSIC_IRQ_PHYS_BASE,
+ .end = INTEL_MSIC_IRQ_PHYS_BASE + 64 - 1,
+ .flags = IORESOURCE_MEM,
+ },
+};
+
+static struct platform_device msic_device = {
+ .name = "intel_msic",
+ .id = -1,
+ .dev = {
+ .platform_data = &msic_pdata,
+ },
+ .num_resources = ARRAY_SIZE(msic_resources),
+ .resource = msic_resources,
+};
+
+static int msic_scu_status_change(struct notifier_block *nb,
+ unsigned long code, void *data)
+{
+ if (code == SCU_DOWN) {
+ platform_device_unregister(&msic_device);
+ return 0;
+ }
+
+ return platform_device_register(&msic_device);
+}
+
+static int __init msic_init(void)
+{
+ static struct notifier_block msic_scu_notifier = {
+ .notifier_call = msic_scu_status_change,
+ };
+
+ /*
+ * We need to be sure that the SCU IPC is ready before MSIC device
+ * can be registered.
+ */
+ if (intel_mid_has_msic())
+ intel_scu_notifier_add(&msic_scu_notifier);
+
+ return 0;
+}
+arch_initcall(msic_init);
+
+/*
+ * msic_generic_platform_data - sets generic platform data for the block
+ * @info: pointer to the SFI device table entry for this block
+ * @block: MSIC block
+ *
+ * Function sets IRQ number from the SFI table entry for given device to
+ * the MSIC platform data.
+ */
+void *msic_generic_platform_data(void *info, enum intel_msic_block block)
+{
+ struct sfi_device_table_entry *entry = info;
+
+ BUG_ON(block < 0 || block >= INTEL_MSIC_BLOCK_LAST);
+ msic_pdata.irq[block] = entry->irq;
+
+ return NULL;
+}
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic.h b/arch/x86/platform/intel-mid/device_libs/platform_msic.h
new file mode 100644
index 000000000..b7be1d041
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic.h
@@ -0,0 +1,19 @@
+/*
+ * platform_msic.h: MSIC platform data header file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+#ifndef _PLATFORM_MSIC_H_
+#define _PLATFORM_MSIC_H_
+
+extern struct intel_msic_platform_data msic_pdata;
+
+void *msic_generic_platform_data(void *info, enum intel_msic_block block);
+
+#endif
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic_audio.c b/arch/x86/platform/intel-mid/device_libs/platform_msic_audio.c
new file mode 100644
index 000000000..d4dc744dd
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic_audio.c
@@ -0,0 +1,46 @@
+/*
+ * platform_msic_audio.c: MSIC audio platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/init.h>
+#include <linux/sfi.h>
+#include <linux/platform_device.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel-mid.h>
+
+#include "platform_msic.h"
+
+static void *msic_audio_platform_data(void *info)
+{
+ struct platform_device *pdev;
+
+ pdev = platform_device_register_simple("sst-platform", -1, NULL, 0);
+
+ if (IS_ERR(pdev)) {
+ pr_err("failed to create audio platform device\n");
+ return NULL;
+ }
+
+ return msic_generic_platform_data(info, INTEL_MSIC_BLOCK_AUDIO);
+}
+
+static const struct devs_id msic_audio_dev_id __initconst = {
+ .name = "msic_audio",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &msic_audio_platform_data,
+};
+
+sfi_device(msic_audio_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic_battery.c b/arch/x86/platform/intel-mid/device_libs/platform_msic_battery.c
new file mode 100644
index 000000000..5c3e99196
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic_battery.c
@@ -0,0 +1,36 @@
+/*
+ * platform_msic_battery.c: MSIC battery platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/init.h>
+#include <linux/sfi.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel-mid.h>
+
+#include "platform_msic.h"
+
+static void __init *msic_battery_platform_data(void *info)
+{
+ return msic_generic_platform_data(info, INTEL_MSIC_BLOCK_BATTERY);
+}
+
+static const struct devs_id msic_battery_dev_id __initconst = {
+ .name = "msic_battery",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &msic_battery_platform_data,
+};
+
+sfi_device(msic_battery_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic_gpio.c b/arch/x86/platform/intel-mid/device_libs/platform_msic_gpio.c
new file mode 100644
index 000000000..9fdb88d46
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic_gpio.c
@@ -0,0 +1,47 @@
+/*
+ * platform_msic_gpio.c: MSIC GPIO platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/sfi.h>
+#include <linux/init.h>
+#include <linux/gpio.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel-mid.h>
+
+#include "platform_msic.h"
+
+static void __init *msic_gpio_platform_data(void *info)
+{
+ static struct intel_msic_gpio_pdata msic_gpio_pdata;
+
+ int gpio = get_gpio_by_name("msic_gpio_base");
+
+ if (gpio < 0)
+ return NULL;
+
+ msic_gpio_pdata.gpio_base = gpio;
+ msic_pdata.gpio = &msic_gpio_pdata;
+
+ return msic_generic_platform_data(info, INTEL_MSIC_BLOCK_GPIO);
+}
+
+static const struct devs_id msic_gpio_dev_id __initconst = {
+ .name = "msic_gpio",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &msic_gpio_platform_data,
+};
+
+sfi_device(msic_gpio_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic_ocd.c b/arch/x86/platform/intel-mid/device_libs/platform_msic_ocd.c
new file mode 100644
index 000000000..7ae37cdbf
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic_ocd.c
@@ -0,0 +1,48 @@
+/*
+ * platform_msic_ocd.c: MSIC OCD platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/sfi.h>
+#include <linux/init.h>
+#include <linux/gpio.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel-mid.h>
+
+#include "platform_msic.h"
+
+static void __init *msic_ocd_platform_data(void *info)
+{
+ static struct intel_msic_ocd_pdata msic_ocd_pdata;
+ int gpio;
+
+ gpio = get_gpio_by_name("ocd_gpio");
+
+ if (gpio < 0)
+ return NULL;
+
+ msic_ocd_pdata.gpio = gpio;
+ msic_pdata.ocd = &msic_ocd_pdata;
+
+ return msic_generic_platform_data(info, INTEL_MSIC_BLOCK_OCD);
+}
+
+static const struct devs_id msic_ocd_dev_id __initconst = {
+ .name = "msic_ocd",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &msic_ocd_platform_data,
+};
+
+sfi_device(msic_ocd_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic_power_btn.c b/arch/x86/platform/intel-mid/device_libs/platform_msic_power_btn.c
new file mode 100644
index 000000000..96809b98c
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic_power_btn.c
@@ -0,0 +1,35 @@
+/*
+ * platform_msic_power_btn.c: MSIC power btn platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/sfi.h>
+#include <linux/init.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel-mid.h>
+
+#include "platform_msic.h"
+
+static void __init *msic_power_btn_platform_data(void *info)
+{
+ return msic_generic_platform_data(info, INTEL_MSIC_BLOCK_POWER_BTN);
+}
+
+static const struct devs_id msic_power_btn_dev_id __initconst = {
+ .name = "msic_power_btn",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &msic_power_btn_platform_data,
+};
+
+sfi_device(msic_power_btn_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_msic_thermal.c b/arch/x86/platform/intel-mid/device_libs/platform_msic_thermal.c
new file mode 100644
index 000000000..3e4167d24
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_msic_thermal.c
@@ -0,0 +1,36 @@
+/*
+ * platform_msic_thermal.c: msic_thermal platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/input.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/gpio.h>
+#include <linux/platform_device.h>
+#include <linux/mfd/intel_msic.h>
+#include <asm/intel-mid.h>
+
+#include "platform_msic.h"
+
+static void __init *msic_thermal_platform_data(void *info)
+{
+ return msic_generic_platform_data(info, INTEL_MSIC_BLOCK_THERMAL);
+}
+
+static const struct devs_id msic_thermal_dev_id __initconst = {
+ .name = "msic_thermal",
+ .type = SFI_DEV_TYPE_IPC,
+ .delay = 1,
+ .msic = 1,
+ .get_platform_data = &msic_thermal_platform_data,
+};
+
+sfi_device(msic_thermal_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_pcal9555a.c b/arch/x86/platform/intel-mid/device_libs/platform_pcal9555a.c
new file mode 100644
index 000000000..429a94192
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_pcal9555a.c
@@ -0,0 +1,99 @@
+/*
+ * PCAL9555a platform data initilization file
+ *
+ * Copyright (C) 2016, Intel Corporation
+ *
+ * Authors: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ * Dan O'Donovan <dan@emutex.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/gpio.h>
+#include <linux/init.h>
+#include <linux/i2c.h>
+#include <linux/platform_data/pca953x.h>
+#include <linux/sfi.h>
+
+#include <asm/intel-mid.h>
+
+#define PCAL9555A_NUM 4
+
+static struct pca953x_platform_data pcal9555a_pdata[PCAL9555A_NUM];
+static int nr;
+
+static void __init *pcal9555a_platform_data(void *info)
+{
+ struct i2c_board_info *i2c_info = info;
+ char *type = i2c_info->type;
+ struct pca953x_platform_data *pcal9555a;
+ char base_pin_name[SFI_NAME_LEN + 1];
+ char intr_pin_name[SFI_NAME_LEN + 1];
+ int gpio_base, intr;
+
+ snprintf(base_pin_name, sizeof(base_pin_name), "%s_base", type);
+ snprintf(intr_pin_name, sizeof(intr_pin_name), "%s_int", type);
+
+ gpio_base = get_gpio_by_name(base_pin_name);
+ intr = get_gpio_by_name(intr_pin_name);
+
+ /* Check if the SFI record valid */
+ if (gpio_base == -1)
+ return NULL;
+
+ if (nr >= PCAL9555A_NUM) {
+ pr_err("%s: Too many instances, only %d supported\n", __func__,
+ PCAL9555A_NUM);
+ return NULL;
+ }
+
+ pcal9555a = &pcal9555a_pdata[nr++];
+ pcal9555a->gpio_base = gpio_base;
+
+ if (intr >= 0) {
+ i2c_info->irq = intr + INTEL_MID_IRQ_OFFSET;
+ pcal9555a->irq_base = gpio_base + INTEL_MID_IRQ_OFFSET;
+ } else {
+ i2c_info->irq = -1;
+ pcal9555a->irq_base = -1;
+ }
+
+ strcpy(type, "pcal9555a");
+ return pcal9555a;
+}
+
+static const struct devs_id pcal9555a_1_dev_id __initconst = {
+ .name = "pcal9555a-1",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &pcal9555a_platform_data,
+};
+
+static const struct devs_id pcal9555a_2_dev_id __initconst = {
+ .name = "pcal9555a-2",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &pcal9555a_platform_data,
+};
+
+static const struct devs_id pcal9555a_3_dev_id __initconst = {
+ .name = "pcal9555a-3",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &pcal9555a_platform_data,
+};
+
+static const struct devs_id pcal9555a_4_dev_id __initconst = {
+ .name = "pcal9555a-4",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &pcal9555a_platform_data,
+};
+
+sfi_device(pcal9555a_1_dev_id);
+sfi_device(pcal9555a_2_dev_id);
+sfi_device(pcal9555a_3_dev_id);
+sfi_device(pcal9555a_4_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_tc35876x.c b/arch/x86/platform/intel-mid/device_libs/platform_tc35876x.c
new file mode 100644
index 000000000..290537655
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_tc35876x.c
@@ -0,0 +1,36 @@
+/*
+ * platform_tc35876x.c: tc35876x platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/gpio.h>
+#include <linux/platform_data/tc35876x.h>
+#include <asm/intel-mid.h>
+
+/*tc35876x DSI_LVDS bridge chip and panel platform data*/
+static void *tc35876x_platform_data(void *data)
+{
+ static struct tc35876x_platform_data pdata;
+
+ /* gpio pins set to -1 will not be used by the driver */
+ pdata.gpio_bridge_reset = get_gpio_by_name("LCMB_RXEN");
+ pdata.gpio_panel_bl_en = get_gpio_by_name("6S6P_BL_EN");
+ pdata.gpio_panel_vadd = get_gpio_by_name("EN_VREG_LCD_V3P3");
+
+ return &pdata;
+}
+
+static const struct devs_id tc35876x_dev_id __initconst = {
+ .name = "i2c_disp_brig",
+ .type = SFI_DEV_TYPE_I2C,
+ .get_platform_data = &tc35876x_platform_data,
+};
+
+sfi_device(tc35876x_dev_id);
diff --git a/arch/x86/platform/intel-mid/device_libs/platform_tca6416.c b/arch/x86/platform/intel-mid/device_libs/platform_tca6416.c
new file mode 100644
index 000000000..4f41372ce
--- /dev/null
+++ b/arch/x86/platform/intel-mid/device_libs/platform_tca6416.c
@@ -0,0 +1,57 @@
+/*
+ * platform_tca6416.c: tca6416 platform data initialization file
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/platform_data/pca953x.h>
+#include <linux/i2c.h>
+#include <linux/gpio.h>
+#include <asm/intel-mid.h>
+
+#define TCA6416_NAME "tca6416"
+#define TCA6416_BASE "tca6416_base"
+#define TCA6416_INTR "tca6416_int"
+
+static void *tca6416_platform_data(void *info)
+{
+ static struct pca953x_platform_data tca6416;
+ struct i2c_board_info *i2c_info = info;
+ int gpio_base, intr;
+ char base_pin_name[SFI_NAME_LEN + 1];
+ char intr_pin_name[SFI_NAME_LEN + 1];
+
+ strcpy(i2c_info->type, TCA6416_NAME);
+ strcpy(base_pin_name, TCA6416_BASE);
+ strcpy(intr_pin_name, TCA6416_INTR);
+
+ gpio_base = get_gpio_by_name(base_pin_name);
+ intr = get_gpio_by_name(intr_pin_name);
+
+ if (gpio_base < 0)
+ return NULL;
+ tca6416.gpio_base = gpio_base;
+ if (intr >= 0) {
+ i2c_info->irq = intr + INTEL_MID_IRQ_OFFSET;
+ tca6416.irq_base = gpio_base + INTEL_MID_IRQ_OFFSET;
+ } else {
+ i2c_info->irq = -1;
+ tca6416.irq_base = -1;
+ }
+ return &tca6416;
+}
+
+static const struct devs_id tca6416_dev_id __initconst = {
+ .name = "tca6416",
+ .type = SFI_DEV_TYPE_I2C,
+ .delay = 1,
+ .get_platform_data = &tca6416_platform_data,
+};
+
+sfi_device(tca6416_dev_id);
diff --git a/arch/x86/platform/intel-mid/intel-mid.c b/arch/x86/platform/intel-mid/intel-mid.c
new file mode 100644
index 000000000..56f66eafb
--- /dev/null
+++ b/arch/x86/platform/intel-mid/intel-mid.c
@@ -0,0 +1,216 @@
+/*
+ * intel-mid.c: Intel MID platform setup code
+ *
+ * (C) Copyright 2008, 2012 Intel Corporation
+ * Author: Jacob Pan (jacob.jun.pan@intel.com)
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#define pr_fmt(fmt) "intel_mid: " fmt
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/regulator/machine.h>
+#include <linux/scatterlist.h>
+#include <linux/sfi.h>
+#include <linux/irq.h>
+#include <linux/export.h>
+#include <linux/notifier.h>
+
+#include <asm/setup.h>
+#include <asm/mpspec_def.h>
+#include <asm/hw_irq.h>
+#include <asm/apic.h>
+#include <asm/io_apic.h>
+#include <asm/intel-mid.h>
+#include <asm/intel_mid_vrtc.h>
+#include <asm/io.h>
+#include <asm/i8259.h>
+#include <asm/intel_scu_ipc.h>
+#include <asm/apb_timer.h>
+#include <asm/reboot.h>
+
+/*
+ * the clockevent devices on Moorestown/Medfield can be APBT or LAPIC clock,
+ * cmdline option x86_intel_mid_timer can be used to override the configuration
+ * to prefer one or the other.
+ * at runtime, there are basically three timer configurations:
+ * 1. per cpu apbt clock only
+ * 2. per cpu always-on lapic clocks only, this is Penwell/Medfield only
+ * 3. per cpu lapic clock (C3STOP) and one apbt clock, with broadcast.
+ *
+ * by default (without cmdline option), platform code first detects cpu type
+ * to see if we are on lincroft or penwell, then set up both lapic or apbt
+ * clocks accordingly.
+ * i.e. by default, medfield uses configuration #2, moorestown uses #1.
+ * config #3 is supported but not recommended on medfield.
+ *
+ * rating and feature summary:
+ * lapic (with C3STOP) --------- 100
+ * apbt (always-on) ------------ 110
+ * lapic (always-on,ARAT) ------ 150
+ */
+
+enum intel_mid_timer_options intel_mid_timer_options;
+
+enum intel_mid_cpu_type __intel_mid_cpu_chip;
+EXPORT_SYMBOL_GPL(__intel_mid_cpu_chip);
+
+static void intel_mid_power_off(void)
+{
+ /* Shut down South Complex via PWRMU */
+ intel_mid_pwr_power_off();
+
+ /* Only for Tangier, the rest will ignore this command */
+ intel_scu_ipc_simple_command(IPCMSG_COLD_OFF, 1);
+};
+
+static void intel_mid_reboot(void)
+{
+ intel_scu_ipc_simple_command(IPCMSG_COLD_RESET, 0);
+}
+
+static void __init intel_mid_setup_bp_timer(void)
+{
+ apbt_time_init();
+ setup_boot_APIC_clock();
+}
+
+static void __init intel_mid_time_init(void)
+{
+ sfi_table_parse(SFI_SIG_MTMR, NULL, NULL, sfi_parse_mtmr);
+
+ switch (intel_mid_timer_options) {
+ case INTEL_MID_TIMER_APBT_ONLY:
+ break;
+ case INTEL_MID_TIMER_LAPIC_APBT:
+ /* Use apbt and local apic */
+ x86_init.timers.setup_percpu_clockev = intel_mid_setup_bp_timer;
+ x86_cpuinit.setup_percpu_clockev = setup_secondary_APIC_clock;
+ return;
+ default:
+ if (!boot_cpu_has(X86_FEATURE_ARAT))
+ break;
+ /* Lapic only, no apbt */
+ x86_init.timers.setup_percpu_clockev = setup_boot_APIC_clock;
+ x86_cpuinit.setup_percpu_clockev = setup_secondary_APIC_clock;
+ return;
+ }
+
+ x86_init.timers.setup_percpu_clockev = apbt_time_init;
+}
+
+static void intel_mid_arch_setup(void)
+{
+ if (boot_cpu_data.x86 != 6) {
+ pr_err("Unknown Intel MID CPU (%d:%d), default to Penwell\n",
+ boot_cpu_data.x86, boot_cpu_data.x86_model);
+ __intel_mid_cpu_chip = INTEL_MID_CPU_CHIP_PENWELL;
+ goto out;
+ }
+
+ switch (boot_cpu_data.x86_model) {
+ case 0x35:
+ __intel_mid_cpu_chip = INTEL_MID_CPU_CHIP_CLOVERVIEW;
+ break;
+ case 0x3C:
+ case 0x4A:
+ __intel_mid_cpu_chip = INTEL_MID_CPU_CHIP_TANGIER;
+ x86_platform.legacy.rtc = 1;
+ break;
+ case 0x27:
+ default:
+ __intel_mid_cpu_chip = INTEL_MID_CPU_CHIP_PENWELL;
+ break;
+ }
+
+out:
+ /*
+ * Intel MID platforms are using explicitly defined regulators.
+ *
+ * Let the regulator core know that we do not have any additional
+ * regulators left. This lets it substitute unprovided regulators with
+ * dummy ones:
+ */
+ regulator_has_full_constraints();
+}
+
+/*
+ * Moorestown does not have external NMI source nor port 0x61 to report
+ * NMI status. The possible NMI sources are from pmu as a result of NMI
+ * watchdog or lock debug. Reading io port 0x61 results in 0xff which
+ * misled NMI handler.
+ */
+static unsigned char intel_mid_get_nmi_reason(void)
+{
+ return 0;
+}
+
+/*
+ * Moorestown specific x86_init function overrides and early setup
+ * calls.
+ */
+void __init x86_intel_mid_early_setup(void)
+{
+ x86_init.resources.probe_roms = x86_init_noop;
+ x86_init.resources.reserve_resources = x86_init_noop;
+
+ x86_init.timers.timer_init = intel_mid_time_init;
+ x86_init.timers.setup_percpu_clockev = x86_init_noop;
+ x86_init.timers.wallclock_init = intel_mid_rtc_init;
+
+ x86_init.irqs.pre_vector_init = x86_init_noop;
+
+ x86_init.oem.arch_setup = intel_mid_arch_setup;
+
+ x86_cpuinit.setup_percpu_clockev = apbt_setup_secondary_clock;
+
+ x86_platform.get_nmi_reason = intel_mid_get_nmi_reason;
+
+ x86_init.pci.arch_init = intel_mid_pci_init;
+ x86_init.pci.fixup_irqs = x86_init_noop;
+
+ legacy_pic = &null_legacy_pic;
+
+ /*
+ * Do nothing for now as everything needed done in
+ * x86_intel_mid_early_setup() below.
+ */
+ x86_init.acpi.reduced_hw_early_init = x86_init_noop;
+
+ pm_power_off = intel_mid_power_off;
+ machine_ops.emergency_restart = intel_mid_reboot;
+
+ /* Avoid searching for BIOS MP tables */
+ x86_init.mpparse.find_smp_config = x86_init_noop;
+ x86_init.mpparse.get_smp_config = x86_init_uint_noop;
+ set_bit(MP_BUS_ISA, mp_bus_not_pci);
+}
+
+/*
+ * if user does not want to use per CPU apb timer, just give it a lower rating
+ * than local apic timer and skip the late per cpu timer init.
+ */
+static inline int __init setup_x86_intel_mid_timer(char *arg)
+{
+ if (!arg)
+ return -EINVAL;
+
+ if (strcmp("apbt_only", arg) == 0)
+ intel_mid_timer_options = INTEL_MID_TIMER_APBT_ONLY;
+ else if (strcmp("lapic_and_apbt", arg) == 0)
+ intel_mid_timer_options = INTEL_MID_TIMER_LAPIC_APBT;
+ else {
+ pr_warn("X86 INTEL_MID timer option %s not recognised use x86_intel_mid_timer=apbt_only or lapic_and_apbt\n",
+ arg);
+ return -EINVAL;
+ }
+ return 0;
+}
+__setup("x86_intel_mid_timer=", setup_x86_intel_mid_timer);
diff --git a/arch/x86/platform/intel-mid/intel_mid_vrtc.c b/arch/x86/platform/intel-mid/intel_mid_vrtc.c
new file mode 100644
index 000000000..a52914aa3
--- /dev/null
+++ b/arch/x86/platform/intel-mid/intel_mid_vrtc.c
@@ -0,0 +1,177 @@
+/*
+ * intel_mid_vrtc.c: Driver for virtual RTC device on Intel MID platform
+ *
+ * (C) Copyright 2009 Intel Corporation
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ *
+ * Note:
+ * VRTC is emulated by system controller firmware, the real HW
+ * RTC is located in the PMIC device. SCU FW shadows PMIC RTC
+ * in a memory mapped IO space that is visible to the host IA
+ * processor.
+ *
+ * This driver is based on RTC CMOS driver.
+ */
+
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/sfi.h>
+#include <linux/platform_device.h>
+#include <linux/mc146818rtc.h>
+
+#include <asm/intel-mid.h>
+#include <asm/intel_mid_vrtc.h>
+#include <asm/time.h>
+#include <asm/fixmap.h>
+
+static unsigned char __iomem *vrtc_virt_base;
+
+unsigned char vrtc_cmos_read(unsigned char reg)
+{
+ unsigned char retval;
+
+ /* vRTC's registers range from 0x0 to 0xD */
+ if (reg > 0xd || !vrtc_virt_base)
+ return 0xff;
+
+ lock_cmos_prefix(reg);
+ retval = __raw_readb(vrtc_virt_base + (reg << 2));
+ lock_cmos_suffix(reg);
+ return retval;
+}
+EXPORT_SYMBOL_GPL(vrtc_cmos_read);
+
+void vrtc_cmos_write(unsigned char val, unsigned char reg)
+{
+ if (reg > 0xd || !vrtc_virt_base)
+ return;
+
+ lock_cmos_prefix(reg);
+ __raw_writeb(val, vrtc_virt_base + (reg << 2));
+ lock_cmos_suffix(reg);
+}
+EXPORT_SYMBOL_GPL(vrtc_cmos_write);
+
+void vrtc_get_time(struct timespec64 *now)
+{
+ u8 sec, min, hour, mday, mon;
+ unsigned long flags;
+ u32 year;
+
+ spin_lock_irqsave(&rtc_lock, flags);
+
+ while ((vrtc_cmos_read(RTC_FREQ_SELECT) & RTC_UIP))
+ cpu_relax();
+
+ sec = vrtc_cmos_read(RTC_SECONDS);
+ min = vrtc_cmos_read(RTC_MINUTES);
+ hour = vrtc_cmos_read(RTC_HOURS);
+ mday = vrtc_cmos_read(RTC_DAY_OF_MONTH);
+ mon = vrtc_cmos_read(RTC_MONTH);
+ year = vrtc_cmos_read(RTC_YEAR);
+
+ spin_unlock_irqrestore(&rtc_lock, flags);
+
+ /* vRTC YEAR reg contains the offset to 1972 */
+ year += 1972;
+
+ pr_info("vRTC: sec: %d min: %d hour: %d day: %d "
+ "mon: %d year: %d\n", sec, min, hour, mday, mon, year);
+
+ now->tv_sec = mktime64(year, mon, mday, hour, min, sec);
+ now->tv_nsec = 0;
+}
+
+int vrtc_set_mmss(const struct timespec64 *now)
+{
+ unsigned long flags;
+ struct rtc_time tm;
+ int year;
+ int retval = 0;
+
+ rtc_time64_to_tm(now->tv_sec, &tm);
+ if (!rtc_valid_tm(&tm) && tm.tm_year >= 72) {
+ /*
+ * tm.year is the number of years since 1900, and the
+ * vrtc need the years since 1972.
+ */
+ year = tm.tm_year - 72;
+ spin_lock_irqsave(&rtc_lock, flags);
+ vrtc_cmos_write(year, RTC_YEAR);
+ vrtc_cmos_write(tm.tm_mon, RTC_MONTH);
+ vrtc_cmos_write(tm.tm_mday, RTC_DAY_OF_MONTH);
+ vrtc_cmos_write(tm.tm_hour, RTC_HOURS);
+ vrtc_cmos_write(tm.tm_min, RTC_MINUTES);
+ vrtc_cmos_write(tm.tm_sec, RTC_SECONDS);
+ spin_unlock_irqrestore(&rtc_lock, flags);
+ } else {
+ pr_err("%s: Invalid vRTC value: write of %llx to vRTC failed\n",
+ __func__, (s64)now->tv_sec);
+ retval = -EINVAL;
+ }
+ return retval;
+}
+
+void __init intel_mid_rtc_init(void)
+{
+ unsigned long vrtc_paddr;
+
+ sfi_table_parse(SFI_SIG_MRTC, NULL, NULL, sfi_parse_mrtc);
+
+ vrtc_paddr = sfi_mrtc_array[0].phys_addr;
+ if (!sfi_mrtc_num || !vrtc_paddr)
+ return;
+
+ vrtc_virt_base = (void __iomem *)set_fixmap_offset_nocache(FIX_LNW_VRTC,
+ vrtc_paddr);
+ x86_platform.get_wallclock = vrtc_get_time;
+ x86_platform.set_wallclock = vrtc_set_mmss;
+}
+
+/*
+ * The Moorestown platform has a memory mapped virtual RTC device that emulates
+ * the programming interface of the RTC.
+ */
+
+static struct resource vrtc_resources[] = {
+ [0] = {
+ .flags = IORESOURCE_MEM,
+ },
+ [1] = {
+ .flags = IORESOURCE_IRQ,
+ }
+};
+
+static struct platform_device vrtc_device = {
+ .name = "rtc_mrst",
+ .id = -1,
+ .resource = vrtc_resources,
+ .num_resources = ARRAY_SIZE(vrtc_resources),
+};
+
+/* Register the RTC device if appropriate */
+static int __init intel_mid_device_create(void)
+{
+ /* No Moorestown, no device */
+ if (!intel_mid_identify_cpu())
+ return -ENODEV;
+ /* No timer, no device */
+ if (!sfi_mrtc_num)
+ return -ENODEV;
+
+ /* iomem resource */
+ vrtc_resources[0].start = sfi_mrtc_array[0].phys_addr;
+ vrtc_resources[0].end = sfi_mrtc_array[0].phys_addr +
+ MRST_VRTC_MAP_SZ;
+ /* irq resource */
+ vrtc_resources[1].start = sfi_mrtc_array[0].irq;
+ vrtc_resources[1].end = sfi_mrtc_array[0].irq;
+
+ return platform_device_register(&vrtc_device);
+}
+device_initcall(intel_mid_device_create);
diff --git a/arch/x86/platform/intel-mid/pwr.c b/arch/x86/platform/intel-mid/pwr.c
new file mode 100644
index 000000000..49ec5b94c
--- /dev/null
+++ b/arch/x86/platform/intel-mid/pwr.c
@@ -0,0 +1,488 @@
+/*
+ * Intel MID Power Management Unit (PWRMU) device driver
+ *
+ * Copyright (C) 2016, Intel Corporation
+ *
+ * Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * Intel MID Power Management Unit device driver handles the South Complex PCI
+ * devices such as GPDMA, SPI, I2C, PWM, and so on. By default PCI core
+ * modifies bits in PMCSR register in the PCI configuration space. This is not
+ * enough on some SoCs like Intel Tangier. In such case PCI core sets a new
+ * power state of the device in question through a PM hook registered in struct
+ * pci_platform_pm_ops (see drivers/pci/pci-mid.c).
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/mutex.h>
+#include <linux/pci.h>
+
+#include <asm/intel-mid.h>
+
+/* Registers */
+#define PM_STS 0x00
+#define PM_CMD 0x04
+#define PM_ICS 0x08
+#define PM_WKC(x) (0x10 + (x) * 4)
+#define PM_WKS(x) (0x18 + (x) * 4)
+#define PM_SSC(x) (0x20 + (x) * 4)
+#define PM_SSS(x) (0x30 + (x) * 4)
+
+/* Bits in PM_STS */
+#define PM_STS_BUSY (1 << 8)
+
+/* Bits in PM_CMD */
+#define PM_CMD_CMD(x) ((x) << 0)
+#define PM_CMD_IOC (1 << 8)
+#define PM_CMD_CM_NOP (0 << 9)
+#define PM_CMD_CM_IMMEDIATE (1 << 9)
+#define PM_CMD_CM_DELAY (2 << 9)
+#define PM_CMD_CM_TRIGGER (3 << 9)
+
+/* System states */
+#define PM_CMD_SYS_STATE_S5 (5 << 16)
+
+/* Trigger variants */
+#define PM_CMD_CFG_TRIGGER_NC (3 << 19)
+
+/* Message to wait for TRIGGER_NC case */
+#define TRIGGER_NC_MSG_2 (2 << 22)
+
+/* List of commands */
+#define CMD_SET_CFG 0x01
+
+/* Bits in PM_ICS */
+#define PM_ICS_INT_STATUS(x) ((x) & 0xff)
+#define PM_ICS_IE (1 << 8)
+#define PM_ICS_IP (1 << 9)
+#define PM_ICS_SW_INT_STS (1 << 10)
+
+/* List of interrupts */
+#define INT_INVALID 0
+#define INT_CMD_COMPLETE 1
+#define INT_CMD_ERR 2
+#define INT_WAKE_EVENT 3
+#define INT_LSS_POWER_ERR 4
+#define INT_S0iX_MSG_ERR 5
+#define INT_NO_C6 6
+#define INT_TRIGGER_ERR 7
+#define INT_INACTIVITY 8
+
+/* South Complex devices */
+#define LSS_MAX_SHARED_DEVS 4
+#define LSS_MAX_DEVS 64
+
+#define LSS_WS_BITS 1 /* wake state width */
+#define LSS_PWS_BITS 2 /* power state width */
+
+/* Supported device IDs */
+#define PCI_DEVICE_ID_PENWELL 0x0828
+#define PCI_DEVICE_ID_TANGIER 0x11a1
+
+struct mid_pwr_dev {
+ struct pci_dev *pdev;
+ pci_power_t state;
+};
+
+struct mid_pwr {
+ struct device *dev;
+ void __iomem *regs;
+ int irq;
+ bool available;
+
+ struct mutex lock;
+ struct mid_pwr_dev lss[LSS_MAX_DEVS][LSS_MAX_SHARED_DEVS];
+};
+
+static struct mid_pwr *midpwr;
+
+static u32 mid_pwr_get_state(struct mid_pwr *pwr, int reg)
+{
+ return readl(pwr->regs + PM_SSS(reg));
+}
+
+static void mid_pwr_set_state(struct mid_pwr *pwr, int reg, u32 value)
+{
+ writel(value, pwr->regs + PM_SSC(reg));
+}
+
+static void mid_pwr_set_wake(struct mid_pwr *pwr, int reg, u32 value)
+{
+ writel(value, pwr->regs + PM_WKC(reg));
+}
+
+static void mid_pwr_interrupt_disable(struct mid_pwr *pwr)
+{
+ writel(~PM_ICS_IE, pwr->regs + PM_ICS);
+}
+
+static bool mid_pwr_is_busy(struct mid_pwr *pwr)
+{
+ return !!(readl(pwr->regs + PM_STS) & PM_STS_BUSY);
+}
+
+/* Wait 500ms that the latest PWRMU command finished */
+static int mid_pwr_wait(struct mid_pwr *pwr)
+{
+ unsigned int count = 500000;
+ bool busy;
+
+ do {
+ busy = mid_pwr_is_busy(pwr);
+ if (!busy)
+ return 0;
+ udelay(1);
+ } while (--count);
+
+ return -EBUSY;
+}
+
+static int mid_pwr_wait_for_cmd(struct mid_pwr *pwr, u8 cmd)
+{
+ writel(PM_CMD_CMD(cmd) | PM_CMD_CM_IMMEDIATE, pwr->regs + PM_CMD);
+ return mid_pwr_wait(pwr);
+}
+
+static int __update_power_state(struct mid_pwr *pwr, int reg, int bit, int new)
+{
+ int curstate;
+ u32 power;
+ int ret;
+
+ /* Check if the device is already in desired state */
+ power = mid_pwr_get_state(pwr, reg);
+ curstate = (power >> bit) & 3;
+ if (curstate == new)
+ return 0;
+
+ /* Update the power state */
+ mid_pwr_set_state(pwr, reg, (power & ~(3 << bit)) | (new << bit));
+
+ /* Send command to SCU */
+ ret = mid_pwr_wait_for_cmd(pwr, CMD_SET_CFG);
+ if (ret)
+ return ret;
+
+ /* Check if the device is already in desired state */
+ power = mid_pwr_get_state(pwr, reg);
+ curstate = (power >> bit) & 3;
+ if (curstate != new)
+ return -EAGAIN;
+
+ return 0;
+}
+
+static pci_power_t __find_weakest_power_state(struct mid_pwr_dev *lss,
+ struct pci_dev *pdev,
+ pci_power_t state)
+{
+ pci_power_t weakest = PCI_D3hot;
+ unsigned int j;
+
+ /* Find device in cache or first free cell */
+ for (j = 0; j < LSS_MAX_SHARED_DEVS; j++) {
+ if (lss[j].pdev == pdev || !lss[j].pdev)
+ break;
+ }
+
+ /* Store the desired state in cache */
+ if (j < LSS_MAX_SHARED_DEVS) {
+ lss[j].pdev = pdev;
+ lss[j].state = state;
+ } else {
+ dev_WARN(&pdev->dev, "No room for device in PWRMU LSS cache\n");
+ weakest = state;
+ }
+
+ /* Find the power state we may use */
+ for (j = 0; j < LSS_MAX_SHARED_DEVS; j++) {
+ if (lss[j].state < weakest)
+ weakest = lss[j].state;
+ }
+
+ return weakest;
+}
+
+static int __set_power_state(struct mid_pwr *pwr, struct pci_dev *pdev,
+ pci_power_t state, int id, int reg, int bit)
+{
+ const char *name;
+ int ret;
+
+ state = __find_weakest_power_state(pwr->lss[id], pdev, state);
+ name = pci_power_name(state);
+
+ ret = __update_power_state(pwr, reg, bit, (__force int)state);
+ if (ret) {
+ dev_warn(&pdev->dev, "Can't set power state %s: %d\n", name, ret);
+ return ret;
+ }
+
+ dev_vdbg(&pdev->dev, "Set power state %s\n", name);
+ return 0;
+}
+
+static int mid_pwr_set_power_state(struct mid_pwr *pwr, struct pci_dev *pdev,
+ pci_power_t state)
+{
+ int id, reg, bit;
+ int ret;
+
+ id = intel_mid_pwr_get_lss_id(pdev);
+ if (id < 0)
+ return id;
+
+ reg = (id * LSS_PWS_BITS) / 32;
+ bit = (id * LSS_PWS_BITS) % 32;
+
+ /* We support states between PCI_D0 and PCI_D3hot */
+ if (state < PCI_D0)
+ state = PCI_D0;
+ if (state > PCI_D3hot)
+ state = PCI_D3hot;
+
+ mutex_lock(&pwr->lock);
+ ret = __set_power_state(pwr, pdev, state, id, reg, bit);
+ mutex_unlock(&pwr->lock);
+ return ret;
+}
+
+int intel_mid_pci_set_power_state(struct pci_dev *pdev, pci_power_t state)
+{
+ struct mid_pwr *pwr = midpwr;
+ int ret = 0;
+
+ might_sleep();
+
+ if (pwr && pwr->available)
+ ret = mid_pwr_set_power_state(pwr, pdev, state);
+ dev_vdbg(&pdev->dev, "set_power_state() returns %d\n", ret);
+
+ return 0;
+}
+
+pci_power_t intel_mid_pci_get_power_state(struct pci_dev *pdev)
+{
+ struct mid_pwr *pwr = midpwr;
+ int id, reg, bit;
+ u32 power;
+
+ if (!pwr || !pwr->available)
+ return PCI_UNKNOWN;
+
+ id = intel_mid_pwr_get_lss_id(pdev);
+ if (id < 0)
+ return PCI_UNKNOWN;
+
+ reg = (id * LSS_PWS_BITS) / 32;
+ bit = (id * LSS_PWS_BITS) % 32;
+ power = mid_pwr_get_state(pwr, reg);
+ return (__force pci_power_t)((power >> bit) & 3);
+}
+
+void intel_mid_pwr_power_off(void)
+{
+ struct mid_pwr *pwr = midpwr;
+ u32 cmd = PM_CMD_SYS_STATE_S5 |
+ PM_CMD_CMD(CMD_SET_CFG) |
+ PM_CMD_CM_TRIGGER |
+ PM_CMD_CFG_TRIGGER_NC |
+ TRIGGER_NC_MSG_2;
+
+ /* Send command to SCU */
+ writel(cmd, pwr->regs + PM_CMD);
+ mid_pwr_wait(pwr);
+}
+
+int intel_mid_pwr_get_lss_id(struct pci_dev *pdev)
+{
+ int vndr;
+ u8 id;
+
+ /*
+ * Mapping to PWRMU index is kept in the Logical SubSystem ID byte of
+ * Vendor capability.
+ */
+ vndr = pci_find_capability(pdev, PCI_CAP_ID_VNDR);
+ if (!vndr)
+ return -EINVAL;
+
+ /* Read the Logical SubSystem ID byte */
+ pci_read_config_byte(pdev, vndr + INTEL_MID_PWR_LSS_OFFSET, &id);
+ if (!(id & INTEL_MID_PWR_LSS_TYPE))
+ return -ENODEV;
+
+ id &= ~INTEL_MID_PWR_LSS_TYPE;
+ if (id >= LSS_MAX_DEVS)
+ return -ERANGE;
+
+ return id;
+}
+
+static irqreturn_t mid_pwr_irq_handler(int irq, void *dev_id)
+{
+ struct mid_pwr *pwr = dev_id;
+ u32 ics;
+
+ ics = readl(pwr->regs + PM_ICS);
+ if (!(ics & PM_ICS_IP))
+ return IRQ_NONE;
+
+ writel(ics | PM_ICS_IP, pwr->regs + PM_ICS);
+
+ dev_warn(pwr->dev, "Unexpected IRQ: %#x\n", PM_ICS_INT_STATUS(ics));
+ return IRQ_HANDLED;
+}
+
+struct mid_pwr_device_info {
+ int (*set_initial_state)(struct mid_pwr *pwr);
+};
+
+static int mid_pwr_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+ struct mid_pwr_device_info *info = (void *)id->driver_data;
+ struct device *dev = &pdev->dev;
+ struct mid_pwr *pwr;
+ int ret;
+
+ ret = pcim_enable_device(pdev);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "error: could not enable device\n");
+ return ret;
+ }
+
+ ret = pcim_iomap_regions(pdev, 1 << 0, pci_name(pdev));
+ if (ret) {
+ dev_err(&pdev->dev, "I/O memory remapping failed\n");
+ return ret;
+ }
+
+ pwr = devm_kzalloc(dev, sizeof(*pwr), GFP_KERNEL);
+ if (!pwr)
+ return -ENOMEM;
+
+ pwr->dev = dev;
+ pwr->regs = pcim_iomap_table(pdev)[0];
+ pwr->irq = pdev->irq;
+
+ mutex_init(&pwr->lock);
+
+ /* Disable interrupts */
+ mid_pwr_interrupt_disable(pwr);
+
+ if (info && info->set_initial_state) {
+ ret = info->set_initial_state(pwr);
+ if (ret)
+ dev_warn(dev, "Can't set initial state: %d\n", ret);
+ }
+
+ ret = devm_request_irq(dev, pdev->irq, mid_pwr_irq_handler,
+ IRQF_NO_SUSPEND, pci_name(pdev), pwr);
+ if (ret)
+ return ret;
+
+ pwr->available = true;
+ midpwr = pwr;
+
+ pci_set_drvdata(pdev, pwr);
+ return 0;
+}
+
+static int mid_set_initial_state(struct mid_pwr *pwr, const u32 *states)
+{
+ unsigned int i, j;
+ int ret;
+
+ /*
+ * Enable wake events.
+ *
+ * PWRMU supports up to 32 sources for wake up the system. Ungate them
+ * all here.
+ */
+ mid_pwr_set_wake(pwr, 0, 0xffffffff);
+ mid_pwr_set_wake(pwr, 1, 0xffffffff);
+
+ /*
+ * Power off South Complex devices.
+ *
+ * There is a map (see a note below) of 64 devices with 2 bits per each
+ * on 32-bit HW registers. The following calls set all devices to one
+ * known initial state, i.e. PCI_D3hot. This is done in conjunction
+ * with PMCSR setting in arch/x86/pci/intel_mid_pci.c.
+ *
+ * NOTE: The actual device mapping is provided by a platform at run
+ * time using vendor capability of PCI configuration space.
+ */
+ mid_pwr_set_state(pwr, 0, states[0]);
+ mid_pwr_set_state(pwr, 1, states[1]);
+ mid_pwr_set_state(pwr, 2, states[2]);
+ mid_pwr_set_state(pwr, 3, states[3]);
+
+ /* Send command to SCU */
+ ret = mid_pwr_wait_for_cmd(pwr, CMD_SET_CFG);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < LSS_MAX_DEVS; i++) {
+ for (j = 0; j < LSS_MAX_SHARED_DEVS; j++)
+ pwr->lss[i][j].state = PCI_D3hot;
+ }
+
+ return 0;
+}
+
+static int pnw_set_initial_state(struct mid_pwr *pwr)
+{
+ /* On Penwell SRAM must stay powered on */
+ static const u32 states[] = {
+ 0xf00fffff, /* PM_SSC(0) */
+ 0xffffffff, /* PM_SSC(1) */
+ 0xffffffff, /* PM_SSC(2) */
+ 0xffffffff, /* PM_SSC(3) */
+ };
+ return mid_set_initial_state(pwr, states);
+}
+
+static int tng_set_initial_state(struct mid_pwr *pwr)
+{
+ static const u32 states[] = {
+ 0xffffffff, /* PM_SSC(0) */
+ 0xffffffff, /* PM_SSC(1) */
+ 0xffffffff, /* PM_SSC(2) */
+ 0xffffffff, /* PM_SSC(3) */
+ };
+ return mid_set_initial_state(pwr, states);
+}
+
+static const struct mid_pwr_device_info pnw_info = {
+ .set_initial_state = pnw_set_initial_state,
+};
+
+static const struct mid_pwr_device_info tng_info = {
+ .set_initial_state = tng_set_initial_state,
+};
+
+/* This table should be in sync with the one in drivers/pci/pci-mid.c */
+static const struct pci_device_id mid_pwr_pci_ids[] = {
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_PENWELL), (kernel_ulong_t)&pnw_info },
+ { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_TANGIER), (kernel_ulong_t)&tng_info },
+ {}
+};
+
+static struct pci_driver mid_pwr_pci_driver = {
+ .name = "intel_mid_pwr",
+ .probe = mid_pwr_probe,
+ .id_table = mid_pwr_pci_ids,
+};
+
+builtin_pci_driver(mid_pwr_pci_driver);
diff --git a/arch/x86/platform/intel-mid/sfi.c b/arch/x86/platform/intel-mid/sfi.c
new file mode 100644
index 000000000..7be1e1fe9
--- /dev/null
+++ b/arch/x86/platform/intel-mid/sfi.c
@@ -0,0 +1,547 @@
+/*
+ * intel_mid_sfi.c: Intel MID SFI initialization code
+ *
+ * (C) Copyright 2013 Intel Corporation
+ * Author: Sathyanarayanan Kuppuswamy <sathyanarayanan.kuppuswamy@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; version 2
+ * of the License.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/interrupt.h>
+#include <linux/scatterlist.h>
+#include <linux/sfi.h>
+#include <linux/spi/spi.h>
+#include <linux/i2c.h>
+#include <linux/skbuff.h>
+#include <linux/gpio.h>
+#include <linux/gpio_keys.h>
+#include <linux/input.h>
+#include <linux/platform_device.h>
+#include <linux/irq.h>
+#include <linux/export.h>
+#include <linux/notifier.h>
+#include <linux/mmc/core.h>
+#include <linux/mmc/card.h>
+#include <linux/blkdev.h>
+
+#include <asm/setup.h>
+#include <asm/mpspec_def.h>
+#include <asm/hw_irq.h>
+#include <asm/apic.h>
+#include <asm/io_apic.h>
+#include <asm/intel-mid.h>
+#include <asm/intel_mid_vrtc.h>
+#include <asm/io.h>
+#include <asm/i8259.h>
+#include <asm/intel_scu_ipc.h>
+#include <asm/apb_timer.h>
+#include <asm/reboot.h>
+
+#define SFI_SIG_OEM0 "OEM0"
+#define MAX_IPCDEVS 24
+#define MAX_SCU_SPI 24
+#define MAX_SCU_I2C 24
+
+static struct platform_device *ipc_devs[MAX_IPCDEVS];
+static struct spi_board_info *spi_devs[MAX_SCU_SPI];
+static struct i2c_board_info *i2c_devs[MAX_SCU_I2C];
+static struct sfi_gpio_table_entry *gpio_table;
+static struct sfi_timer_table_entry sfi_mtimer_array[SFI_MTMR_MAX_NUM];
+static int ipc_next_dev;
+static int spi_next_dev;
+static int i2c_next_dev;
+static int i2c_bus[MAX_SCU_I2C];
+static int gpio_num_entry;
+static u32 sfi_mtimer_usage[SFI_MTMR_MAX_NUM];
+int sfi_mrtc_num;
+int sfi_mtimer_num;
+
+struct sfi_rtc_table_entry sfi_mrtc_array[SFI_MRTC_MAX];
+EXPORT_SYMBOL_GPL(sfi_mrtc_array);
+
+struct blocking_notifier_head intel_scu_notifier =
+ BLOCKING_NOTIFIER_INIT(intel_scu_notifier);
+EXPORT_SYMBOL_GPL(intel_scu_notifier);
+
+#define intel_mid_sfi_get_pdata(dev, priv) \
+ ((dev)->get_platform_data ? (dev)->get_platform_data(priv) : NULL)
+
+/* parse all the mtimer info to a static mtimer array */
+int __init sfi_parse_mtmr(struct sfi_table_header *table)
+{
+ struct sfi_table_simple *sb;
+ struct sfi_timer_table_entry *pentry;
+ struct mpc_intsrc mp_irq;
+ int totallen;
+
+ sb = (struct sfi_table_simple *)table;
+ if (!sfi_mtimer_num) {
+ sfi_mtimer_num = SFI_GET_NUM_ENTRIES(sb,
+ struct sfi_timer_table_entry);
+ pentry = (struct sfi_timer_table_entry *) sb->pentry;
+ totallen = sfi_mtimer_num * sizeof(*pentry);
+ memcpy(sfi_mtimer_array, pentry, totallen);
+ }
+
+ pr_debug("SFI MTIMER info (num = %d):\n", sfi_mtimer_num);
+ pentry = sfi_mtimer_array;
+ for (totallen = 0; totallen < sfi_mtimer_num; totallen++, pentry++) {
+ pr_debug("timer[%d]: paddr = 0x%08x, freq = %dHz, irq = %d\n",
+ totallen, (u32)pentry->phys_addr,
+ pentry->freq_hz, pentry->irq);
+ mp_irq.type = MP_INTSRC;
+ mp_irq.irqtype = mp_INT;
+ mp_irq.irqflag = MP_IRQTRIG_EDGE | MP_IRQPOL_ACTIVE_HIGH;
+ mp_irq.srcbus = MP_BUS_ISA;
+ mp_irq.srcbusirq = pentry->irq; /* IRQ */
+ mp_irq.dstapic = MP_APIC_ALL;
+ mp_irq.dstirq = pentry->irq;
+ mp_save_irq(&mp_irq);
+ mp_map_gsi_to_irq(pentry->irq, IOAPIC_MAP_ALLOC, NULL);
+ }
+
+ return 0;
+}
+
+struct sfi_timer_table_entry *sfi_get_mtmr(int hint)
+{
+ int i;
+ if (hint < sfi_mtimer_num) {
+ if (!sfi_mtimer_usage[hint]) {
+ pr_debug("hint taken for timer %d irq %d\n",
+ hint, sfi_mtimer_array[hint].irq);
+ sfi_mtimer_usage[hint] = 1;
+ return &sfi_mtimer_array[hint];
+ }
+ }
+ /* take the first timer available */
+ for (i = 0; i < sfi_mtimer_num;) {
+ if (!sfi_mtimer_usage[i]) {
+ sfi_mtimer_usage[i] = 1;
+ return &sfi_mtimer_array[i];
+ }
+ i++;
+ }
+ return NULL;
+}
+
+void sfi_free_mtmr(struct sfi_timer_table_entry *mtmr)
+{
+ int i;
+ for (i = 0; i < sfi_mtimer_num;) {
+ if (mtmr->irq == sfi_mtimer_array[i].irq) {
+ sfi_mtimer_usage[i] = 0;
+ return;
+ }
+ i++;
+ }
+}
+
+/* parse all the mrtc info to a global mrtc array */
+int __init sfi_parse_mrtc(struct sfi_table_header *table)
+{
+ struct sfi_table_simple *sb;
+ struct sfi_rtc_table_entry *pentry;
+ struct mpc_intsrc mp_irq;
+
+ int totallen;
+
+ sb = (struct sfi_table_simple *)table;
+ if (!sfi_mrtc_num) {
+ sfi_mrtc_num = SFI_GET_NUM_ENTRIES(sb,
+ struct sfi_rtc_table_entry);
+ pentry = (struct sfi_rtc_table_entry *)sb->pentry;
+ totallen = sfi_mrtc_num * sizeof(*pentry);
+ memcpy(sfi_mrtc_array, pentry, totallen);
+ }
+
+ pr_debug("SFI RTC info (num = %d):\n", sfi_mrtc_num);
+ pentry = sfi_mrtc_array;
+ for (totallen = 0; totallen < sfi_mrtc_num; totallen++, pentry++) {
+ pr_debug("RTC[%d]: paddr = 0x%08x, irq = %d\n",
+ totallen, (u32)pentry->phys_addr, pentry->irq);
+ mp_irq.type = MP_INTSRC;
+ mp_irq.irqtype = mp_INT;
+ mp_irq.irqflag = MP_IRQTRIG_LEVEL | MP_IRQPOL_ACTIVE_LOW;
+ mp_irq.srcbus = MP_BUS_ISA;
+ mp_irq.srcbusirq = pentry->irq; /* IRQ */
+ mp_irq.dstapic = MP_APIC_ALL;
+ mp_irq.dstirq = pentry->irq;
+ mp_save_irq(&mp_irq);
+ mp_map_gsi_to_irq(pentry->irq, IOAPIC_MAP_ALLOC, NULL);
+ }
+ return 0;
+}
+
+
+/*
+ * Parsing GPIO table first, since the DEVS table will need this table
+ * to map the pin name to the actual pin.
+ */
+static int __init sfi_parse_gpio(struct sfi_table_header *table)
+{
+ struct sfi_table_simple *sb;
+ struct sfi_gpio_table_entry *pentry;
+ int num, i;
+
+ if (gpio_table)
+ return 0;
+ sb = (struct sfi_table_simple *)table;
+ num = SFI_GET_NUM_ENTRIES(sb, struct sfi_gpio_table_entry);
+ pentry = (struct sfi_gpio_table_entry *)sb->pentry;
+
+ gpio_table = kmemdup(pentry, num * sizeof(*pentry), GFP_KERNEL);
+ if (!gpio_table)
+ return -1;
+ gpio_num_entry = num;
+
+ pr_debug("GPIO pin info:\n");
+ for (i = 0; i < num; i++, pentry++)
+ pr_debug("info[%2d]: controller = %16.16s, pin_name = %16.16s,"
+ " pin = %d\n", i,
+ pentry->controller_name,
+ pentry->pin_name,
+ pentry->pin_no);
+ return 0;
+}
+
+int get_gpio_by_name(const char *name)
+{
+ struct sfi_gpio_table_entry *pentry = gpio_table;
+ int i;
+
+ if (!pentry)
+ return -1;
+ for (i = 0; i < gpio_num_entry; i++, pentry++) {
+ if (!strncmp(name, pentry->pin_name, SFI_NAME_LEN))
+ return pentry->pin_no;
+ }
+ return -EINVAL;
+}
+
+static void __init intel_scu_ipc_device_register(struct platform_device *pdev)
+{
+ if (ipc_next_dev == MAX_IPCDEVS)
+ pr_err("too many SCU IPC devices");
+ else
+ ipc_devs[ipc_next_dev++] = pdev;
+}
+
+static void __init intel_scu_spi_device_register(struct spi_board_info *sdev)
+{
+ struct spi_board_info *new_dev;
+
+ if (spi_next_dev == MAX_SCU_SPI) {
+ pr_err("too many SCU SPI devices");
+ return;
+ }
+
+ new_dev = kzalloc(sizeof(*sdev), GFP_KERNEL);
+ if (!new_dev) {
+ pr_err("failed to alloc mem for delayed spi dev %s\n",
+ sdev->modalias);
+ return;
+ }
+ *new_dev = *sdev;
+
+ spi_devs[spi_next_dev++] = new_dev;
+}
+
+static void __init intel_scu_i2c_device_register(int bus,
+ struct i2c_board_info *idev)
+{
+ struct i2c_board_info *new_dev;
+
+ if (i2c_next_dev == MAX_SCU_I2C) {
+ pr_err("too many SCU I2C devices");
+ return;
+ }
+
+ new_dev = kzalloc(sizeof(*idev), GFP_KERNEL);
+ if (!new_dev) {
+ pr_err("failed to alloc mem for delayed i2c dev %s\n",
+ idev->type);
+ return;
+ }
+ *new_dev = *idev;
+
+ i2c_bus[i2c_next_dev] = bus;
+ i2c_devs[i2c_next_dev++] = new_dev;
+}
+
+/* Called by IPC driver */
+void intel_scu_devices_create(void)
+{
+ int i;
+
+ for (i = 0; i < ipc_next_dev; i++)
+ platform_device_add(ipc_devs[i]);
+
+ for (i = 0; i < spi_next_dev; i++)
+ spi_register_board_info(spi_devs[i], 1);
+
+ for (i = 0; i < i2c_next_dev; i++) {
+ struct i2c_adapter *adapter;
+ struct i2c_client *client;
+
+ adapter = i2c_get_adapter(i2c_bus[i]);
+ if (adapter) {
+ client = i2c_new_device(adapter, i2c_devs[i]);
+ if (!client)
+ pr_err("can't create i2c device %s\n",
+ i2c_devs[i]->type);
+ } else
+ i2c_register_board_info(i2c_bus[i], i2c_devs[i], 1);
+ }
+ intel_scu_notifier_post(SCU_AVAILABLE, NULL);
+}
+EXPORT_SYMBOL_GPL(intel_scu_devices_create);
+
+/* Called by IPC driver */
+void intel_scu_devices_destroy(void)
+{
+ int i;
+
+ intel_scu_notifier_post(SCU_DOWN, NULL);
+
+ for (i = 0; i < ipc_next_dev; i++)
+ platform_device_del(ipc_devs[i]);
+}
+EXPORT_SYMBOL_GPL(intel_scu_devices_destroy);
+
+static void __init install_irq_resource(struct platform_device *pdev, int irq)
+{
+ /* Single threaded */
+ static struct resource res __initdata = {
+ .name = "IRQ",
+ .flags = IORESOURCE_IRQ,
+ };
+ res.start = irq;
+ platform_device_add_resources(pdev, &res, 1);
+}
+
+static void __init sfi_handle_ipc_dev(struct sfi_device_table_entry *pentry,
+ struct devs_id *dev)
+{
+ struct platform_device *pdev;
+ void *pdata = NULL;
+
+ pr_debug("IPC bus, name = %16.16s, irq = 0x%2x\n",
+ pentry->name, pentry->irq);
+
+ /*
+ * We need to call platform init of IPC devices to fill misc_pdata
+ * structure. It will be used in msic_init for initialization.
+ */
+ pdata = intel_mid_sfi_get_pdata(dev, pentry);
+ if (IS_ERR(pdata))
+ return;
+
+ /*
+ * On Medfield the platform device creation is handled by the MSIC
+ * MFD driver so we don't need to do it here.
+ */
+ if (dev->msic && intel_mid_has_msic())
+ return;
+
+ pdev = platform_device_alloc(pentry->name, 0);
+ if (pdev == NULL) {
+ pr_err("out of memory for SFI platform device '%s'.\n",
+ pentry->name);
+ return;
+ }
+ install_irq_resource(pdev, pentry->irq);
+
+ pdev->dev.platform_data = pdata;
+ if (dev->delay)
+ intel_scu_ipc_device_register(pdev);
+ else
+ platform_device_add(pdev);
+}
+
+static void __init sfi_handle_spi_dev(struct sfi_device_table_entry *pentry,
+ struct devs_id *dev)
+{
+ struct spi_board_info spi_info;
+ void *pdata = NULL;
+
+ memset(&spi_info, 0, sizeof(spi_info));
+ strncpy(spi_info.modalias, pentry->name, SFI_NAME_LEN);
+ spi_info.irq = ((pentry->irq == (u8)0xff) ? 0 : pentry->irq);
+ spi_info.bus_num = pentry->host_num;
+ spi_info.chip_select = pentry->addr;
+ spi_info.max_speed_hz = pentry->max_freq;
+ pr_debug("SPI bus=%d, name=%16.16s, irq=0x%2x, max_freq=%d, cs=%d\n",
+ spi_info.bus_num,
+ spi_info.modalias,
+ spi_info.irq,
+ spi_info.max_speed_hz,
+ spi_info.chip_select);
+
+ pdata = intel_mid_sfi_get_pdata(dev, &spi_info);
+ if (IS_ERR(pdata))
+ return;
+
+ spi_info.platform_data = pdata;
+ if (dev->delay)
+ intel_scu_spi_device_register(&spi_info);
+ else
+ spi_register_board_info(&spi_info, 1);
+}
+
+static void __init sfi_handle_i2c_dev(struct sfi_device_table_entry *pentry,
+ struct devs_id *dev)
+{
+ struct i2c_board_info i2c_info;
+ void *pdata = NULL;
+
+ memset(&i2c_info, 0, sizeof(i2c_info));
+ strncpy(i2c_info.type, pentry->name, SFI_NAME_LEN);
+ i2c_info.irq = ((pentry->irq == (u8)0xff) ? 0 : pentry->irq);
+ i2c_info.addr = pentry->addr;
+ pr_debug("I2C bus = %d, name = %16.16s, irq = 0x%2x, addr = 0x%x\n",
+ pentry->host_num,
+ i2c_info.type,
+ i2c_info.irq,
+ i2c_info.addr);
+ pdata = intel_mid_sfi_get_pdata(dev, &i2c_info);
+ i2c_info.platform_data = pdata;
+ if (IS_ERR(pdata))
+ return;
+
+ if (dev->delay)
+ intel_scu_i2c_device_register(pentry->host_num, &i2c_info);
+ else
+ i2c_register_board_info(pentry->host_num, &i2c_info, 1);
+}
+
+static void __init sfi_handle_sd_dev(struct sfi_device_table_entry *pentry,
+ struct devs_id *dev)
+{
+ struct mid_sd_board_info sd_info;
+ void *pdata;
+
+ memset(&sd_info, 0, sizeof(sd_info));
+ strncpy(sd_info.name, pentry->name, SFI_NAME_LEN);
+ sd_info.bus_num = pentry->host_num;
+ sd_info.max_clk = pentry->max_freq;
+ sd_info.addr = pentry->addr;
+ pr_debug("SD bus = %d, name = %16.16s, max_clk = %d, addr = 0x%x\n",
+ sd_info.bus_num,
+ sd_info.name,
+ sd_info.max_clk,
+ sd_info.addr);
+ pdata = intel_mid_sfi_get_pdata(dev, &sd_info);
+ if (IS_ERR(pdata))
+ return;
+
+ /* Nothing we can do with this for now */
+ sd_info.platform_data = pdata;
+
+ pr_debug("Successfully registered %16.16s", sd_info.name);
+}
+
+extern struct devs_id *const __x86_intel_mid_dev_start[],
+ *const __x86_intel_mid_dev_end[];
+
+static struct devs_id __init *get_device_id(u8 type, char *name)
+{
+ struct devs_id *const *dev_table;
+
+ for (dev_table = __x86_intel_mid_dev_start;
+ dev_table < __x86_intel_mid_dev_end; dev_table++) {
+ struct devs_id *dev = *dev_table;
+ if (dev->type == type &&
+ !strncmp(dev->name, name, SFI_NAME_LEN)) {
+ return dev;
+ }
+ }
+
+ return NULL;
+}
+
+static int __init sfi_parse_devs(struct sfi_table_header *table)
+{
+ struct sfi_table_simple *sb;
+ struct sfi_device_table_entry *pentry;
+ struct devs_id *dev = NULL;
+ int num, i, ret;
+ int polarity;
+ struct irq_alloc_info info;
+
+ sb = (struct sfi_table_simple *)table;
+ num = SFI_GET_NUM_ENTRIES(sb, struct sfi_device_table_entry);
+ pentry = (struct sfi_device_table_entry *)sb->pentry;
+
+ for (i = 0; i < num; i++, pentry++) {
+ int irq = pentry->irq;
+
+ if (irq != (u8)0xff) { /* native RTE case */
+ /* these SPI2 devices are not exposed to system as PCI
+ * devices, but they have separate RTE entry in IOAPIC
+ * so we have to enable them one by one here
+ */
+ if (intel_mid_identify_cpu() ==
+ INTEL_MID_CPU_CHIP_TANGIER) {
+ if (!strncmp(pentry->name, "r69001-ts-i2c", 13))
+ /* active low */
+ polarity = 1;
+ else if (!strncmp(pentry->name,
+ "synaptics_3202", 14))
+ /* active low */
+ polarity = 1;
+ else if (irq == 41)
+ /* fast_int_1 */
+ polarity = 1;
+ else
+ /* active high */
+ polarity = 0;
+ } else {
+ /* PNW and CLV go with active low */
+ polarity = 1;
+ }
+
+ ioapic_set_alloc_attr(&info, NUMA_NO_NODE, 1, polarity);
+ ret = mp_map_gsi_to_irq(irq, IOAPIC_MAP_ALLOC, &info);
+ WARN_ON(ret < 0);
+ }
+
+ dev = get_device_id(pentry->type, pentry->name);
+
+ if (!dev)
+ continue;
+
+ switch (pentry->type) {
+ case SFI_DEV_TYPE_IPC:
+ sfi_handle_ipc_dev(pentry, dev);
+ break;
+ case SFI_DEV_TYPE_SPI:
+ sfi_handle_spi_dev(pentry, dev);
+ break;
+ case SFI_DEV_TYPE_I2C:
+ sfi_handle_i2c_dev(pentry, dev);
+ break;
+ case SFI_DEV_TYPE_SD:
+ sfi_handle_sd_dev(pentry, dev);
+ break;
+ case SFI_DEV_TYPE_UART:
+ case SFI_DEV_TYPE_HSI:
+ default:
+ break;
+ }
+ }
+ return 0;
+}
+
+static int __init intel_mid_platform_init(void)
+{
+ sfi_table_parse(SFI_SIG_GPIO, NULL, NULL, sfi_parse_gpio);
+ sfi_table_parse(SFI_SIG_DEVS, NULL, NULL, sfi_parse_devs);
+ return 0;
+}
+arch_initcall(intel_mid_platform_init);
diff --git a/arch/x86/platform/intel-quark/Makefile b/arch/x86/platform/intel-quark/Makefile
new file mode 100644
index 000000000..9cc57ed36
--- /dev/null
+++ b/arch/x86/platform/intel-quark/Makefile
@@ -0,0 +1,2 @@
+obj-$(CONFIG_INTEL_IMR) += imr.o
+obj-$(CONFIG_DEBUG_IMR_SELFTEST) += imr_selftest.o
diff --git a/arch/x86/platform/intel-quark/imr.c b/arch/x86/platform/intel-quark/imr.c
new file mode 100644
index 000000000..49828c270
--- /dev/null
+++ b/arch/x86/platform/intel-quark/imr.c
@@ -0,0 +1,600 @@
+/**
+ * imr.c -- Intel Isolated Memory Region driver
+ *
+ * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2015 Bryan O'Donoghue <pure.logic@nexus-software.ie>
+ *
+ * IMR registers define an isolated region of memory that can
+ * be masked to prohibit certain system agents from accessing memory.
+ * When a device behind a masked port performs an access - snooped or
+ * not, an IMR may optionally prevent that transaction from changing
+ * the state of memory or from getting correct data in response to the
+ * operation.
+ *
+ * Write data will be dropped and reads will return 0xFFFFFFFF, the
+ * system will reset and system BIOS will print out an error message to
+ * inform the user that an IMR has been violated.
+ *
+ * This code is based on the Linux MTRR code and reference code from
+ * Intel's Quark BSP EFI, Linux and grub code.
+ *
+ * See quark-x1000-datasheet.pdf for register definitions.
+ * http://www.intel.com/content/dam/www/public/us/en/documents/datasheets/quark-x1000-datasheet.pdf
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <asm-generic/sections.h>
+#include <asm/cpu_device_id.h>
+#include <asm/imr.h>
+#include <asm/iosf_mbi.h>
+#include <linux/debugfs.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+
+struct imr_device {
+ struct dentry *file;
+ bool init;
+ struct mutex lock;
+ int max_imr;
+ int reg_base;
+};
+
+static struct imr_device imr_dev;
+
+/*
+ * IMR read/write mask control registers.
+ * See quark-x1000-datasheet.pdf sections 12.7.4.5 and 12.7.4.6 for
+ * bit definitions.
+ *
+ * addr_hi
+ * 31 Lock bit
+ * 30:24 Reserved
+ * 23:2 1 KiB aligned lo address
+ * 1:0 Reserved
+ *
+ * addr_hi
+ * 31:24 Reserved
+ * 23:2 1 KiB aligned hi address
+ * 1:0 Reserved
+ */
+#define IMR_LOCK BIT(31)
+
+struct imr_regs {
+ u32 addr_lo;
+ u32 addr_hi;
+ u32 rmask;
+ u32 wmask;
+};
+
+#define IMR_NUM_REGS (sizeof(struct imr_regs)/sizeof(u32))
+#define IMR_SHIFT 8
+#define imr_to_phys(x) ((x) << IMR_SHIFT)
+#define phys_to_imr(x) ((x) >> IMR_SHIFT)
+
+/**
+ * imr_is_enabled - true if an IMR is enabled false otherwise.
+ *
+ * Determines if an IMR is enabled based on address range and read/write
+ * mask. An IMR set with an address range set to zero and a read/write
+ * access mask set to all is considered to be disabled. An IMR in any
+ * other state - for example set to zero but without read/write access
+ * all is considered to be enabled. This definition of disabled is how
+ * firmware switches off an IMR and is maintained in kernel for
+ * consistency.
+ *
+ * @imr: pointer to IMR descriptor.
+ * @return: true if IMR enabled false if disabled.
+ */
+static inline int imr_is_enabled(struct imr_regs *imr)
+{
+ return !(imr->rmask == IMR_READ_ACCESS_ALL &&
+ imr->wmask == IMR_WRITE_ACCESS_ALL &&
+ imr_to_phys(imr->addr_lo) == 0 &&
+ imr_to_phys(imr->addr_hi) == 0);
+}
+
+/**
+ * imr_read - read an IMR at a given index.
+ *
+ * Requires caller to hold imr mutex.
+ *
+ * @idev: pointer to imr_device structure.
+ * @imr_id: IMR entry to read.
+ * @imr: IMR structure representing address and access masks.
+ * @return: 0 on success or error code passed from mbi_iosf on failure.
+ */
+static int imr_read(struct imr_device *idev, u32 imr_id, struct imr_regs *imr)
+{
+ u32 reg = imr_id * IMR_NUM_REGS + idev->reg_base;
+ int ret;
+
+ ret = iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->addr_lo);
+ if (ret)
+ return ret;
+
+ ret = iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->addr_hi);
+ if (ret)
+ return ret;
+
+ ret = iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->rmask);
+ if (ret)
+ return ret;
+
+ return iosf_mbi_read(QRK_MBI_UNIT_MM, MBI_REG_READ, reg++, &imr->wmask);
+}
+
+/**
+ * imr_write - write an IMR at a given index.
+ *
+ * Requires caller to hold imr mutex.
+ * Note lock bits need to be written independently of address bits.
+ *
+ * @idev: pointer to imr_device structure.
+ * @imr_id: IMR entry to write.
+ * @imr: IMR structure representing address and access masks.
+ * @return: 0 on success or error code passed from mbi_iosf on failure.
+ */
+static int imr_write(struct imr_device *idev, u32 imr_id, struct imr_regs *imr)
+{
+ unsigned long flags;
+ u32 reg = imr_id * IMR_NUM_REGS + idev->reg_base;
+ int ret;
+
+ local_irq_save(flags);
+
+ ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->addr_lo);
+ if (ret)
+ goto failed;
+
+ ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->addr_hi);
+ if (ret)
+ goto failed;
+
+ ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->rmask);
+ if (ret)
+ goto failed;
+
+ ret = iosf_mbi_write(QRK_MBI_UNIT_MM, MBI_REG_WRITE, reg++, imr->wmask);
+ if (ret)
+ goto failed;
+
+ local_irq_restore(flags);
+ return 0;
+failed:
+ /*
+ * If writing to the IOSF failed then we're in an unknown state,
+ * likely a very bad state. An IMR in an invalid state will almost
+ * certainly lead to a memory access violation.
+ */
+ local_irq_restore(flags);
+ WARN(ret, "IOSF-MBI write fail range 0x%08x-0x%08x unreliable\n",
+ imr_to_phys(imr->addr_lo), imr_to_phys(imr->addr_hi) + IMR_MASK);
+
+ return ret;
+}
+
+/**
+ * imr_dbgfs_state_show - print state of IMR registers.
+ *
+ * @s: pointer to seq_file for output.
+ * @unused: unused parameter.
+ * @return: 0 on success or error code passed from mbi_iosf on failure.
+ */
+static int imr_dbgfs_state_show(struct seq_file *s, void *unused)
+{
+ phys_addr_t base;
+ phys_addr_t end;
+ int i;
+ struct imr_device *idev = s->private;
+ struct imr_regs imr;
+ size_t size;
+ int ret = -ENODEV;
+
+ mutex_lock(&idev->lock);
+
+ for (i = 0; i < idev->max_imr; i++) {
+
+ ret = imr_read(idev, i, &imr);
+ if (ret)
+ break;
+
+ /*
+ * Remember to add IMR_ALIGN bytes to size to indicate the
+ * inherent IMR_ALIGN size bytes contained in the masked away
+ * lower ten bits.
+ */
+ if (imr_is_enabled(&imr)) {
+ base = imr_to_phys(imr.addr_lo);
+ end = imr_to_phys(imr.addr_hi) + IMR_MASK;
+ size = end - base + 1;
+ } else {
+ base = 0;
+ end = 0;
+ size = 0;
+ }
+ seq_printf(s, "imr%02i: base=%pa, end=%pa, size=0x%08zx "
+ "rmask=0x%08x, wmask=0x%08x, %s, %s\n", i,
+ &base, &end, size, imr.rmask, imr.wmask,
+ imr_is_enabled(&imr) ? "enabled " : "disabled",
+ imr.addr_lo & IMR_LOCK ? "locked" : "unlocked");
+ }
+
+ mutex_unlock(&idev->lock);
+ return ret;
+}
+DEFINE_SHOW_ATTRIBUTE(imr_dbgfs_state);
+
+/**
+ * imr_debugfs_register - register debugfs hooks.
+ *
+ * @idev: pointer to imr_device structure.
+ * @return: 0 on success - errno on failure.
+ */
+static int imr_debugfs_register(struct imr_device *idev)
+{
+ idev->file = debugfs_create_file("imr_state", 0444, NULL, idev,
+ &imr_dbgfs_state_fops);
+ return PTR_ERR_OR_ZERO(idev->file);
+}
+
+/**
+ * imr_check_params - check passed address range IMR alignment and non-zero size
+ *
+ * @base: base address of intended IMR.
+ * @size: size of intended IMR.
+ * @return: zero on valid range -EINVAL on unaligned base/size.
+ */
+static int imr_check_params(phys_addr_t base, size_t size)
+{
+ if ((base & IMR_MASK) || (size & IMR_MASK)) {
+ pr_err("base %pa size 0x%08zx must align to 1KiB\n",
+ &base, size);
+ return -EINVAL;
+ }
+ if (size == 0)
+ return -EINVAL;
+
+ return 0;
+}
+
+/**
+ * imr_raw_size - account for the IMR_ALIGN bytes that addr_hi appends.
+ *
+ * IMR addr_hi has a built in offset of plus IMR_ALIGN (0x400) bytes from the
+ * value in the register. We need to subtract IMR_ALIGN bytes from input sizes
+ * as a result.
+ *
+ * @size: input size bytes.
+ * @return: reduced size.
+ */
+static inline size_t imr_raw_size(size_t size)
+{
+ return size - IMR_ALIGN;
+}
+
+/**
+ * imr_address_overlap - detects an address overlap.
+ *
+ * @addr: address to check against an existing IMR.
+ * @imr: imr being checked.
+ * @return: true for overlap false for no overlap.
+ */
+static inline int imr_address_overlap(phys_addr_t addr, struct imr_regs *imr)
+{
+ return addr >= imr_to_phys(imr->addr_lo) && addr <= imr_to_phys(imr->addr_hi);
+}
+
+/**
+ * imr_add_range - add an Isolated Memory Region.
+ *
+ * @base: physical base address of region aligned to 1KiB.
+ * @size: physical size of region in bytes must be aligned to 1KiB.
+ * @read_mask: read access mask.
+ * @write_mask: write access mask.
+ * @return: zero on success or negative value indicating error.
+ */
+int imr_add_range(phys_addr_t base, size_t size,
+ unsigned int rmask, unsigned int wmask)
+{
+ phys_addr_t end;
+ unsigned int i;
+ struct imr_device *idev = &imr_dev;
+ struct imr_regs imr;
+ size_t raw_size;
+ int reg;
+ int ret;
+
+ if (WARN_ONCE(idev->init == false, "driver not initialized"))
+ return -ENODEV;
+
+ ret = imr_check_params(base, size);
+ if (ret)
+ return ret;
+
+ /* Tweak the size value. */
+ raw_size = imr_raw_size(size);
+ end = base + raw_size;
+
+ /*
+ * Check for reserved IMR value common to firmware, kernel and grub
+ * indicating a disabled IMR.
+ */
+ imr.addr_lo = phys_to_imr(base);
+ imr.addr_hi = phys_to_imr(end);
+ imr.rmask = rmask;
+ imr.wmask = wmask;
+ if (!imr_is_enabled(&imr))
+ return -ENOTSUPP;
+
+ mutex_lock(&idev->lock);
+
+ /*
+ * Find a free IMR while checking for an existing overlapping range.
+ * Note there's no restriction in silicon to prevent IMR overlaps.
+ * For the sake of simplicity and ease in defining/debugging an IMR
+ * memory map we exclude IMR overlaps.
+ */
+ reg = -1;
+ for (i = 0; i < idev->max_imr; i++) {
+ ret = imr_read(idev, i, &imr);
+ if (ret)
+ goto failed;
+
+ /* Find overlap @ base or end of requested range. */
+ ret = -EINVAL;
+ if (imr_is_enabled(&imr)) {
+ if (imr_address_overlap(base, &imr))
+ goto failed;
+ if (imr_address_overlap(end, &imr))
+ goto failed;
+ } else {
+ reg = i;
+ }
+ }
+
+ /* Error out if we have no free IMR entries. */
+ if (reg == -1) {
+ ret = -ENOMEM;
+ goto failed;
+ }
+
+ pr_debug("add %d phys %pa-%pa size %zx mask 0x%08x wmask 0x%08x\n",
+ reg, &base, &end, raw_size, rmask, wmask);
+
+ /* Enable IMR at specified range and access mask. */
+ imr.addr_lo = phys_to_imr(base);
+ imr.addr_hi = phys_to_imr(end);
+ imr.rmask = rmask;
+ imr.wmask = wmask;
+
+ ret = imr_write(idev, reg, &imr);
+ if (ret < 0) {
+ /*
+ * In the highly unlikely event iosf_mbi_write failed
+ * attempt to rollback the IMR setup skipping the trapping
+ * of further IOSF write failures.
+ */
+ imr.addr_lo = 0;
+ imr.addr_hi = 0;
+ imr.rmask = IMR_READ_ACCESS_ALL;
+ imr.wmask = IMR_WRITE_ACCESS_ALL;
+ imr_write(idev, reg, &imr);
+ }
+failed:
+ mutex_unlock(&idev->lock);
+ return ret;
+}
+EXPORT_SYMBOL_GPL(imr_add_range);
+
+/**
+ * __imr_remove_range - delete an Isolated Memory Region.
+ *
+ * This function allows you to delete an IMR by its index specified by reg or
+ * by address range specified by base and size respectively. If you specify an
+ * index on its own the base and size parameters are ignored.
+ * imr_remove_range(0, base, size); delete IMR at index 0 base/size ignored.
+ * imr_remove_range(-1, base, size); delete IMR from base to base+size.
+ *
+ * @reg: imr index to remove.
+ * @base: physical base address of region aligned to 1 KiB.
+ * @size: physical size of region in bytes aligned to 1 KiB.
+ * @return: -EINVAL on invalid range or out or range id
+ * -ENODEV if reg is valid but no IMR exists or is locked
+ * 0 on success.
+ */
+static int __imr_remove_range(int reg, phys_addr_t base, size_t size)
+{
+ phys_addr_t end;
+ bool found = false;
+ unsigned int i;
+ struct imr_device *idev = &imr_dev;
+ struct imr_regs imr;
+ size_t raw_size;
+ int ret = 0;
+
+ if (WARN_ONCE(idev->init == false, "driver not initialized"))
+ return -ENODEV;
+
+ /*
+ * Validate address range if deleting by address, else we are
+ * deleting by index where base and size will be ignored.
+ */
+ if (reg == -1) {
+ ret = imr_check_params(base, size);
+ if (ret)
+ return ret;
+ }
+
+ /* Tweak the size value. */
+ raw_size = imr_raw_size(size);
+ end = base + raw_size;
+
+ mutex_lock(&idev->lock);
+
+ if (reg >= 0) {
+ /* If a specific IMR is given try to use it. */
+ ret = imr_read(idev, reg, &imr);
+ if (ret)
+ goto failed;
+
+ if (!imr_is_enabled(&imr) || imr.addr_lo & IMR_LOCK) {
+ ret = -ENODEV;
+ goto failed;
+ }
+ found = true;
+ } else {
+ /* Search for match based on address range. */
+ for (i = 0; i < idev->max_imr; i++) {
+ ret = imr_read(idev, i, &imr);
+ if (ret)
+ goto failed;
+
+ if (!imr_is_enabled(&imr) || imr.addr_lo & IMR_LOCK)
+ continue;
+
+ if ((imr_to_phys(imr.addr_lo) == base) &&
+ (imr_to_phys(imr.addr_hi) == end)) {
+ found = true;
+ reg = i;
+ break;
+ }
+ }
+ }
+
+ if (!found) {
+ ret = -ENODEV;
+ goto failed;
+ }
+
+ pr_debug("remove %d phys %pa-%pa size %zx\n", reg, &base, &end, raw_size);
+
+ /* Tear down the IMR. */
+ imr.addr_lo = 0;
+ imr.addr_hi = 0;
+ imr.rmask = IMR_READ_ACCESS_ALL;
+ imr.wmask = IMR_WRITE_ACCESS_ALL;
+
+ ret = imr_write(idev, reg, &imr);
+
+failed:
+ mutex_unlock(&idev->lock);
+ return ret;
+}
+
+/**
+ * imr_remove_range - delete an Isolated Memory Region by address
+ *
+ * This function allows you to delete an IMR by an address range specified
+ * by base and size respectively.
+ * imr_remove_range(base, size); delete IMR from base to base+size.
+ *
+ * @base: physical base address of region aligned to 1 KiB.
+ * @size: physical size of region in bytes aligned to 1 KiB.
+ * @return: -EINVAL on invalid range or out or range id
+ * -ENODEV if reg is valid but no IMR exists or is locked
+ * 0 on success.
+ */
+int imr_remove_range(phys_addr_t base, size_t size)
+{
+ return __imr_remove_range(-1, base, size);
+}
+EXPORT_SYMBOL_GPL(imr_remove_range);
+
+/**
+ * imr_clear - delete an Isolated Memory Region by index
+ *
+ * This function allows you to delete an IMR by an address range specified
+ * by the index of the IMR. Useful for initial sanitization of the IMR
+ * address map.
+ * imr_ge(base, size); delete IMR from base to base+size.
+ *
+ * @reg: imr index to remove.
+ * @return: -EINVAL on invalid range or out or range id
+ * -ENODEV if reg is valid but no IMR exists or is locked
+ * 0 on success.
+ */
+static inline int imr_clear(int reg)
+{
+ return __imr_remove_range(reg, 0, 0);
+}
+
+/**
+ * imr_fixup_memmap - Tear down IMRs used during bootup.
+ *
+ * BIOS and Grub both setup IMRs around compressed kernel, initrd memory
+ * that need to be removed before the kernel hands out one of the IMR
+ * encased addresses to a downstream DMA agent such as the SD or Ethernet.
+ * IMRs on Galileo are setup to immediately reset the system on violation.
+ * As a result if you're running a root filesystem from SD - you'll need
+ * the boot-time IMRs torn down or you'll find seemingly random resets when
+ * using your filesystem.
+ *
+ * @idev: pointer to imr_device structure.
+ * @return:
+ */
+static void __init imr_fixup_memmap(struct imr_device *idev)
+{
+ phys_addr_t base = virt_to_phys(&_text);
+ size_t size = virt_to_phys(&__end_rodata) - base;
+ unsigned long start, end;
+ int i;
+ int ret;
+
+ /* Tear down all existing unlocked IMRs. */
+ for (i = 0; i < idev->max_imr; i++)
+ imr_clear(i);
+
+ start = (unsigned long)_text;
+ end = (unsigned long)__end_rodata - 1;
+
+ /*
+ * Setup an unlocked IMR around the physical extent of the kernel
+ * from the beginning of the .text secton to the end of the
+ * .rodata section as one physically contiguous block.
+ *
+ * We don't round up @size since it is already PAGE_SIZE aligned.
+ * See vmlinux.lds.S for details.
+ */
+ ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+ if (ret < 0) {
+ pr_err("unable to setup IMR for kernel: %zu KiB (%lx - %lx)\n",
+ size / 1024, start, end);
+ } else {
+ pr_info("protecting kernel .text - .rodata: %zu KiB (%lx - %lx)\n",
+ size / 1024, start, end);
+ }
+
+}
+
+static const struct x86_cpu_id imr_ids[] __initconst = {
+ { X86_VENDOR_INTEL, 5, 9 }, /* Intel Quark SoC X1000. */
+ {}
+};
+
+/**
+ * imr_init - entry point for IMR driver.
+ *
+ * return: -ENODEV for no IMR support 0 if good to go.
+ */
+static int __init imr_init(void)
+{
+ struct imr_device *idev = &imr_dev;
+ int ret;
+
+ if (!x86_match_cpu(imr_ids) || !iosf_mbi_available())
+ return -ENODEV;
+
+ idev->max_imr = QUARK_X1000_IMR_MAX;
+ idev->reg_base = QUARK_X1000_IMR_REGBASE;
+ idev->init = true;
+
+ mutex_init(&idev->lock);
+ ret = imr_debugfs_register(idev);
+ if (ret != 0)
+ pr_warn("debugfs register failed!\n");
+ imr_fixup_memmap(idev);
+ return 0;
+}
+device_initcall(imr_init);
diff --git a/arch/x86/platform/intel-quark/imr_selftest.c b/arch/x86/platform/intel-quark/imr_selftest.c
new file mode 100644
index 000000000..42f879b75
--- /dev/null
+++ b/arch/x86/platform/intel-quark/imr_selftest.c
@@ -0,0 +1,127 @@
+// SPDX-License-Identifier: GPL-2.0
+/**
+ * imr_selftest.c -- Intel Isolated Memory Region self-test driver
+ *
+ * Copyright(c) 2013 Intel Corporation.
+ * Copyright(c) 2015 Bryan O'Donoghue <pure.logic@nexus-software.ie>
+ *
+ * IMR self test. The purpose of this module is to run a set of tests on the
+ * IMR API to validate it's sanity. We check for overlapping, reserved
+ * addresses and setup/teardown sanity.
+ *
+ */
+
+#include <asm-generic/sections.h>
+#include <asm/cpu_device_id.h>
+#include <asm/imr.h>
+#include <linux/init.h>
+#include <linux/mm.h>
+#include <linux/types.h>
+
+#define SELFTEST KBUILD_MODNAME ": "
+/**
+ * imr_self_test_result - Print result string for self test.
+ *
+ * @res: result code - true if test passed false otherwise.
+ * @fmt: format string.
+ * ... variadic argument list.
+ */
+static __printf(2, 3)
+void __init imr_self_test_result(int res, const char *fmt, ...)
+{
+ va_list vlist;
+
+ /* Print pass/fail. */
+ if (res)
+ pr_info(SELFTEST "pass ");
+ else
+ pr_info(SELFTEST "fail ");
+
+ /* Print variable string. */
+ va_start(vlist, fmt);
+ vprintk(fmt, vlist);
+ va_end(vlist);
+
+ /* Optional warning. */
+ WARN(res == 0, "test failed");
+}
+#undef SELFTEST
+
+/**
+ * imr_self_test
+ *
+ * Verify IMR self_test with some simple tests to verify overlap,
+ * zero sized allocations and 1 KiB sized areas.
+ *
+ */
+static void __init imr_self_test(void)
+{
+ phys_addr_t base = virt_to_phys(&_text);
+ size_t size = virt_to_phys(&__end_rodata) - base;
+ const char *fmt_over = "overlapped IMR @ (0x%08lx - 0x%08lx)\n";
+ int ret;
+
+ /* Test zero zero. */
+ ret = imr_add_range(0, 0, 0, 0);
+ imr_self_test_result(ret < 0, "zero sized IMR\n");
+
+ /* Test exact overlap. */
+ ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+ imr_self_test_result(ret < 0, fmt_over, __va(base), __va(base + size));
+
+ /* Test overlap with base inside of existing. */
+ base += size - IMR_ALIGN;
+ ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+ imr_self_test_result(ret < 0, fmt_over, __va(base), __va(base + size));
+
+ /* Test overlap with end inside of existing. */
+ base -= size + IMR_ALIGN * 2;
+ ret = imr_add_range(base, size, IMR_CPU, IMR_CPU);
+ imr_self_test_result(ret < 0, fmt_over, __va(base), __va(base + size));
+
+ /* Test that a 1 KiB IMR @ zero with read/write all will bomb out. */
+ ret = imr_add_range(0, IMR_ALIGN, IMR_READ_ACCESS_ALL,
+ IMR_WRITE_ACCESS_ALL);
+ imr_self_test_result(ret < 0, "1KiB IMR @ 0x00000000 - access-all\n");
+
+ /* Test that a 1 KiB IMR @ zero with CPU only will work. */
+ ret = imr_add_range(0, IMR_ALIGN, IMR_CPU, IMR_CPU);
+ imr_self_test_result(ret >= 0, "1KiB IMR @ 0x00000000 - cpu-access\n");
+ if (ret >= 0) {
+ ret = imr_remove_range(0, IMR_ALIGN);
+ imr_self_test_result(ret == 0, "teardown - cpu-access\n");
+ }
+
+ /* Test 2 KiB works. */
+ size = IMR_ALIGN * 2;
+ ret = imr_add_range(0, size, IMR_READ_ACCESS_ALL, IMR_WRITE_ACCESS_ALL);
+ imr_self_test_result(ret >= 0, "2KiB IMR @ 0x00000000\n");
+ if (ret >= 0) {
+ ret = imr_remove_range(0, size);
+ imr_self_test_result(ret == 0, "teardown 2KiB\n");
+ }
+}
+
+static const struct x86_cpu_id imr_ids[] __initconst = {
+ { X86_VENDOR_INTEL, 5, 9 }, /* Intel Quark SoC X1000. */
+ {}
+};
+
+/**
+ * imr_self_test_init - entry point for IMR driver.
+ *
+ * return: -ENODEV for no IMR support 0 if good to go.
+ */
+static int __init imr_self_test_init(void)
+{
+ if (x86_match_cpu(imr_ids))
+ imr_self_test();
+ return 0;
+}
+
+/**
+ * imr_self_test_exit - exit point for IMR code.
+ *
+ * return:
+ */
+device_initcall(imr_self_test_init);
diff --git a/arch/x86/platform/intel/Makefile b/arch/x86/platform/intel/Makefile
new file mode 100644
index 000000000..b878032fb
--- /dev/null
+++ b/arch/x86/platform/intel/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_IOSF_MBI) += iosf_mbi.o
diff --git a/arch/x86/platform/intel/iosf_mbi.c b/arch/x86/platform/intel/iosf_mbi.c
new file mode 100644
index 000000000..6f37a2137
--- /dev/null
+++ b/arch/x86/platform/intel/iosf_mbi.c
@@ -0,0 +1,391 @@
+/*
+ * IOSF-SB MailBox Interface Driver
+ * Copyright (c) 2013, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ *
+ * The IOSF-SB is a fabric bus available on Atom based SOC's that uses a
+ * mailbox interface (MBI) to communicate with mutiple devices. This
+ * driver implements access to this interface for those platforms that can
+ * enumerate the device using PCI.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/pci.h>
+#include <linux/debugfs.h>
+#include <linux/capability.h>
+
+#include <asm/iosf_mbi.h>
+
+#define PCI_DEVICE_ID_BAYTRAIL 0x0F00
+#define PCI_DEVICE_ID_BRASWELL 0x2280
+#define PCI_DEVICE_ID_QUARK_X1000 0x0958
+#define PCI_DEVICE_ID_TANGIER 0x1170
+
+static struct pci_dev *mbi_pdev;
+static DEFINE_SPINLOCK(iosf_mbi_lock);
+static DEFINE_MUTEX(iosf_mbi_punit_mutex);
+static BLOCKING_NOTIFIER_HEAD(iosf_mbi_pmic_bus_access_notifier);
+
+static inline u32 iosf_mbi_form_mcr(u8 op, u8 port, u8 offset)
+{
+ return (op << 24) | (port << 16) | (offset << 8) | MBI_ENABLE;
+}
+
+static int iosf_mbi_pci_read_mdr(u32 mcrx, u32 mcr, u32 *mdr)
+{
+ int result;
+
+ if (!mbi_pdev)
+ return -ENODEV;
+
+ if (mcrx) {
+ result = pci_write_config_dword(mbi_pdev, MBI_MCRX_OFFSET,
+ mcrx);
+ if (result < 0)
+ goto fail_read;
+ }
+
+ result = pci_write_config_dword(mbi_pdev, MBI_MCR_OFFSET, mcr);
+ if (result < 0)
+ goto fail_read;
+
+ result = pci_read_config_dword(mbi_pdev, MBI_MDR_OFFSET, mdr);
+ if (result < 0)
+ goto fail_read;
+
+ return 0;
+
+fail_read:
+ dev_err(&mbi_pdev->dev, "PCI config access failed with %d\n", result);
+ return result;
+}
+
+static int iosf_mbi_pci_write_mdr(u32 mcrx, u32 mcr, u32 mdr)
+{
+ int result;
+
+ if (!mbi_pdev)
+ return -ENODEV;
+
+ result = pci_write_config_dword(mbi_pdev, MBI_MDR_OFFSET, mdr);
+ if (result < 0)
+ goto fail_write;
+
+ if (mcrx) {
+ result = pci_write_config_dword(mbi_pdev, MBI_MCRX_OFFSET,
+ mcrx);
+ if (result < 0)
+ goto fail_write;
+ }
+
+ result = pci_write_config_dword(mbi_pdev, MBI_MCR_OFFSET, mcr);
+ if (result < 0)
+ goto fail_write;
+
+ return 0;
+
+fail_write:
+ dev_err(&mbi_pdev->dev, "PCI config access failed with %d\n", result);
+ return result;
+}
+
+int iosf_mbi_read(u8 port, u8 opcode, u32 offset, u32 *mdr)
+{
+ u32 mcr, mcrx;
+ unsigned long flags;
+ int ret;
+
+ /* Access to the GFX unit is handled by GPU code */
+ if (port == BT_MBI_UNIT_GFX) {
+ WARN_ON(1);
+ return -EPERM;
+ }
+
+ mcr = iosf_mbi_form_mcr(opcode, port, offset & MBI_MASK_LO);
+ mcrx = offset & MBI_MASK_HI;
+
+ spin_lock_irqsave(&iosf_mbi_lock, flags);
+ ret = iosf_mbi_pci_read_mdr(mcrx, mcr, mdr);
+ spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+
+ return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_read);
+
+int iosf_mbi_write(u8 port, u8 opcode, u32 offset, u32 mdr)
+{
+ u32 mcr, mcrx;
+ unsigned long flags;
+ int ret;
+
+ /* Access to the GFX unit is handled by GPU code */
+ if (port == BT_MBI_UNIT_GFX) {
+ WARN_ON(1);
+ return -EPERM;
+ }
+
+ mcr = iosf_mbi_form_mcr(opcode, port, offset & MBI_MASK_LO);
+ mcrx = offset & MBI_MASK_HI;
+
+ spin_lock_irqsave(&iosf_mbi_lock, flags);
+ ret = iosf_mbi_pci_write_mdr(mcrx, mcr, mdr);
+ spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+
+ return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_write);
+
+int iosf_mbi_modify(u8 port, u8 opcode, u32 offset, u32 mdr, u32 mask)
+{
+ u32 mcr, mcrx;
+ u32 value;
+ unsigned long flags;
+ int ret;
+
+ /* Access to the GFX unit is handled by GPU code */
+ if (port == BT_MBI_UNIT_GFX) {
+ WARN_ON(1);
+ return -EPERM;
+ }
+
+ mcr = iosf_mbi_form_mcr(opcode, port, offset & MBI_MASK_LO);
+ mcrx = offset & MBI_MASK_HI;
+
+ spin_lock_irqsave(&iosf_mbi_lock, flags);
+
+ /* Read current mdr value */
+ ret = iosf_mbi_pci_read_mdr(mcrx, mcr & MBI_RD_MASK, &value);
+ if (ret < 0) {
+ spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+ return ret;
+ }
+
+ /* Apply mask */
+ value &= ~mask;
+ mdr &= mask;
+ value |= mdr;
+
+ /* Write back */
+ ret = iosf_mbi_pci_write_mdr(mcrx, mcr | MBI_WR_MASK, value);
+
+ spin_unlock_irqrestore(&iosf_mbi_lock, flags);
+
+ return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_modify);
+
+bool iosf_mbi_available(void)
+{
+ /* Mbi isn't hot-pluggable. No remove routine is provided */
+ return mbi_pdev;
+}
+EXPORT_SYMBOL(iosf_mbi_available);
+
+void iosf_mbi_punit_acquire(void)
+{
+ mutex_lock(&iosf_mbi_punit_mutex);
+}
+EXPORT_SYMBOL(iosf_mbi_punit_acquire);
+
+void iosf_mbi_punit_release(void)
+{
+ mutex_unlock(&iosf_mbi_punit_mutex);
+}
+EXPORT_SYMBOL(iosf_mbi_punit_release);
+
+int iosf_mbi_register_pmic_bus_access_notifier(struct notifier_block *nb)
+{
+ int ret;
+
+ /* Wait for the bus to go inactive before registering */
+ mutex_lock(&iosf_mbi_punit_mutex);
+ ret = blocking_notifier_chain_register(
+ &iosf_mbi_pmic_bus_access_notifier, nb);
+ mutex_unlock(&iosf_mbi_punit_mutex);
+
+ return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_register_pmic_bus_access_notifier);
+
+int iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(
+ struct notifier_block *nb)
+{
+ iosf_mbi_assert_punit_acquired();
+
+ return blocking_notifier_chain_unregister(
+ &iosf_mbi_pmic_bus_access_notifier, nb);
+}
+EXPORT_SYMBOL(iosf_mbi_unregister_pmic_bus_access_notifier_unlocked);
+
+int iosf_mbi_unregister_pmic_bus_access_notifier(struct notifier_block *nb)
+{
+ int ret;
+
+ /* Wait for the bus to go inactive before unregistering */
+ mutex_lock(&iosf_mbi_punit_mutex);
+ ret = iosf_mbi_unregister_pmic_bus_access_notifier_unlocked(nb);
+ mutex_unlock(&iosf_mbi_punit_mutex);
+
+ return ret;
+}
+EXPORT_SYMBOL(iosf_mbi_unregister_pmic_bus_access_notifier);
+
+int iosf_mbi_call_pmic_bus_access_notifier_chain(unsigned long val, void *v)
+{
+ return blocking_notifier_call_chain(
+ &iosf_mbi_pmic_bus_access_notifier, val, v);
+}
+EXPORT_SYMBOL(iosf_mbi_call_pmic_bus_access_notifier_chain);
+
+void iosf_mbi_assert_punit_acquired(void)
+{
+ WARN_ON(!mutex_is_locked(&iosf_mbi_punit_mutex));
+}
+EXPORT_SYMBOL(iosf_mbi_assert_punit_acquired);
+
+#ifdef CONFIG_IOSF_MBI_DEBUG
+static u32 dbg_mdr;
+static u32 dbg_mcr;
+static u32 dbg_mcrx;
+
+static int mcr_get(void *data, u64 *val)
+{
+ *val = *(u32 *)data;
+ return 0;
+}
+
+static int mcr_set(void *data, u64 val)
+{
+ u8 command = ((u32)val & 0xFF000000) >> 24,
+ port = ((u32)val & 0x00FF0000) >> 16,
+ offset = ((u32)val & 0x0000FF00) >> 8;
+ int err;
+
+ *(u32 *)data = val;
+
+ if (!capable(CAP_SYS_RAWIO))
+ return -EACCES;
+
+ if (command & 1u)
+ err = iosf_mbi_write(port,
+ command,
+ dbg_mcrx | offset,
+ dbg_mdr);
+ else
+ err = iosf_mbi_read(port,
+ command,
+ dbg_mcrx | offset,
+ &dbg_mdr);
+
+ return err;
+}
+DEFINE_SIMPLE_ATTRIBUTE(iosf_mcr_fops, mcr_get, mcr_set , "%llx\n");
+
+static struct dentry *iosf_dbg;
+
+static void iosf_sideband_debug_init(void)
+{
+ struct dentry *d;
+
+ iosf_dbg = debugfs_create_dir("iosf_sb", NULL);
+ if (IS_ERR_OR_NULL(iosf_dbg))
+ return;
+
+ /* mdr */
+ d = debugfs_create_x32("mdr", 0660, iosf_dbg, &dbg_mdr);
+ if (!d)
+ goto cleanup;
+
+ /* mcrx */
+ d = debugfs_create_x32("mcrx", 0660, iosf_dbg, &dbg_mcrx);
+ if (!d)
+ goto cleanup;
+
+ /* mcr - initiates mailbox tranaction */
+ d = debugfs_create_file("mcr", 0660, iosf_dbg, &dbg_mcr, &iosf_mcr_fops);
+ if (!d)
+ goto cleanup;
+
+ return;
+
+cleanup:
+ debugfs_remove_recursive(d);
+}
+
+static void iosf_debugfs_init(void)
+{
+ iosf_sideband_debug_init();
+}
+
+static void iosf_debugfs_remove(void)
+{
+ debugfs_remove_recursive(iosf_dbg);
+}
+#else
+static inline void iosf_debugfs_init(void) { }
+static inline void iosf_debugfs_remove(void) { }
+#endif /* CONFIG_IOSF_MBI_DEBUG */
+
+static int iosf_mbi_probe(struct pci_dev *pdev,
+ const struct pci_device_id *unused)
+{
+ int ret;
+
+ ret = pci_enable_device(pdev);
+ if (ret < 0) {
+ dev_err(&pdev->dev, "error: could not enable device\n");
+ return ret;
+ }
+
+ mbi_pdev = pci_dev_get(pdev);
+ return 0;
+}
+
+static const struct pci_device_id iosf_mbi_pci_ids[] = {
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_BAYTRAIL) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_BRASWELL) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_QUARK_X1000) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_TANGIER) },
+ { 0, },
+};
+MODULE_DEVICE_TABLE(pci, iosf_mbi_pci_ids);
+
+static struct pci_driver iosf_mbi_pci_driver = {
+ .name = "iosf_mbi_pci",
+ .probe = iosf_mbi_probe,
+ .id_table = iosf_mbi_pci_ids,
+};
+
+static int __init iosf_mbi_init(void)
+{
+ iosf_debugfs_init();
+
+ return pci_register_driver(&iosf_mbi_pci_driver);
+}
+
+static void __exit iosf_mbi_exit(void)
+{
+ iosf_debugfs_remove();
+
+ pci_unregister_driver(&iosf_mbi_pci_driver);
+ pci_dev_put(mbi_pdev);
+ mbi_pdev = NULL;
+}
+
+module_init(iosf_mbi_init);
+module_exit(iosf_mbi_exit);
+
+MODULE_AUTHOR("David E. Box <david.e.box@linux.intel.com>");
+MODULE_DESCRIPTION("IOSF Mailbox Interface accessor");
+MODULE_LICENSE("GPL v2");
diff --git a/arch/x86/platform/iris/Makefile b/arch/x86/platform/iris/Makefile
new file mode 100644
index 000000000..db921983a
--- /dev/null
+++ b/arch/x86/platform/iris/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_X86_32_IRIS) += iris.o
diff --git a/arch/x86/platform/iris/iris.c b/arch/x86/platform/iris/iris.c
new file mode 100644
index 000000000..735ba21ef
--- /dev/null
+++ b/arch/x86/platform/iris/iris.c
@@ -0,0 +1,136 @@
+/*
+ * Eurobraille/Iris power off support.
+ *
+ * Eurobraille's Iris machine is a PC with no APM or ACPI support.
+ * It is shutdown by a special I/O sequence which this module provides.
+ *
+ * Copyright (C) Shérab <Sebastien.Hinderer@ens-lyon.org>
+ *
+ * This program is free software ; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation ; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY ; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with the program ; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/moduleparam.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/delay.h>
+#include <linux/pm.h>
+#include <asm/io.h>
+
+#define IRIS_GIO_BASE 0x340
+#define IRIS_GIO_INPUT IRIS_GIO_BASE
+#define IRIS_GIO_OUTPUT (IRIS_GIO_BASE + 1)
+#define IRIS_GIO_PULSE 0x80 /* First byte to send */
+#define IRIS_GIO_REST 0x00 /* Second byte to send */
+#define IRIS_GIO_NODEV 0xff /* Likely not an Iris */
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Sébastien Hinderer <Sebastien.Hinderer@ens-lyon.org>");
+MODULE_DESCRIPTION("A power_off handler for Iris devices from EuroBraille");
+MODULE_SUPPORTED_DEVICE("Eurobraille/Iris");
+
+static bool force;
+
+module_param(force, bool, 0);
+MODULE_PARM_DESC(force, "Set to one to force poweroff handler installation.");
+
+static void (*old_pm_power_off)(void);
+
+static void iris_power_off(void)
+{
+ outb(IRIS_GIO_PULSE, IRIS_GIO_OUTPUT);
+ msleep(850);
+ outb(IRIS_GIO_REST, IRIS_GIO_OUTPUT);
+}
+
+/*
+ * Before installing the power_off handler, try to make sure the OS is
+ * running on an Iris. Since Iris does not support DMI, this is done
+ * by reading its input port and seeing whether the read value is
+ * meaningful.
+ */
+static int iris_probe(struct platform_device *pdev)
+{
+ unsigned char status = inb(IRIS_GIO_INPUT);
+ if (status == IRIS_GIO_NODEV) {
+ printk(KERN_ERR "This machine does not seem to be an Iris. "
+ "Power off handler not installed.\n");
+ return -ENODEV;
+ }
+ old_pm_power_off = pm_power_off;
+ pm_power_off = &iris_power_off;
+ printk(KERN_INFO "Iris power_off handler installed.\n");
+ return 0;
+}
+
+static int iris_remove(struct platform_device *pdev)
+{
+ pm_power_off = old_pm_power_off;
+ printk(KERN_INFO "Iris power_off handler uninstalled.\n");
+ return 0;
+}
+
+static struct platform_driver iris_driver = {
+ .driver = {
+ .name = "iris",
+ },
+ .probe = iris_probe,
+ .remove = iris_remove,
+};
+
+static struct resource iris_resources[] = {
+ {
+ .start = IRIS_GIO_BASE,
+ .end = IRIS_GIO_OUTPUT,
+ .flags = IORESOURCE_IO,
+ .name = "address"
+ }
+};
+
+static struct platform_device *iris_device;
+
+static int iris_init(void)
+{
+ int ret;
+ if (force != 1) {
+ printk(KERN_ERR "The force parameter has not been set to 1."
+ " The Iris poweroff handler will not be installed.\n");
+ return -ENODEV;
+ }
+ ret = platform_driver_register(&iris_driver);
+ if (ret < 0) {
+ printk(KERN_ERR "Failed to register iris platform driver: %d\n",
+ ret);
+ return ret;
+ }
+ iris_device = platform_device_register_simple("iris", (-1),
+ iris_resources, ARRAY_SIZE(iris_resources));
+ if (IS_ERR(iris_device)) {
+ printk(KERN_ERR "Failed to register iris platform device\n");
+ platform_driver_unregister(&iris_driver);
+ return PTR_ERR(iris_device);
+ }
+ return 0;
+}
+
+static void iris_exit(void)
+{
+ platform_device_unregister(iris_device);
+ platform_driver_unregister(&iris_driver);
+}
+
+module_init(iris_init);
+module_exit(iris_exit);
diff --git a/arch/x86/platform/olpc/Makefile b/arch/x86/platform/olpc/Makefile
new file mode 100644
index 000000000..049f92a93
--- /dev/null
+++ b/arch/x86/platform/olpc/Makefile
@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0
+obj-$(CONFIG_OLPC) += olpc.o olpc_ofw.o olpc_dt.o
+obj-$(CONFIG_OLPC_XO1_PM) += olpc-xo1-pm.o xo1-wakeup.o
+obj-$(CONFIG_OLPC_XO1_RTC) += olpc-xo1-rtc.o
+obj-$(CONFIG_OLPC_XO1_SCI) += olpc-xo1-sci.o
+obj-$(CONFIG_OLPC_XO15_SCI) += olpc-xo15-sci.o
diff --git a/arch/x86/platform/olpc/olpc-xo1-pm.c b/arch/x86/platform/olpc/olpc-xo1-pm.c
new file mode 100644
index 000000000..0668aaff8
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc-xo1-pm.c
@@ -0,0 +1,200 @@
+/*
+ * Support for power management features of the OLPC XO-1 laptop
+ *
+ * Copyright (C) 2010 Andres Salomon <dilinger@queued.net>
+ * Copyright (C) 2010 One Laptop per Child
+ * Copyright (C) 2006 Red Hat, Inc.
+ * Copyright (C) 2006 Advanced Micro Devices, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/cs5535.h>
+#include <linux/platform_device.h>
+#include <linux/export.h>
+#include <linux/pm.h>
+#include <linux/mfd/core.h>
+#include <linux/suspend.h>
+#include <linux/olpc-ec.h>
+
+#include <asm/io.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME "olpc-xo1-pm"
+
+static unsigned long acpi_base;
+static unsigned long pms_base;
+
+static u16 wakeup_mask = CS5536_PM_PWRBTN;
+
+static struct {
+ unsigned long address;
+ unsigned short segment;
+} ofw_bios_entry = { 0xF0000 + PAGE_OFFSET, __KERNEL_CS };
+
+/* Set bits in the wakeup mask */
+void olpc_xo1_pm_wakeup_set(u16 value)
+{
+ wakeup_mask |= value;
+}
+EXPORT_SYMBOL_GPL(olpc_xo1_pm_wakeup_set);
+
+/* Clear bits in the wakeup mask */
+void olpc_xo1_pm_wakeup_clear(u16 value)
+{
+ wakeup_mask &= ~value;
+}
+EXPORT_SYMBOL_GPL(olpc_xo1_pm_wakeup_clear);
+
+static int xo1_power_state_enter(suspend_state_t pm_state)
+{
+ unsigned long saved_sci_mask;
+
+ /* Only STR is supported */
+ if (pm_state != PM_SUSPEND_MEM)
+ return -EINVAL;
+
+ /*
+ * Save SCI mask (this gets lost since PM1_EN is used as a mask for
+ * wakeup events, which is not necessarily the same event set)
+ */
+ saved_sci_mask = inl(acpi_base + CS5536_PM1_STS);
+ saved_sci_mask &= 0xffff0000;
+
+ /* Save CPU state */
+ do_olpc_suspend_lowlevel();
+
+ /* Resume path starts here */
+
+ /* Restore SCI mask (using dword access to CS5536_PM1_EN) */
+ outl(saved_sci_mask, acpi_base + CS5536_PM1_STS);
+
+ return 0;
+}
+
+asmlinkage __visible int xo1_do_sleep(u8 sleep_state)
+{
+ void *pgd_addr = __va(read_cr3_pa());
+
+ /* Program wakeup mask (using dword access to CS5536_PM1_EN) */
+ outl(wakeup_mask << 16, acpi_base + CS5536_PM1_STS);
+
+ __asm__("movl %0,%%eax" : : "r" (pgd_addr));
+ __asm__("call *(%%edi); cld"
+ : : "D" (&ofw_bios_entry));
+ __asm__("movb $0x34, %al\n\t"
+ "outb %al, $0x70\n\t"
+ "movb $0x30, %al\n\t"
+ "outb %al, $0x71\n\t");
+ return 0;
+}
+
+static void xo1_power_off(void)
+{
+ printk(KERN_INFO "OLPC XO-1 power off sequence...\n");
+
+ /* Enable all of these controls with 0 delay */
+ outl(0x40000000, pms_base + CS5536_PM_SCLK);
+ outl(0x40000000, pms_base + CS5536_PM_IN_SLPCTL);
+ outl(0x40000000, pms_base + CS5536_PM_WKXD);
+ outl(0x40000000, pms_base + CS5536_PM_WKD);
+
+ /* Clear status bits (possibly unnecessary) */
+ outl(0x0002ffff, pms_base + CS5536_PM_SSC);
+ outl(0xffffffff, acpi_base + CS5536_PM_GPE0_STS);
+
+ /* Write SLP_EN bit to start the machinery */
+ outl(0x00002000, acpi_base + CS5536_PM1_CNT);
+}
+
+static int xo1_power_state_valid(suspend_state_t pm_state)
+{
+ /* suspend-to-RAM only */
+ return pm_state == PM_SUSPEND_MEM;
+}
+
+static const struct platform_suspend_ops xo1_suspend_ops = {
+ .valid = xo1_power_state_valid,
+ .enter = xo1_power_state_enter,
+};
+
+static int xo1_pm_probe(struct platform_device *pdev)
+{
+ struct resource *res;
+ int err;
+
+ /* don't run on non-XOs */
+ if (!machine_is_olpc())
+ return -ENODEV;
+
+ err = mfd_cell_enable(pdev);
+ if (err)
+ return err;
+
+ res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "can't fetch device resource info\n");
+ return -EIO;
+ }
+ if (strcmp(pdev->name, "cs5535-pms") == 0)
+ pms_base = res->start;
+ else if (strcmp(pdev->name, "olpc-xo1-pm-acpi") == 0)
+ acpi_base = res->start;
+
+ /* If we have both addresses, we can override the poweroff hook */
+ if (pms_base && acpi_base) {
+ suspend_set_ops(&xo1_suspend_ops);
+ pm_power_off = xo1_power_off;
+ printk(KERN_INFO "OLPC XO-1 support registered\n");
+ }
+
+ return 0;
+}
+
+static int xo1_pm_remove(struct platform_device *pdev)
+{
+ mfd_cell_disable(pdev);
+
+ if (strcmp(pdev->name, "cs5535-pms") == 0)
+ pms_base = 0;
+ else if (strcmp(pdev->name, "olpc-xo1-pm-acpi") == 0)
+ acpi_base = 0;
+
+ pm_power_off = NULL;
+ return 0;
+}
+
+static struct platform_driver cs5535_pms_driver = {
+ .driver = {
+ .name = "cs5535-pms",
+ },
+ .probe = xo1_pm_probe,
+ .remove = xo1_pm_remove,
+};
+
+static struct platform_driver cs5535_acpi_driver = {
+ .driver = {
+ .name = "olpc-xo1-pm-acpi",
+ },
+ .probe = xo1_pm_probe,
+ .remove = xo1_pm_remove,
+};
+
+static int __init xo1_pm_init(void)
+{
+ int r;
+
+ r = platform_driver_register(&cs5535_pms_driver);
+ if (r)
+ return r;
+
+ r = platform_driver_register(&cs5535_acpi_driver);
+ if (r)
+ platform_driver_unregister(&cs5535_pms_driver);
+
+ return r;
+}
+arch_initcall(xo1_pm_init);
diff --git a/arch/x86/platform/olpc/olpc-xo1-rtc.c b/arch/x86/platform/olpc/olpc-xo1-rtc.c
new file mode 100644
index 000000000..8e7ddd7e3
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc-xo1-rtc.c
@@ -0,0 +1,84 @@
+/*
+ * Support for OLPC XO-1 Real Time Clock (RTC)
+ *
+ * Copyright (C) 2011 One Laptop per Child
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/mc146818rtc.h>
+#include <linux/platform_device.h>
+#include <linux/rtc.h>
+#include <linux/of.h>
+
+#include <asm/msr.h>
+#include <asm/olpc.h>
+#include <asm/x86_init.h>
+
+static void rtc_wake_on(struct device *dev)
+{
+ olpc_xo1_pm_wakeup_set(CS5536_PM_RTC);
+}
+
+static void rtc_wake_off(struct device *dev)
+{
+ olpc_xo1_pm_wakeup_clear(CS5536_PM_RTC);
+}
+
+static struct resource rtc_platform_resource[] = {
+ [0] = {
+ .start = RTC_PORT(0),
+ .end = RTC_PORT(1),
+ .flags = IORESOURCE_IO,
+ },
+ [1] = {
+ .start = RTC_IRQ,
+ .end = RTC_IRQ,
+ .flags = IORESOURCE_IRQ,
+ }
+};
+
+static struct cmos_rtc_board_info rtc_info = {
+ .rtc_day_alarm = 0,
+ .rtc_mon_alarm = 0,
+ .rtc_century = 0,
+ .wake_on = rtc_wake_on,
+ .wake_off = rtc_wake_off,
+};
+
+static struct platform_device xo1_rtc_device = {
+ .name = "rtc_cmos",
+ .id = -1,
+ .num_resources = ARRAY_SIZE(rtc_platform_resource),
+ .dev.platform_data = &rtc_info,
+ .resource = rtc_platform_resource,
+};
+
+static int __init xo1_rtc_init(void)
+{
+ int r;
+ struct device_node *node;
+
+ node = of_find_compatible_node(NULL, NULL, "olpc,xo1-rtc");
+ if (!node)
+ return 0;
+ of_node_put(node);
+
+ pr_info("olpc-xo1-rtc: Initializing OLPC XO-1 RTC\n");
+ rdmsrl(MSR_RTC_DOMA_OFFSET, rtc_info.rtc_day_alarm);
+ rdmsrl(MSR_RTC_MONA_OFFSET, rtc_info.rtc_mon_alarm);
+ rdmsrl(MSR_RTC_CEN_OFFSET, rtc_info.rtc_century);
+
+ r = platform_device_register(&xo1_rtc_device);
+ if (r)
+ return r;
+
+ x86_platform.legacy.rtc = 0;
+
+ device_init_wakeup(&xo1_rtc_device.dev, 1);
+ return 0;
+}
+arch_initcall(xo1_rtc_init);
diff --git a/arch/x86/platform/olpc/olpc-xo1-sci.c b/arch/x86/platform/olpc/olpc-xo1-sci.c
new file mode 100644
index 000000000..7fa8b3b53
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc-xo1-sci.c
@@ -0,0 +1,642 @@
+/*
+ * Support for OLPC XO-1 System Control Interrupts (SCI)
+ *
+ * Copyright (C) 2010 One Laptop per Child
+ * Copyright (C) 2006 Red Hat, Inc.
+ * Copyright (C) 2006 Advanced Micro Devices, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/cs5535.h>
+#include <linux/device.h>
+#include <linux/gpio.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+#include <linux/pm_wakeup.h>
+#include <linux/mfd/core.h>
+#include <linux/power_supply.h>
+#include <linux/suspend.h>
+#include <linux/workqueue.h>
+#include <linux/olpc-ec.h>
+
+#include <asm/io.h>
+#include <asm/msr.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME "olpc-xo1-sci"
+#define PFX DRV_NAME ": "
+
+static unsigned long acpi_base;
+static struct input_dev *power_button_idev;
+static struct input_dev *ebook_switch_idev;
+static struct input_dev *lid_switch_idev;
+
+static int sci_irq;
+
+static bool lid_open;
+static bool lid_inverted;
+static int lid_wake_mode;
+
+enum lid_wake_modes {
+ LID_WAKE_ALWAYS,
+ LID_WAKE_OPEN,
+ LID_WAKE_CLOSE,
+};
+
+static const char * const lid_wake_mode_names[] = {
+ [LID_WAKE_ALWAYS] = "always",
+ [LID_WAKE_OPEN] = "open",
+ [LID_WAKE_CLOSE] = "close",
+};
+
+static void battery_status_changed(void)
+{
+ struct power_supply *psy = power_supply_get_by_name("olpc-battery");
+
+ if (psy) {
+ power_supply_changed(psy);
+ power_supply_put(psy);
+ }
+}
+
+static void ac_status_changed(void)
+{
+ struct power_supply *psy = power_supply_get_by_name("olpc-ac");
+
+ if (psy) {
+ power_supply_changed(psy);
+ power_supply_put(psy);
+ }
+}
+
+/* Report current ebook switch state through input layer */
+static void send_ebook_state(void)
+{
+ unsigned char state;
+
+ if (olpc_ec_cmd(EC_READ_EB_MODE, NULL, 0, &state, 1)) {
+ pr_err(PFX "failed to get ebook state\n");
+ return;
+ }
+
+ if (!!test_bit(SW_TABLET_MODE, ebook_switch_idev->sw) == state)
+ return; /* Nothing new to report. */
+
+ input_report_switch(ebook_switch_idev, SW_TABLET_MODE, state);
+ input_sync(ebook_switch_idev);
+ pm_wakeup_event(&ebook_switch_idev->dev, 0);
+}
+
+static void flip_lid_inverter(void)
+{
+ /* gpio is high; invert so we'll get l->h event interrupt */
+ if (lid_inverted)
+ cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_INPUT_INVERT);
+ else
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_INPUT_INVERT);
+ lid_inverted = !lid_inverted;
+}
+
+static void detect_lid_state(void)
+{
+ /*
+ * the edge detector hookup on the gpio inputs on the geode is
+ * odd, to say the least. See http://dev.laptop.org/ticket/5703
+ * for details, but in a nutshell: we don't use the edge
+ * detectors. instead, we make use of an anomoly: with the both
+ * edge detectors turned off, we still get an edge event on a
+ * positive edge transition. to take advantage of this, we use the
+ * front-end inverter to ensure that that's the edge we're always
+ * going to see next.
+ */
+
+ int state;
+
+ state = cs5535_gpio_isset(OLPC_GPIO_LID, GPIO_READ_BACK);
+ lid_open = !state ^ !lid_inverted; /* x ^^ y */
+ if (!state)
+ return;
+
+ flip_lid_inverter();
+}
+
+/* Report current lid switch state through input layer */
+static void send_lid_state(void)
+{
+ if (!!test_bit(SW_LID, lid_switch_idev->sw) == !lid_open)
+ return; /* Nothing new to report. */
+
+ input_report_switch(lid_switch_idev, SW_LID, !lid_open);
+ input_sync(lid_switch_idev);
+ pm_wakeup_event(&lid_switch_idev->dev, 0);
+}
+
+static ssize_t lid_wake_mode_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ const char *mode = lid_wake_mode_names[lid_wake_mode];
+ return sprintf(buf, "%s\n", mode);
+}
+static ssize_t lid_wake_mode_set(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ int i;
+ for (i = 0; i < ARRAY_SIZE(lid_wake_mode_names); i++) {
+ const char *mode = lid_wake_mode_names[i];
+ if (strlen(mode) != count || strncasecmp(mode, buf, count))
+ continue;
+
+ lid_wake_mode = i;
+ return count;
+ }
+ return -EINVAL;
+}
+static DEVICE_ATTR(lid_wake_mode, S_IWUSR | S_IRUGO, lid_wake_mode_show,
+ lid_wake_mode_set);
+
+/*
+ * Process all items in the EC's SCI queue.
+ *
+ * This is handled in a workqueue because olpc_ec_cmd can be slow (and
+ * can even timeout).
+ *
+ * If propagate_events is false, the queue is drained without events being
+ * generated for the interrupts.
+ */
+static void process_sci_queue(bool propagate_events)
+{
+ int r;
+ u16 data;
+
+ do {
+ r = olpc_ec_sci_query(&data);
+ if (r || !data)
+ break;
+
+ pr_debug(PFX "SCI 0x%x received\n", data);
+
+ switch (data) {
+ case EC_SCI_SRC_BATERR:
+ case EC_SCI_SRC_BATSOC:
+ case EC_SCI_SRC_BATTERY:
+ case EC_SCI_SRC_BATCRIT:
+ battery_status_changed();
+ break;
+ case EC_SCI_SRC_ACPWR:
+ ac_status_changed();
+ break;
+ }
+
+ if (data == EC_SCI_SRC_EBOOK && propagate_events)
+ send_ebook_state();
+ } while (data);
+
+ if (r)
+ pr_err(PFX "Failed to clear SCI queue");
+}
+
+static void process_sci_queue_work(struct work_struct *work)
+{
+ process_sci_queue(true);
+}
+
+static DECLARE_WORK(sci_work, process_sci_queue_work);
+
+static irqreturn_t xo1_sci_intr(int irq, void *dev_id)
+{
+ struct platform_device *pdev = dev_id;
+ u32 sts;
+ u32 gpe;
+
+ sts = inl(acpi_base + CS5536_PM1_STS);
+ outl(sts | 0xffff, acpi_base + CS5536_PM1_STS);
+
+ gpe = inl(acpi_base + CS5536_PM_GPE0_STS);
+ outl(0xffffffff, acpi_base + CS5536_PM_GPE0_STS);
+
+ dev_dbg(&pdev->dev, "sts %x gpe %x\n", sts, gpe);
+
+ if (sts & CS5536_PWRBTN_FLAG) {
+ if (!(sts & CS5536_WAK_FLAG)) {
+ /* Only report power button input when it was pressed
+ * during regular operation (as opposed to when it
+ * was used to wake the system). */
+ input_report_key(power_button_idev, KEY_POWER, 1);
+ input_sync(power_button_idev);
+ input_report_key(power_button_idev, KEY_POWER, 0);
+ input_sync(power_button_idev);
+ }
+ /* Report the wakeup event in all cases. */
+ pm_wakeup_event(&power_button_idev->dev, 0);
+ }
+
+ if ((sts & (CS5536_RTC_FLAG | CS5536_WAK_FLAG)) ==
+ (CS5536_RTC_FLAG | CS5536_WAK_FLAG)) {
+ /* When the system is woken by the RTC alarm, report the
+ * event on the rtc device. */
+ struct device *rtc = bus_find_device_by_name(
+ &platform_bus_type, NULL, "rtc_cmos");
+ if (rtc) {
+ pm_wakeup_event(rtc, 0);
+ put_device(rtc);
+ }
+ }
+
+ if (gpe & CS5536_GPIOM7_PME_FLAG) { /* EC GPIO */
+ cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_NEGATIVE_EDGE_STS);
+ schedule_work(&sci_work);
+ }
+
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_STS);
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_STS);
+ detect_lid_state();
+ send_lid_state();
+
+ return IRQ_HANDLED;
+}
+
+static int xo1_sci_suspend(struct platform_device *pdev, pm_message_t state)
+{
+ if (device_may_wakeup(&power_button_idev->dev))
+ olpc_xo1_pm_wakeup_set(CS5536_PM_PWRBTN);
+ else
+ olpc_xo1_pm_wakeup_clear(CS5536_PM_PWRBTN);
+
+ if (device_may_wakeup(&ebook_switch_idev->dev))
+ olpc_ec_wakeup_set(EC_SCI_SRC_EBOOK);
+ else
+ olpc_ec_wakeup_clear(EC_SCI_SRC_EBOOK);
+
+ if (!device_may_wakeup(&lid_switch_idev->dev)) {
+ cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+ } else if ((lid_open && lid_wake_mode == LID_WAKE_OPEN) ||
+ (!lid_open && lid_wake_mode == LID_WAKE_CLOSE)) {
+ flip_lid_inverter();
+
+ /* we may have just caused an event */
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_STS);
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_STS);
+
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+ }
+
+ return 0;
+}
+
+static int xo1_sci_resume(struct platform_device *pdev)
+{
+ /*
+ * We don't know what may have happened while we were asleep.
+ * Reestablish our lid setup so we're sure to catch all transitions.
+ */
+ detect_lid_state();
+ send_lid_state();
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+
+ /* Enable all EC events */
+ olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+ /* Power/battery status might have changed too */
+ battery_status_changed();
+ ac_status_changed();
+ return 0;
+}
+
+static int setup_sci_interrupt(struct platform_device *pdev)
+{
+ u32 lo, hi;
+ u32 sts;
+ int r;
+
+ rdmsr(0x51400020, lo, hi);
+ sci_irq = (lo >> 20) & 15;
+
+ if (sci_irq) {
+ dev_info(&pdev->dev, "SCI is mapped to IRQ %d\n", sci_irq);
+ } else {
+ /* Zero means masked */
+ dev_info(&pdev->dev, "SCI unmapped. Mapping to IRQ 3\n");
+ sci_irq = 3;
+ lo |= 0x00300000;
+ wrmsrl(0x51400020, lo);
+ }
+
+ /* Select level triggered in PIC */
+ if (sci_irq < 8) {
+ lo = inb(CS5536_PIC_INT_SEL1);
+ lo |= 1 << sci_irq;
+ outb(lo, CS5536_PIC_INT_SEL1);
+ } else {
+ lo = inb(CS5536_PIC_INT_SEL2);
+ lo |= 1 << (sci_irq - 8);
+ outb(lo, CS5536_PIC_INT_SEL2);
+ }
+
+ /* Enable interesting SCI events, and clear pending interrupts */
+ sts = inl(acpi_base + CS5536_PM1_STS);
+ outl(((CS5536_PM_PWRBTN | CS5536_PM_RTC) << 16) | 0xffff,
+ acpi_base + CS5536_PM1_STS);
+
+ r = request_irq(sci_irq, xo1_sci_intr, 0, DRV_NAME, pdev);
+ if (r)
+ dev_err(&pdev->dev, "can't request interrupt\n");
+
+ return r;
+}
+
+static int setup_ec_sci(void)
+{
+ int r;
+
+ r = gpio_request(OLPC_GPIO_ECSCI, "OLPC-ECSCI");
+ if (r)
+ return r;
+
+ gpio_direction_input(OLPC_GPIO_ECSCI);
+
+ /* Clear pending EC SCI events */
+ cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_NEGATIVE_EDGE_STS);
+ cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_POSITIVE_EDGE_STS);
+
+ /*
+ * Enable EC SCI events, and map them to both a PME and the SCI
+ * interrupt.
+ *
+ * Ordinarily, in addition to functioning as GPIOs, Geode GPIOs can
+ * be mapped to regular interrupts *or* Geode-specific Power
+ * Management Events (PMEs) - events that bring the system out of
+ * suspend. In this case, we want both of those things - the system
+ * wakeup, *and* the ability to get an interrupt when an event occurs.
+ *
+ * To achieve this, we map the GPIO to a PME, and then we use one
+ * of the many generic knobs on the CS5535 PIC to additionally map the
+ * PME to the regular SCI interrupt line.
+ */
+ cs5535_gpio_set(OLPC_GPIO_ECSCI, GPIO_EVENTS_ENABLE);
+
+ /* Set the SCI to cause a PME event on group 7 */
+ cs5535_gpio_setup_event(OLPC_GPIO_ECSCI, 7, 1);
+
+ /* And have group 7 also fire the SCI interrupt */
+ cs5535_pic_unreqz_select_high(7, sci_irq);
+
+ return 0;
+}
+
+static void free_ec_sci(void)
+{
+ gpio_free(OLPC_GPIO_ECSCI);
+}
+
+static int setup_lid_events(void)
+{
+ int r;
+
+ r = gpio_request(OLPC_GPIO_LID, "OLPC-LID");
+ if (r)
+ return r;
+
+ gpio_direction_input(OLPC_GPIO_LID);
+
+ cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_INPUT_INVERT);
+ lid_inverted = 0;
+
+ /* Clear edge detection and event enable for now */
+ cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+ cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_EN);
+ cs5535_gpio_clear(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_EN);
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_NEGATIVE_EDGE_STS);
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_POSITIVE_EDGE_STS);
+
+ /* Set the LID to cause an PME event on group 6 */
+ cs5535_gpio_setup_event(OLPC_GPIO_LID, 6, 1);
+
+ /* Set PME group 6 to fire the SCI interrupt */
+ cs5535_gpio_set_irq(6, sci_irq);
+
+ /* Enable the event */
+ cs5535_gpio_set(OLPC_GPIO_LID, GPIO_EVENTS_ENABLE);
+
+ return 0;
+}
+
+static void free_lid_events(void)
+{
+ gpio_free(OLPC_GPIO_LID);
+}
+
+static int setup_power_button(struct platform_device *pdev)
+{
+ int r;
+
+ power_button_idev = input_allocate_device();
+ if (!power_button_idev)
+ return -ENOMEM;
+
+ power_button_idev->name = "Power Button";
+ power_button_idev->phys = DRV_NAME "/input0";
+ set_bit(EV_KEY, power_button_idev->evbit);
+ set_bit(KEY_POWER, power_button_idev->keybit);
+
+ power_button_idev->dev.parent = &pdev->dev;
+ device_init_wakeup(&power_button_idev->dev, 1);
+
+ r = input_register_device(power_button_idev);
+ if (r) {
+ dev_err(&pdev->dev, "failed to register power button: %d\n", r);
+ input_free_device(power_button_idev);
+ }
+
+ return r;
+}
+
+static void free_power_button(void)
+{
+ input_unregister_device(power_button_idev);
+}
+
+static int setup_ebook_switch(struct platform_device *pdev)
+{
+ int r;
+
+ ebook_switch_idev = input_allocate_device();
+ if (!ebook_switch_idev)
+ return -ENOMEM;
+
+ ebook_switch_idev->name = "EBook Switch";
+ ebook_switch_idev->phys = DRV_NAME "/input1";
+ set_bit(EV_SW, ebook_switch_idev->evbit);
+ set_bit(SW_TABLET_MODE, ebook_switch_idev->swbit);
+
+ ebook_switch_idev->dev.parent = &pdev->dev;
+ device_set_wakeup_capable(&ebook_switch_idev->dev, true);
+
+ r = input_register_device(ebook_switch_idev);
+ if (r) {
+ dev_err(&pdev->dev, "failed to register ebook switch: %d\n", r);
+ input_free_device(ebook_switch_idev);
+ }
+
+ return r;
+}
+
+static void free_ebook_switch(void)
+{
+ input_unregister_device(ebook_switch_idev);
+}
+
+static int setup_lid_switch(struct platform_device *pdev)
+{
+ int r;
+
+ lid_switch_idev = input_allocate_device();
+ if (!lid_switch_idev)
+ return -ENOMEM;
+
+ lid_switch_idev->name = "Lid Switch";
+ lid_switch_idev->phys = DRV_NAME "/input2";
+ set_bit(EV_SW, lid_switch_idev->evbit);
+ set_bit(SW_LID, lid_switch_idev->swbit);
+
+ lid_switch_idev->dev.parent = &pdev->dev;
+ device_set_wakeup_capable(&lid_switch_idev->dev, true);
+
+ r = input_register_device(lid_switch_idev);
+ if (r) {
+ dev_err(&pdev->dev, "failed to register lid switch: %d\n", r);
+ goto err_register;
+ }
+
+ r = device_create_file(&lid_switch_idev->dev, &dev_attr_lid_wake_mode);
+ if (r) {
+ dev_err(&pdev->dev, "failed to create wake mode attr: %d\n", r);
+ goto err_create_attr;
+ }
+
+ return 0;
+
+err_create_attr:
+ input_unregister_device(lid_switch_idev);
+ lid_switch_idev = NULL;
+err_register:
+ input_free_device(lid_switch_idev);
+ return r;
+}
+
+static void free_lid_switch(void)
+{
+ device_remove_file(&lid_switch_idev->dev, &dev_attr_lid_wake_mode);
+ input_unregister_device(lid_switch_idev);
+}
+
+static int xo1_sci_probe(struct platform_device *pdev)
+{
+ struct resource *res;
+ int r;
+
+ /* don't run on non-XOs */
+ if (!machine_is_olpc())
+ return -ENODEV;
+
+ r = mfd_cell_enable(pdev);
+ if (r)
+ return r;
+
+ res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+ if (!res) {
+ dev_err(&pdev->dev, "can't fetch device resource info\n");
+ return -EIO;
+ }
+ acpi_base = res->start;
+
+ r = setup_power_button(pdev);
+ if (r)
+ return r;
+
+ r = setup_ebook_switch(pdev);
+ if (r)
+ goto err_ebook;
+
+ r = setup_lid_switch(pdev);
+ if (r)
+ goto err_lid;
+
+ r = setup_lid_events();
+ if (r)
+ goto err_lidevt;
+
+ r = setup_ec_sci();
+ if (r)
+ goto err_ecsci;
+
+ /* Enable PME generation for EC-generated events */
+ outl(CS5536_GPIOM6_PME_EN | CS5536_GPIOM7_PME_EN,
+ acpi_base + CS5536_PM_GPE0_EN);
+
+ /* Clear pending events */
+ outl(0xffffffff, acpi_base + CS5536_PM_GPE0_STS);
+ process_sci_queue(false);
+
+ /* Initial sync */
+ send_ebook_state();
+ detect_lid_state();
+ send_lid_state();
+
+ r = setup_sci_interrupt(pdev);
+ if (r)
+ goto err_sci;
+
+ /* Enable all EC events */
+ olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+ return r;
+
+err_sci:
+ free_ec_sci();
+err_ecsci:
+ free_lid_events();
+err_lidevt:
+ free_lid_switch();
+err_lid:
+ free_ebook_switch();
+err_ebook:
+ free_power_button();
+ return r;
+}
+
+static int xo1_sci_remove(struct platform_device *pdev)
+{
+ mfd_cell_disable(pdev);
+ free_irq(sci_irq, pdev);
+ cancel_work_sync(&sci_work);
+ free_ec_sci();
+ free_lid_events();
+ free_lid_switch();
+ free_ebook_switch();
+ free_power_button();
+ acpi_base = 0;
+ return 0;
+}
+
+static struct platform_driver xo1_sci_driver = {
+ .driver = {
+ .name = "olpc-xo1-sci-acpi",
+ },
+ .probe = xo1_sci_probe,
+ .remove = xo1_sci_remove,
+ .suspend = xo1_sci_suspend,
+ .resume = xo1_sci_resume,
+};
+
+static int __init xo1_sci_init(void)
+{
+ return platform_driver_register(&xo1_sci_driver);
+}
+arch_initcall(xo1_sci_init);
diff --git a/arch/x86/platform/olpc/olpc-xo15-sci.c b/arch/x86/platform/olpc/olpc-xo15-sci.c
new file mode 100644
index 000000000..c0533fbc3
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc-xo15-sci.c
@@ -0,0 +1,235 @@
+/*
+ * Support for OLPC XO-1.5 System Control Interrupts (SCI)
+ *
+ * Copyright (C) 2009-2010 One Laptop per Child
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/device.h>
+#include <linux/slab.h>
+#include <linux/workqueue.h>
+#include <linux/power_supply.h>
+#include <linux/olpc-ec.h>
+
+#include <linux/acpi.h>
+#include <asm/olpc.h>
+
+#define DRV_NAME "olpc-xo15-sci"
+#define PFX DRV_NAME ": "
+#define XO15_SCI_CLASS DRV_NAME
+#define XO15_SCI_DEVICE_NAME "OLPC XO-1.5 SCI"
+
+static unsigned long xo15_sci_gpe;
+static bool lid_wake_on_close;
+
+/*
+ * The normal ACPI LID wakeup behavior is wake-on-open, but not
+ * wake-on-close. This is implemented as standard by the XO-1.5 DSDT.
+ *
+ * We provide here a sysfs attribute that will additionally enable
+ * wake-on-close behavior. This is useful (e.g.) when we oportunistically
+ * suspend with the display running; if the lid is then closed, we want to
+ * wake up to turn the display off.
+ *
+ * This is controlled through a custom method in the XO-1.5 DSDT.
+ */
+static int set_lid_wake_behavior(bool wake_on_close)
+{
+ acpi_status status;
+
+ status = acpi_execute_simple_method(NULL, "\\_SB.PCI0.LID.LIDW", wake_on_close);
+ if (ACPI_FAILURE(status)) {
+ pr_warning(PFX "failed to set lid behavior\n");
+ return 1;
+ }
+
+ lid_wake_on_close = wake_on_close;
+
+ return 0;
+}
+
+static ssize_t
+lid_wake_on_close_show(struct kobject *s, struct kobj_attribute *attr, char *buf)
+{
+ return sprintf(buf, "%u\n", lid_wake_on_close);
+}
+
+static ssize_t lid_wake_on_close_store(struct kobject *s,
+ struct kobj_attribute *attr,
+ const char *buf, size_t n)
+{
+ unsigned int val;
+
+ if (sscanf(buf, "%u", &val) != 1)
+ return -EINVAL;
+
+ set_lid_wake_behavior(!!val);
+
+ return n;
+}
+
+static struct kobj_attribute lid_wake_on_close_attr =
+ __ATTR(lid_wake_on_close, 0644,
+ lid_wake_on_close_show,
+ lid_wake_on_close_store);
+
+static void battery_status_changed(void)
+{
+ struct power_supply *psy = power_supply_get_by_name("olpc-battery");
+
+ if (psy) {
+ power_supply_changed(psy);
+ power_supply_put(psy);
+ }
+}
+
+static void ac_status_changed(void)
+{
+ struct power_supply *psy = power_supply_get_by_name("olpc-ac");
+
+ if (psy) {
+ power_supply_changed(psy);
+ power_supply_put(psy);
+ }
+}
+
+static void process_sci_queue(void)
+{
+ u16 data;
+ int r;
+
+ do {
+ r = olpc_ec_sci_query(&data);
+ if (r || !data)
+ break;
+
+ pr_debug(PFX "SCI 0x%x received\n", data);
+
+ switch (data) {
+ case EC_SCI_SRC_BATERR:
+ case EC_SCI_SRC_BATSOC:
+ case EC_SCI_SRC_BATTERY:
+ case EC_SCI_SRC_BATCRIT:
+ battery_status_changed();
+ break;
+ case EC_SCI_SRC_ACPWR:
+ ac_status_changed();
+ break;
+ }
+ } while (data);
+
+ if (r)
+ pr_err(PFX "Failed to clear SCI queue");
+}
+
+static void process_sci_queue_work(struct work_struct *work)
+{
+ process_sci_queue();
+}
+
+static DECLARE_WORK(sci_work, process_sci_queue_work);
+
+static u32 xo15_sci_gpe_handler(acpi_handle gpe_device, u32 gpe, void *context)
+{
+ schedule_work(&sci_work);
+ return ACPI_INTERRUPT_HANDLED | ACPI_REENABLE_GPE;
+}
+
+static int xo15_sci_add(struct acpi_device *device)
+{
+ unsigned long long tmp;
+ acpi_status status;
+ int r;
+
+ if (!device)
+ return -EINVAL;
+
+ strcpy(acpi_device_name(device), XO15_SCI_DEVICE_NAME);
+ strcpy(acpi_device_class(device), XO15_SCI_CLASS);
+
+ /* Get GPE bit assignment (EC events). */
+ status = acpi_evaluate_integer(device->handle, "_GPE", NULL, &tmp);
+ if (ACPI_FAILURE(status))
+ return -EINVAL;
+
+ xo15_sci_gpe = tmp;
+ status = acpi_install_gpe_handler(NULL, xo15_sci_gpe,
+ ACPI_GPE_EDGE_TRIGGERED,
+ xo15_sci_gpe_handler, device);
+ if (ACPI_FAILURE(status))
+ return -ENODEV;
+
+ dev_info(&device->dev, "Initialized, GPE = 0x%lx\n", xo15_sci_gpe);
+
+ r = sysfs_create_file(&device->dev.kobj, &lid_wake_on_close_attr.attr);
+ if (r)
+ goto err_sysfs;
+
+ /* Flush queue, and enable all SCI events */
+ process_sci_queue();
+ olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+ acpi_enable_gpe(NULL, xo15_sci_gpe);
+
+ /* Enable wake-on-EC */
+ if (device->wakeup.flags.valid)
+ device_init_wakeup(&device->dev, true);
+
+ return 0;
+
+err_sysfs:
+ acpi_remove_gpe_handler(NULL, xo15_sci_gpe, xo15_sci_gpe_handler);
+ cancel_work_sync(&sci_work);
+ return r;
+}
+
+static int xo15_sci_remove(struct acpi_device *device)
+{
+ acpi_disable_gpe(NULL, xo15_sci_gpe);
+ acpi_remove_gpe_handler(NULL, xo15_sci_gpe, xo15_sci_gpe_handler);
+ cancel_work_sync(&sci_work);
+ sysfs_remove_file(&device->dev.kobj, &lid_wake_on_close_attr.attr);
+ return 0;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int xo15_sci_resume(struct device *dev)
+{
+ /* Enable all EC events */
+ olpc_ec_mask_write(EC_SCI_SRC_ALL);
+
+ /* Power/battery status might have changed */
+ battery_status_changed();
+ ac_status_changed();
+
+ return 0;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(xo15_sci_pm, NULL, xo15_sci_resume);
+
+static const struct acpi_device_id xo15_sci_device_ids[] = {
+ {"XO15EC", 0},
+ {"", 0},
+};
+
+static struct acpi_driver xo15_sci_drv = {
+ .name = DRV_NAME,
+ .class = XO15_SCI_CLASS,
+ .ids = xo15_sci_device_ids,
+ .ops = {
+ .add = xo15_sci_add,
+ .remove = xo15_sci_remove,
+ },
+ .drv.pm = &xo15_sci_pm,
+};
+
+static int __init xo15_sci_init(void)
+{
+ return acpi_bus_register_driver(&xo15_sci_drv);
+}
+device_initcall(xo15_sci_init);
diff --git a/arch/x86/platform/olpc/olpc.c b/arch/x86/platform/olpc/olpc.c
new file mode 100644
index 000000000..f0e920fb9
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc.c
@@ -0,0 +1,408 @@
+/*
+ * Support for the OLPC DCON and OLPC EC access
+ *
+ * Copyright © 2006 Advanced Micro Devices, Inc.
+ * Copyright © 2007-2008 Andres Salomon <dilinger@debian.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/export.h>
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/string.h>
+#include <linux/platform_device.h>
+#include <linux/of.h>
+#include <linux/syscore_ops.h>
+#include <linux/mutex.h>
+#include <linux/olpc-ec.h>
+
+#include <asm/geode.h>
+#include <asm/setup.h>
+#include <asm/olpc.h>
+#include <asm/olpc_ofw.h>
+
+struct olpc_platform_t olpc_platform_info;
+EXPORT_SYMBOL_GPL(olpc_platform_info);
+
+/* EC event mask to be applied during suspend (defining wakeup sources). */
+static u16 ec_wakeup_mask;
+
+/* what the timeout *should* be (in ms) */
+#define EC_BASE_TIMEOUT 20
+
+/* the timeout that bugs in the EC might force us to actually use */
+static int ec_timeout = EC_BASE_TIMEOUT;
+
+static int __init olpc_ec_timeout_set(char *str)
+{
+ if (get_option(&str, &ec_timeout) != 1) {
+ ec_timeout = EC_BASE_TIMEOUT;
+ printk(KERN_ERR "olpc-ec: invalid argument to "
+ "'olpc_ec_timeout=', ignoring!\n");
+ }
+ printk(KERN_DEBUG "olpc-ec: using %d ms delay for EC commands.\n",
+ ec_timeout);
+ return 1;
+}
+__setup("olpc_ec_timeout=", olpc_ec_timeout_set);
+
+/*
+ * These {i,o}bf_status functions return whether the buffers are full or not.
+ */
+
+static inline unsigned int ibf_status(unsigned int port)
+{
+ return !!(inb(port) & 0x02);
+}
+
+static inline unsigned int obf_status(unsigned int port)
+{
+ return inb(port) & 0x01;
+}
+
+#define wait_on_ibf(p, d) __wait_on_ibf(__LINE__, (p), (d))
+static int __wait_on_ibf(unsigned int line, unsigned int port, int desired)
+{
+ unsigned int timeo;
+ int state = ibf_status(port);
+
+ for (timeo = ec_timeout; state != desired && timeo; timeo--) {
+ mdelay(1);
+ state = ibf_status(port);
+ }
+
+ if ((state == desired) && (ec_timeout > EC_BASE_TIMEOUT) &&
+ timeo < (ec_timeout - EC_BASE_TIMEOUT)) {
+ printk(KERN_WARNING "olpc-ec: %d: waited %u ms for IBF!\n",
+ line, ec_timeout - timeo);
+ }
+
+ return !(state == desired);
+}
+
+#define wait_on_obf(p, d) __wait_on_obf(__LINE__, (p), (d))
+static int __wait_on_obf(unsigned int line, unsigned int port, int desired)
+{
+ unsigned int timeo;
+ int state = obf_status(port);
+
+ for (timeo = ec_timeout; state != desired && timeo; timeo--) {
+ mdelay(1);
+ state = obf_status(port);
+ }
+
+ if ((state == desired) && (ec_timeout > EC_BASE_TIMEOUT) &&
+ timeo < (ec_timeout - EC_BASE_TIMEOUT)) {
+ printk(KERN_WARNING "olpc-ec: %d: waited %u ms for OBF!\n",
+ line, ec_timeout - timeo);
+ }
+
+ return !(state == desired);
+}
+
+/*
+ * This allows the kernel to run Embedded Controller commands. The EC is
+ * documented at <http://wiki.laptop.org/go/Embedded_controller>, and the
+ * available EC commands are here:
+ * <http://wiki.laptop.org/go/Ec_specification>. Unfortunately, while
+ * OpenFirmware's source is available, the EC's is not.
+ */
+static int olpc_xo1_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *outbuf,
+ size_t outlen, void *arg)
+{
+ int ret = -EIO;
+ int i;
+ int restarts = 0;
+
+ /* Clear OBF */
+ for (i = 0; i < 10 && (obf_status(0x6c) == 1); i++)
+ inb(0x68);
+ if (i == 10) {
+ printk(KERN_ERR "olpc-ec: timeout while attempting to "
+ "clear OBF flag!\n");
+ goto err;
+ }
+
+ if (wait_on_ibf(0x6c, 0)) {
+ printk(KERN_ERR "olpc-ec: timeout waiting for EC to "
+ "quiesce!\n");
+ goto err;
+ }
+
+restart:
+ /*
+ * Note that if we time out during any IBF checks, that's a failure;
+ * we have to return. There's no way for the kernel to clear that.
+ *
+ * If we time out during an OBF check, we can restart the command;
+ * reissuing it will clear the OBF flag, and we should be alright.
+ * The OBF flag will sometimes misbehave due to what we believe
+ * is a hardware quirk..
+ */
+ pr_devel("olpc-ec: running cmd 0x%x\n", cmd);
+ outb(cmd, 0x6c);
+
+ if (wait_on_ibf(0x6c, 0)) {
+ printk(KERN_ERR "olpc-ec: timeout waiting for EC to read "
+ "command!\n");
+ goto err;
+ }
+
+ if (inbuf && inlen) {
+ /* write data to EC */
+ for (i = 0; i < inlen; i++) {
+ pr_devel("olpc-ec: sending cmd arg 0x%x\n", inbuf[i]);
+ outb(inbuf[i], 0x68);
+ if (wait_on_ibf(0x6c, 0)) {
+ printk(KERN_ERR "olpc-ec: timeout waiting for"
+ " EC accept data!\n");
+ goto err;
+ }
+ }
+ }
+ if (outbuf && outlen) {
+ /* read data from EC */
+ for (i = 0; i < outlen; i++) {
+ if (wait_on_obf(0x6c, 1)) {
+ printk(KERN_ERR "olpc-ec: timeout waiting for"
+ " EC to provide data!\n");
+ if (restarts++ < 10)
+ goto restart;
+ goto err;
+ }
+ outbuf[i] = inb(0x68);
+ pr_devel("olpc-ec: received 0x%x\n", outbuf[i]);
+ }
+ }
+
+ ret = 0;
+err:
+ return ret;
+}
+
+void olpc_ec_wakeup_set(u16 value)
+{
+ ec_wakeup_mask |= value;
+}
+EXPORT_SYMBOL_GPL(olpc_ec_wakeup_set);
+
+void olpc_ec_wakeup_clear(u16 value)
+{
+ ec_wakeup_mask &= ~value;
+}
+EXPORT_SYMBOL_GPL(olpc_ec_wakeup_clear);
+
+/*
+ * Returns true if the compile and runtime configurations allow for EC events
+ * to wake the system.
+ */
+bool olpc_ec_wakeup_available(void)
+{
+ if (!machine_is_olpc())
+ return false;
+
+ /*
+ * XO-1 EC wakeups are available when olpc-xo1-sci driver is
+ * compiled in
+ */
+#ifdef CONFIG_OLPC_XO1_SCI
+ if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) /* XO-1 */
+ return true;
+#endif
+
+ /*
+ * XO-1.5 EC wakeups are available when olpc-xo15-sci driver is
+ * compiled in
+ */
+#ifdef CONFIG_OLPC_XO15_SCI
+ if (olpc_platform_info.boardrev >= olpc_board_pre(0xd0)) /* XO-1.5 */
+ return true;
+#endif
+
+ return false;
+}
+EXPORT_SYMBOL_GPL(olpc_ec_wakeup_available);
+
+int olpc_ec_mask_write(u16 bits)
+{
+ if (olpc_platform_info.flags & OLPC_F_EC_WIDE_SCI) {
+ __be16 ec_word = cpu_to_be16(bits);
+ return olpc_ec_cmd(EC_WRITE_EXT_SCI_MASK, (void *) &ec_word, 2,
+ NULL, 0);
+ } else {
+ unsigned char ec_byte = bits & 0xff;
+ return olpc_ec_cmd(EC_WRITE_SCI_MASK, &ec_byte, 1, NULL, 0);
+ }
+}
+EXPORT_SYMBOL_GPL(olpc_ec_mask_write);
+
+int olpc_ec_sci_query(u16 *sci_value)
+{
+ int ret;
+
+ if (olpc_platform_info.flags & OLPC_F_EC_WIDE_SCI) {
+ __be16 ec_word;
+ ret = olpc_ec_cmd(EC_EXT_SCI_QUERY,
+ NULL, 0, (void *) &ec_word, 2);
+ if (ret == 0)
+ *sci_value = be16_to_cpu(ec_word);
+ } else {
+ unsigned char ec_byte;
+ ret = olpc_ec_cmd(EC_SCI_QUERY, NULL, 0, &ec_byte, 1);
+ if (ret == 0)
+ *sci_value = ec_byte;
+ }
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(olpc_ec_sci_query);
+
+static bool __init check_ofw_architecture(struct device_node *root)
+{
+ const char *olpc_arch;
+ int propsize;
+
+ olpc_arch = of_get_property(root, "architecture", &propsize);
+ return propsize == 5 && strncmp("OLPC", olpc_arch, 5) == 0;
+}
+
+static u32 __init get_board_revision(struct device_node *root)
+{
+ int propsize;
+ const __be32 *rev;
+
+ rev = of_get_property(root, "board-revision-int", &propsize);
+ if (propsize != 4)
+ return 0;
+
+ return be32_to_cpu(*rev);
+}
+
+static bool __init platform_detect(void)
+{
+ struct device_node *root = of_find_node_by_path("/");
+ bool success;
+
+ if (!root)
+ return false;
+
+ success = check_ofw_architecture(root);
+ if (success) {
+ olpc_platform_info.boardrev = get_board_revision(root);
+ olpc_platform_info.flags |= OLPC_F_PRESENT;
+ }
+
+ of_node_put(root);
+ return success;
+}
+
+static int __init add_xo1_platform_devices(void)
+{
+ struct platform_device *pdev;
+
+ pdev = platform_device_register_simple("xo1-rfkill", -1, NULL, 0);
+ if (IS_ERR(pdev))
+ return PTR_ERR(pdev);
+
+ pdev = platform_device_register_simple("olpc-xo1", -1, NULL, 0);
+
+ return PTR_ERR_OR_ZERO(pdev);
+}
+
+static int olpc_xo1_ec_probe(struct platform_device *pdev)
+{
+ /* get the EC revision */
+ olpc_ec_cmd(EC_FIRMWARE_REV, NULL, 0,
+ (unsigned char *) &olpc_platform_info.ecver, 1);
+
+ /* EC version 0x5f adds support for wide SCI mask */
+ if (olpc_platform_info.ecver >= 0x5f)
+ olpc_platform_info.flags |= OLPC_F_EC_WIDE_SCI;
+
+ pr_info("OLPC board revision %s%X (EC=%x)\n",
+ ((olpc_platform_info.boardrev & 0xf) < 8) ? "pre" : "",
+ olpc_platform_info.boardrev >> 4,
+ olpc_platform_info.ecver);
+
+ return 0;
+}
+static int olpc_xo1_ec_suspend(struct platform_device *pdev)
+{
+ olpc_ec_mask_write(ec_wakeup_mask);
+
+ /*
+ * Squelch SCIs while suspended. This is a fix for
+ * <http://dev.laptop.org/ticket/1835>.
+ */
+ return olpc_ec_cmd(EC_SET_SCI_INHIBIT, NULL, 0, NULL, 0);
+}
+
+static int olpc_xo1_ec_resume(struct platform_device *pdev)
+{
+ /* Tell the EC to stop inhibiting SCIs */
+ olpc_ec_cmd(EC_SET_SCI_INHIBIT_RELEASE, NULL, 0, NULL, 0);
+
+ /*
+ * Tell the wireless module to restart USB communication.
+ * Must be done twice.
+ */
+ olpc_ec_cmd(EC_WAKE_UP_WLAN, NULL, 0, NULL, 0);
+ olpc_ec_cmd(EC_WAKE_UP_WLAN, NULL, 0, NULL, 0);
+
+ return 0;
+}
+
+static struct olpc_ec_driver ec_xo1_driver = {
+ .probe = olpc_xo1_ec_probe,
+ .suspend = olpc_xo1_ec_suspend,
+ .resume = olpc_xo1_ec_resume,
+ .ec_cmd = olpc_xo1_ec_cmd,
+};
+
+static struct olpc_ec_driver ec_xo1_5_driver = {
+ .probe = olpc_xo1_ec_probe,
+ .ec_cmd = olpc_xo1_ec_cmd,
+};
+
+static int __init olpc_init(void)
+{
+ int r = 0;
+
+ if (!olpc_ofw_present() || !platform_detect())
+ return 0;
+
+ /* register the XO-1 and 1.5-specific EC handler */
+ if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) /* XO-1 */
+ olpc_ec_driver_register(&ec_xo1_driver, NULL);
+ else
+ olpc_ec_driver_register(&ec_xo1_5_driver, NULL);
+ platform_device_register_simple("olpc-ec", -1, NULL, 0);
+
+ /* assume B1 and above models always have a DCON */
+ if (olpc_board_at_least(olpc_board(0xb1)))
+ olpc_platform_info.flags |= OLPC_F_DCON;
+
+#ifdef CONFIG_PCI_OLPC
+ /* If the VSA exists let it emulate PCI, if not emulate in kernel.
+ * XO-1 only. */
+ if (olpc_platform_info.boardrev < olpc_board_pre(0xd0) &&
+ !cs5535_has_vsa2())
+ x86_init.pci.arch_init = pci_olpc_init;
+#endif
+
+ if (olpc_platform_info.boardrev < olpc_board_pre(0xd0)) { /* XO-1 */
+ r = add_xo1_platform_devices();
+ if (r)
+ return r;
+ }
+
+ return 0;
+}
+
+postcore_initcall(olpc_init);
diff --git a/arch/x86/platform/olpc/olpc_dt.c b/arch/x86/platform/olpc/olpc_dt.c
new file mode 100644
index 000000000..d6ee92986
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc_dt.c
@@ -0,0 +1,304 @@
+/*
+ * OLPC-specific OFW device tree support code.
+ *
+ * Paul Mackerras August 1996.
+ * Copyright (C) 1996-2005 Paul Mackerras.
+ *
+ * Adapted for 64bit PowerPC by Dave Engebretsen and Peter Bergner.
+ * {engebret|bergner}@us.ibm.com
+ *
+ * Adapted for sparc by David S. Miller davem@davemloft.net
+ * Adapted for x86/OLPC by Andres Salomon <dilinger@queued.net>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/bootmem.h>
+#include <linux/of.h>
+#include <linux/of_platform.h>
+#include <linux/of_pdt.h>
+#include <asm/olpc.h>
+#include <asm/olpc_ofw.h>
+
+static phandle __init olpc_dt_getsibling(phandle node)
+{
+ const void *args[] = { (void *)node };
+ void *res[] = { &node };
+
+ if ((s32)node == -1)
+ return 0;
+
+ if (olpc_ofw("peer", args, res) || (s32)node == -1)
+ return 0;
+
+ return node;
+}
+
+static phandle __init olpc_dt_getchild(phandle node)
+{
+ const void *args[] = { (void *)node };
+ void *res[] = { &node };
+
+ if ((s32)node == -1)
+ return 0;
+
+ if (olpc_ofw("child", args, res) || (s32)node == -1) {
+ pr_err("PROM: %s: fetching child failed!\n", __func__);
+ return 0;
+ }
+
+ return node;
+}
+
+static int __init olpc_dt_getproplen(phandle node, const char *prop)
+{
+ const void *args[] = { (void *)node, prop };
+ int len;
+ void *res[] = { &len };
+
+ if ((s32)node == -1)
+ return -1;
+
+ if (olpc_ofw("getproplen", args, res)) {
+ pr_err("PROM: %s: getproplen failed!\n", __func__);
+ return -1;
+ }
+
+ return len;
+}
+
+static int __init olpc_dt_getproperty(phandle node, const char *prop,
+ char *buf, int bufsize)
+{
+ int plen;
+
+ plen = olpc_dt_getproplen(node, prop);
+ if (plen > bufsize || plen < 1) {
+ return -1;
+ } else {
+ const void *args[] = { (void *)node, prop, buf, (void *)plen };
+ void *res[] = { &plen };
+
+ if (olpc_ofw("getprop", args, res)) {
+ pr_err("PROM: %s: getprop failed!\n", __func__);
+ return -1;
+ }
+ }
+
+ return plen;
+}
+
+static int __init olpc_dt_nextprop(phandle node, char *prev, char *buf)
+{
+ const void *args[] = { (void *)node, prev, buf };
+ int success;
+ void *res[] = { &success };
+
+ buf[0] = '\0';
+
+ if ((s32)node == -1)
+ return -1;
+
+ if (olpc_ofw("nextprop", args, res) || success != 1)
+ return -1;
+
+ return 0;
+}
+
+static int __init olpc_dt_pkg2path(phandle node, char *buf,
+ const int buflen, int *len)
+{
+ const void *args[] = { (void *)node, buf, (void *)buflen };
+ void *res[] = { len };
+
+ if ((s32)node == -1)
+ return -1;
+
+ if (olpc_ofw("package-to-path", args, res) || *len < 1)
+ return -1;
+
+ return 0;
+}
+
+static unsigned int prom_early_allocated __initdata;
+
+void * __init prom_early_alloc(unsigned long size)
+{
+ static u8 *mem;
+ static size_t free_mem;
+ void *res;
+
+ if (free_mem < size) {
+ const size_t chunk_size = max(PAGE_SIZE, size);
+
+ /*
+ * To mimimize the number of allocations, grab at least
+ * PAGE_SIZE of memory (that's an arbitrary choice that's
+ * fast enough on the platforms we care about while minimizing
+ * wasted bootmem) and hand off chunks of it to callers.
+ */
+ res = alloc_bootmem(chunk_size);
+ BUG_ON(!res);
+ prom_early_allocated += chunk_size;
+ memset(res, 0, chunk_size);
+ free_mem = chunk_size;
+ mem = res;
+ }
+
+ /* allocate from the local cache */
+ free_mem -= size;
+ res = mem;
+ mem += size;
+ return res;
+}
+
+static struct of_pdt_ops prom_olpc_ops __initdata = {
+ .nextprop = olpc_dt_nextprop,
+ .getproplen = olpc_dt_getproplen,
+ .getproperty = olpc_dt_getproperty,
+ .getchild = olpc_dt_getchild,
+ .getsibling = olpc_dt_getsibling,
+ .pkg2path = olpc_dt_pkg2path,
+};
+
+static phandle __init olpc_dt_finddevice(const char *path)
+{
+ phandle node;
+ const void *args[] = { path };
+ void *res[] = { &node };
+
+ if (olpc_ofw("finddevice", args, res)) {
+ pr_err("olpc_dt: finddevice failed!\n");
+ return 0;
+ }
+
+ if ((s32) node == -1)
+ return 0;
+
+ return node;
+}
+
+static int __init olpc_dt_interpret(const char *words)
+{
+ int result;
+ const void *args[] = { words };
+ void *res[] = { &result };
+
+ if (olpc_ofw("interpret", args, res)) {
+ pr_err("olpc_dt: interpret failed!\n");
+ return -1;
+ }
+
+ return result;
+}
+
+/*
+ * Extract board revision directly from OFW device tree.
+ * We can't use olpc_platform_info because that hasn't been set up yet.
+ */
+static u32 __init olpc_dt_get_board_revision(void)
+{
+ phandle node;
+ __be32 rev;
+ int r;
+
+ node = olpc_dt_finddevice("/");
+ if (!node)
+ return 0;
+
+ r = olpc_dt_getproperty(node, "board-revision-int",
+ (char *) &rev, sizeof(rev));
+ if (r < 0)
+ return 0;
+
+ return be32_to_cpu(rev);
+}
+
+void __init olpc_dt_fixup(void)
+{
+ int r;
+ char buf[64];
+ phandle node;
+ u32 board_rev;
+
+ node = olpc_dt_finddevice("/battery@0");
+ if (!node)
+ return;
+
+ /*
+ * If the battery node has a compatible property, we are running a new
+ * enough firmware and don't have fixups to make.
+ */
+ r = olpc_dt_getproperty(node, "compatible", buf, sizeof(buf));
+ if (r > 0)
+ return;
+
+ pr_info("PROM DT: Old firmware detected, applying fixes\n");
+
+ /* Add olpc,xo1-battery compatible marker to battery node */
+ olpc_dt_interpret("\" /battery@0\" find-device"
+ " \" olpc,xo1-battery\" +compatible"
+ " device-end");
+
+ board_rev = olpc_dt_get_board_revision();
+ if (!board_rev)
+ return;
+
+ if (board_rev >= olpc_board_pre(0xd0)) {
+ /* XO-1.5: add dcon device */
+ olpc_dt_interpret("\" /pci/display@1\" find-device"
+ " new-device"
+ " \" dcon\" device-name \" olpc,xo1-dcon\" +compatible"
+ " finish-device device-end");
+ } else {
+ /* XO-1: add dcon device, mark RTC as olpc,xo1-rtc */
+ olpc_dt_interpret("\" /pci/display@1,1\" find-device"
+ " new-device"
+ " \" dcon\" device-name \" olpc,xo1-dcon\" +compatible"
+ " finish-device device-end"
+ " \" /rtc\" find-device"
+ " \" olpc,xo1-rtc\" +compatible"
+ " device-end");
+ }
+}
+
+void __init olpc_dt_build_devicetree(void)
+{
+ phandle root;
+
+ if (!olpc_ofw_is_installed())
+ return;
+
+ olpc_dt_fixup();
+
+ root = olpc_dt_getsibling(0);
+ if (!root) {
+ pr_err("PROM: unable to get root node from OFW!\n");
+ return;
+ }
+ of_pdt_build_devicetree(root, &prom_olpc_ops);
+
+ pr_info("PROM DT: Built device tree with %u bytes of memory.\n",
+ prom_early_allocated);
+}
+
+/* A list of DT node/bus matches that we want to expose as platform devices */
+static struct of_device_id __initdata of_ids[] = {
+ { .compatible = "olpc,xo1-battery" },
+ { .compatible = "olpc,xo1-dcon" },
+ { .compatible = "olpc,xo1-rtc" },
+ {},
+};
+
+static int __init olpc_create_platform_devices(void)
+{
+ if (machine_is_olpc())
+ return of_platform_bus_probe(NULL, of_ids, NULL);
+ else
+ return 0;
+}
+device_initcall(olpc_create_platform_devices);
diff --git a/arch/x86/platform/olpc/olpc_ofw.c b/arch/x86/platform/olpc/olpc_ofw.c
new file mode 100644
index 000000000..f1aab8cdb
--- /dev/null
+++ b/arch/x86/platform/olpc/olpc_ofw.c
@@ -0,0 +1,120 @@
+#include <linux/kernel.h>
+#include <linux/export.h>
+#include <linux/spinlock_types.h>
+#include <linux/init.h>
+#include <asm/page.h>
+#include <asm/setup.h>
+#include <asm/io.h>
+#include <asm/cpufeature.h>
+#include <asm/special_insns.h>
+#include <asm/pgtable.h>
+#include <asm/olpc_ofw.h>
+
+/* address of OFW callback interface; will be NULL if OFW isn't found */
+static int (*olpc_ofw_cif)(int *);
+
+/* page dir entry containing OFW's pgdir table; filled in by head_32.S */
+u32 olpc_ofw_pgd __initdata;
+
+static DEFINE_SPINLOCK(ofw_lock);
+
+#define MAXARGS 10
+
+void __init setup_olpc_ofw_pgd(void)
+{
+ pgd_t *base, *ofw_pde;
+
+ if (!olpc_ofw_cif)
+ return;
+
+ /* fetch OFW's PDE */
+ base = early_ioremap(olpc_ofw_pgd, sizeof(olpc_ofw_pgd) * PTRS_PER_PGD);
+ if (!base) {
+ printk(KERN_ERR "failed to remap OFW's pgd - disabling OFW!\n");
+ olpc_ofw_cif = NULL;
+ return;
+ }
+ ofw_pde = &base[OLPC_OFW_PDE_NR];
+
+ /* install OFW's PDE permanently into the kernel's pgtable */
+ set_pgd(&swapper_pg_dir[OLPC_OFW_PDE_NR], *ofw_pde);
+ /* implicit optimization barrier here due to uninline function return */
+
+ early_iounmap(base, sizeof(olpc_ofw_pgd) * PTRS_PER_PGD);
+}
+
+int __olpc_ofw(const char *name, int nr_args, const void **args, int nr_res,
+ void **res)
+{
+ int ofw_args[MAXARGS + 3];
+ unsigned long flags;
+ int ret, i, *p;
+
+ BUG_ON(nr_args + nr_res > MAXARGS);
+
+ if (!olpc_ofw_cif)
+ return -EIO;
+
+ ofw_args[0] = (int)name;
+ ofw_args[1] = nr_args;
+ ofw_args[2] = nr_res;
+
+ p = &ofw_args[3];
+ for (i = 0; i < nr_args; i++, p++)
+ *p = (int)args[i];
+
+ /* call into ofw */
+ spin_lock_irqsave(&ofw_lock, flags);
+ ret = olpc_ofw_cif(ofw_args);
+ spin_unlock_irqrestore(&ofw_lock, flags);
+
+ if (!ret) {
+ for (i = 0; i < nr_res; i++, p++)
+ *((int *)res[i]) = *p;
+ }
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(__olpc_ofw);
+
+bool olpc_ofw_present(void)
+{
+ return olpc_ofw_cif != NULL;
+}
+EXPORT_SYMBOL_GPL(olpc_ofw_present);
+
+/* OFW cif _should_ be above this address */
+#define OFW_MIN 0xff000000
+
+/* OFW starts on a 1MB boundary */
+#define OFW_BOUND (1<<20)
+
+void __init olpc_ofw_detect(void)
+{
+ struct olpc_ofw_header *hdr = &boot_params.olpc_ofw_header;
+ unsigned long start;
+
+ /* ensure OFW booted us by checking for "OFW " string */
+ if (hdr->ofw_magic != OLPC_OFW_SIG)
+ return;
+
+ olpc_ofw_cif = (int (*)(int *))hdr->cif_handler;
+
+ if ((unsigned long)olpc_ofw_cif < OFW_MIN) {
+ printk(KERN_ERR "OFW detected, but cif has invalid address 0x%lx - disabling.\n",
+ (unsigned long)olpc_ofw_cif);
+ olpc_ofw_cif = NULL;
+ return;
+ }
+
+ /* determine where OFW starts in memory */
+ start = round_down((unsigned long)olpc_ofw_cif, OFW_BOUND);
+ printk(KERN_INFO "OFW detected in memory, cif @ 0x%lx (reserving top %ldMB)\n",
+ (unsigned long)olpc_ofw_cif, (-start) >> 20);
+ reserve_top_address(-start);
+}
+
+bool __init olpc_ofw_is_installed(void)
+{
+ return olpc_ofw_cif != NULL;
+}
diff --git a/arch/x86/platform/olpc/xo1-wakeup.S b/arch/x86/platform/olpc/xo1-wakeup.S
new file mode 100644
index 000000000..5fee3a2c2
--- /dev/null
+++ b/arch/x86/platform/olpc/xo1-wakeup.S
@@ -0,0 +1,125 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+.text
+#include <linux/linkage.h>
+#include <asm/segment.h>
+#include <asm/page.h>
+#include <asm/pgtable_32.h>
+
+ .macro writepost,value
+ movb $0x34, %al
+ outb %al, $0x70
+ movb $\value, %al
+ outb %al, $0x71
+ .endm
+
+wakeup_start:
+ # OFW lands us here, running in protected mode, with a
+ # kernel-compatible GDT already setup.
+
+ # Clear any dangerous flags
+ pushl $0
+ popfl
+
+ writepost 0x31
+
+ # Set up %cr3
+ movl $initial_page_table - __PAGE_OFFSET, %eax
+ movl %eax, %cr3
+
+ movl saved_cr4, %eax
+ movl %eax, %cr4
+
+ movl saved_cr0, %eax
+ movl %eax, %cr0
+
+ # Control registers were modified, pipeline resync is needed
+ jmp 1f
+1:
+
+ movw $__KERNEL_DS, %ax
+ movw %ax, %ss
+ movw %ax, %ds
+ movw %ax, %es
+ movw %ax, %fs
+ movw %ax, %gs
+
+ lgdt saved_gdt
+ lidt saved_idt
+ lldt saved_ldt
+ ljmp $(__KERNEL_CS),$1f
+1:
+ movl %cr3, %eax
+ movl %eax, %cr3
+ wbinvd
+
+ # Go back to the return point
+ jmp ret_point
+
+save_registers:
+ sgdt saved_gdt
+ sidt saved_idt
+ sldt saved_ldt
+
+ pushl %edx
+ movl %cr4, %edx
+ movl %edx, saved_cr4
+
+ movl %cr0, %edx
+ movl %edx, saved_cr0
+
+ popl %edx
+
+ movl %ebx, saved_context_ebx
+ movl %ebp, saved_context_ebp
+ movl %esi, saved_context_esi
+ movl %edi, saved_context_edi
+
+ pushfl
+ popl saved_context_eflags
+
+ ret
+
+restore_registers:
+ movl saved_context_ebp, %ebp
+ movl saved_context_ebx, %ebx
+ movl saved_context_esi, %esi
+ movl saved_context_edi, %edi
+
+ pushl saved_context_eflags
+ popfl
+
+ ret
+
+ENTRY(do_olpc_suspend_lowlevel)
+ call save_processor_state
+ call save_registers
+
+ # This is the stack context we want to remember
+ movl %esp, saved_context_esp
+
+ pushl $3
+ call xo1_do_sleep
+
+ jmp wakeup_start
+ .p2align 4,,7
+ret_point:
+ movl saved_context_esp, %esp
+
+ writepost 0x32
+
+ call restore_registers
+ call restore_processor_state
+ ret
+
+.data
+saved_gdt: .long 0,0
+saved_idt: .long 0,0
+saved_ldt: .long 0
+saved_cr4: .long 0
+saved_cr0: .long 0
+saved_context_esp: .long 0
+saved_context_edi: .long 0
+saved_context_esi: .long 0
+saved_context_ebx: .long 0
+saved_context_ebp: .long 0
+saved_context_eflags: .long 0
diff --git a/arch/x86/platform/scx200/Makefile b/arch/x86/platform/scx200/Makefile
new file mode 100644
index 000000000..762b4c7f4
--- /dev/null
+++ b/arch/x86/platform/scx200/Makefile
@@ -0,0 +1,2 @@
+obj-$(CONFIG_SCx200) += scx200.o
+scx200-y += scx200_32.o
diff --git a/arch/x86/platform/scx200/scx200_32.c b/arch/x86/platform/scx200/scx200_32.c
new file mode 100644
index 000000000..3dc9aee41
--- /dev/null
+++ b/arch/x86/platform/scx200/scx200_32.c
@@ -0,0 +1,129 @@
+/*
+ * Copyright (c) 2001,2002 Christer Weinigel <wingel@nano-system.com>
+ *
+ * National Semiconductor SCx200 support.
+ */
+
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/mutex.h>
+#include <linux/pci.h>
+
+#include <linux/scx200.h>
+#include <linux/scx200_gpio.h>
+
+/* Verify that the configuration block really is there */
+#define scx200_cb_probe(base) (inw((base) + SCx200_CBA) == (base))
+
+MODULE_AUTHOR("Christer Weinigel <wingel@nano-system.com>");
+MODULE_DESCRIPTION("NatSemi SCx200 Driver");
+MODULE_LICENSE("GPL");
+
+unsigned scx200_gpio_base = 0;
+unsigned long scx200_gpio_shadow[2];
+
+unsigned scx200_cb_base = 0;
+
+static struct pci_device_id scx200_tbl[] = {
+ { PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SCx200_BRIDGE) },
+ { PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SC1100_BRIDGE) },
+ { PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SCx200_XBUS) },
+ { PCI_VDEVICE(NS, PCI_DEVICE_ID_NS_SC1100_XBUS) },
+ { },
+};
+MODULE_DEVICE_TABLE(pci,scx200_tbl);
+
+static int scx200_probe(struct pci_dev *, const struct pci_device_id *);
+
+static struct pci_driver scx200_pci_driver = {
+ .name = "scx200",
+ .id_table = scx200_tbl,
+ .probe = scx200_probe,
+};
+
+static DEFINE_MUTEX(scx200_gpio_config_lock);
+
+static void scx200_init_shadow(void)
+{
+ int bank;
+
+ /* read the current values driven on the GPIO signals */
+ for (bank = 0; bank < 2; ++bank)
+ scx200_gpio_shadow[bank] = inl(scx200_gpio_base + 0x10 * bank);
+}
+
+static int scx200_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+ unsigned base;
+
+ if (pdev->device == PCI_DEVICE_ID_NS_SCx200_BRIDGE ||
+ pdev->device == PCI_DEVICE_ID_NS_SC1100_BRIDGE) {
+ base = pci_resource_start(pdev, 0);
+ pr_info("GPIO base 0x%x\n", base);
+
+ if (!request_region(base, SCx200_GPIO_SIZE,
+ "NatSemi SCx200 GPIO")) {
+ pr_err("can't allocate I/O for GPIOs\n");
+ return -EBUSY;
+ }
+
+ scx200_gpio_base = base;
+ scx200_init_shadow();
+
+ } else {
+ /* find the base of the Configuration Block */
+ if (scx200_cb_probe(SCx200_CB_BASE_FIXED)) {
+ scx200_cb_base = SCx200_CB_BASE_FIXED;
+ } else {
+ pci_read_config_dword(pdev, SCx200_CBA_SCRATCH, &base);
+ if (scx200_cb_probe(base)) {
+ scx200_cb_base = base;
+ } else {
+ pr_warn("Configuration Block not found\n");
+ return -ENODEV;
+ }
+ }
+ pr_info("Configuration Block base 0x%x\n", scx200_cb_base);
+ }
+
+ return 0;
+}
+
+u32 scx200_gpio_configure(unsigned index, u32 mask, u32 bits)
+{
+ u32 config, new_config;
+
+ mutex_lock(&scx200_gpio_config_lock);
+
+ outl(index, scx200_gpio_base + 0x20);
+ config = inl(scx200_gpio_base + 0x24);
+
+ new_config = (config & mask) | bits;
+ outl(new_config, scx200_gpio_base + 0x24);
+
+ mutex_unlock(&scx200_gpio_config_lock);
+
+ return config;
+}
+
+static int __init scx200_init(void)
+{
+ pr_info("NatSemi SCx200 Driver\n");
+ return pci_register_driver(&scx200_pci_driver);
+}
+
+static void __exit scx200_cleanup(void)
+{
+ pci_unregister_driver(&scx200_pci_driver);
+ release_region(scx200_gpio_base, SCx200_GPIO_SIZE);
+}
+
+module_init(scx200_init);
+module_exit(scx200_cleanup);
+
+EXPORT_SYMBOL(scx200_gpio_base);
+EXPORT_SYMBOL(scx200_gpio_shadow);
+EXPORT_SYMBOL(scx200_gpio_configure);
+EXPORT_SYMBOL(scx200_cb_base);
diff --git a/arch/x86/platform/sfi/Makefile b/arch/x86/platform/sfi/Makefile
new file mode 100644
index 000000000..cc5db1168
--- /dev/null
+++ b/arch/x86/platform/sfi/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_SFI) += sfi.o
diff --git a/arch/x86/platform/sfi/sfi.c b/arch/x86/platform/sfi/sfi.c
new file mode 100644
index 000000000..6c7111bbd
--- /dev/null
+++ b/arch/x86/platform/sfi/sfi.c
@@ -0,0 +1,114 @@
+/*
+ * sfi.c - x86 architecture SFI support.
+ *
+ * Copyright (c) 2009, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ *
+ */
+
+#define KMSG_COMPONENT "SFI"
+#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
+
+#include <linux/acpi.h>
+#include <linux/init.h>
+#include <linux/sfi.h>
+#include <linux/io.h>
+
+#include <asm/irqdomain.h>
+#include <asm/io_apic.h>
+#include <asm/mpspec.h>
+#include <asm/setup.h>
+#include <asm/apic.h>
+
+#ifdef CONFIG_X86_LOCAL_APIC
+static unsigned long sfi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE;
+
+/* All CPUs enumerated by SFI must be present and enabled */
+static void __init mp_sfi_register_lapic(u8 id)
+{
+ if (MAX_LOCAL_APIC - id <= 0) {
+ pr_warning("Processor #%d invalid (max %d)\n",
+ id, MAX_LOCAL_APIC);
+ return;
+ }
+
+ pr_info("registering lapic[%d]\n", id);
+
+ generic_processor_info(id, GET_APIC_VERSION(apic_read(APIC_LVR)));
+}
+
+static int __init sfi_parse_cpus(struct sfi_table_header *table)
+{
+ struct sfi_table_simple *sb;
+ struct sfi_cpu_table_entry *pentry;
+ int i;
+ int cpu_num;
+
+ sb = (struct sfi_table_simple *)table;
+ cpu_num = SFI_GET_NUM_ENTRIES(sb, struct sfi_cpu_table_entry);
+ pentry = (struct sfi_cpu_table_entry *)sb->pentry;
+
+ for (i = 0; i < cpu_num; i++) {
+ mp_sfi_register_lapic(pentry->apic_id);
+ pentry++;
+ }
+
+ smp_found_config = 1;
+ return 0;
+}
+#endif /* CONFIG_X86_LOCAL_APIC */
+
+#ifdef CONFIG_X86_IO_APIC
+
+static int __init sfi_parse_ioapic(struct sfi_table_header *table)
+{
+ struct sfi_table_simple *sb;
+ struct sfi_apic_table_entry *pentry;
+ int i, num;
+ struct ioapic_domain_cfg cfg = {
+ .type = IOAPIC_DOMAIN_STRICT,
+ .ops = &mp_ioapic_irqdomain_ops,
+ };
+
+ sb = (struct sfi_table_simple *)table;
+ num = SFI_GET_NUM_ENTRIES(sb, struct sfi_apic_table_entry);
+ pentry = (struct sfi_apic_table_entry *)sb->pentry;
+
+ for (i = 0; i < num; i++) {
+ mp_register_ioapic(i, pentry->phys_addr, gsi_top, &cfg);
+ pentry++;
+ }
+
+ WARN(pic_mode, KERN_WARNING
+ "SFI: pic_mod shouldn't be 1 when IOAPIC table is present\n");
+ pic_mode = 0;
+ return 0;
+}
+#endif /* CONFIG_X86_IO_APIC */
+
+/*
+ * sfi_platform_init(): register lapics & io-apics
+ */
+int __init sfi_platform_init(void)
+{
+#ifdef CONFIG_X86_LOCAL_APIC
+ register_lapic_address(sfi_lapic_addr);
+ sfi_table_parse(SFI_SIG_CPUS, NULL, NULL, sfi_parse_cpus);
+#endif
+#ifdef CONFIG_X86_IO_APIC
+ sfi_table_parse(SFI_SIG_APIC, NULL, NULL, sfi_parse_ioapic);
+#endif
+ return 0;
+}
diff --git a/arch/x86/platform/ts5500/Makefile b/arch/x86/platform/ts5500/Makefile
new file mode 100644
index 000000000..c54e348c9
--- /dev/null
+++ b/arch/x86/platform/ts5500/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_TS5500) += ts5500.o
diff --git a/arch/x86/platform/ts5500/ts5500.c b/arch/x86/platform/ts5500/ts5500.c
new file mode 100644
index 000000000..fd39301f2
--- /dev/null
+++ b/arch/x86/platform/ts5500/ts5500.c
@@ -0,0 +1,347 @@
+/*
+ * Technologic Systems TS-5500 Single Board Computer support
+ *
+ * Copyright (C) 2013-2014 Savoir-faire Linux Inc.
+ * Vivien Didelot <vivien.didelot@savoirfairelinux.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU General Public License as published by the Free Software
+ * Foundation; either version 2 of the License, or (at your option) any later
+ * version.
+ *
+ *
+ * This driver registers the Technologic Systems TS-5500 Single Board Computer
+ * (SBC) and its devices, and exposes information to userspace such as jumpers'
+ * state or available options. For further information about sysfs entries, see
+ * Documentation/ABI/testing/sysfs-platform-ts5500.
+ *
+ * This code may be extended to support similar x86-based platforms.
+ * Actually, the TS-5500 and TS-5400 are supported.
+ */
+
+#include <linux/delay.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/leds.h>
+#include <linux/init.h>
+#include <linux/platform_data/gpio-ts5500.h>
+#include <linux/platform_data/max197.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+/* Product code register */
+#define TS5500_PRODUCT_CODE_ADDR 0x74
+#define TS5500_PRODUCT_CODE 0x60 /* TS-5500 product code */
+#define TS5400_PRODUCT_CODE 0x40 /* TS-5400 product code */
+
+/* SRAM/RS-485/ADC options, and RS-485 RTS/Automatic RS-485 flags register */
+#define TS5500_SRAM_RS485_ADC_ADDR 0x75
+#define TS5500_SRAM BIT(0) /* SRAM option */
+#define TS5500_RS485 BIT(1) /* RS-485 option */
+#define TS5500_ADC BIT(2) /* A/D converter option */
+#define TS5500_RS485_RTS BIT(6) /* RTS for RS-485 */
+#define TS5500_RS485_AUTO BIT(7) /* Automatic RS-485 */
+
+/* External Reset/Industrial Temperature Range options register */
+#define TS5500_ERESET_ITR_ADDR 0x76
+#define TS5500_ERESET BIT(0) /* External Reset option */
+#define TS5500_ITR BIT(1) /* Indust. Temp. Range option */
+
+/* LED/Jumpers register */
+#define TS5500_LED_JP_ADDR 0x77
+#define TS5500_LED BIT(0) /* LED flag */
+#define TS5500_JP1 BIT(1) /* Automatic CMOS */
+#define TS5500_JP2 BIT(2) /* Enable Serial Console */
+#define TS5500_JP3 BIT(3) /* Write Enable Drive A */
+#define TS5500_JP4 BIT(4) /* Fast Console (115K baud) */
+#define TS5500_JP5 BIT(5) /* User Jumper */
+#define TS5500_JP6 BIT(6) /* Console on COM1 (req. JP2) */
+#define TS5500_JP7 BIT(7) /* Undocumented (Unused) */
+
+/* A/D Converter registers */
+#define TS5500_ADC_CONV_BUSY_ADDR 0x195 /* Conversion state register */
+#define TS5500_ADC_CONV_BUSY BIT(0)
+#define TS5500_ADC_CONV_INIT_LSB_ADDR 0x196 /* Start conv. / LSB register */
+#define TS5500_ADC_CONV_MSB_ADDR 0x197 /* MSB register */
+#define TS5500_ADC_CONV_DELAY 12 /* usec */
+
+/**
+ * struct ts5500_sbc - TS-5500 board description
+ * @name: Board model name.
+ * @id: Board product ID.
+ * @sram: Flag for SRAM option.
+ * @rs485: Flag for RS-485 option.
+ * @adc: Flag for Analog/Digital converter option.
+ * @ereset: Flag for External Reset option.
+ * @itr: Flag for Industrial Temperature Range option.
+ * @jumpers: Bitfield for jumpers' state.
+ */
+struct ts5500_sbc {
+ const char *name;
+ int id;
+ bool sram;
+ bool rs485;
+ bool adc;
+ bool ereset;
+ bool itr;
+ u8 jumpers;
+};
+
+/* Board signatures in BIOS shadow RAM */
+static const struct {
+ const char * const string;
+ const ssize_t offset;
+} ts5500_signatures[] __initconst = {
+ { "TS-5x00 AMD Elan", 0xb14 },
+};
+
+static int __init ts5500_check_signature(void)
+{
+ void __iomem *bios;
+ int i, ret = -ENODEV;
+
+ bios = ioremap(0xf0000, 0x10000);
+ if (!bios)
+ return -ENOMEM;
+
+ for (i = 0; i < ARRAY_SIZE(ts5500_signatures); i++) {
+ if (check_signature(bios + ts5500_signatures[i].offset,
+ ts5500_signatures[i].string,
+ strlen(ts5500_signatures[i].string))) {
+ ret = 0;
+ break;
+ }
+ }
+
+ iounmap(bios);
+ return ret;
+}
+
+static int __init ts5500_detect_config(struct ts5500_sbc *sbc)
+{
+ u8 tmp;
+ int ret = 0;
+
+ if (!request_region(TS5500_PRODUCT_CODE_ADDR, 4, "ts5500"))
+ return -EBUSY;
+
+ sbc->id = inb(TS5500_PRODUCT_CODE_ADDR);
+ if (sbc->id == TS5500_PRODUCT_CODE) {
+ sbc->name = "TS-5500";
+ } else if (sbc->id == TS5400_PRODUCT_CODE) {
+ sbc->name = "TS-5400";
+ } else {
+ pr_err("ts5500: unknown product code 0x%x\n", sbc->id);
+ ret = -ENODEV;
+ goto cleanup;
+ }
+
+ tmp = inb(TS5500_SRAM_RS485_ADC_ADDR);
+ sbc->sram = tmp & TS5500_SRAM;
+ sbc->rs485 = tmp & TS5500_RS485;
+ sbc->adc = tmp & TS5500_ADC;
+
+ tmp = inb(TS5500_ERESET_ITR_ADDR);
+ sbc->ereset = tmp & TS5500_ERESET;
+ sbc->itr = tmp & TS5500_ITR;
+
+ tmp = inb(TS5500_LED_JP_ADDR);
+ sbc->jumpers = tmp & ~TS5500_LED;
+
+cleanup:
+ release_region(TS5500_PRODUCT_CODE_ADDR, 4);
+ return ret;
+}
+
+static ssize_t name_show(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ struct ts5500_sbc *sbc = dev_get_drvdata(dev);
+
+ return sprintf(buf, "%s\n", sbc->name);
+}
+static DEVICE_ATTR_RO(name);
+
+static ssize_t id_show(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ struct ts5500_sbc *sbc = dev_get_drvdata(dev);
+
+ return sprintf(buf, "0x%.2x\n", sbc->id);
+}
+static DEVICE_ATTR_RO(id);
+
+static ssize_t jumpers_show(struct device *dev, struct device_attribute *attr,
+ char *buf)
+{
+ struct ts5500_sbc *sbc = dev_get_drvdata(dev);
+
+ return sprintf(buf, "0x%.2x\n", sbc->jumpers >> 1);
+}
+static DEVICE_ATTR_RO(jumpers);
+
+#define TS5500_ATTR_BOOL(_field) \
+ static ssize_t _field##_show(struct device *dev, \
+ struct device_attribute *attr, char *buf) \
+ { \
+ struct ts5500_sbc *sbc = dev_get_drvdata(dev); \
+ \
+ return sprintf(buf, "%d\n", sbc->_field); \
+ } \
+ static DEVICE_ATTR_RO(_field)
+
+TS5500_ATTR_BOOL(sram);
+TS5500_ATTR_BOOL(rs485);
+TS5500_ATTR_BOOL(adc);
+TS5500_ATTR_BOOL(ereset);
+TS5500_ATTR_BOOL(itr);
+
+static struct attribute *ts5500_attributes[] = {
+ &dev_attr_id.attr,
+ &dev_attr_name.attr,
+ &dev_attr_jumpers.attr,
+ &dev_attr_sram.attr,
+ &dev_attr_rs485.attr,
+ &dev_attr_adc.attr,
+ &dev_attr_ereset.attr,
+ &dev_attr_itr.attr,
+ NULL
+};
+
+static const struct attribute_group ts5500_attr_group = {
+ .attrs = ts5500_attributes,
+};
+
+static struct resource ts5500_dio1_resource[] = {
+ DEFINE_RES_IRQ_NAMED(7, "DIO1 interrupt"),
+};
+
+static struct platform_device ts5500_dio1_pdev = {
+ .name = "ts5500-dio1",
+ .id = -1,
+ .resource = ts5500_dio1_resource,
+ .num_resources = 1,
+};
+
+static struct resource ts5500_dio2_resource[] = {
+ DEFINE_RES_IRQ_NAMED(6, "DIO2 interrupt"),
+};
+
+static struct platform_device ts5500_dio2_pdev = {
+ .name = "ts5500-dio2",
+ .id = -1,
+ .resource = ts5500_dio2_resource,
+ .num_resources = 1,
+};
+
+static void ts5500_led_set(struct led_classdev *led_cdev,
+ enum led_brightness brightness)
+{
+ outb(!!brightness, TS5500_LED_JP_ADDR);
+}
+
+static enum led_brightness ts5500_led_get(struct led_classdev *led_cdev)
+{
+ return (inb(TS5500_LED_JP_ADDR) & TS5500_LED) ? LED_FULL : LED_OFF;
+}
+
+static struct led_classdev ts5500_led_cdev = {
+ .name = "ts5500:green:",
+ .brightness_set = ts5500_led_set,
+ .brightness_get = ts5500_led_get,
+};
+
+static int ts5500_adc_convert(u8 ctrl)
+{
+ u8 lsb, msb;
+
+ /* Start conversion (ensure the 3 MSB are set to 0) */
+ outb(ctrl & 0x1f, TS5500_ADC_CONV_INIT_LSB_ADDR);
+
+ /*
+ * The platform has CPLD logic driving the A/D converter.
+ * The conversion must complete within 11 microseconds,
+ * otherwise we have to re-initiate a conversion.
+ */
+ udelay(TS5500_ADC_CONV_DELAY);
+ if (inb(TS5500_ADC_CONV_BUSY_ADDR) & TS5500_ADC_CONV_BUSY)
+ return -EBUSY;
+
+ /* Read the raw data */
+ lsb = inb(TS5500_ADC_CONV_INIT_LSB_ADDR);
+ msb = inb(TS5500_ADC_CONV_MSB_ADDR);
+
+ return (msb << 8) | lsb;
+}
+
+static struct max197_platform_data ts5500_adc_pdata = {
+ .convert = ts5500_adc_convert,
+};
+
+static struct platform_device ts5500_adc_pdev = {
+ .name = "max197",
+ .id = -1,
+ .dev = {
+ .platform_data = &ts5500_adc_pdata,
+ },
+};
+
+static int __init ts5500_init(void)
+{
+ struct platform_device *pdev;
+ struct ts5500_sbc *sbc;
+ int err;
+
+ /*
+ * There is no DMI available or PCI bridge subvendor info,
+ * only the BIOS provides a 16-bit identification call.
+ * It is safer to find a signature in the BIOS shadow RAM.
+ */
+ err = ts5500_check_signature();
+ if (err)
+ return err;
+
+ pdev = platform_device_register_simple("ts5500", -1, NULL, 0);
+ if (IS_ERR(pdev))
+ return PTR_ERR(pdev);
+
+ sbc = devm_kzalloc(&pdev->dev, sizeof(struct ts5500_sbc), GFP_KERNEL);
+ if (!sbc) {
+ err = -ENOMEM;
+ goto error;
+ }
+
+ err = ts5500_detect_config(sbc);
+ if (err)
+ goto error;
+
+ platform_set_drvdata(pdev, sbc);
+
+ err = sysfs_create_group(&pdev->dev.kobj, &ts5500_attr_group);
+ if (err)
+ goto error;
+
+ if (sbc->id == TS5500_PRODUCT_CODE) {
+ ts5500_dio1_pdev.dev.parent = &pdev->dev;
+ if (platform_device_register(&ts5500_dio1_pdev))
+ dev_warn(&pdev->dev, "DIO1 block registration failed\n");
+ ts5500_dio2_pdev.dev.parent = &pdev->dev;
+ if (platform_device_register(&ts5500_dio2_pdev))
+ dev_warn(&pdev->dev, "DIO2 block registration failed\n");
+ }
+
+ if (led_classdev_register(&pdev->dev, &ts5500_led_cdev))
+ dev_warn(&pdev->dev, "LED registration failed\n");
+
+ if (sbc->adc) {
+ ts5500_adc_pdev.dev.parent = &pdev->dev;
+ if (platform_device_register(&ts5500_adc_pdev))
+ dev_warn(&pdev->dev, "ADC registration failed\n");
+ }
+
+ return 0;
+error:
+ platform_device_unregister(pdev);
+ return err;
+}
+device_initcall(ts5500_init);
diff --git a/arch/x86/platform/uv/Makefile b/arch/x86/platform/uv/Makefile
new file mode 100644
index 000000000..52079bebd
--- /dev/null
+++ b/arch/x86/platform/uv/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_X86_UV) += tlb_uv.o bios_uv.o uv_irq.o uv_sysfs.o uv_time.o uv_nmi.o
diff --git a/arch/x86/platform/uv/bios_uv.c b/arch/x86/platform/uv/bios_uv.c
new file mode 100644
index 000000000..eb33432f2
--- /dev/null
+++ b/arch/x86/platform/uv/bios_uv.c
@@ -0,0 +1,240 @@
+/*
+ * BIOS run time interface routines.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Copyright (c) 2008-2009 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) Russ Anderson <rja@sgi.com>
+ */
+
+#include <linux/efi.h>
+#include <linux/export.h>
+#include <linux/slab.h>
+#include <asm/efi.h>
+#include <linux/io.h>
+#include <asm/uv/bios.h>
+#include <asm/uv/uv_hub.h>
+
+struct uv_systab *uv_systab;
+
+static s64 __uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
+ u64 a4, u64 a5)
+{
+ struct uv_systab *tab = uv_systab;
+ s64 ret;
+
+ if (!tab || !tab->function)
+ /*
+ * BIOS does not support UV systab
+ */
+ return BIOS_STATUS_UNIMPLEMENTED;
+
+ /*
+ * If EFI_OLD_MEMMAP is set, we need to fall back to using our old EFI
+ * callback method, which uses efi_call() directly, with the kernel page tables:
+ */
+ if (unlikely(test_bit(EFI_OLD_MEMMAP, &efi.flags)))
+ ret = efi_call((void *)__va(tab->function), (u64)which, a1, a2, a3, a4, a5);
+ else
+ ret = efi_call_virt_pointer(tab, function, (u64)which, a1, a2, a3, a4, a5);
+
+ return ret;
+}
+
+s64 uv_bios_call(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3, u64 a4, u64 a5)
+{
+ s64 ret;
+
+ if (down_interruptible(&__efi_uv_runtime_lock))
+ return BIOS_STATUS_ABORT;
+
+ ret = __uv_bios_call(which, a1, a2, a3, a4, a5);
+ up(&__efi_uv_runtime_lock);
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(uv_bios_call);
+
+s64 uv_bios_call_irqsave(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
+ u64 a4, u64 a5)
+{
+ unsigned long bios_flags;
+ s64 ret;
+
+ if (down_interruptible(&__efi_uv_runtime_lock))
+ return BIOS_STATUS_ABORT;
+
+ local_irq_save(bios_flags);
+ ret = __uv_bios_call(which, a1, a2, a3, a4, a5);
+ local_irq_restore(bios_flags);
+
+ up(&__efi_uv_runtime_lock);
+
+ return ret;
+}
+
+s64 uv_bios_call_reentrant(enum uv_bios_cmd which, u64 a1, u64 a2, u64 a3,
+ u64 a4, u64 a5)
+{
+ s64 ret;
+
+ preempt_disable();
+ ret = uv_bios_call(which, a1, a2, a3, a4, a5);
+ preempt_enable();
+
+ return ret;
+}
+
+
+long sn_partition_id;
+EXPORT_SYMBOL_GPL(sn_partition_id);
+long sn_coherency_id;
+EXPORT_SYMBOL_GPL(sn_coherency_id);
+long sn_region_size;
+EXPORT_SYMBOL_GPL(sn_region_size);
+long system_serial_number;
+EXPORT_SYMBOL_GPL(system_serial_number);
+int uv_type;
+EXPORT_SYMBOL_GPL(uv_type);
+
+
+s64 uv_bios_get_sn_info(int fc, int *uvtype, long *partid, long *coher,
+ long *region, long *ssn)
+{
+ s64 ret;
+ u64 v0, v1;
+ union partition_info_u part;
+
+ ret = uv_bios_call_irqsave(UV_BIOS_GET_SN_INFO, fc,
+ (u64)(&v0), (u64)(&v1), 0, 0);
+ if (ret != BIOS_STATUS_SUCCESS)
+ return ret;
+
+ part.val = v0;
+ if (uvtype)
+ *uvtype = part.hub_version;
+ if (partid)
+ *partid = part.partition_id;
+ if (coher)
+ *coher = part.coherence_id;
+ if (region)
+ *region = part.region_size;
+ if (ssn)
+ *ssn = v1;
+ return ret;
+}
+EXPORT_SYMBOL_GPL(uv_bios_get_sn_info);
+
+int
+uv_bios_mq_watchlist_alloc(unsigned long addr, unsigned int mq_size,
+ unsigned long *intr_mmr_offset)
+{
+ u64 watchlist;
+ s64 ret;
+
+ /*
+ * bios returns watchlist number or negative error number.
+ */
+ ret = (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_ALLOC, addr,
+ mq_size, (u64)intr_mmr_offset,
+ (u64)&watchlist, 0);
+ if (ret < BIOS_STATUS_SUCCESS)
+ return ret;
+
+ return watchlist;
+}
+EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_alloc);
+
+int
+uv_bios_mq_watchlist_free(int blade, int watchlist_num)
+{
+ return (int)uv_bios_call_irqsave(UV_BIOS_WATCHLIST_FREE,
+ blade, watchlist_num, 0, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_mq_watchlist_free);
+
+s64
+uv_bios_change_memprotect(u64 paddr, u64 len, enum uv_memprotect perms)
+{
+ return uv_bios_call_irqsave(UV_BIOS_MEMPROTECT, paddr, len,
+ perms, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_change_memprotect);
+
+s64
+uv_bios_reserved_page_pa(u64 buf, u64 *cookie, u64 *addr, u64 *len)
+{
+ return uv_bios_call_irqsave(UV_BIOS_GET_PARTITION_ADDR, (u64)cookie,
+ (u64)addr, buf, (u64)len, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_reserved_page_pa);
+
+s64 uv_bios_freq_base(u64 clock_type, u64 *ticks_per_second)
+{
+ return uv_bios_call(UV_BIOS_FREQ_BASE, clock_type,
+ (u64)ticks_per_second, 0, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_freq_base);
+
+/*
+ * uv_bios_set_legacy_vga_target - Set Legacy VGA I/O Target
+ * @decode: true to enable target, false to disable target
+ * @domain: PCI domain number
+ * @bus: PCI bus number
+ *
+ * Returns:
+ * 0: Success
+ * -EINVAL: Invalid domain or bus number
+ * -ENOSYS: Capability not available
+ * -EBUSY: Legacy VGA I/O cannot be retargeted at this time
+ */
+int uv_bios_set_legacy_vga_target(bool decode, int domain, int bus)
+{
+ return uv_bios_call(UV_BIOS_SET_LEGACY_VGA_TARGET,
+ (u64)decode, (u64)domain, (u64)bus, 0, 0);
+}
+EXPORT_SYMBOL_GPL(uv_bios_set_legacy_vga_target);
+
+#ifdef CONFIG_EFI
+void uv_bios_init(void)
+{
+ uv_systab = NULL;
+ if ((efi.uv_systab == EFI_INVALID_TABLE_ADDR) ||
+ !efi.uv_systab || efi_runtime_disabled()) {
+ pr_crit("UV: UVsystab: missing\n");
+ return;
+ }
+
+ uv_systab = ioremap(efi.uv_systab, sizeof(struct uv_systab));
+ if (!uv_systab || strncmp(uv_systab->signature, UV_SYSTAB_SIG, 4)) {
+ pr_err("UV: UVsystab: bad signature!\n");
+ iounmap(uv_systab);
+ return;
+ }
+
+ /* Starting with UV4 the UV systab size is variable */
+ if (uv_systab->revision >= UV_SYSTAB_VERSION_UV4) {
+ int size = uv_systab->size;
+
+ iounmap(uv_systab);
+ uv_systab = ioremap(efi.uv_systab, size);
+ if (!uv_systab) {
+ pr_err("UV: UVsystab: ioremap(%d) failed!\n", size);
+ return;
+ }
+ }
+ pr_info("UV: UVsystab: Revision:%x\n", uv_systab->revision);
+}
+#endif
diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c
new file mode 100644
index 000000000..a4130b84d
--- /dev/null
+++ b/arch/x86/platform/uv/tlb_uv.c
@@ -0,0 +1,2284 @@
+/*
+ * SGI UltraViolet TLB flush routines.
+ *
+ * (c) 2008-2014 Cliff Wickman <cpw@sgi.com>, SGI.
+ *
+ * This code is released under the GNU General Public License version 2 or
+ * later.
+ */
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+#include <linux/debugfs.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+
+#include <asm/mmu_context.h>
+#include <asm/uv/uv.h>
+#include <asm/uv/uv_mmrs.h>
+#include <asm/uv/uv_hub.h>
+#include <asm/uv/uv_bau.h>
+#include <asm/apic.h>
+#include <asm/tsc.h>
+#include <asm/irq_vectors.h>
+#include <asm/timer.h>
+
+static struct bau_operations ops __ro_after_init;
+
+/* timeouts in nanoseconds (indexed by UVH_AGING_PRESCALE_SEL urgency7 30:28) */
+static const int timeout_base_ns[] = {
+ 20,
+ 160,
+ 1280,
+ 10240,
+ 81920,
+ 655360,
+ 5242880,
+ 167772160
+};
+
+static int timeout_us;
+static bool nobau = true;
+static int nobau_perm;
+
+/* tunables: */
+static int max_concurr = MAX_BAU_CONCURRENT;
+static int max_concurr_const = MAX_BAU_CONCURRENT;
+static int plugged_delay = PLUGGED_DELAY;
+static int plugsb4reset = PLUGSB4RESET;
+static int giveup_limit = GIVEUP_LIMIT;
+static int timeoutsb4reset = TIMEOUTSB4RESET;
+static int ipi_reset_limit = IPI_RESET_LIMIT;
+static int complete_threshold = COMPLETE_THRESHOLD;
+static int congested_respns_us = CONGESTED_RESPONSE_US;
+static int congested_reps = CONGESTED_REPS;
+static int disabled_period = DISABLED_PERIOD;
+
+static struct tunables tunables[] = {
+ {&max_concurr, MAX_BAU_CONCURRENT}, /* must be [0] */
+ {&plugged_delay, PLUGGED_DELAY},
+ {&plugsb4reset, PLUGSB4RESET},
+ {&timeoutsb4reset, TIMEOUTSB4RESET},
+ {&ipi_reset_limit, IPI_RESET_LIMIT},
+ {&complete_threshold, COMPLETE_THRESHOLD},
+ {&congested_respns_us, CONGESTED_RESPONSE_US},
+ {&congested_reps, CONGESTED_REPS},
+ {&disabled_period, DISABLED_PERIOD},
+ {&giveup_limit, GIVEUP_LIMIT}
+};
+
+static struct dentry *tunables_dir;
+static struct dentry *tunables_file;
+
+/* these correspond to the statistics printed by ptc_seq_show() */
+static char *stat_description[] = {
+ "sent: number of shootdown messages sent",
+ "stime: time spent sending messages",
+ "numuvhubs: number of hubs targeted with shootdown",
+ "numuvhubs16: number times 16 or more hubs targeted",
+ "numuvhubs8: number times 8 or more hubs targeted",
+ "numuvhubs4: number times 4 or more hubs targeted",
+ "numuvhubs2: number times 2 or more hubs targeted",
+ "numuvhubs1: number times 1 hub targeted",
+ "numcpus: number of cpus targeted with shootdown",
+ "dto: number of destination timeouts",
+ "retries: destination timeout retries sent",
+ "rok: : destination timeouts successfully retried",
+ "resetp: ipi-style resource resets for plugs",
+ "resett: ipi-style resource resets for timeouts",
+ "giveup: fall-backs to ipi-style shootdowns",
+ "sto: number of source timeouts",
+ "bz: number of stay-busy's",
+ "throt: number times spun in throttle",
+ "swack: image of UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE",
+ "recv: shootdown messages received",
+ "rtime: time spent processing messages",
+ "all: shootdown all-tlb messages",
+ "one: shootdown one-tlb messages",
+ "mult: interrupts that found multiple messages",
+ "none: interrupts that found no messages",
+ "retry: number of retry messages processed",
+ "canc: number messages canceled by retries",
+ "nocan: number retries that found nothing to cancel",
+ "reset: number of ipi-style reset requests processed",
+ "rcan: number messages canceled by reset requests",
+ "disable: number times use of the BAU was disabled",
+ "enable: number times use of the BAU was re-enabled"
+};
+
+static int __init setup_bau(char *arg)
+{
+ int result;
+
+ if (!arg)
+ return -EINVAL;
+
+ result = strtobool(arg, &nobau);
+ if (result)
+ return result;
+
+ /* we need to flip the logic here, so that bau=y sets nobau to false */
+ nobau = !nobau;
+
+ if (!nobau)
+ pr_info("UV BAU Enabled\n");
+ else
+ pr_info("UV BAU Disabled\n");
+
+ return 0;
+}
+early_param("bau", setup_bau);
+
+/* base pnode in this partition */
+static int uv_base_pnode __read_mostly;
+
+static DEFINE_PER_CPU(struct ptc_stats, ptcstats);
+static DEFINE_PER_CPU(struct bau_control, bau_control);
+static DEFINE_PER_CPU(cpumask_var_t, uv_flush_tlb_mask);
+
+static void
+set_bau_on(void)
+{
+ int cpu;
+ struct bau_control *bcp;
+
+ if (nobau_perm) {
+ pr_info("BAU not initialized; cannot be turned on\n");
+ return;
+ }
+ nobau = false;
+ for_each_present_cpu(cpu) {
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->nobau = false;
+ }
+ pr_info("BAU turned on\n");
+ return;
+}
+
+static void
+set_bau_off(void)
+{
+ int cpu;
+ struct bau_control *bcp;
+
+ nobau = true;
+ for_each_present_cpu(cpu) {
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->nobau = true;
+ }
+ pr_info("BAU turned off\n");
+ return;
+}
+
+/*
+ * Determine the first node on a uvhub. 'Nodes' are used for kernel
+ * memory allocation.
+ */
+static int __init uvhub_to_first_node(int uvhub)
+{
+ int node, b;
+
+ for_each_online_node(node) {
+ b = uv_node_to_blade_id(node);
+ if (uvhub == b)
+ return node;
+ }
+ return -1;
+}
+
+/*
+ * Determine the apicid of the first cpu on a uvhub.
+ */
+static int __init uvhub_to_first_apicid(int uvhub)
+{
+ int cpu;
+
+ for_each_present_cpu(cpu)
+ if (uvhub == uv_cpu_to_blade_id(cpu))
+ return per_cpu(x86_cpu_to_apicid, cpu);
+ return -1;
+}
+
+/*
+ * Free a software acknowledge hardware resource by clearing its Pending
+ * bit. This will return a reply to the sender.
+ * If the message has timed out, a reply has already been sent by the
+ * hardware but the resource has not been released. In that case our
+ * clear of the Timeout bit (as well) will free the resource. No reply will
+ * be sent (the hardware will only do one reply per message).
+ */
+static void reply_to_message(struct msg_desc *mdp, struct bau_control *bcp,
+ int do_acknowledge)
+{
+ unsigned long dw;
+ struct bau_pq_entry *msg;
+
+ msg = mdp->msg;
+ if (!msg->canceled && do_acknowledge) {
+ dw = (msg->swack_vec << UV_SW_ACK_NPENDING) | msg->swack_vec;
+ ops.write_l_sw_ack(dw);
+ }
+ msg->replied_to = 1;
+ msg->swack_vec = 0;
+}
+
+/*
+ * Process the receipt of a RETRY message
+ */
+static void bau_process_retry_msg(struct msg_desc *mdp,
+ struct bau_control *bcp)
+{
+ int i;
+ int cancel_count = 0;
+ unsigned long msg_res;
+ unsigned long mmr = 0;
+ struct bau_pq_entry *msg = mdp->msg;
+ struct bau_pq_entry *msg2;
+ struct ptc_stats *stat = bcp->statp;
+
+ stat->d_retries++;
+ /*
+ * cancel any message from msg+1 to the retry itself
+ */
+ for (msg2 = msg+1, i = 0; i < DEST_Q_SIZE; msg2++, i++) {
+ if (msg2 > mdp->queue_last)
+ msg2 = mdp->queue_first;
+ if (msg2 == msg)
+ break;
+
+ /* same conditions for cancellation as do_reset */
+ if ((msg2->replied_to == 0) && (msg2->canceled == 0) &&
+ (msg2->swack_vec) && ((msg2->swack_vec &
+ msg->swack_vec) == 0) &&
+ (msg2->sending_cpu == msg->sending_cpu) &&
+ (msg2->msg_type != MSG_NOOP)) {
+ mmr = ops.read_l_sw_ack();
+ msg_res = msg2->swack_vec;
+ /*
+ * This is a message retry; clear the resources held
+ * by the previous message only if they timed out.
+ * If it has not timed out we have an unexpected
+ * situation to report.
+ */
+ if (mmr & (msg_res << UV_SW_ACK_NPENDING)) {
+ unsigned long mr;
+ /*
+ * Is the resource timed out?
+ * Make everyone ignore the cancelled message.
+ */
+ msg2->canceled = 1;
+ stat->d_canceled++;
+ cancel_count++;
+ mr = (msg_res << UV_SW_ACK_NPENDING) | msg_res;
+ ops.write_l_sw_ack(mr);
+ }
+ }
+ }
+ if (!cancel_count)
+ stat->d_nocanceled++;
+}
+
+/*
+ * Do all the things a cpu should do for a TLB shootdown message.
+ * Other cpu's may come here at the same time for this message.
+ */
+static void bau_process_message(struct msg_desc *mdp, struct bau_control *bcp,
+ int do_acknowledge)
+{
+ short socket_ack_count = 0;
+ short *sp;
+ struct atomic_short *asp;
+ struct ptc_stats *stat = bcp->statp;
+ struct bau_pq_entry *msg = mdp->msg;
+ struct bau_control *smaster = bcp->socket_master;
+
+ /*
+ * This must be a normal message, or retry of a normal message
+ */
+ if (msg->address == TLB_FLUSH_ALL) {
+ local_flush_tlb();
+ stat->d_alltlb++;
+ } else {
+ __flush_tlb_one_user(msg->address);
+ stat->d_onetlb++;
+ }
+ stat->d_requestee++;
+
+ /*
+ * One cpu on each uvhub has the additional job on a RETRY
+ * of releasing the resource held by the message that is
+ * being retried. That message is identified by sending
+ * cpu number.
+ */
+ if (msg->msg_type == MSG_RETRY && bcp == bcp->uvhub_master)
+ bau_process_retry_msg(mdp, bcp);
+
+ /*
+ * This is a swack message, so we have to reply to it.
+ * Count each responding cpu on the socket. This avoids
+ * pinging the count's cache line back and forth between
+ * the sockets.
+ */
+ sp = &smaster->socket_acknowledge_count[mdp->msg_slot];
+ asp = (struct atomic_short *)sp;
+ socket_ack_count = atom_asr(1, asp);
+ if (socket_ack_count == bcp->cpus_in_socket) {
+ int msg_ack_count;
+ /*
+ * Both sockets dump their completed count total into
+ * the message's count.
+ */
+ *sp = 0;
+ asp = (struct atomic_short *)&msg->acknowledge_count;
+ msg_ack_count = atom_asr(socket_ack_count, asp);
+
+ if (msg_ack_count == bcp->cpus_in_uvhub) {
+ /*
+ * All cpus in uvhub saw it; reply
+ * (unless we are in the UV2 workaround)
+ */
+ reply_to_message(mdp, bcp, do_acknowledge);
+ }
+ }
+
+ return;
+}
+
+/*
+ * Determine the first cpu on a pnode.
+ */
+static int pnode_to_first_cpu(int pnode, struct bau_control *smaster)
+{
+ int cpu;
+ struct hub_and_pnode *hpp;
+
+ for_each_present_cpu(cpu) {
+ hpp = &smaster->thp[cpu];
+ if (pnode == hpp->pnode)
+ return cpu;
+ }
+ return -1;
+}
+
+/*
+ * Last resort when we get a large number of destination timeouts is
+ * to clear resources held by a given cpu.
+ * Do this with IPI so that all messages in the BAU message queue
+ * can be identified by their nonzero swack_vec field.
+ *
+ * This is entered for a single cpu on the uvhub.
+ * The sender want's this uvhub to free a specific message's
+ * swack resources.
+ */
+static void do_reset(void *ptr)
+{
+ int i;
+ struct bau_control *bcp = &per_cpu(bau_control, smp_processor_id());
+ struct reset_args *rap = (struct reset_args *)ptr;
+ struct bau_pq_entry *msg;
+ struct ptc_stats *stat = bcp->statp;
+
+ stat->d_resets++;
+ /*
+ * We're looking for the given sender, and
+ * will free its swack resource.
+ * If all cpu's finally responded after the timeout, its
+ * message 'replied_to' was set.
+ */
+ for (msg = bcp->queue_first, i = 0; i < DEST_Q_SIZE; msg++, i++) {
+ unsigned long msg_res;
+ /* do_reset: same conditions for cancellation as
+ bau_process_retry_msg() */
+ if ((msg->replied_to == 0) &&
+ (msg->canceled == 0) &&
+ (msg->sending_cpu == rap->sender) &&
+ (msg->swack_vec) &&
+ (msg->msg_type != MSG_NOOP)) {
+ unsigned long mmr;
+ unsigned long mr;
+ /*
+ * make everyone else ignore this message
+ */
+ msg->canceled = 1;
+ /*
+ * only reset the resource if it is still pending
+ */
+ mmr = ops.read_l_sw_ack();
+ msg_res = msg->swack_vec;
+ mr = (msg_res << UV_SW_ACK_NPENDING) | msg_res;
+ if (mmr & msg_res) {
+ stat->d_rcanceled++;
+ ops.write_l_sw_ack(mr);
+ }
+ }
+ }
+ return;
+}
+
+/*
+ * Use IPI to get all target uvhubs to release resources held by
+ * a given sending cpu number.
+ */
+static void reset_with_ipi(struct pnmask *distribution, struct bau_control *bcp)
+{
+ int pnode;
+ int apnode;
+ int maskbits;
+ int sender = bcp->cpu;
+ cpumask_t *mask = bcp->uvhub_master->cpumask;
+ struct bau_control *smaster = bcp->socket_master;
+ struct reset_args reset_args;
+
+ reset_args.sender = sender;
+ cpumask_clear(mask);
+ /* find a single cpu for each uvhub in this distribution mask */
+ maskbits = sizeof(struct pnmask) * BITSPERBYTE;
+ /* each bit is a pnode relative to the partition base pnode */
+ for (pnode = 0; pnode < maskbits; pnode++) {
+ int cpu;
+ if (!bau_uvhub_isset(pnode, distribution))
+ continue;
+ apnode = pnode + bcp->partition_base_pnode;
+ cpu = pnode_to_first_cpu(apnode, smaster);
+ cpumask_set_cpu(cpu, mask);
+ }
+
+ /* IPI all cpus; preemption is already disabled */
+ smp_call_function_many(mask, do_reset, (void *)&reset_args, 1);
+ return;
+}
+
+/*
+ * Not to be confused with cycles_2_ns() from tsc.c; this gives a relative
+ * number, not an absolute. It converts a duration in cycles to a duration in
+ * ns.
+ */
+static inline unsigned long long cycles_2_ns(unsigned long long cyc)
+{
+ struct cyc2ns_data data;
+ unsigned long long ns;
+
+ cyc2ns_read_begin(&data);
+ ns = mul_u64_u32_shr(cyc, data.cyc2ns_mul, data.cyc2ns_shift);
+ cyc2ns_read_end();
+
+ return ns;
+}
+
+/*
+ * The reverse of the above; converts a duration in ns to a duration in cycles.
+ */
+static inline unsigned long long ns_2_cycles(unsigned long long ns)
+{
+ struct cyc2ns_data data;
+ unsigned long long cyc;
+
+ cyc2ns_read_begin(&data);
+ cyc = (ns << data.cyc2ns_shift) / data.cyc2ns_mul;
+ cyc2ns_read_end();
+
+ return cyc;
+}
+
+static inline unsigned long cycles_2_us(unsigned long long cyc)
+{
+ return cycles_2_ns(cyc) / NSEC_PER_USEC;
+}
+
+static inline cycles_t sec_2_cycles(unsigned long sec)
+{
+ return ns_2_cycles(sec * NSEC_PER_SEC);
+}
+
+static inline unsigned long long usec_2_cycles(unsigned long usec)
+{
+ return ns_2_cycles(usec * NSEC_PER_USEC);
+}
+
+/*
+ * wait for all cpus on this hub to finish their sends and go quiet
+ * leaves uvhub_quiesce set so that no new broadcasts are started by
+ * bau_flush_send_and_wait()
+ */
+static inline void quiesce_local_uvhub(struct bau_control *hmaster)
+{
+ atom_asr(1, (struct atomic_short *)&hmaster->uvhub_quiesce);
+}
+
+/*
+ * mark this quiet-requestor as done
+ */
+static inline void end_uvhub_quiesce(struct bau_control *hmaster)
+{
+ atom_asr(-1, (struct atomic_short *)&hmaster->uvhub_quiesce);
+}
+
+static unsigned long uv1_read_status(unsigned long mmr_offset, int right_shift)
+{
+ unsigned long descriptor_status;
+
+ descriptor_status = uv_read_local_mmr(mmr_offset);
+ descriptor_status >>= right_shift;
+ descriptor_status &= UV_ACT_STATUS_MASK;
+ return descriptor_status;
+}
+
+/*
+ * Wait for completion of a broadcast software ack message
+ * return COMPLETE, RETRY(PLUGGED or TIMEOUT) or GIVEUP
+ */
+static int uv1_wait_completion(struct bau_desc *bau_desc,
+ struct bau_control *bcp, long try)
+{
+ unsigned long descriptor_status;
+ cycles_t ttm;
+ u64 mmr_offset = bcp->status_mmr;
+ int right_shift = bcp->status_index;
+ struct ptc_stats *stat = bcp->statp;
+
+ descriptor_status = uv1_read_status(mmr_offset, right_shift);
+ /* spin on the status MMR, waiting for it to go idle */
+ while ((descriptor_status != DS_IDLE)) {
+ /*
+ * Our software ack messages may be blocked because
+ * there are no swack resources available. As long
+ * as none of them has timed out hardware will NACK
+ * our message and its state will stay IDLE.
+ */
+ if (descriptor_status == DS_SOURCE_TIMEOUT) {
+ stat->s_stimeout++;
+ return FLUSH_GIVEUP;
+ } else if (descriptor_status == DS_DESTINATION_TIMEOUT) {
+ stat->s_dtimeout++;
+ ttm = get_cycles();
+
+ /*
+ * Our retries may be blocked by all destination
+ * swack resources being consumed, and a timeout
+ * pending. In that case hardware returns the
+ * ERROR that looks like a destination timeout.
+ */
+ if (cycles_2_us(ttm - bcp->send_message) < timeout_us) {
+ bcp->conseccompletes = 0;
+ return FLUSH_RETRY_PLUGGED;
+ }
+
+ bcp->conseccompletes = 0;
+ return FLUSH_RETRY_TIMEOUT;
+ } else {
+ /*
+ * descriptor_status is still BUSY
+ */
+ cpu_relax();
+ }
+ descriptor_status = uv1_read_status(mmr_offset, right_shift);
+ }
+ bcp->conseccompletes++;
+ return FLUSH_COMPLETE;
+}
+
+/*
+ * UV2 could have an extra bit of status in the ACTIVATION_STATUS_2 register.
+ * But not currently used.
+ */
+static unsigned long uv2_3_read_status(unsigned long offset, int rshft, int desc)
+{
+ return ((read_lmmr(offset) >> rshft) & UV_ACT_STATUS_MASK) << 1;
+}
+
+/*
+ * Entered when a bau descriptor has gone into a permanent busy wait because
+ * of a hardware bug.
+ * Workaround the bug.
+ */
+static int handle_uv2_busy(struct bau_control *bcp)
+{
+ struct ptc_stats *stat = bcp->statp;
+
+ stat->s_uv2_wars++;
+ bcp->busy = 1;
+ return FLUSH_GIVEUP;
+}
+
+static int uv2_3_wait_completion(struct bau_desc *bau_desc,
+ struct bau_control *bcp, long try)
+{
+ unsigned long descriptor_stat;
+ cycles_t ttm;
+ u64 mmr_offset = bcp->status_mmr;
+ int right_shift = bcp->status_index;
+ int desc = bcp->uvhub_cpu;
+ long busy_reps = 0;
+ struct ptc_stats *stat = bcp->statp;
+
+ descriptor_stat = uv2_3_read_status(mmr_offset, right_shift, desc);
+
+ /* spin on the status MMR, waiting for it to go idle */
+ while (descriptor_stat != UV2H_DESC_IDLE) {
+ if (descriptor_stat == UV2H_DESC_SOURCE_TIMEOUT) {
+ /*
+ * A h/w bug on the destination side may
+ * have prevented the message being marked
+ * pending, thus it doesn't get replied to
+ * and gets continually nacked until it times
+ * out with a SOURCE_TIMEOUT.
+ */
+ stat->s_stimeout++;
+ return FLUSH_GIVEUP;
+ } else if (descriptor_stat == UV2H_DESC_DEST_TIMEOUT) {
+ ttm = get_cycles();
+
+ /*
+ * Our retries may be blocked by all destination
+ * swack resources being consumed, and a timeout
+ * pending. In that case hardware returns the
+ * ERROR that looks like a destination timeout.
+ * Without using the extended status we have to
+ * deduce from the short time that this was a
+ * strong nack.
+ */
+ if (cycles_2_us(ttm - bcp->send_message) < timeout_us) {
+ bcp->conseccompletes = 0;
+ stat->s_plugged++;
+ /* FLUSH_RETRY_PLUGGED causes hang on boot */
+ return FLUSH_GIVEUP;
+ }
+ stat->s_dtimeout++;
+ bcp->conseccompletes = 0;
+ /* FLUSH_RETRY_TIMEOUT causes hang on boot */
+ return FLUSH_GIVEUP;
+ } else {
+ busy_reps++;
+ if (busy_reps > 1000000) {
+ /* not to hammer on the clock */
+ busy_reps = 0;
+ ttm = get_cycles();
+ if ((ttm - bcp->send_message) > bcp->timeout_interval)
+ return handle_uv2_busy(bcp);
+ }
+ /*
+ * descriptor_stat is still BUSY
+ */
+ cpu_relax();
+ }
+ descriptor_stat = uv2_3_read_status(mmr_offset, right_shift, desc);
+ }
+ bcp->conseccompletes++;
+ return FLUSH_COMPLETE;
+}
+
+/*
+ * Returns the status of current BAU message for cpu desc as a bit field
+ * [Error][Busy][Aux]
+ */
+static u64 read_status(u64 status_mmr, int index, int desc)
+{
+ u64 stat;
+
+ stat = ((read_lmmr(status_mmr) >> index) & UV_ACT_STATUS_MASK) << 1;
+ stat |= (read_lmmr(UVH_LB_BAU_SB_ACTIVATION_STATUS_2) >> desc) & 0x1;
+
+ return stat;
+}
+
+static int uv4_wait_completion(struct bau_desc *bau_desc,
+ struct bau_control *bcp, long try)
+{
+ struct ptc_stats *stat = bcp->statp;
+ u64 descriptor_stat;
+ u64 mmr = bcp->status_mmr;
+ int index = bcp->status_index;
+ int desc = bcp->uvhub_cpu;
+
+ descriptor_stat = read_status(mmr, index, desc);
+
+ /* spin on the status MMR, waiting for it to go idle */
+ while (descriptor_stat != UV2H_DESC_IDLE) {
+ switch (descriptor_stat) {
+ case UV2H_DESC_SOURCE_TIMEOUT:
+ stat->s_stimeout++;
+ return FLUSH_GIVEUP;
+
+ case UV2H_DESC_DEST_TIMEOUT:
+ stat->s_dtimeout++;
+ bcp->conseccompletes = 0;
+ return FLUSH_RETRY_TIMEOUT;
+
+ case UV2H_DESC_DEST_STRONG_NACK:
+ stat->s_plugged++;
+ bcp->conseccompletes = 0;
+ return FLUSH_RETRY_PLUGGED;
+
+ case UV2H_DESC_DEST_PUT_ERR:
+ bcp->conseccompletes = 0;
+ return FLUSH_GIVEUP;
+
+ default:
+ /* descriptor_stat is still BUSY */
+ cpu_relax();
+ }
+ descriptor_stat = read_status(mmr, index, desc);
+ }
+ bcp->conseccompletes++;
+ return FLUSH_COMPLETE;
+}
+
+/*
+ * Our retries are blocked by all destination sw ack resources being
+ * in use, and a timeout is pending. In that case hardware immediately
+ * returns the ERROR that looks like a destination timeout.
+ */
+static void destination_plugged(struct bau_desc *bau_desc,
+ struct bau_control *bcp,
+ struct bau_control *hmaster, struct ptc_stats *stat)
+{
+ udelay(bcp->plugged_delay);
+ bcp->plugged_tries++;
+
+ if (bcp->plugged_tries >= bcp->plugsb4reset) {
+ bcp->plugged_tries = 0;
+
+ quiesce_local_uvhub(hmaster);
+
+ spin_lock(&hmaster->queue_lock);
+ reset_with_ipi(&bau_desc->distribution, bcp);
+ spin_unlock(&hmaster->queue_lock);
+
+ end_uvhub_quiesce(hmaster);
+
+ bcp->ipi_attempts++;
+ stat->s_resets_plug++;
+ }
+}
+
+static void destination_timeout(struct bau_desc *bau_desc,
+ struct bau_control *bcp, struct bau_control *hmaster,
+ struct ptc_stats *stat)
+{
+ hmaster->max_concurr = 1;
+ bcp->timeout_tries++;
+ if (bcp->timeout_tries >= bcp->timeoutsb4reset) {
+ bcp->timeout_tries = 0;
+
+ quiesce_local_uvhub(hmaster);
+
+ spin_lock(&hmaster->queue_lock);
+ reset_with_ipi(&bau_desc->distribution, bcp);
+ spin_unlock(&hmaster->queue_lock);
+
+ end_uvhub_quiesce(hmaster);
+
+ bcp->ipi_attempts++;
+ stat->s_resets_timeout++;
+ }
+}
+
+/*
+ * Stop all cpus on a uvhub from using the BAU for a period of time.
+ * This is reversed by check_enable.
+ */
+static void disable_for_period(struct bau_control *bcp, struct ptc_stats *stat)
+{
+ int tcpu;
+ struct bau_control *tbcp;
+ struct bau_control *hmaster;
+ cycles_t tm1;
+
+ hmaster = bcp->uvhub_master;
+ spin_lock(&hmaster->disable_lock);
+ if (!bcp->baudisabled) {
+ stat->s_bau_disabled++;
+ tm1 = get_cycles();
+ for_each_present_cpu(tcpu) {
+ tbcp = &per_cpu(bau_control, tcpu);
+ if (tbcp->uvhub_master == hmaster) {
+ tbcp->baudisabled = 1;
+ tbcp->set_bau_on_time =
+ tm1 + bcp->disabled_period;
+ }
+ }
+ }
+ spin_unlock(&hmaster->disable_lock);
+}
+
+static void count_max_concurr(int stat, struct bau_control *bcp,
+ struct bau_control *hmaster)
+{
+ bcp->plugged_tries = 0;
+ bcp->timeout_tries = 0;
+ if (stat != FLUSH_COMPLETE)
+ return;
+ if (bcp->conseccompletes <= bcp->complete_threshold)
+ return;
+ if (hmaster->max_concurr >= hmaster->max_concurr_const)
+ return;
+ hmaster->max_concurr++;
+}
+
+static void record_send_stats(cycles_t time1, cycles_t time2,
+ struct bau_control *bcp, struct ptc_stats *stat,
+ int completion_status, int try)
+{
+ cycles_t elapsed;
+
+ if (time2 > time1) {
+ elapsed = time2 - time1;
+ stat->s_time += elapsed;
+
+ if ((completion_status == FLUSH_COMPLETE) && (try == 1)) {
+ bcp->period_requests++;
+ bcp->period_time += elapsed;
+ if ((elapsed > usec_2_cycles(bcp->cong_response_us)) &&
+ (bcp->period_requests > bcp->cong_reps) &&
+ ((bcp->period_time / bcp->period_requests) >
+ usec_2_cycles(bcp->cong_response_us))) {
+ stat->s_congested++;
+ disable_for_period(bcp, stat);
+ }
+ }
+ } else
+ stat->s_requestor--;
+
+ if (completion_status == FLUSH_COMPLETE && try > 1)
+ stat->s_retriesok++;
+ else if (completion_status == FLUSH_GIVEUP) {
+ stat->s_giveup++;
+ if (get_cycles() > bcp->period_end)
+ bcp->period_giveups = 0;
+ bcp->period_giveups++;
+ if (bcp->period_giveups == 1)
+ bcp->period_end = get_cycles() + bcp->disabled_period;
+ if (bcp->period_giveups > bcp->giveup_limit) {
+ disable_for_period(bcp, stat);
+ stat->s_giveuplimit++;
+ }
+ }
+}
+
+/*
+ * Because of a uv1 hardware bug only a limited number of concurrent
+ * requests can be made.
+ */
+static void uv1_throttle(struct bau_control *hmaster, struct ptc_stats *stat)
+{
+ spinlock_t *lock = &hmaster->uvhub_lock;
+ atomic_t *v;
+
+ v = &hmaster->active_descriptor_count;
+ if (!atomic_inc_unless_ge(lock, v, hmaster->max_concurr)) {
+ stat->s_throttles++;
+ do {
+ cpu_relax();
+ } while (!atomic_inc_unless_ge(lock, v, hmaster->max_concurr));
+ }
+}
+
+/*
+ * Handle the completion status of a message send.
+ */
+static void handle_cmplt(int completion_status, struct bau_desc *bau_desc,
+ struct bau_control *bcp, struct bau_control *hmaster,
+ struct ptc_stats *stat)
+{
+ if (completion_status == FLUSH_RETRY_PLUGGED)
+ destination_plugged(bau_desc, bcp, hmaster, stat);
+ else if (completion_status == FLUSH_RETRY_TIMEOUT)
+ destination_timeout(bau_desc, bcp, hmaster, stat);
+}
+
+/*
+ * Send a broadcast and wait for it to complete.
+ *
+ * The flush_mask contains the cpus the broadcast is to be sent to including
+ * cpus that are on the local uvhub.
+ *
+ * Returns 0 if all flushing represented in the mask was done.
+ * Returns 1 if it gives up entirely and the original cpu mask is to be
+ * returned to the kernel.
+ */
+static int uv_flush_send_and_wait(struct cpumask *flush_mask,
+ struct bau_control *bcp,
+ struct bau_desc *bau_desc)
+{
+ int seq_number = 0;
+ int completion_stat = 0;
+ int uv1 = 0;
+ long try = 0;
+ unsigned long index;
+ cycles_t time1;
+ cycles_t time2;
+ struct ptc_stats *stat = bcp->statp;
+ struct bau_control *hmaster = bcp->uvhub_master;
+ struct uv1_bau_msg_header *uv1_hdr = NULL;
+ struct uv2_3_bau_msg_header *uv2_3_hdr = NULL;
+
+ if (bcp->uvhub_version == UV_BAU_V1) {
+ uv1 = 1;
+ uv1_throttle(hmaster, stat);
+ }
+
+ while (hmaster->uvhub_quiesce)
+ cpu_relax();
+
+ time1 = get_cycles();
+ if (uv1)
+ uv1_hdr = &bau_desc->header.uv1_hdr;
+ else
+ /* uv2 and uv3 */
+ uv2_3_hdr = &bau_desc->header.uv2_3_hdr;
+
+ do {
+ if (try == 0) {
+ if (uv1)
+ uv1_hdr->msg_type = MSG_REGULAR;
+ else
+ uv2_3_hdr->msg_type = MSG_REGULAR;
+ seq_number = bcp->message_number++;
+ } else {
+ if (uv1)
+ uv1_hdr->msg_type = MSG_RETRY;
+ else
+ uv2_3_hdr->msg_type = MSG_RETRY;
+ stat->s_retry_messages++;
+ }
+
+ if (uv1)
+ uv1_hdr->sequence = seq_number;
+ else
+ uv2_3_hdr->sequence = seq_number;
+ index = (1UL << AS_PUSH_SHIFT) | bcp->uvhub_cpu;
+ bcp->send_message = get_cycles();
+
+ write_mmr_activation(index);
+
+ try++;
+ completion_stat = ops.wait_completion(bau_desc, bcp, try);
+
+ handle_cmplt(completion_stat, bau_desc, bcp, hmaster, stat);
+
+ if (bcp->ipi_attempts >= bcp->ipi_reset_limit) {
+ bcp->ipi_attempts = 0;
+ stat->s_overipilimit++;
+ completion_stat = FLUSH_GIVEUP;
+ break;
+ }
+ cpu_relax();
+ } while ((completion_stat == FLUSH_RETRY_PLUGGED) ||
+ (completion_stat == FLUSH_RETRY_TIMEOUT));
+
+ time2 = get_cycles();
+
+ count_max_concurr(completion_stat, bcp, hmaster);
+
+ while (hmaster->uvhub_quiesce)
+ cpu_relax();
+
+ atomic_dec(&hmaster->active_descriptor_count);
+
+ record_send_stats(time1, time2, bcp, stat, completion_stat, try);
+
+ if (completion_stat == FLUSH_GIVEUP)
+ /* FLUSH_GIVEUP will fall back to using IPI's for tlb flush */
+ return 1;
+ return 0;
+}
+
+/*
+ * The BAU is disabled for this uvhub. When the disabled time period has
+ * expired re-enable it.
+ * Return 0 if it is re-enabled for all cpus on this uvhub.
+ */
+static int check_enable(struct bau_control *bcp, struct ptc_stats *stat)
+{
+ int tcpu;
+ struct bau_control *tbcp;
+ struct bau_control *hmaster;
+
+ hmaster = bcp->uvhub_master;
+ spin_lock(&hmaster->disable_lock);
+ if (bcp->baudisabled && (get_cycles() >= bcp->set_bau_on_time)) {
+ stat->s_bau_reenabled++;
+ for_each_present_cpu(tcpu) {
+ tbcp = &per_cpu(bau_control, tcpu);
+ if (tbcp->uvhub_master == hmaster) {
+ tbcp->baudisabled = 0;
+ tbcp->period_requests = 0;
+ tbcp->period_time = 0;
+ tbcp->period_giveups = 0;
+ }
+ }
+ spin_unlock(&hmaster->disable_lock);
+ return 0;
+ }
+ spin_unlock(&hmaster->disable_lock);
+ return -1;
+}
+
+static void record_send_statistics(struct ptc_stats *stat, int locals, int hubs,
+ int remotes, struct bau_desc *bau_desc)
+{
+ stat->s_requestor++;
+ stat->s_ntargcpu += remotes + locals;
+ stat->s_ntargremotes += remotes;
+ stat->s_ntarglocals += locals;
+
+ /* uvhub statistics */
+ hubs = bau_uvhub_weight(&bau_desc->distribution);
+ if (locals) {
+ stat->s_ntarglocaluvhub++;
+ stat->s_ntargremoteuvhub += (hubs - 1);
+ } else
+ stat->s_ntargremoteuvhub += hubs;
+
+ stat->s_ntarguvhub += hubs;
+
+ if (hubs >= 16)
+ stat->s_ntarguvhub16++;
+ else if (hubs >= 8)
+ stat->s_ntarguvhub8++;
+ else if (hubs >= 4)
+ stat->s_ntarguvhub4++;
+ else if (hubs >= 2)
+ stat->s_ntarguvhub2++;
+ else
+ stat->s_ntarguvhub1++;
+}
+
+/*
+ * Translate a cpu mask to the uvhub distribution mask in the BAU
+ * activation descriptor.
+ */
+static int set_distrib_bits(struct cpumask *flush_mask, struct bau_control *bcp,
+ struct bau_desc *bau_desc, int *localsp, int *remotesp)
+{
+ int cpu;
+ int pnode;
+ int cnt = 0;
+ struct hub_and_pnode *hpp;
+
+ for_each_cpu(cpu, flush_mask) {
+ /*
+ * The distribution vector is a bit map of pnodes, relative
+ * to the partition base pnode (and the partition base nasid
+ * in the header).
+ * Translate cpu to pnode and hub using a local memory array.
+ */
+ hpp = &bcp->socket_master->thp[cpu];
+ pnode = hpp->pnode - bcp->partition_base_pnode;
+ bau_uvhub_set(pnode, &bau_desc->distribution);
+ cnt++;
+ if (hpp->uvhub == bcp->uvhub)
+ (*localsp)++;
+ else
+ (*remotesp)++;
+ }
+ if (!cnt)
+ return 1;
+ return 0;
+}
+
+/*
+ * globally purge translation cache of a virtual address or all TLB's
+ * @cpumask: mask of all cpu's in which the address is to be removed
+ * @mm: mm_struct containing virtual address range
+ * @start: start virtual address to be removed from TLB
+ * @end: end virtual address to be remove from TLB
+ * @cpu: the current cpu
+ *
+ * This is the entry point for initiating any UV global TLB shootdown.
+ *
+ * Purges the translation caches of all specified processors of the given
+ * virtual address, or purges all TLB's on specified processors.
+ *
+ * The caller has derived the cpumask from the mm_struct. This function
+ * is called only if there are bits set in the mask. (e.g. flush_tlb_page())
+ *
+ * The cpumask is converted into a uvhubmask of the uvhubs containing
+ * those cpus.
+ *
+ * Note that this function should be called with preemption disabled.
+ *
+ * Returns NULL if all remote flushing was done.
+ * Returns pointer to cpumask if some remote flushing remains to be
+ * done. The returned pointer is valid till preemption is re-enabled.
+ */
+const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask,
+ const struct flush_tlb_info *info)
+{
+ unsigned int cpu = smp_processor_id();
+ int locals = 0, remotes = 0, hubs = 0;
+ struct bau_desc *bau_desc;
+ struct cpumask *flush_mask;
+ struct ptc_stats *stat;
+ struct bau_control *bcp;
+ unsigned long descriptor_status, status, address;
+
+ bcp = &per_cpu(bau_control, cpu);
+
+ if (bcp->nobau)
+ return cpumask;
+
+ stat = bcp->statp;
+ stat->s_enters++;
+
+ if (bcp->busy) {
+ descriptor_status =
+ read_lmmr(UVH_LB_BAU_SB_ACTIVATION_STATUS_0);
+ status = ((descriptor_status >> (bcp->uvhub_cpu *
+ UV_ACT_STATUS_SIZE)) & UV_ACT_STATUS_MASK) << 1;
+ if (status == UV2H_DESC_BUSY)
+ return cpumask;
+ bcp->busy = 0;
+ }
+
+ /* bau was disabled due to slow response */
+ if (bcp->baudisabled) {
+ if (check_enable(bcp, stat)) {
+ stat->s_ipifordisabled++;
+ return cpumask;
+ }
+ }
+
+ /*
+ * Each sending cpu has a per-cpu mask which it fills from the caller's
+ * cpu mask. All cpus are converted to uvhubs and copied to the
+ * activation descriptor.
+ */
+ flush_mask = (struct cpumask *)per_cpu(uv_flush_tlb_mask, cpu);
+ /* don't actually do a shootdown of the local cpu */
+ cpumask_andnot(flush_mask, cpumask, cpumask_of(cpu));
+
+ if (cpumask_test_cpu(cpu, cpumask))
+ stat->s_ntargself++;
+
+ bau_desc = bcp->descriptor_base;
+ bau_desc += (ITEMS_PER_DESC * bcp->uvhub_cpu);
+ bau_uvhubs_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE);
+ if (set_distrib_bits(flush_mask, bcp, bau_desc, &locals, &remotes))
+ return NULL;
+
+ record_send_statistics(stat, locals, hubs, remotes, bau_desc);
+
+ if (!info->end || (info->end - info->start) <= PAGE_SIZE)
+ address = info->start;
+ else
+ address = TLB_FLUSH_ALL;
+
+ switch (bcp->uvhub_version) {
+ case UV_BAU_V1:
+ case UV_BAU_V2:
+ case UV_BAU_V3:
+ bau_desc->payload.uv1_2_3.address = address;
+ bau_desc->payload.uv1_2_3.sending_cpu = cpu;
+ break;
+ case UV_BAU_V4:
+ bau_desc->payload.uv4.address = address;
+ bau_desc->payload.uv4.sending_cpu = cpu;
+ bau_desc->payload.uv4.qualifier = BAU_DESC_QUALIFIER;
+ break;
+ }
+
+ /*
+ * uv_flush_send_and_wait returns 0 if all cpu's were messaged,
+ * or 1 if it gave up and the original cpumask should be returned.
+ */
+ if (!uv_flush_send_and_wait(flush_mask, bcp, bau_desc))
+ return NULL;
+ else
+ return cpumask;
+}
+
+/*
+ * Search the message queue for any 'other' unprocessed message with the
+ * same software acknowledge resource bit vector as the 'msg' message.
+ */
+static struct bau_pq_entry *find_another_by_swack(struct bau_pq_entry *msg,
+ struct bau_control *bcp)
+{
+ struct bau_pq_entry *msg_next = msg + 1;
+ unsigned char swack_vec = msg->swack_vec;
+
+ if (msg_next > bcp->queue_last)
+ msg_next = bcp->queue_first;
+ while (msg_next != msg) {
+ if ((msg_next->canceled == 0) && (msg_next->replied_to == 0) &&
+ (msg_next->swack_vec == swack_vec))
+ return msg_next;
+ msg_next++;
+ if (msg_next > bcp->queue_last)
+ msg_next = bcp->queue_first;
+ }
+ return NULL;
+}
+
+/*
+ * UV2 needs to work around a bug in which an arriving message has not
+ * set a bit in the UVH_LB_BAU_INTD_SOFTWARE_ACKNOWLEDGE register.
+ * Such a message must be ignored.
+ */
+static void process_uv2_message(struct msg_desc *mdp, struct bau_control *bcp)
+{
+ unsigned long mmr_image;
+ unsigned char swack_vec;
+ struct bau_pq_entry *msg = mdp->msg;
+ struct bau_pq_entry *other_msg;
+
+ mmr_image = ops.read_l_sw_ack();
+ swack_vec = msg->swack_vec;
+
+ if ((swack_vec & mmr_image) == 0) {
+ /*
+ * This message was assigned a swack resource, but no
+ * reserved acknowlegment is pending.
+ * The bug has prevented this message from setting the MMR.
+ */
+ /*
+ * Some message has set the MMR 'pending' bit; it might have
+ * been another message. Look for that message.
+ */
+ other_msg = find_another_by_swack(msg, bcp);
+ if (other_msg) {
+ /*
+ * There is another. Process this one but do not
+ * ack it.
+ */
+ bau_process_message(mdp, bcp, 0);
+ /*
+ * Let the natural processing of that other message
+ * acknowledge it. Don't get the processing of sw_ack's
+ * out of order.
+ */
+ return;
+ }
+ }
+
+ /*
+ * Either the MMR shows this one pending a reply or there is no
+ * other message using this sw_ack, so it is safe to acknowledge it.
+ */
+ bau_process_message(mdp, bcp, 1);
+
+ return;
+}
+
+/*
+ * The BAU message interrupt comes here. (registered by set_intr_gate)
+ * See entry_64.S
+ *
+ * We received a broadcast assist message.
+ *
+ * Interrupts are disabled; this interrupt could represent
+ * the receipt of several messages.
+ *
+ * All cores/threads on this hub get this interrupt.
+ * The last one to see it does the software ack.
+ * (the resource will not be freed until noninterruptable cpus see this
+ * interrupt; hardware may timeout the s/w ack and reply ERROR)
+ */
+void uv_bau_message_interrupt(struct pt_regs *regs)
+{
+ int count = 0;
+ cycles_t time_start;
+ struct bau_pq_entry *msg;
+ struct bau_control *bcp;
+ struct ptc_stats *stat;
+ struct msg_desc msgdesc;
+
+ ack_APIC_irq();
+ kvm_set_cpu_l1tf_flush_l1d();
+ time_start = get_cycles();
+
+ bcp = &per_cpu(bau_control, smp_processor_id());
+ stat = bcp->statp;
+
+ msgdesc.queue_first = bcp->queue_first;
+ msgdesc.queue_last = bcp->queue_last;
+
+ msg = bcp->bau_msg_head;
+ while (msg->swack_vec) {
+ count++;
+
+ msgdesc.msg_slot = msg - msgdesc.queue_first;
+ msgdesc.msg = msg;
+ if (bcp->uvhub_version == UV_BAU_V2)
+ process_uv2_message(&msgdesc, bcp);
+ else
+ /* no error workaround for uv1 or uv3 */
+ bau_process_message(&msgdesc, bcp, 1);
+
+ msg++;
+ if (msg > msgdesc.queue_last)
+ msg = msgdesc.queue_first;
+ bcp->bau_msg_head = msg;
+ }
+ stat->d_time += (get_cycles() - time_start);
+ if (!count)
+ stat->d_nomsg++;
+ else if (count > 1)
+ stat->d_multmsg++;
+}
+
+/*
+ * Each target uvhub (i.e. a uvhub that has cpu's) needs to have
+ * shootdown message timeouts enabled. The timeout does not cause
+ * an interrupt, but causes an error message to be returned to
+ * the sender.
+ */
+static void __init enable_timeouts(void)
+{
+ int uvhub;
+ int nuvhubs;
+ int pnode;
+ unsigned long mmr_image;
+
+ nuvhubs = uv_num_possible_blades();
+
+ for (uvhub = 0; uvhub < nuvhubs; uvhub++) {
+ if (!uv_blade_nr_possible_cpus(uvhub))
+ continue;
+
+ pnode = uv_blade_to_pnode(uvhub);
+ mmr_image = read_mmr_misc_control(pnode);
+ /*
+ * Set the timeout period and then lock it in, in three
+ * steps; captures and locks in the period.
+ *
+ * To program the period, the SOFT_ACK_MODE must be off.
+ */
+ mmr_image &= ~(1L << SOFTACK_MSHIFT);
+ write_mmr_misc_control(pnode, mmr_image);
+ /*
+ * Set the 4-bit period.
+ */
+ mmr_image &= ~((unsigned long)0xf << SOFTACK_PSHIFT);
+ mmr_image |= (SOFTACK_TIMEOUT_PERIOD << SOFTACK_PSHIFT);
+ write_mmr_misc_control(pnode, mmr_image);
+ /*
+ * UV1:
+ * Subsequent reversals of the timebase bit (3) cause an
+ * immediate timeout of one or all INTD resources as
+ * indicated in bits 2:0 (7 causes all of them to timeout).
+ */
+ mmr_image |= (1L << SOFTACK_MSHIFT);
+ if (is_uv2_hub()) {
+ /* do not touch the legacy mode bit */
+ /* hw bug workaround; do not use extended status */
+ mmr_image &= ~(1L << UV2_EXT_SHFT);
+ } else if (is_uv3_hub()) {
+ mmr_image &= ~(1L << PREFETCH_HINT_SHFT);
+ mmr_image |= (1L << SB_STATUS_SHFT);
+ }
+ write_mmr_misc_control(pnode, mmr_image);
+ }
+}
+
+static void *ptc_seq_start(struct seq_file *file, loff_t *offset)
+{
+ if (*offset < num_possible_cpus())
+ return offset;
+ return NULL;
+}
+
+static void *ptc_seq_next(struct seq_file *file, void *data, loff_t *offset)
+{
+ (*offset)++;
+ if (*offset < num_possible_cpus())
+ return offset;
+ return NULL;
+}
+
+static void ptc_seq_stop(struct seq_file *file, void *data)
+{
+}
+
+/*
+ * Display the statistics thru /proc/sgi_uv/ptc_statistics
+ * 'data' points to the cpu number
+ * Note: see the descriptions in stat_description[].
+ */
+static int ptc_seq_show(struct seq_file *file, void *data)
+{
+ struct ptc_stats *stat;
+ struct bau_control *bcp;
+ int cpu;
+
+ cpu = *(loff_t *)data;
+ if (!cpu) {
+ seq_puts(file,
+ "# cpu bauoff sent stime self locals remotes ncpus localhub ");
+ seq_puts(file, "remotehub numuvhubs numuvhubs16 numuvhubs8 ");
+ seq_puts(file,
+ "numuvhubs4 numuvhubs2 numuvhubs1 dto snacks retries ");
+ seq_puts(file,
+ "rok resetp resett giveup sto bz throt disable ");
+ seq_puts(file,
+ "enable wars warshw warwaits enters ipidis plugged ");
+ seq_puts(file,
+ "ipiover glim cong swack recv rtime all one mult ");
+ seq_puts(file, "none retry canc nocan reset rcan\n");
+ }
+ if (cpu < num_possible_cpus() && cpu_online(cpu)) {
+ bcp = &per_cpu(bau_control, cpu);
+ if (bcp->nobau) {
+ seq_printf(file, "cpu %d bau disabled\n", cpu);
+ return 0;
+ }
+ stat = bcp->statp;
+ /* source side statistics */
+ seq_printf(file,
+ "cpu %d %d %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld ",
+ cpu, bcp->nobau, stat->s_requestor,
+ cycles_2_us(stat->s_time),
+ stat->s_ntargself, stat->s_ntarglocals,
+ stat->s_ntargremotes, stat->s_ntargcpu,
+ stat->s_ntarglocaluvhub, stat->s_ntargremoteuvhub,
+ stat->s_ntarguvhub, stat->s_ntarguvhub16);
+ seq_printf(file, "%ld %ld %ld %ld %ld %ld ",
+ stat->s_ntarguvhub8, stat->s_ntarguvhub4,
+ stat->s_ntarguvhub2, stat->s_ntarguvhub1,
+ stat->s_dtimeout, stat->s_strongnacks);
+ seq_printf(file, "%ld %ld %ld %ld %ld %ld %ld %ld ",
+ stat->s_retry_messages, stat->s_retriesok,
+ stat->s_resets_plug, stat->s_resets_timeout,
+ stat->s_giveup, stat->s_stimeout,
+ stat->s_busy, stat->s_throttles);
+ seq_printf(file, "%ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld ",
+ stat->s_bau_disabled, stat->s_bau_reenabled,
+ stat->s_uv2_wars, stat->s_uv2_wars_hw,
+ stat->s_uv2_war_waits, stat->s_enters,
+ stat->s_ipifordisabled, stat->s_plugged,
+ stat->s_overipilimit, stat->s_giveuplimit,
+ stat->s_congested);
+
+ /* destination side statistics */
+ seq_printf(file,
+ "%lx %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld %ld\n",
+ ops.read_g_sw_ack(uv_cpu_to_pnode(cpu)),
+ stat->d_requestee, cycles_2_us(stat->d_time),
+ stat->d_alltlb, stat->d_onetlb, stat->d_multmsg,
+ stat->d_nomsg, stat->d_retries, stat->d_canceled,
+ stat->d_nocanceled, stat->d_resets,
+ stat->d_rcanceled);
+ }
+ return 0;
+}
+
+/*
+ * Display the tunables thru debugfs
+ */
+static ssize_t tunables_read(struct file *file, char __user *userbuf,
+ size_t count, loff_t *ppos)
+{
+ char *buf;
+ int ret;
+
+ buf = kasprintf(GFP_KERNEL, "%s %s %s\n%d %d %d %d %d %d %d %d %d %d\n",
+ "max_concur plugged_delay plugsb4reset timeoutsb4reset",
+ "ipi_reset_limit complete_threshold congested_response_us",
+ "congested_reps disabled_period giveup_limit",
+ max_concurr, plugged_delay, plugsb4reset,
+ timeoutsb4reset, ipi_reset_limit, complete_threshold,
+ congested_respns_us, congested_reps, disabled_period,
+ giveup_limit);
+
+ if (!buf)
+ return -ENOMEM;
+
+ ret = simple_read_from_buffer(userbuf, count, ppos, buf, strlen(buf));
+ kfree(buf);
+ return ret;
+}
+
+/*
+ * handle a write to /proc/sgi_uv/ptc_statistics
+ * -1: reset the statistics
+ * 0: display meaning of the statistics
+ */
+static ssize_t ptc_proc_write(struct file *file, const char __user *user,
+ size_t count, loff_t *data)
+{
+ int cpu;
+ int i;
+ int elements;
+ long input_arg;
+ char optstr[64];
+ struct ptc_stats *stat;
+
+ if (count == 0 || count > sizeof(optstr))
+ return -EINVAL;
+ if (copy_from_user(optstr, user, count))
+ return -EFAULT;
+ optstr[count - 1] = '\0';
+
+ if (!strcmp(optstr, "on")) {
+ set_bau_on();
+ return count;
+ } else if (!strcmp(optstr, "off")) {
+ set_bau_off();
+ return count;
+ }
+
+ if (kstrtol(optstr, 10, &input_arg) < 0) {
+ pr_debug("%s is invalid\n", optstr);
+ return -EINVAL;
+ }
+
+ if (input_arg == 0) {
+ elements = ARRAY_SIZE(stat_description);
+ pr_debug("# cpu: cpu number\n");
+ pr_debug("Sender statistics:\n");
+ for (i = 0; i < elements; i++)
+ pr_debug("%s\n", stat_description[i]);
+ } else if (input_arg == -1) {
+ for_each_present_cpu(cpu) {
+ stat = &per_cpu(ptcstats, cpu);
+ memset(stat, 0, sizeof(struct ptc_stats));
+ }
+ }
+
+ return count;
+}
+
+static int local_atoi(const char *name)
+{
+ int val = 0;
+
+ for (;; name++) {
+ switch (*name) {
+ case '0' ... '9':
+ val = 10*val+(*name-'0');
+ break;
+ default:
+ return val;
+ }
+ }
+}
+
+/*
+ * Parse the values written to /sys/kernel/debug/sgi_uv/bau_tunables.
+ * Zero values reset them to defaults.
+ */
+static int parse_tunables_write(struct bau_control *bcp, char *instr,
+ int count)
+{
+ char *p;
+ char *q;
+ int cnt = 0;
+ int val;
+ int e = ARRAY_SIZE(tunables);
+
+ p = instr + strspn(instr, WHITESPACE);
+ q = p;
+ for (; *p; p = q + strspn(q, WHITESPACE)) {
+ q = p + strcspn(p, WHITESPACE);
+ cnt++;
+ if (q == p)
+ break;
+ }
+ if (cnt != e) {
+ pr_info("bau tunable error: should be %d values\n", e);
+ return -EINVAL;
+ }
+
+ p = instr + strspn(instr, WHITESPACE);
+ q = p;
+ for (cnt = 0; *p; p = q + strspn(q, WHITESPACE), cnt++) {
+ q = p + strcspn(p, WHITESPACE);
+ val = local_atoi(p);
+ switch (cnt) {
+ case 0:
+ if (val == 0) {
+ max_concurr = MAX_BAU_CONCURRENT;
+ max_concurr_const = MAX_BAU_CONCURRENT;
+ continue;
+ }
+ if (val < 1 || val > bcp->cpus_in_uvhub) {
+ pr_debug(
+ "Error: BAU max concurrent %d is invalid\n",
+ val);
+ return -EINVAL;
+ }
+ max_concurr = val;
+ max_concurr_const = val;
+ continue;
+ default:
+ if (val == 0)
+ *tunables[cnt].tunp = tunables[cnt].deflt;
+ else
+ *tunables[cnt].tunp = val;
+ continue;
+ }
+ }
+ return 0;
+}
+
+/*
+ * Handle a write to debugfs. (/sys/kernel/debug/sgi_uv/bau_tunables)
+ */
+static ssize_t tunables_write(struct file *file, const char __user *user,
+ size_t count, loff_t *data)
+{
+ int cpu;
+ int ret;
+ char instr[100];
+ struct bau_control *bcp;
+
+ if (count == 0 || count > sizeof(instr)-1)
+ return -EINVAL;
+ if (copy_from_user(instr, user, count))
+ return -EFAULT;
+
+ instr[count] = '\0';
+
+ cpu = get_cpu();
+ bcp = &per_cpu(bau_control, cpu);
+ ret = parse_tunables_write(bcp, instr, count);
+ put_cpu();
+ if (ret)
+ return ret;
+
+ for_each_present_cpu(cpu) {
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->max_concurr = max_concurr;
+ bcp->max_concurr_const = max_concurr;
+ bcp->plugged_delay = plugged_delay;
+ bcp->plugsb4reset = plugsb4reset;
+ bcp->timeoutsb4reset = timeoutsb4reset;
+ bcp->ipi_reset_limit = ipi_reset_limit;
+ bcp->complete_threshold = complete_threshold;
+ bcp->cong_response_us = congested_respns_us;
+ bcp->cong_reps = congested_reps;
+ bcp->disabled_period = sec_2_cycles(disabled_period);
+ bcp->giveup_limit = giveup_limit;
+ }
+ return count;
+}
+
+static const struct seq_operations uv_ptc_seq_ops = {
+ .start = ptc_seq_start,
+ .next = ptc_seq_next,
+ .stop = ptc_seq_stop,
+ .show = ptc_seq_show
+};
+
+static int ptc_proc_open(struct inode *inode, struct file *file)
+{
+ return seq_open(file, &uv_ptc_seq_ops);
+}
+
+static int tunables_open(struct inode *inode, struct file *file)
+{
+ return 0;
+}
+
+static const struct file_operations proc_uv_ptc_operations = {
+ .open = ptc_proc_open,
+ .read = seq_read,
+ .write = ptc_proc_write,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+static const struct file_operations tunables_fops = {
+ .open = tunables_open,
+ .read = tunables_read,
+ .write = tunables_write,
+ .llseek = default_llseek,
+};
+
+static int __init uv_ptc_init(void)
+{
+ struct proc_dir_entry *proc_uv_ptc;
+
+ if (!is_uv_system())
+ return 0;
+
+ proc_uv_ptc = proc_create(UV_PTC_BASENAME, 0444, NULL,
+ &proc_uv_ptc_operations);
+ if (!proc_uv_ptc) {
+ pr_err("unable to create %s proc entry\n",
+ UV_PTC_BASENAME);
+ return -EINVAL;
+ }
+
+ tunables_dir = debugfs_create_dir(UV_BAU_TUNABLES_DIR, NULL);
+ if (!tunables_dir) {
+ pr_err("unable to create debugfs directory %s\n",
+ UV_BAU_TUNABLES_DIR);
+ return -EINVAL;
+ }
+ tunables_file = debugfs_create_file(UV_BAU_TUNABLES_FILE, 0600,
+ tunables_dir, NULL, &tunables_fops);
+ if (!tunables_file) {
+ pr_err("unable to create debugfs file %s\n",
+ UV_BAU_TUNABLES_FILE);
+ return -EINVAL;
+ }
+ return 0;
+}
+
+/*
+ * Initialize the sending side's sending buffers.
+ */
+static void activation_descriptor_init(int node, int pnode, int base_pnode)
+{
+ int i;
+ int cpu;
+ int uv1 = 0;
+ unsigned long gpa;
+ unsigned long m;
+ unsigned long n;
+ size_t dsize;
+ struct bau_desc *bau_desc;
+ struct bau_desc *bd2;
+ struct uv1_bau_msg_header *uv1_hdr;
+ struct uv2_3_bau_msg_header *uv2_3_hdr;
+ struct bau_control *bcp;
+
+ /*
+ * each bau_desc is 64 bytes; there are 8 (ITEMS_PER_DESC)
+ * per cpu; and one per cpu on the uvhub (ADP_SZ)
+ */
+ dsize = sizeof(struct bau_desc) * ADP_SZ * ITEMS_PER_DESC;
+ bau_desc = kmalloc_node(dsize, GFP_KERNEL, node);
+ BUG_ON(!bau_desc);
+
+ gpa = uv_gpa(bau_desc);
+ n = uv_gpa_to_gnode(gpa);
+ m = ops.bau_gpa_to_offset(gpa);
+ if (is_uv1_hub())
+ uv1 = 1;
+
+ /* the 14-bit pnode */
+ write_mmr_descriptor_base(pnode,
+ (n << UVH_LB_BAU_SB_DESCRIPTOR_BASE_NODE_ID_SHFT | m));
+ /*
+ * Initializing all 8 (ITEMS_PER_DESC) descriptors for each
+ * cpu even though we only use the first one; one descriptor can
+ * describe a broadcast to 256 uv hubs.
+ */
+ for (i = 0, bd2 = bau_desc; i < (ADP_SZ * ITEMS_PER_DESC); i++, bd2++) {
+ memset(bd2, 0, sizeof(struct bau_desc));
+ if (uv1) {
+ uv1_hdr = &bd2->header.uv1_hdr;
+ uv1_hdr->swack_flag = 1;
+ /*
+ * The base_dest_nasid set in the message header
+ * is the nasid of the first uvhub in the partition.
+ * The bit map will indicate destination pnode numbers
+ * relative to that base. They may not be consecutive
+ * if nasid striding is being used.
+ */
+ uv1_hdr->base_dest_nasid =
+ UV_PNODE_TO_NASID(base_pnode);
+ uv1_hdr->dest_subnodeid = UV_LB_SUBNODEID;
+ uv1_hdr->command = UV_NET_ENDPOINT_INTD;
+ uv1_hdr->int_both = 1;
+ /*
+ * all others need to be set to zero:
+ * fairness chaining multilevel count replied_to
+ */
+ } else {
+ /*
+ * BIOS uses legacy mode, but uv2 and uv3 hardware always
+ * uses native mode for selective broadcasts.
+ */
+ uv2_3_hdr = &bd2->header.uv2_3_hdr;
+ uv2_3_hdr->swack_flag = 1;
+ uv2_3_hdr->base_dest_nasid =
+ UV_PNODE_TO_NASID(base_pnode);
+ uv2_3_hdr->dest_subnodeid = UV_LB_SUBNODEID;
+ uv2_3_hdr->command = UV_NET_ENDPOINT_INTD;
+ }
+ }
+ for_each_present_cpu(cpu) {
+ if (pnode != uv_blade_to_pnode(uv_cpu_to_blade_id(cpu)))
+ continue;
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->descriptor_base = bau_desc;
+ }
+}
+
+/*
+ * initialize the destination side's receiving buffers
+ * entered for each uvhub in the partition
+ * - node is first node (kernel memory notion) on the uvhub
+ * - pnode is the uvhub's physical identifier
+ */
+static void pq_init(int node, int pnode)
+{
+ int cpu;
+ size_t plsize;
+ char *cp;
+ void *vp;
+ unsigned long gnode, first, last, tail;
+ struct bau_pq_entry *pqp;
+ struct bau_control *bcp;
+
+ plsize = (DEST_Q_SIZE + 1) * sizeof(struct bau_pq_entry);
+ vp = kmalloc_node(plsize, GFP_KERNEL, node);
+ pqp = (struct bau_pq_entry *)vp;
+ BUG_ON(!pqp);
+
+ cp = (char *)pqp + 31;
+ pqp = (struct bau_pq_entry *)(((unsigned long)cp >> 5) << 5);
+
+ for_each_present_cpu(cpu) {
+ if (pnode != uv_cpu_to_pnode(cpu))
+ continue;
+ /* for every cpu on this pnode: */
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->queue_first = pqp;
+ bcp->bau_msg_head = pqp;
+ bcp->queue_last = pqp + (DEST_Q_SIZE - 1);
+ }
+
+ first = ops.bau_gpa_to_offset(uv_gpa(pqp));
+ last = ops.bau_gpa_to_offset(uv_gpa(pqp + (DEST_Q_SIZE - 1)));
+
+ /*
+ * Pre UV4, the gnode is required to locate the payload queue
+ * and the payload queue tail must be maintained by the kernel.
+ */
+ bcp = &per_cpu(bau_control, smp_processor_id());
+ if (bcp->uvhub_version <= UV_BAU_V3) {
+ tail = first;
+ gnode = uv_gpa_to_gnode(uv_gpa(pqp));
+ first = (gnode << UV_PAYLOADQ_GNODE_SHIFT) | tail;
+ write_mmr_payload_tail(pnode, tail);
+ }
+
+ ops.write_payload_first(pnode, first);
+ ops.write_payload_last(pnode, last);
+
+ /* in effect, all msg_type's are set to MSG_NOOP */
+ memset(pqp, 0, sizeof(struct bau_pq_entry) * DEST_Q_SIZE);
+}
+
+/*
+ * Initialization of each UV hub's structures
+ */
+static void __init init_uvhub(int uvhub, int vector, int base_pnode)
+{
+ int node;
+ int pnode;
+ unsigned long apicid;
+
+ node = uvhub_to_first_node(uvhub);
+ pnode = uv_blade_to_pnode(uvhub);
+
+ activation_descriptor_init(node, pnode, base_pnode);
+
+ pq_init(node, pnode);
+ /*
+ * The below initialization can't be in firmware because the
+ * messaging IRQ will be determined by the OS.
+ */
+ apicid = uvhub_to_first_apicid(uvhub) | uv_apicid_hibits;
+ write_mmr_data_config(pnode, ((apicid << 32) | vector));
+}
+
+/*
+ * We will set BAU_MISC_CONTROL with a timeout period.
+ * But the BIOS has set UVH_AGING_PRESCALE_SEL and UVH_TRANSACTION_TIMEOUT.
+ * So the destination timeout period has to be calculated from them.
+ */
+static int calculate_destination_timeout(void)
+{
+ unsigned long mmr_image;
+ int mult1;
+ int mult2;
+ int index;
+ int base;
+ int ret;
+ unsigned long ts_ns;
+
+ if (is_uv1_hub()) {
+ mult1 = SOFTACK_TIMEOUT_PERIOD & BAU_MISC_CONTROL_MULT_MASK;
+ mmr_image = uv_read_local_mmr(UVH_AGING_PRESCALE_SEL);
+ index = (mmr_image >> BAU_URGENCY_7_SHIFT) & BAU_URGENCY_7_MASK;
+ mmr_image = uv_read_local_mmr(UVH_TRANSACTION_TIMEOUT);
+ mult2 = (mmr_image >> BAU_TRANS_SHIFT) & BAU_TRANS_MASK;
+ ts_ns = timeout_base_ns[index];
+ ts_ns *= (mult1 * mult2);
+ ret = ts_ns / 1000;
+ } else {
+ /* same destination timeout for uv2 and uv3 */
+ /* 4 bits 0/1 for 10/80us base, 3 bits of multiplier */
+ mmr_image = uv_read_local_mmr(UVH_LB_BAU_MISC_CONTROL);
+ mmr_image = (mmr_image & UV_SA_MASK) >> UV_SA_SHFT;
+ if (mmr_image & (1L << UV2_ACK_UNITS_SHFT))
+ base = 80;
+ else
+ base = 10;
+ mult1 = mmr_image & UV2_ACK_MASK;
+ ret = mult1 * base;
+ }
+ return ret;
+}
+
+static void __init init_per_cpu_tunables(void)
+{
+ int cpu;
+ struct bau_control *bcp;
+
+ for_each_present_cpu(cpu) {
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->baudisabled = 0;
+ if (nobau)
+ bcp->nobau = true;
+ bcp->statp = &per_cpu(ptcstats, cpu);
+ /* time interval to catch a hardware stay-busy bug */
+ bcp->timeout_interval = usec_2_cycles(2*timeout_us);
+ bcp->max_concurr = max_concurr;
+ bcp->max_concurr_const = max_concurr;
+ bcp->plugged_delay = plugged_delay;
+ bcp->plugsb4reset = plugsb4reset;
+ bcp->timeoutsb4reset = timeoutsb4reset;
+ bcp->ipi_reset_limit = ipi_reset_limit;
+ bcp->complete_threshold = complete_threshold;
+ bcp->cong_response_us = congested_respns_us;
+ bcp->cong_reps = congested_reps;
+ bcp->disabled_period = sec_2_cycles(disabled_period);
+ bcp->giveup_limit = giveup_limit;
+ spin_lock_init(&bcp->queue_lock);
+ spin_lock_init(&bcp->uvhub_lock);
+ spin_lock_init(&bcp->disable_lock);
+ }
+}
+
+/*
+ * Scan all cpus to collect blade and socket summaries.
+ */
+static int __init get_cpu_topology(int base_pnode,
+ struct uvhub_desc *uvhub_descs,
+ unsigned char *uvhub_mask)
+{
+ int cpu;
+ int pnode;
+ int uvhub;
+ int socket;
+ struct bau_control *bcp;
+ struct uvhub_desc *bdp;
+ struct socket_desc *sdp;
+
+ for_each_present_cpu(cpu) {
+ bcp = &per_cpu(bau_control, cpu);
+
+ memset(bcp, 0, sizeof(struct bau_control));
+
+ pnode = uv_cpu_hub_info(cpu)->pnode;
+ if ((pnode - base_pnode) >= UV_DISTRIBUTION_SIZE) {
+ pr_emerg(
+ "cpu %d pnode %d-%d beyond %d; BAU disabled\n",
+ cpu, pnode, base_pnode, UV_DISTRIBUTION_SIZE);
+ return 1;
+ }
+
+ bcp->osnode = cpu_to_node(cpu);
+ bcp->partition_base_pnode = base_pnode;
+
+ uvhub = uv_cpu_hub_info(cpu)->numa_blade_id;
+ *(uvhub_mask + (uvhub/8)) |= (1 << (uvhub%8));
+ bdp = &uvhub_descs[uvhub];
+
+ bdp->num_cpus++;
+ bdp->uvhub = uvhub;
+ bdp->pnode = pnode;
+
+ /* kludge: 'assuming' one node per socket, and assuming that
+ disabling a socket just leaves a gap in node numbers */
+ socket = bcp->osnode & 1;
+ bdp->socket_mask |= (1 << socket);
+ sdp = &bdp->socket[socket];
+ sdp->cpu_number[sdp->num_cpus] = cpu;
+ sdp->num_cpus++;
+ if (sdp->num_cpus > MAX_CPUS_PER_SOCKET) {
+ pr_emerg("%d cpus per socket invalid\n",
+ sdp->num_cpus);
+ return 1;
+ }
+ }
+ return 0;
+}
+
+/*
+ * Each socket is to get a local array of pnodes/hubs.
+ */
+static void make_per_cpu_thp(struct bau_control *smaster)
+{
+ int cpu;
+ size_t hpsz = sizeof(struct hub_and_pnode) * num_possible_cpus();
+
+ smaster->thp = kmalloc_node(hpsz, GFP_KERNEL, smaster->osnode);
+ memset(smaster->thp, 0, hpsz);
+ for_each_present_cpu(cpu) {
+ smaster->thp[cpu].pnode = uv_cpu_hub_info(cpu)->pnode;
+ smaster->thp[cpu].uvhub = uv_cpu_hub_info(cpu)->numa_blade_id;
+ }
+}
+
+/*
+ * Each uvhub is to get a local cpumask.
+ */
+static void make_per_hub_cpumask(struct bau_control *hmaster)
+{
+ int sz = sizeof(cpumask_t);
+
+ hmaster->cpumask = kzalloc_node(sz, GFP_KERNEL, hmaster->osnode);
+}
+
+/*
+ * Initialize all the per_cpu information for the cpu's on a given socket,
+ * given what has been gathered into the socket_desc struct.
+ * And reports the chosen hub and socket masters back to the caller.
+ */
+static int scan_sock(struct socket_desc *sdp, struct uvhub_desc *bdp,
+ struct bau_control **smasterp,
+ struct bau_control **hmasterp)
+{
+ int i, cpu, uvhub_cpu;
+ struct bau_control *bcp;
+
+ for (i = 0; i < sdp->num_cpus; i++) {
+ cpu = sdp->cpu_number[i];
+ bcp = &per_cpu(bau_control, cpu);
+ bcp->cpu = cpu;
+ if (i == 0) {
+ *smasterp = bcp;
+ if (!(*hmasterp))
+ *hmasterp = bcp;
+ }
+ bcp->cpus_in_uvhub = bdp->num_cpus;
+ bcp->cpus_in_socket = sdp->num_cpus;
+ bcp->socket_master = *smasterp;
+ bcp->uvhub = bdp->uvhub;
+ if (is_uv1_hub())
+ bcp->uvhub_version = UV_BAU_V1;
+ else if (is_uv2_hub())
+ bcp->uvhub_version = UV_BAU_V2;
+ else if (is_uv3_hub())
+ bcp->uvhub_version = UV_BAU_V3;
+ else if (is_uv4_hub())
+ bcp->uvhub_version = UV_BAU_V4;
+ else {
+ pr_emerg("uvhub version not 1, 2, 3, or 4\n");
+ return 1;
+ }
+ bcp->uvhub_master = *hmasterp;
+ uvhub_cpu = uv_cpu_blade_processor_id(cpu);
+ bcp->uvhub_cpu = uvhub_cpu;
+
+ /*
+ * The ERROR and BUSY status registers are located pairwise over
+ * the STATUS_0 and STATUS_1 mmrs; each an array[32] of 2 bits.
+ */
+ if (uvhub_cpu < UV_CPUS_PER_AS) {
+ bcp->status_mmr = UVH_LB_BAU_SB_ACTIVATION_STATUS_0;
+ bcp->status_index = uvhub_cpu * UV_ACT_STATUS_SIZE;
+ } else {
+ bcp->status_mmr = UVH_LB_BAU_SB_ACTIVATION_STATUS_1;
+ bcp->status_index = (uvhub_cpu - UV_CPUS_PER_AS)
+ * UV_ACT_STATUS_SIZE;
+ }
+
+ if (bcp->uvhub_cpu >= MAX_CPUS_PER_UVHUB) {
+ pr_emerg("%d cpus per uvhub invalid\n",
+ bcp->uvhub_cpu);
+ return 1;
+ }
+ }
+ return 0;
+}
+
+/*
+ * Summarize the blade and socket topology into the per_cpu structures.
+ */
+static int __init summarize_uvhub_sockets(int nuvhubs,
+ struct uvhub_desc *uvhub_descs,
+ unsigned char *uvhub_mask)
+{
+ int socket;
+ int uvhub;
+ unsigned short socket_mask;
+
+ for (uvhub = 0; uvhub < nuvhubs; uvhub++) {
+ struct uvhub_desc *bdp;
+ struct bau_control *smaster = NULL;
+ struct bau_control *hmaster = NULL;
+
+ if (!(*(uvhub_mask + (uvhub/8)) & (1 << (uvhub%8))))
+ continue;
+
+ bdp = &uvhub_descs[uvhub];
+ socket_mask = bdp->socket_mask;
+ socket = 0;
+ while (socket_mask) {
+ struct socket_desc *sdp;
+ if ((socket_mask & 1)) {
+ sdp = &bdp->socket[socket];
+ if (scan_sock(sdp, bdp, &smaster, &hmaster))
+ return 1;
+ make_per_cpu_thp(smaster);
+ }
+ socket++;
+ socket_mask = (socket_mask >> 1);
+ }
+ make_per_hub_cpumask(hmaster);
+ }
+ return 0;
+}
+
+/*
+ * initialize the bau_control structure for each cpu
+ */
+static int __init init_per_cpu(int nuvhubs, int base_part_pnode)
+{
+ unsigned char *uvhub_mask;
+ void *vp;
+ struct uvhub_desc *uvhub_descs;
+
+ if (is_uv3_hub() || is_uv2_hub() || is_uv1_hub())
+ timeout_us = calculate_destination_timeout();
+
+ vp = kmalloc_array(nuvhubs, sizeof(struct uvhub_desc), GFP_KERNEL);
+ uvhub_descs = (struct uvhub_desc *)vp;
+ memset(uvhub_descs, 0, nuvhubs * sizeof(struct uvhub_desc));
+ uvhub_mask = kzalloc((nuvhubs+7)/8, GFP_KERNEL);
+
+ if (get_cpu_topology(base_part_pnode, uvhub_descs, uvhub_mask))
+ goto fail;
+
+ if (summarize_uvhub_sockets(nuvhubs, uvhub_descs, uvhub_mask))
+ goto fail;
+
+ kfree(uvhub_descs);
+ kfree(uvhub_mask);
+ init_per_cpu_tunables();
+ return 0;
+
+fail:
+ kfree(uvhub_descs);
+ kfree(uvhub_mask);
+ return 1;
+}
+
+static const struct bau_operations uv1_bau_ops __initconst = {
+ .bau_gpa_to_offset = uv_gpa_to_offset,
+ .read_l_sw_ack = read_mmr_sw_ack,
+ .read_g_sw_ack = read_gmmr_sw_ack,
+ .write_l_sw_ack = write_mmr_sw_ack,
+ .write_g_sw_ack = write_gmmr_sw_ack,
+ .write_payload_first = write_mmr_payload_first,
+ .write_payload_last = write_mmr_payload_last,
+ .wait_completion = uv1_wait_completion,
+};
+
+static const struct bau_operations uv2_3_bau_ops __initconst = {
+ .bau_gpa_to_offset = uv_gpa_to_offset,
+ .read_l_sw_ack = read_mmr_sw_ack,
+ .read_g_sw_ack = read_gmmr_sw_ack,
+ .write_l_sw_ack = write_mmr_sw_ack,
+ .write_g_sw_ack = write_gmmr_sw_ack,
+ .write_payload_first = write_mmr_payload_first,
+ .write_payload_last = write_mmr_payload_last,
+ .wait_completion = uv2_3_wait_completion,
+};
+
+static const struct bau_operations uv4_bau_ops __initconst = {
+ .bau_gpa_to_offset = uv_gpa_to_soc_phys_ram,
+ .read_l_sw_ack = read_mmr_proc_sw_ack,
+ .read_g_sw_ack = read_gmmr_proc_sw_ack,
+ .write_l_sw_ack = write_mmr_proc_sw_ack,
+ .write_g_sw_ack = write_gmmr_proc_sw_ack,
+ .write_payload_first = write_mmr_proc_payload_first,
+ .write_payload_last = write_mmr_proc_payload_last,
+ .wait_completion = uv4_wait_completion,
+};
+
+/*
+ * Initialization of BAU-related structures
+ */
+static int __init uv_bau_init(void)
+{
+ int uvhub;
+ int pnode;
+ int nuvhubs;
+ int cur_cpu;
+ int cpus;
+ int vector;
+ cpumask_var_t *mask;
+
+ if (!is_uv_system())
+ return 0;
+
+ if (is_uv4_hub())
+ ops = uv4_bau_ops;
+ else if (is_uv3_hub())
+ ops = uv2_3_bau_ops;
+ else if (is_uv2_hub())
+ ops = uv2_3_bau_ops;
+ else if (is_uv1_hub())
+ ops = uv1_bau_ops;
+
+ nuvhubs = uv_num_possible_blades();
+ if (nuvhubs < 2) {
+ pr_crit("UV: BAU disabled - insufficient hub count\n");
+ goto err_bau_disable;
+ }
+
+ for_each_possible_cpu(cur_cpu) {
+ mask = &per_cpu(uv_flush_tlb_mask, cur_cpu);
+ zalloc_cpumask_var_node(mask, GFP_KERNEL, cpu_to_node(cur_cpu));
+ }
+
+ uv_base_pnode = 0x7fffffff;
+ for (uvhub = 0; uvhub < nuvhubs; uvhub++) {
+ cpus = uv_blade_nr_possible_cpus(uvhub);
+ if (cpus && (uv_blade_to_pnode(uvhub) < uv_base_pnode))
+ uv_base_pnode = uv_blade_to_pnode(uvhub);
+ }
+
+ /* software timeouts are not supported on UV4 */
+ if (is_uv3_hub() || is_uv2_hub() || is_uv1_hub())
+ enable_timeouts();
+
+ if (init_per_cpu(nuvhubs, uv_base_pnode)) {
+ pr_crit("UV: BAU disabled - per CPU init failed\n");
+ goto err_bau_disable;
+ }
+
+ vector = UV_BAU_MESSAGE;
+ for_each_possible_blade(uvhub) {
+ if (uv_blade_nr_possible_cpus(uvhub))
+ init_uvhub(uvhub, vector, uv_base_pnode);
+ }
+
+ for_each_possible_blade(uvhub) {
+ if (uv_blade_nr_possible_cpus(uvhub)) {
+ unsigned long val;
+ unsigned long mmr;
+ pnode = uv_blade_to_pnode(uvhub);
+ /* INIT the bau */
+ val = 1L << 63;
+ write_gmmr_activation(pnode, val);
+ mmr = 1; /* should be 1 to broadcast to both sockets */
+ if (!is_uv1_hub())
+ write_mmr_data_broadcast(pnode, mmr);
+ }
+ }
+
+ return 0;
+
+err_bau_disable:
+
+ for_each_possible_cpu(cur_cpu)
+ free_cpumask_var(per_cpu(uv_flush_tlb_mask, cur_cpu));
+
+ set_bau_off();
+ nobau_perm = 1;
+
+ return -EINVAL;
+}
+core_initcall(uv_bau_init);
+fs_initcall(uv_ptc_init);
diff --git a/arch/x86/platform/uv/uv_irq.c b/arch/x86/platform/uv/uv_irq.c
new file mode 100644
index 000000000..abb607539
--- /dev/null
+++ b/arch/x86/platform/uv/uv_irq.c
@@ -0,0 +1,217 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * SGI UV IRQ functions
+ *
+ * Copyright (C) 2008 Silicon Graphics, Inc. All rights reserved.
+ */
+
+#include <linux/export.h>
+#include <linux/rbtree.h>
+#include <linux/slab.h>
+#include <linux/irq.h>
+
+#include <asm/irqdomain.h>
+#include <asm/apic.h>
+#include <asm/uv/uv_irq.h>
+#include <asm/uv/uv_hub.h>
+
+/* MMR offset and pnode of hub sourcing interrupts for a given irq */
+struct uv_irq_2_mmr_pnode {
+ unsigned long offset;
+ int pnode;
+};
+
+static void uv_program_mmr(struct irq_cfg *cfg, struct uv_irq_2_mmr_pnode *info)
+{
+ unsigned long mmr_value;
+ struct uv_IO_APIC_route_entry *entry;
+
+ BUILD_BUG_ON(sizeof(struct uv_IO_APIC_route_entry) !=
+ sizeof(unsigned long));
+
+ mmr_value = 0;
+ entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
+ entry->vector = cfg->vector;
+ entry->delivery_mode = apic->irq_delivery_mode;
+ entry->dest_mode = apic->irq_dest_mode;
+ entry->polarity = 0;
+ entry->trigger = 0;
+ entry->mask = 0;
+ entry->dest = cfg->dest_apicid;
+
+ uv_write_global_mmr64(info->pnode, info->offset, mmr_value);
+}
+
+static void uv_noop(struct irq_data *data) { }
+
+static int
+uv_set_irq_affinity(struct irq_data *data, const struct cpumask *mask,
+ bool force)
+{
+ struct irq_data *parent = data->parent_data;
+ struct irq_cfg *cfg = irqd_cfg(data);
+ int ret;
+
+ ret = parent->chip->irq_set_affinity(parent, mask, force);
+ if (ret >= 0) {
+ uv_program_mmr(cfg, data->chip_data);
+ send_cleanup_vector(cfg);
+ }
+
+ return ret;
+}
+
+static struct irq_chip uv_irq_chip = {
+ .name = "UV-CORE",
+ .irq_mask = uv_noop,
+ .irq_unmask = uv_noop,
+ .irq_eoi = apic_ack_irq,
+ .irq_set_affinity = uv_set_irq_affinity,
+};
+
+static int uv_domain_alloc(struct irq_domain *domain, unsigned int virq,
+ unsigned int nr_irqs, void *arg)
+{
+ struct uv_irq_2_mmr_pnode *chip_data;
+ struct irq_alloc_info *info = arg;
+ struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+ int ret;
+
+ if (nr_irqs > 1 || !info || info->type != X86_IRQ_ALLOC_TYPE_UV)
+ return -EINVAL;
+
+ chip_data = kmalloc_node(sizeof(*chip_data), GFP_KERNEL,
+ irq_data_get_node(irq_data));
+ if (!chip_data)
+ return -ENOMEM;
+
+ ret = irq_domain_alloc_irqs_parent(domain, virq, nr_irqs, arg);
+ if (ret >= 0) {
+ if (info->uv_limit == UV_AFFINITY_CPU)
+ irq_set_status_flags(virq, IRQ_NO_BALANCING);
+ else
+ irq_set_status_flags(virq, IRQ_MOVE_PCNTXT);
+
+ chip_data->pnode = uv_blade_to_pnode(info->uv_blade);
+ chip_data->offset = info->uv_offset;
+ irq_domain_set_info(domain, virq, virq, &uv_irq_chip, chip_data,
+ handle_percpu_irq, NULL, info->uv_name);
+ } else {
+ kfree(chip_data);
+ }
+
+ return ret;
+}
+
+static void uv_domain_free(struct irq_domain *domain, unsigned int virq,
+ unsigned int nr_irqs)
+{
+ struct irq_data *irq_data = irq_domain_get_irq_data(domain, virq);
+
+ BUG_ON(nr_irqs != 1);
+ kfree(irq_data->chip_data);
+ irq_clear_status_flags(virq, IRQ_MOVE_PCNTXT);
+ irq_clear_status_flags(virq, IRQ_NO_BALANCING);
+ irq_domain_free_irqs_top(domain, virq, nr_irqs);
+}
+
+/*
+ * Re-target the irq to the specified CPU and enable the specified MMR located
+ * on the specified blade to allow the sending of MSIs to the specified CPU.
+ */
+static int uv_domain_activate(struct irq_domain *domain,
+ struct irq_data *irq_data, bool reserve)
+{
+ uv_program_mmr(irqd_cfg(irq_data), irq_data->chip_data);
+ return 0;
+}
+
+/*
+ * Disable the specified MMR located on the specified blade so that MSIs are
+ * longer allowed to be sent.
+ */
+static void uv_domain_deactivate(struct irq_domain *domain,
+ struct irq_data *irq_data)
+{
+ unsigned long mmr_value;
+ struct uv_IO_APIC_route_entry *entry;
+
+ mmr_value = 0;
+ entry = (struct uv_IO_APIC_route_entry *)&mmr_value;
+ entry->mask = 1;
+ uv_program_mmr(irqd_cfg(irq_data), irq_data->chip_data);
+}
+
+static const struct irq_domain_ops uv_domain_ops = {
+ .alloc = uv_domain_alloc,
+ .free = uv_domain_free,
+ .activate = uv_domain_activate,
+ .deactivate = uv_domain_deactivate,
+};
+
+static struct irq_domain *uv_get_irq_domain(void)
+{
+ static struct irq_domain *uv_domain;
+ static DEFINE_MUTEX(uv_lock);
+ struct fwnode_handle *fn;
+
+ mutex_lock(&uv_lock);
+ if (uv_domain)
+ goto out;
+
+ fn = irq_domain_alloc_named_fwnode("UV-CORE");
+ if (!fn)
+ goto out;
+
+ uv_domain = irq_domain_create_tree(fn, &uv_domain_ops, NULL);
+ if (uv_domain)
+ uv_domain->parent = x86_vector_domain;
+ else
+ irq_domain_free_fwnode(fn);
+out:
+ mutex_unlock(&uv_lock);
+
+ return uv_domain;
+}
+
+/*
+ * Set up a mapping of an available irq and vector, and enable the specified
+ * MMR that defines the MSI that is to be sent to the specified CPU when an
+ * interrupt is raised.
+ */
+int uv_setup_irq(char *irq_name, int cpu, int mmr_blade,
+ unsigned long mmr_offset, int limit)
+{
+ struct irq_alloc_info info;
+ struct irq_domain *domain = uv_get_irq_domain();
+
+ if (!domain)
+ return -ENOMEM;
+
+ init_irq_alloc_info(&info, cpumask_of(cpu));
+ info.type = X86_IRQ_ALLOC_TYPE_UV;
+ info.uv_limit = limit;
+ info.uv_blade = mmr_blade;
+ info.uv_offset = mmr_offset;
+ info.uv_name = irq_name;
+
+ return irq_domain_alloc_irqs(domain, 1,
+ uv_blade_to_memory_nid(mmr_blade), &info);
+}
+EXPORT_SYMBOL_GPL(uv_setup_irq);
+
+/*
+ * Tear down a mapping of an irq and vector, and disable the specified MMR that
+ * defined the MSI that was to be sent to the specified CPU when an interrupt
+ * was raised.
+ *
+ * Set mmr_blade and mmr_offset to what was passed in on uv_setup_irq().
+ */
+void uv_teardown_irq(unsigned int irq)
+{
+ irq_domain_free_irqs(irq, 1);
+}
+EXPORT_SYMBOL_GPL(uv_teardown_irq);
diff --git a/arch/x86/platform/uv/uv_nmi.c b/arch/x86/platform/uv/uv_nmi.c
new file mode 100644
index 000000000..5f64f3087
--- /dev/null
+++ b/arch/x86/platform/uv/uv_nmi.c
@@ -0,0 +1,1066 @@
+/*
+ * SGI NMI support routines
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Copyright (c) 2009-2013 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) Mike Travis
+ */
+
+#include <linux/cpu.h>
+#include <linux/delay.h>
+#include <linux/kdb.h>
+#include <linux/kexec.h>
+#include <linux/kgdb.h>
+#include <linux/moduleparam.h>
+#include <linux/nmi.h>
+#include <linux/sched.h>
+#include <linux/sched/debug.h>
+#include <linux/slab.h>
+#include <linux/clocksource.h>
+
+#include <asm/apic.h>
+#include <asm/current.h>
+#include <asm/kdebug.h>
+#include <asm/local64.h>
+#include <asm/nmi.h>
+#include <asm/traps.h>
+#include <asm/uv/uv.h>
+#include <asm/uv/uv_hub.h>
+#include <asm/uv/uv_mmrs.h>
+
+/*
+ * UV handler for NMI
+ *
+ * Handle system-wide NMI events generated by the global 'power nmi' command.
+ *
+ * Basic operation is to field the NMI interrupt on each CPU and wait
+ * until all CPU's have arrived into the nmi handler. If some CPU's do not
+ * make it into the handler, try and force them in with the IPI(NMI) signal.
+ *
+ * We also have to lessen UV Hub MMR accesses as much as possible as this
+ * disrupts the UV Hub's primary mission of directing NumaLink traffic and
+ * can cause system problems to occur.
+ *
+ * To do this we register our primary NMI notifier on the NMI_UNKNOWN
+ * chain. This reduces the number of false NMI calls when the perf
+ * tools are running which generate an enormous number of NMIs per
+ * second (~4M/s for 1024 CPU threads). Our secondary NMI handler is
+ * very short as it only checks that if it has been "pinged" with the
+ * IPI(NMI) signal as mentioned above, and does not read the UV Hub's MMR.
+ *
+ */
+
+static struct uv_hub_nmi_s **uv_hub_nmi_list;
+
+DEFINE_PER_CPU(struct uv_cpu_nmi_s, uv_cpu_nmi);
+
+/* UV hubless values */
+#define NMI_CONTROL_PORT 0x70
+#define NMI_DUMMY_PORT 0x71
+#define PAD_OWN_GPP_D_0 0x2c
+#define GPI_NMI_STS_GPP_D_0 0x164
+#define GPI_NMI_ENA_GPP_D_0 0x174
+#define STS_GPP_D_0_MASK 0x1
+#define PAD_CFG_DW0_GPP_D_0 0x4c0
+#define GPIROUTNMI (1ul << 17)
+#define PCH_PCR_GPIO_1_BASE 0xfdae0000ul
+#define PCH_PCR_GPIO_ADDRESS(offset) (int *)((u64)(pch_base) | (u64)(offset))
+
+static u64 *pch_base;
+static unsigned long nmi_mmr;
+static unsigned long nmi_mmr_clear;
+static unsigned long nmi_mmr_pending;
+
+static atomic_t uv_in_nmi;
+static atomic_t uv_nmi_cpu = ATOMIC_INIT(-1);
+static atomic_t uv_nmi_cpus_in_nmi = ATOMIC_INIT(-1);
+static atomic_t uv_nmi_slave_continue;
+static cpumask_var_t uv_nmi_cpu_mask;
+
+/* Values for uv_nmi_slave_continue */
+#define SLAVE_CLEAR 0
+#define SLAVE_CONTINUE 1
+#define SLAVE_EXIT 2
+
+/*
+ * Default is all stack dumps go to the console and buffer.
+ * Lower level to send to log buffer only.
+ */
+static int uv_nmi_loglevel = CONSOLE_LOGLEVEL_DEFAULT;
+module_param_named(dump_loglevel, uv_nmi_loglevel, int, 0644);
+
+/*
+ * The following values show statistics on how perf events are affecting
+ * this system.
+ */
+static int param_get_local64(char *buffer, const struct kernel_param *kp)
+{
+ return sprintf(buffer, "%lu\n", local64_read((local64_t *)kp->arg));
+}
+
+static int param_set_local64(const char *val, const struct kernel_param *kp)
+{
+ /* Clear on any write */
+ local64_set((local64_t *)kp->arg, 0);
+ return 0;
+}
+
+static const struct kernel_param_ops param_ops_local64 = {
+ .get = param_get_local64,
+ .set = param_set_local64,
+};
+#define param_check_local64(name, p) __param_check(name, p, local64_t)
+
+static local64_t uv_nmi_count;
+module_param_named(nmi_count, uv_nmi_count, local64, 0644);
+
+static local64_t uv_nmi_misses;
+module_param_named(nmi_misses, uv_nmi_misses, local64, 0644);
+
+static local64_t uv_nmi_ping_count;
+module_param_named(ping_count, uv_nmi_ping_count, local64, 0644);
+
+static local64_t uv_nmi_ping_misses;
+module_param_named(ping_misses, uv_nmi_ping_misses, local64, 0644);
+
+/*
+ * Following values allow tuning for large systems under heavy loading
+ */
+static int uv_nmi_initial_delay = 100;
+module_param_named(initial_delay, uv_nmi_initial_delay, int, 0644);
+
+static int uv_nmi_slave_delay = 100;
+module_param_named(slave_delay, uv_nmi_slave_delay, int, 0644);
+
+static int uv_nmi_loop_delay = 100;
+module_param_named(loop_delay, uv_nmi_loop_delay, int, 0644);
+
+static int uv_nmi_trigger_delay = 10000;
+module_param_named(trigger_delay, uv_nmi_trigger_delay, int, 0644);
+
+static int uv_nmi_wait_count = 100;
+module_param_named(wait_count, uv_nmi_wait_count, int, 0644);
+
+static int uv_nmi_retry_count = 500;
+module_param_named(retry_count, uv_nmi_retry_count, int, 0644);
+
+static bool uv_pch_intr_enable = true;
+static bool uv_pch_intr_now_enabled;
+module_param_named(pch_intr_enable, uv_pch_intr_enable, bool, 0644);
+
+static bool uv_pch_init_enable = true;
+module_param_named(pch_init_enable, uv_pch_init_enable, bool, 0644);
+
+static int uv_nmi_debug;
+module_param_named(debug, uv_nmi_debug, int, 0644);
+
+#define nmi_debug(fmt, ...) \
+ do { \
+ if (uv_nmi_debug) \
+ pr_info(fmt, ##__VA_ARGS__); \
+ } while (0)
+
+/* Valid NMI Actions */
+#define ACTION_LEN 16
+static struct nmi_action {
+ char *action;
+ char *desc;
+} valid_acts[] = {
+ { "kdump", "do kernel crash dump" },
+ { "dump", "dump process stack for each cpu" },
+ { "ips", "dump Inst Ptr info for each cpu" },
+ { "kdb", "enter KDB (needs kgdboc= assignment)" },
+ { "kgdb", "enter KGDB (needs gdb target remote)" },
+ { "health", "check if CPUs respond to NMI" },
+};
+typedef char action_t[ACTION_LEN];
+static action_t uv_nmi_action = { "dump" };
+
+static int param_get_action(char *buffer, const struct kernel_param *kp)
+{
+ return sprintf(buffer, "%s\n", uv_nmi_action);
+}
+
+static int param_set_action(const char *val, const struct kernel_param *kp)
+{
+ int i;
+ int n = ARRAY_SIZE(valid_acts);
+ char arg[ACTION_LEN], *p;
+
+ /* (remove possible '\n') */
+ strncpy(arg, val, ACTION_LEN - 1);
+ arg[ACTION_LEN - 1] = '\0';
+ p = strchr(arg, '\n');
+ if (p)
+ *p = '\0';
+
+ for (i = 0; i < n; i++)
+ if (!strcmp(arg, valid_acts[i].action))
+ break;
+
+ if (i < n) {
+ strcpy(uv_nmi_action, arg);
+ pr_info("UV: New NMI action:%s\n", uv_nmi_action);
+ return 0;
+ }
+
+ pr_err("UV: Invalid NMI action:%s, valid actions are:\n", arg);
+ for (i = 0; i < n; i++)
+ pr_err("UV: %-8s - %s\n",
+ valid_acts[i].action, valid_acts[i].desc);
+ return -EINVAL;
+}
+
+static const struct kernel_param_ops param_ops_action = {
+ .get = param_get_action,
+ .set = param_set_action,
+};
+#define param_check_action(name, p) __param_check(name, p, action_t)
+
+module_param_named(action, uv_nmi_action, action, 0644);
+
+static inline bool uv_nmi_action_is(const char *action)
+{
+ return (strncmp(uv_nmi_action, action, strlen(action)) == 0);
+}
+
+/* Setup which NMI support is present in system */
+static void uv_nmi_setup_mmrs(void)
+{
+ if (uv_read_local_mmr(UVH_NMI_MMRX_SUPPORTED)) {
+ uv_write_local_mmr(UVH_NMI_MMRX_REQ,
+ 1UL << UVH_NMI_MMRX_REQ_SHIFT);
+ nmi_mmr = UVH_NMI_MMRX;
+ nmi_mmr_clear = UVH_NMI_MMRX_CLEAR;
+ nmi_mmr_pending = 1UL << UVH_NMI_MMRX_SHIFT;
+ pr_info("UV: SMI NMI support: %s\n", UVH_NMI_MMRX_TYPE);
+ } else {
+ nmi_mmr = UVH_NMI_MMR;
+ nmi_mmr_clear = UVH_NMI_MMR_CLEAR;
+ nmi_mmr_pending = 1UL << UVH_NMI_MMR_SHIFT;
+ pr_info("UV: SMI NMI support: %s\n", UVH_NMI_MMR_TYPE);
+ }
+}
+
+/* Read NMI MMR and check if NMI flag was set by BMC. */
+static inline int uv_nmi_test_mmr(struct uv_hub_nmi_s *hub_nmi)
+{
+ hub_nmi->nmi_value = uv_read_local_mmr(nmi_mmr);
+ atomic_inc(&hub_nmi->read_mmr_count);
+ return !!(hub_nmi->nmi_value & nmi_mmr_pending);
+}
+
+static inline void uv_local_mmr_clear_nmi(void)
+{
+ uv_write_local_mmr(nmi_mmr_clear, nmi_mmr_pending);
+}
+
+/*
+ * UV hubless NMI handler functions
+ */
+static inline void uv_reassert_nmi(void)
+{
+ /* (from arch/x86/include/asm/mach_traps.h) */
+ outb(0x8f, NMI_CONTROL_PORT);
+ inb(NMI_DUMMY_PORT); /* dummy read */
+ outb(0x0f, NMI_CONTROL_PORT);
+ inb(NMI_DUMMY_PORT); /* dummy read */
+}
+
+static void uv_init_hubless_pch_io(int offset, int mask, int data)
+{
+ int *addr = PCH_PCR_GPIO_ADDRESS(offset);
+ int readd = readl(addr);
+
+ if (mask) { /* OR in new data */
+ int writed = (readd & ~mask) | data;
+
+ nmi_debug("UV:PCH: %p = %x & %x | %x (%x)\n",
+ addr, readd, ~mask, data, writed);
+ writel(writed, addr);
+ } else if (readd & data) { /* clear status bit */
+ nmi_debug("UV:PCH: %p = %x\n", addr, data);
+ writel(data, addr);
+ }
+
+ (void)readl(addr); /* flush write data */
+}
+
+static void uv_nmi_setup_hubless_intr(void)
+{
+ uv_pch_intr_now_enabled = uv_pch_intr_enable;
+
+ uv_init_hubless_pch_io(
+ PAD_CFG_DW0_GPP_D_0, GPIROUTNMI,
+ uv_pch_intr_now_enabled ? GPIROUTNMI : 0);
+
+ nmi_debug("UV:NMI: GPP_D_0 interrupt %s\n",
+ uv_pch_intr_now_enabled ? "enabled" : "disabled");
+}
+
+static struct init_nmi {
+ unsigned int offset;
+ unsigned int mask;
+ unsigned int data;
+} init_nmi[] = {
+ { /* HOSTSW_OWN_GPP_D_0 */
+ .offset = 0x84,
+ .mask = 0x1,
+ .data = 0x0, /* ACPI Mode */
+ },
+
+/* Clear status: */
+ { /* GPI_INT_STS_GPP_D_0 */
+ .offset = 0x104,
+ .mask = 0x0,
+ .data = 0x1, /* Clear Status */
+ },
+ { /* GPI_GPE_STS_GPP_D_0 */
+ .offset = 0x124,
+ .mask = 0x0,
+ .data = 0x1, /* Clear Status */
+ },
+ { /* GPI_SMI_STS_GPP_D_0 */
+ .offset = 0x144,
+ .mask = 0x0,
+ .data = 0x1, /* Clear Status */
+ },
+ { /* GPI_NMI_STS_GPP_D_0 */
+ .offset = 0x164,
+ .mask = 0x0,
+ .data = 0x1, /* Clear Status */
+ },
+
+/* Disable interrupts: */
+ { /* GPI_INT_EN_GPP_D_0 */
+ .offset = 0x114,
+ .mask = 0x1,
+ .data = 0x0, /* Disable interrupt generation */
+ },
+ { /* GPI_GPE_EN_GPP_D_0 */
+ .offset = 0x134,
+ .mask = 0x1,
+ .data = 0x0, /* Disable interrupt generation */
+ },
+ { /* GPI_SMI_EN_GPP_D_0 */
+ .offset = 0x154,
+ .mask = 0x1,
+ .data = 0x0, /* Disable interrupt generation */
+ },
+ { /* GPI_NMI_EN_GPP_D_0 */
+ .offset = 0x174,
+ .mask = 0x1,
+ .data = 0x0, /* Disable interrupt generation */
+ },
+
+/* Setup GPP_D_0 Pad Config: */
+ { /* PAD_CFG_DW0_GPP_D_0 */
+ .offset = 0x4c0,
+ .mask = 0xffffffff,
+ .data = 0x82020100,
+/*
+ * 31:30 Pad Reset Config (PADRSTCFG): = 2h # PLTRST# (default)
+ *
+ * 29 RX Pad State Select (RXPADSTSEL): = 0 # Raw RX pad state directly
+ * from RX buffer (default)
+ *
+ * 28 RX Raw Override to '1' (RXRAW1): = 0 # No Override
+ *
+ * 26:25 RX Level/Edge Configuration (RXEVCFG):
+ * = 0h # Level
+ * = 1h # Edge
+ *
+ * 23 RX Invert (RXINV): = 0 # No Inversion (signal active high)
+ *
+ * 20 GPIO Input Route IOxAPIC (GPIROUTIOXAPIC):
+ * = 0 # Routing does not cause peripheral IRQ...
+ * # (we want an NMI not an IRQ)
+ *
+ * 19 GPIO Input Route SCI (GPIROUTSCI): = 0 # Routing does not cause SCI.
+ * 18 GPIO Input Route SMI (GPIROUTSMI): = 0 # Routing does not cause SMI.
+ * 17 GPIO Input Route NMI (GPIROUTNMI): = 1 # Routing can cause NMI.
+ *
+ * 11:10 Pad Mode (PMODE1/0): = 0h = GPIO control the Pad.
+ * 9 GPIO RX Disable (GPIORXDIS):
+ * = 0 # Enable the input buffer (active low enable)
+ *
+ * 8 GPIO TX Disable (GPIOTXDIS):
+ * = 1 # Disable the output buffer; i.e. Hi-Z
+ *
+ * 1 GPIO RX State (GPIORXSTATE): This is the current internal RX pad state..
+ * 0 GPIO TX State (GPIOTXSTATE):
+ * = 0 # (Leave at default)
+ */
+ },
+
+/* Pad Config DW1 */
+ { /* PAD_CFG_DW1_GPP_D_0 */
+ .offset = 0x4c4,
+ .mask = 0x3c00,
+ .data = 0, /* Termination = none (default) */
+ },
+};
+
+static void uv_init_hubless_pch_d0(void)
+{
+ int i, read;
+
+ read = *PCH_PCR_GPIO_ADDRESS(PAD_OWN_GPP_D_0);
+ if (read != 0) {
+ pr_info("UV: Hubless NMI already configured\n");
+ return;
+ }
+
+ nmi_debug("UV: Initializing UV Hubless NMI on PCH\n");
+ for (i = 0; i < ARRAY_SIZE(init_nmi); i++) {
+ uv_init_hubless_pch_io(init_nmi[i].offset,
+ init_nmi[i].mask,
+ init_nmi[i].data);
+ }
+}
+
+static int uv_nmi_test_hubless(struct uv_hub_nmi_s *hub_nmi)
+{
+ int *pstat = PCH_PCR_GPIO_ADDRESS(GPI_NMI_STS_GPP_D_0);
+ int status = *pstat;
+
+ hub_nmi->nmi_value = status;
+ atomic_inc(&hub_nmi->read_mmr_count);
+
+ if (!(status & STS_GPP_D_0_MASK)) /* Not a UV external NMI */
+ return 0;
+
+ *pstat = STS_GPP_D_0_MASK; /* Is a UV NMI: clear GPP_D_0 status */
+ (void)*pstat; /* Flush write */
+
+ return 1;
+}
+
+static int uv_test_nmi(struct uv_hub_nmi_s *hub_nmi)
+{
+ if (hub_nmi->hub_present)
+ return uv_nmi_test_mmr(hub_nmi);
+
+ if (hub_nmi->pch_owner) /* Only PCH owner can check status */
+ return uv_nmi_test_hubless(hub_nmi);
+
+ return -1;
+}
+
+/*
+ * If first CPU in on this hub, set hub_nmi "in_nmi" and "owner" values and
+ * return true. If first CPU in on the system, set global "in_nmi" flag.
+ */
+static int uv_set_in_nmi(int cpu, struct uv_hub_nmi_s *hub_nmi)
+{
+ int first = atomic_add_unless(&hub_nmi->in_nmi, 1, 1);
+
+ if (first) {
+ atomic_set(&hub_nmi->cpu_owner, cpu);
+ if (atomic_add_unless(&uv_in_nmi, 1, 1))
+ atomic_set(&uv_nmi_cpu, cpu);
+
+ atomic_inc(&hub_nmi->nmi_count);
+ }
+ return first;
+}
+
+/* Check if this is a system NMI event */
+static int uv_check_nmi(struct uv_hub_nmi_s *hub_nmi)
+{
+ int cpu = smp_processor_id();
+ int nmi = 0;
+ int nmi_detected = 0;
+
+ local64_inc(&uv_nmi_count);
+ this_cpu_inc(uv_cpu_nmi.queries);
+
+ do {
+ nmi = atomic_read(&hub_nmi->in_nmi);
+ if (nmi)
+ break;
+
+ if (raw_spin_trylock(&hub_nmi->nmi_lock)) {
+ nmi_detected = uv_test_nmi(hub_nmi);
+
+ /* Check flag for UV external NMI */
+ if (nmi_detected > 0) {
+ uv_set_in_nmi(cpu, hub_nmi);
+ nmi = 1;
+ break;
+ }
+
+ /* A non-PCH node in a hubless system waits for NMI */
+ else if (nmi_detected < 0)
+ goto slave_wait;
+
+ /* MMR/PCH NMI flag is clear */
+ raw_spin_unlock(&hub_nmi->nmi_lock);
+
+ } else {
+
+ /* Wait a moment for the HUB NMI locker to set flag */
+slave_wait: cpu_relax();
+ udelay(uv_nmi_slave_delay);
+
+ /* Re-check hub in_nmi flag */
+ nmi = atomic_read(&hub_nmi->in_nmi);
+ if (nmi)
+ break;
+ }
+
+ /*
+ * Check if this BMC missed setting the MMR NMI flag (or)
+ * UV hubless system where only PCH owner can check flag
+ */
+ if (!nmi) {
+ nmi = atomic_read(&uv_in_nmi);
+ if (nmi)
+ uv_set_in_nmi(cpu, hub_nmi);
+ }
+
+ /* If we're holding the hub lock, release it now */
+ if (nmi_detected < 0)
+ raw_spin_unlock(&hub_nmi->nmi_lock);
+
+ } while (0);
+
+ if (!nmi)
+ local64_inc(&uv_nmi_misses);
+
+ return nmi;
+}
+
+/* Need to reset the NMI MMR register, but only once per hub. */
+static inline void uv_clear_nmi(int cpu)
+{
+ struct uv_hub_nmi_s *hub_nmi = uv_hub_nmi;
+
+ if (cpu == atomic_read(&hub_nmi->cpu_owner)) {
+ atomic_set(&hub_nmi->cpu_owner, -1);
+ atomic_set(&hub_nmi->in_nmi, 0);
+ if (hub_nmi->hub_present)
+ uv_local_mmr_clear_nmi();
+ else
+ uv_reassert_nmi();
+ raw_spin_unlock(&hub_nmi->nmi_lock);
+ }
+}
+
+/* Ping non-responding CPU's attemping to force them into the NMI handler */
+static void uv_nmi_nr_cpus_ping(void)
+{
+ int cpu;
+
+ for_each_cpu(cpu, uv_nmi_cpu_mask)
+ uv_cpu_nmi_per(cpu).pinging = 1;
+
+ apic->send_IPI_mask(uv_nmi_cpu_mask, APIC_DM_NMI);
+}
+
+/* Clean up flags for CPU's that ignored both NMI and ping */
+static void uv_nmi_cleanup_mask(void)
+{
+ int cpu;
+
+ for_each_cpu(cpu, uv_nmi_cpu_mask) {
+ uv_cpu_nmi_per(cpu).pinging = 0;
+ uv_cpu_nmi_per(cpu).state = UV_NMI_STATE_OUT;
+ cpumask_clear_cpu(cpu, uv_nmi_cpu_mask);
+ }
+}
+
+/* Loop waiting as CPU's enter NMI handler */
+static int uv_nmi_wait_cpus(int first)
+{
+ int i, j, k, n = num_online_cpus();
+ int last_k = 0, waiting = 0;
+ int cpu = smp_processor_id();
+
+ if (first) {
+ cpumask_copy(uv_nmi_cpu_mask, cpu_online_mask);
+ k = 0;
+ } else {
+ k = n - cpumask_weight(uv_nmi_cpu_mask);
+ }
+
+ /* PCH NMI causes only one CPU to respond */
+ if (first && uv_pch_intr_now_enabled) {
+ cpumask_clear_cpu(cpu, uv_nmi_cpu_mask);
+ return n - k - 1;
+ }
+
+ udelay(uv_nmi_initial_delay);
+ for (i = 0; i < uv_nmi_retry_count; i++) {
+ int loop_delay = uv_nmi_loop_delay;
+
+ for_each_cpu(j, uv_nmi_cpu_mask) {
+ if (uv_cpu_nmi_per(j).state) {
+ cpumask_clear_cpu(j, uv_nmi_cpu_mask);
+ if (++k >= n)
+ break;
+ }
+ }
+ if (k >= n) { /* all in? */
+ k = n;
+ break;
+ }
+ if (last_k != k) { /* abort if no new CPU's coming in */
+ last_k = k;
+ waiting = 0;
+ } else if (++waiting > uv_nmi_wait_count)
+ break;
+
+ /* Extend delay if waiting only for CPU 0: */
+ if (waiting && (n - k) == 1 &&
+ cpumask_test_cpu(0, uv_nmi_cpu_mask))
+ loop_delay *= 100;
+
+ udelay(loop_delay);
+ }
+ atomic_set(&uv_nmi_cpus_in_nmi, k);
+ return n - k;
+}
+
+/* Wait until all slave CPU's have entered UV NMI handler */
+static void uv_nmi_wait(int master)
+{
+ /* Indicate this CPU is in: */
+ this_cpu_write(uv_cpu_nmi.state, UV_NMI_STATE_IN);
+
+ /* If not the first CPU in (the master), then we are a slave CPU */
+ if (!master)
+ return;
+
+ do {
+ /* Wait for all other CPU's to gather here */
+ if (!uv_nmi_wait_cpus(1))
+ break;
+
+ /* If not all made it in, send IPI NMI to them */
+ pr_alert("UV: Sending NMI IPI to %d CPUs: %*pbl\n",
+ cpumask_weight(uv_nmi_cpu_mask),
+ cpumask_pr_args(uv_nmi_cpu_mask));
+
+ uv_nmi_nr_cpus_ping();
+
+ /* If all CPU's are in, then done */
+ if (!uv_nmi_wait_cpus(0))
+ break;
+
+ pr_alert("UV: %d CPUs not in NMI loop: %*pbl\n",
+ cpumask_weight(uv_nmi_cpu_mask),
+ cpumask_pr_args(uv_nmi_cpu_mask));
+ } while (0);
+
+ pr_alert("UV: %d of %d CPUs in NMI\n",
+ atomic_read(&uv_nmi_cpus_in_nmi), num_online_cpus());
+}
+
+/* Dump Instruction Pointer header */
+static void uv_nmi_dump_cpu_ip_hdr(void)
+{
+ pr_info("\nUV: %4s %6s %-32s %s (Note: PID 0 not listed)\n",
+ "CPU", "PID", "COMMAND", "IP");
+}
+
+/* Dump Instruction Pointer info */
+static void uv_nmi_dump_cpu_ip(int cpu, struct pt_regs *regs)
+{
+ pr_info("UV: %4d %6d %-32.32s %pS",
+ cpu, current->pid, current->comm, (void *)regs->ip);
+}
+
+/*
+ * Dump this CPU's state. If action was set to "kdump" and the crash_kexec
+ * failed, then we provide "dump" as an alternate action. Action "dump" now
+ * also includes the show "ips" (instruction pointers) action whereas the
+ * action "ips" only displays instruction pointers for the non-idle CPU's.
+ * This is an abbreviated form of the "ps" command.
+ */
+static void uv_nmi_dump_state_cpu(int cpu, struct pt_regs *regs)
+{
+ const char *dots = " ................................. ";
+
+ if (cpu == 0)
+ uv_nmi_dump_cpu_ip_hdr();
+
+ if (current->pid != 0 || !uv_nmi_action_is("ips"))
+ uv_nmi_dump_cpu_ip(cpu, regs);
+
+ if (uv_nmi_action_is("dump")) {
+ pr_info("UV:%sNMI process trace for CPU %d\n", dots, cpu);
+ show_regs(regs);
+ }
+
+ this_cpu_write(uv_cpu_nmi.state, UV_NMI_STATE_DUMP_DONE);
+}
+
+/* Trigger a slave CPU to dump it's state */
+static void uv_nmi_trigger_dump(int cpu)
+{
+ int retry = uv_nmi_trigger_delay;
+
+ if (uv_cpu_nmi_per(cpu).state != UV_NMI_STATE_IN)
+ return;
+
+ uv_cpu_nmi_per(cpu).state = UV_NMI_STATE_DUMP;
+ do {
+ cpu_relax();
+ udelay(10);
+ if (uv_cpu_nmi_per(cpu).state
+ != UV_NMI_STATE_DUMP)
+ return;
+ } while (--retry > 0);
+
+ pr_crit("UV: CPU %d stuck in process dump function\n", cpu);
+ uv_cpu_nmi_per(cpu).state = UV_NMI_STATE_DUMP_DONE;
+}
+
+/* Wait until all CPU's ready to exit */
+static void uv_nmi_sync_exit(int master)
+{
+ atomic_dec(&uv_nmi_cpus_in_nmi);
+ if (master) {
+ while (atomic_read(&uv_nmi_cpus_in_nmi) > 0)
+ cpu_relax();
+ atomic_set(&uv_nmi_slave_continue, SLAVE_CLEAR);
+ } else {
+ while (atomic_read(&uv_nmi_slave_continue))
+ cpu_relax();
+ }
+}
+
+/* Current "health" check is to check which CPU's are responsive */
+static void uv_nmi_action_health(int cpu, struct pt_regs *regs, int master)
+{
+ if (master) {
+ int in = atomic_read(&uv_nmi_cpus_in_nmi);
+ int out = num_online_cpus() - in;
+
+ pr_alert("UV: NMI CPU health check (non-responding:%d)\n", out);
+ atomic_set(&uv_nmi_slave_continue, SLAVE_EXIT);
+ } else {
+ while (!atomic_read(&uv_nmi_slave_continue))
+ cpu_relax();
+ }
+ uv_nmi_sync_exit(master);
+}
+
+/* Walk through CPU list and dump state of each */
+static void uv_nmi_dump_state(int cpu, struct pt_regs *regs, int master)
+{
+ if (master) {
+ int tcpu;
+ int ignored = 0;
+ int saved_console_loglevel = console_loglevel;
+
+ pr_alert("UV: tracing %s for %d CPUs from CPU %d\n",
+ uv_nmi_action_is("ips") ? "IPs" : "processes",
+ atomic_read(&uv_nmi_cpus_in_nmi), cpu);
+
+ console_loglevel = uv_nmi_loglevel;
+ atomic_set(&uv_nmi_slave_continue, SLAVE_EXIT);
+ for_each_online_cpu(tcpu) {
+ if (cpumask_test_cpu(tcpu, uv_nmi_cpu_mask))
+ ignored++;
+ else if (tcpu == cpu)
+ uv_nmi_dump_state_cpu(tcpu, regs);
+ else
+ uv_nmi_trigger_dump(tcpu);
+ }
+ if (ignored)
+ pr_alert("UV: %d CPUs ignored NMI\n", ignored);
+
+ console_loglevel = saved_console_loglevel;
+ pr_alert("UV: process trace complete\n");
+ } else {
+ while (!atomic_read(&uv_nmi_slave_continue))
+ cpu_relax();
+ while (this_cpu_read(uv_cpu_nmi.state) != UV_NMI_STATE_DUMP)
+ cpu_relax();
+ uv_nmi_dump_state_cpu(cpu, regs);
+ }
+ uv_nmi_sync_exit(master);
+}
+
+static void uv_nmi_touch_watchdogs(void)
+{
+ touch_softlockup_watchdog_sync();
+ clocksource_touch_watchdog();
+ rcu_cpu_stall_reset();
+ touch_nmi_watchdog();
+}
+
+static atomic_t uv_nmi_kexec_failed;
+
+#if defined(CONFIG_KEXEC_CORE)
+static void uv_nmi_kdump(int cpu, int master, struct pt_regs *regs)
+{
+ /* Call crash to dump system state */
+ if (master) {
+ pr_emerg("UV: NMI executing crash_kexec on CPU%d\n", cpu);
+ crash_kexec(regs);
+
+ pr_emerg("UV: crash_kexec unexpectedly returned, ");
+ atomic_set(&uv_nmi_kexec_failed, 1);
+ if (!kexec_crash_image) {
+ pr_cont("crash kernel not loaded\n");
+ return;
+ }
+ pr_cont("kexec busy, stalling cpus while waiting\n");
+ }
+
+ /* If crash exec fails the slaves should return, otherwise stall */
+ while (atomic_read(&uv_nmi_kexec_failed) == 0)
+ mdelay(10);
+}
+
+#else /* !CONFIG_KEXEC_CORE */
+static inline void uv_nmi_kdump(int cpu, int master, struct pt_regs *regs)
+{
+ if (master)
+ pr_err("UV: NMI kdump: KEXEC not supported in this kernel\n");
+ atomic_set(&uv_nmi_kexec_failed, 1);
+}
+#endif /* !CONFIG_KEXEC_CORE */
+
+#ifdef CONFIG_KGDB
+#ifdef CONFIG_KGDB_KDB
+static inline int uv_nmi_kdb_reason(void)
+{
+ return KDB_REASON_SYSTEM_NMI;
+}
+#else /* !CONFIG_KGDB_KDB */
+static inline int uv_nmi_kdb_reason(void)
+{
+ /* Ensure user is expecting to attach gdb remote */
+ if (uv_nmi_action_is("kgdb"))
+ return 0;
+
+ pr_err("UV: NMI error: KDB is not enabled in this kernel\n");
+ return -1;
+}
+#endif /* CONFIG_KGDB_KDB */
+
+/*
+ * Call KGDB/KDB from NMI handler
+ *
+ * Note that if both KGDB and KDB are configured, then the action of 'kgdb' or
+ * 'kdb' has no affect on which is used. See the KGDB documention for further
+ * information.
+ */
+static void uv_call_kgdb_kdb(int cpu, struct pt_regs *regs, int master)
+{
+ if (master) {
+ int reason = uv_nmi_kdb_reason();
+ int ret;
+
+ if (reason < 0)
+ return;
+
+ /* Call KGDB NMI handler as MASTER */
+ ret = kgdb_nmicallin(cpu, X86_TRAP_NMI, regs, reason,
+ &uv_nmi_slave_continue);
+ if (ret) {
+ pr_alert("KGDB returned error, is kgdboc set?\n");
+ atomic_set(&uv_nmi_slave_continue, SLAVE_EXIT);
+ }
+ } else {
+ /* Wait for KGDB signal that it's ready for slaves to enter */
+ int sig;
+
+ do {
+ cpu_relax();
+ sig = atomic_read(&uv_nmi_slave_continue);
+ } while (!sig);
+
+ /* Call KGDB as slave */
+ if (sig == SLAVE_CONTINUE)
+ kgdb_nmicallback(cpu, regs);
+ }
+ uv_nmi_sync_exit(master);
+}
+
+#else /* !CONFIG_KGDB */
+static inline void uv_call_kgdb_kdb(int cpu, struct pt_regs *regs, int master)
+{
+ pr_err("UV: NMI error: KGDB is not enabled in this kernel\n");
+}
+#endif /* !CONFIG_KGDB */
+
+/*
+ * UV NMI handler
+ */
+static int uv_handle_nmi(unsigned int reason, struct pt_regs *regs)
+{
+ struct uv_hub_nmi_s *hub_nmi = uv_hub_nmi;
+ int cpu = smp_processor_id();
+ int master = 0;
+ unsigned long flags;
+
+ local_irq_save(flags);
+
+ /* If not a UV System NMI, ignore */
+ if (!this_cpu_read(uv_cpu_nmi.pinging) && !uv_check_nmi(hub_nmi)) {
+ local_irq_restore(flags);
+ return NMI_DONE;
+ }
+
+ /* Indicate we are the first CPU into the NMI handler */
+ master = (atomic_read(&uv_nmi_cpu) == cpu);
+
+ /* If NMI action is "kdump", then attempt to do it */
+ if (uv_nmi_action_is("kdump")) {
+ uv_nmi_kdump(cpu, master, regs);
+
+ /* Unexpected return, revert action to "dump" */
+ if (master)
+ strncpy(uv_nmi_action, "dump", strlen(uv_nmi_action));
+ }
+
+ /* Pause as all CPU's enter the NMI handler */
+ uv_nmi_wait(master);
+
+ /* Process actions other than "kdump": */
+ if (uv_nmi_action_is("health")) {
+ uv_nmi_action_health(cpu, regs, master);
+ } else if (uv_nmi_action_is("ips") || uv_nmi_action_is("dump")) {
+ uv_nmi_dump_state(cpu, regs, master);
+ } else if (uv_nmi_action_is("kdb") || uv_nmi_action_is("kgdb")) {
+ uv_call_kgdb_kdb(cpu, regs, master);
+ } else {
+ if (master)
+ pr_alert("UV: unknown NMI action: %s\n", uv_nmi_action);
+ uv_nmi_sync_exit(master);
+ }
+
+ /* Clear per_cpu "in_nmi" flag */
+ this_cpu_write(uv_cpu_nmi.state, UV_NMI_STATE_OUT);
+
+ /* Clear MMR NMI flag on each hub */
+ uv_clear_nmi(cpu);
+
+ /* Clear global flags */
+ if (master) {
+ if (cpumask_weight(uv_nmi_cpu_mask))
+ uv_nmi_cleanup_mask();
+ atomic_set(&uv_nmi_cpus_in_nmi, -1);
+ atomic_set(&uv_nmi_cpu, -1);
+ atomic_set(&uv_in_nmi, 0);
+ atomic_set(&uv_nmi_kexec_failed, 0);
+ atomic_set(&uv_nmi_slave_continue, SLAVE_CLEAR);
+ }
+
+ uv_nmi_touch_watchdogs();
+ local_irq_restore(flags);
+
+ return NMI_HANDLED;
+}
+
+/*
+ * NMI handler for pulling in CPU's when perf events are grabbing our NMI
+ */
+static int uv_handle_nmi_ping(unsigned int reason, struct pt_regs *regs)
+{
+ int ret;
+
+ this_cpu_inc(uv_cpu_nmi.queries);
+ if (!this_cpu_read(uv_cpu_nmi.pinging)) {
+ local64_inc(&uv_nmi_ping_misses);
+ return NMI_DONE;
+ }
+
+ this_cpu_inc(uv_cpu_nmi.pings);
+ local64_inc(&uv_nmi_ping_count);
+ ret = uv_handle_nmi(reason, regs);
+ this_cpu_write(uv_cpu_nmi.pinging, 0);
+ return ret;
+}
+
+static void uv_register_nmi_notifier(void)
+{
+ if (register_nmi_handler(NMI_UNKNOWN, uv_handle_nmi, 0, "uv"))
+ pr_warn("UV: NMI handler failed to register\n");
+
+ if (register_nmi_handler(NMI_LOCAL, uv_handle_nmi_ping, 0, "uvping"))
+ pr_warn("UV: PING NMI handler failed to register\n");
+}
+
+void uv_nmi_init(void)
+{
+ unsigned int value;
+
+ /*
+ * Unmask NMI on all CPU's
+ */
+ value = apic_read(APIC_LVT1) | APIC_DM_NMI;
+ value &= ~APIC_LVT_MASKED;
+ apic_write(APIC_LVT1, value);
+}
+
+/* Setup HUB NMI info */
+static void __init uv_nmi_setup_common(bool hubbed)
+{
+ int size = sizeof(void *) * (1 << NODES_SHIFT);
+ int cpu;
+
+ uv_hub_nmi_list = kzalloc(size, GFP_KERNEL);
+ nmi_debug("UV: NMI hub list @ 0x%p (%d)\n", uv_hub_nmi_list, size);
+ BUG_ON(!uv_hub_nmi_list);
+ size = sizeof(struct uv_hub_nmi_s);
+ for_each_present_cpu(cpu) {
+ int nid = cpu_to_node(cpu);
+ if (uv_hub_nmi_list[nid] == NULL) {
+ uv_hub_nmi_list[nid] = kzalloc_node(size,
+ GFP_KERNEL, nid);
+ BUG_ON(!uv_hub_nmi_list[nid]);
+ raw_spin_lock_init(&(uv_hub_nmi_list[nid]->nmi_lock));
+ atomic_set(&uv_hub_nmi_list[nid]->cpu_owner, -1);
+ uv_hub_nmi_list[nid]->hub_present = hubbed;
+ uv_hub_nmi_list[nid]->pch_owner = (nid == 0);
+ }
+ uv_hub_nmi_per(cpu) = uv_hub_nmi_list[nid];
+ }
+ BUG_ON(!alloc_cpumask_var(&uv_nmi_cpu_mask, GFP_KERNEL));
+}
+
+/* Setup for UV Hub systems */
+void __init uv_nmi_setup(void)
+{
+ uv_nmi_setup_mmrs();
+ uv_nmi_setup_common(true);
+ uv_register_nmi_notifier();
+ pr_info("UV: Hub NMI enabled\n");
+}
+
+/* Setup for UV Hubless systems */
+void __init uv_nmi_setup_hubless(void)
+{
+ uv_nmi_setup_common(false);
+ pch_base = xlate_dev_mem_ptr(PCH_PCR_GPIO_1_BASE);
+ nmi_debug("UV: PCH base:%p from 0x%lx, GPP_D_0\n",
+ pch_base, PCH_PCR_GPIO_1_BASE);
+ if (uv_pch_init_enable)
+ uv_init_hubless_pch_d0();
+ uv_init_hubless_pch_io(GPI_NMI_ENA_GPP_D_0,
+ STS_GPP_D_0_MASK, STS_GPP_D_0_MASK);
+ uv_nmi_setup_hubless_intr();
+ /* Ensure NMI enabled in Processor Interface Reg: */
+ uv_reassert_nmi();
+ uv_register_nmi_notifier();
+ pr_info("UV: Hubless NMI enabled\n");
+}
diff --git a/arch/x86/platform/uv/uv_sysfs.c b/arch/x86/platform/uv/uv_sysfs.c
new file mode 100644
index 000000000..e9da9ebd9
--- /dev/null
+++ b/arch/x86/platform/uv/uv_sysfs.c
@@ -0,0 +1,76 @@
+/*
+ * This file supports the /sys/firmware/sgi_uv interfaces for SGI UV.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Copyright (c) 2008 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) Russ Anderson
+ */
+
+#include <linux/device.h>
+#include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+
+struct kobject *sgi_uv_kobj;
+
+static ssize_t partition_id_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%ld\n", sn_partition_id);
+}
+
+static ssize_t coherence_id_show(struct kobject *kobj,
+ struct kobj_attribute *attr, char *buf)
+{
+ return snprintf(buf, PAGE_SIZE, "%ld\n", uv_partition_coherence_id());
+}
+
+static struct kobj_attribute partition_id_attr =
+ __ATTR(partition_id, S_IRUGO, partition_id_show, NULL);
+
+static struct kobj_attribute coherence_id_attr =
+ __ATTR(coherence_id, S_IRUGO, coherence_id_show, NULL);
+
+
+static int __init sgi_uv_sysfs_init(void)
+{
+ unsigned long ret;
+
+ if (!is_uv_system())
+ return -ENODEV;
+
+ if (!sgi_uv_kobj)
+ sgi_uv_kobj = kobject_create_and_add("sgi_uv", firmware_kobj);
+ if (!sgi_uv_kobj) {
+ printk(KERN_WARNING "kobject_create_and_add sgi_uv failed\n");
+ return -EINVAL;
+ }
+
+ ret = sysfs_create_file(sgi_uv_kobj, &partition_id_attr.attr);
+ if (ret) {
+ printk(KERN_WARNING "sysfs_create_file partition_id failed\n");
+ return ret;
+ }
+
+ ret = sysfs_create_file(sgi_uv_kobj, &coherence_id_attr.attr);
+ if (ret) {
+ printk(KERN_WARNING "sysfs_create_file coherence_id failed\n");
+ return ret;
+ }
+
+ return 0;
+}
+
+device_initcall(sgi_uv_sysfs_init);
diff --git a/arch/x86/platform/uv/uv_time.c b/arch/x86/platform/uv/uv_time.c
new file mode 100644
index 000000000..a36b368ee
--- /dev/null
+++ b/arch/x86/platform/uv/uv_time.c
@@ -0,0 +1,416 @@
+/*
+ * SGI RTC clock/timer routines.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Copyright (c) 2009-2013 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) Dimitri Sivanich
+ */
+#include <linux/clockchips.h>
+#include <linux/slab.h>
+
+#include <asm/uv/uv_mmrs.h>
+#include <asm/uv/uv_hub.h>
+#include <asm/uv/bios.h>
+#include <asm/uv/uv.h>
+#include <asm/apic.h>
+#include <asm/cpu.h>
+
+#define RTC_NAME "sgi_rtc"
+
+static u64 uv_read_rtc(struct clocksource *cs);
+static int uv_rtc_next_event(unsigned long, struct clock_event_device *);
+static int uv_rtc_shutdown(struct clock_event_device *evt);
+
+static struct clocksource clocksource_uv = {
+ .name = RTC_NAME,
+ .rating = 299,
+ .read = uv_read_rtc,
+ .mask = (u64)UVH_RTC_REAL_TIME_CLOCK_MASK,
+ .flags = CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static struct clock_event_device clock_event_device_uv = {
+ .name = RTC_NAME,
+ .features = CLOCK_EVT_FEAT_ONESHOT,
+ .shift = 20,
+ .rating = 400,
+ .irq = -1,
+ .set_next_event = uv_rtc_next_event,
+ .set_state_shutdown = uv_rtc_shutdown,
+ .event_handler = NULL,
+};
+
+static DEFINE_PER_CPU(struct clock_event_device, cpu_ced);
+
+/* There is one of these allocated per node */
+struct uv_rtc_timer_head {
+ spinlock_t lock;
+ /* next cpu waiting for timer, local node relative: */
+ int next_cpu;
+ /* number of cpus on this node: */
+ int ncpus;
+ struct {
+ int lcpu; /* systemwide logical cpu number */
+ u64 expires; /* next timer expiration for this cpu */
+ } cpu[1];
+};
+
+/*
+ * Access to uv_rtc_timer_head via blade id.
+ */
+static struct uv_rtc_timer_head **blade_info __read_mostly;
+
+static int uv_rtc_evt_enable;
+
+/*
+ * Hardware interface routines
+ */
+
+/* Send IPIs to another node */
+static void uv_rtc_send_IPI(int cpu)
+{
+ unsigned long apicid, val;
+ int pnode;
+
+ apicid = cpu_physical_id(cpu);
+ pnode = uv_apicid_to_pnode(apicid);
+ apicid |= uv_apicid_hibits;
+ val = (1UL << UVH_IPI_INT_SEND_SHFT) |
+ (apicid << UVH_IPI_INT_APIC_ID_SHFT) |
+ (X86_PLATFORM_IPI_VECTOR << UVH_IPI_INT_VECTOR_SHFT);
+
+ uv_write_global_mmr64(pnode, UVH_IPI_INT, val);
+}
+
+/* Check for an RTC interrupt pending */
+static int uv_intr_pending(int pnode)
+{
+ if (is_uv1_hub())
+ return uv_read_global_mmr64(pnode, UVH_EVENT_OCCURRED0) &
+ UV1H_EVENT_OCCURRED0_RTC1_MASK;
+ else if (is_uvx_hub())
+ return uv_read_global_mmr64(pnode, UVXH_EVENT_OCCURRED2) &
+ UVXH_EVENT_OCCURRED2_RTC_1_MASK;
+ return 0;
+}
+
+/* Setup interrupt and return non-zero if early expiration occurred. */
+static int uv_setup_intr(int cpu, u64 expires)
+{
+ u64 val;
+ unsigned long apicid = cpu_physical_id(cpu) | uv_apicid_hibits;
+ int pnode = uv_cpu_to_pnode(cpu);
+
+ uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG,
+ UVH_RTC1_INT_CONFIG_M_MASK);
+ uv_write_global_mmr64(pnode, UVH_INT_CMPB, -1L);
+
+ if (is_uv1_hub())
+ uv_write_global_mmr64(pnode, UVH_EVENT_OCCURRED0_ALIAS,
+ UV1H_EVENT_OCCURRED0_RTC1_MASK);
+ else
+ uv_write_global_mmr64(pnode, UVXH_EVENT_OCCURRED2_ALIAS,
+ UVXH_EVENT_OCCURRED2_RTC_1_MASK);
+
+ val = (X86_PLATFORM_IPI_VECTOR << UVH_RTC1_INT_CONFIG_VECTOR_SHFT) |
+ ((u64)apicid << UVH_RTC1_INT_CONFIG_APIC_ID_SHFT);
+
+ /* Set configuration */
+ uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG, val);
+ /* Initialize comparator value */
+ uv_write_global_mmr64(pnode, UVH_INT_CMPB, expires);
+
+ if (uv_read_rtc(NULL) <= expires)
+ return 0;
+
+ return !uv_intr_pending(pnode);
+}
+
+/*
+ * Per-cpu timer tracking routines
+ */
+
+static __init void uv_rtc_deallocate_timers(void)
+{
+ int bid;
+
+ for_each_possible_blade(bid) {
+ kfree(blade_info[bid]);
+ }
+ kfree(blade_info);
+}
+
+/* Allocate per-node list of cpu timer expiration times. */
+static __init int uv_rtc_allocate_timers(void)
+{
+ int cpu;
+
+ blade_info = kcalloc(uv_possible_blades, sizeof(void *), GFP_KERNEL);
+ if (!blade_info)
+ return -ENOMEM;
+
+ for_each_present_cpu(cpu) {
+ int nid = cpu_to_node(cpu);
+ int bid = uv_cpu_to_blade_id(cpu);
+ int bcpu = uv_cpu_blade_processor_id(cpu);
+ struct uv_rtc_timer_head *head = blade_info[bid];
+
+ if (!head) {
+ head = kmalloc_node(sizeof(struct uv_rtc_timer_head) +
+ (uv_blade_nr_possible_cpus(bid) *
+ 2 * sizeof(u64)),
+ GFP_KERNEL, nid);
+ if (!head) {
+ uv_rtc_deallocate_timers();
+ return -ENOMEM;
+ }
+ spin_lock_init(&head->lock);
+ head->ncpus = uv_blade_nr_possible_cpus(bid);
+ head->next_cpu = -1;
+ blade_info[bid] = head;
+ }
+
+ head->cpu[bcpu].lcpu = cpu;
+ head->cpu[bcpu].expires = ULLONG_MAX;
+ }
+
+ return 0;
+}
+
+/* Find and set the next expiring timer. */
+static void uv_rtc_find_next_timer(struct uv_rtc_timer_head *head, int pnode)
+{
+ u64 lowest = ULLONG_MAX;
+ int c, bcpu = -1;
+
+ head->next_cpu = -1;
+ for (c = 0; c < head->ncpus; c++) {
+ u64 exp = head->cpu[c].expires;
+ if (exp < lowest) {
+ bcpu = c;
+ lowest = exp;
+ }
+ }
+ if (bcpu >= 0) {
+ head->next_cpu = bcpu;
+ c = head->cpu[bcpu].lcpu;
+ if (uv_setup_intr(c, lowest))
+ /* If we didn't set it up in time, trigger */
+ uv_rtc_send_IPI(c);
+ } else {
+ uv_write_global_mmr64(pnode, UVH_RTC1_INT_CONFIG,
+ UVH_RTC1_INT_CONFIG_M_MASK);
+ }
+}
+
+/*
+ * Set expiration time for current cpu.
+ *
+ * Returns 1 if we missed the expiration time.
+ */
+static int uv_rtc_set_timer(int cpu, u64 expires)
+{
+ int pnode = uv_cpu_to_pnode(cpu);
+ int bid = uv_cpu_to_blade_id(cpu);
+ struct uv_rtc_timer_head *head = blade_info[bid];
+ int bcpu = uv_cpu_blade_processor_id(cpu);
+ u64 *t = &head->cpu[bcpu].expires;
+ unsigned long flags;
+ int next_cpu;
+
+ spin_lock_irqsave(&head->lock, flags);
+
+ next_cpu = head->next_cpu;
+ *t = expires;
+
+ /* Will this one be next to go off? */
+ if (next_cpu < 0 || bcpu == next_cpu ||
+ expires < head->cpu[next_cpu].expires) {
+ head->next_cpu = bcpu;
+ if (uv_setup_intr(cpu, expires)) {
+ *t = ULLONG_MAX;
+ uv_rtc_find_next_timer(head, pnode);
+ spin_unlock_irqrestore(&head->lock, flags);
+ return -ETIME;
+ }
+ }
+
+ spin_unlock_irqrestore(&head->lock, flags);
+ return 0;
+}
+
+/*
+ * Unset expiration time for current cpu.
+ *
+ * Returns 1 if this timer was pending.
+ */
+static int uv_rtc_unset_timer(int cpu, int force)
+{
+ int pnode = uv_cpu_to_pnode(cpu);
+ int bid = uv_cpu_to_blade_id(cpu);
+ struct uv_rtc_timer_head *head = blade_info[bid];
+ int bcpu = uv_cpu_blade_processor_id(cpu);
+ u64 *t = &head->cpu[bcpu].expires;
+ unsigned long flags;
+ int rc = 0;
+
+ spin_lock_irqsave(&head->lock, flags);
+
+ if ((head->next_cpu == bcpu && uv_read_rtc(NULL) >= *t) || force)
+ rc = 1;
+
+ if (rc) {
+ *t = ULLONG_MAX;
+ /* Was the hardware setup for this timer? */
+ if (head->next_cpu == bcpu)
+ uv_rtc_find_next_timer(head, pnode);
+ }
+
+ spin_unlock_irqrestore(&head->lock, flags);
+
+ return rc;
+}
+
+
+/*
+ * Kernel interface routines.
+ */
+
+/*
+ * Read the RTC.
+ *
+ * Starting with HUB rev 2.0, the UV RTC register is replicated across all
+ * cachelines of it's own page. This allows faster simultaneous reads
+ * from a given socket.
+ */
+static u64 uv_read_rtc(struct clocksource *cs)
+{
+ unsigned long offset;
+
+ if (uv_get_min_hub_revision_id() == 1)
+ offset = 0;
+ else
+ offset = (uv_blade_processor_id() * L1_CACHE_BYTES) % PAGE_SIZE;
+
+ return (u64)uv_read_local_mmr(UVH_RTC | offset);
+}
+
+/*
+ * Program the next event, relative to now
+ */
+static int uv_rtc_next_event(unsigned long delta,
+ struct clock_event_device *ced)
+{
+ int ced_cpu = cpumask_first(ced->cpumask);
+
+ return uv_rtc_set_timer(ced_cpu, delta + uv_read_rtc(NULL));
+}
+
+/*
+ * Shutdown the RTC timer
+ */
+static int uv_rtc_shutdown(struct clock_event_device *evt)
+{
+ int ced_cpu = cpumask_first(evt->cpumask);
+
+ uv_rtc_unset_timer(ced_cpu, 1);
+ return 0;
+}
+
+static void uv_rtc_interrupt(void)
+{
+ int cpu = smp_processor_id();
+ struct clock_event_device *ced = &per_cpu(cpu_ced, cpu);
+
+ if (!ced || !ced->event_handler)
+ return;
+
+ if (uv_rtc_unset_timer(cpu, 0) != 1)
+ return;
+
+ ced->event_handler(ced);
+}
+
+static int __init uv_enable_evt_rtc(char *str)
+{
+ uv_rtc_evt_enable = 1;
+
+ return 1;
+}
+__setup("uvrtcevt", uv_enable_evt_rtc);
+
+static __init void uv_rtc_register_clockevents(struct work_struct *dummy)
+{
+ struct clock_event_device *ced = this_cpu_ptr(&cpu_ced);
+
+ *ced = clock_event_device_uv;
+ ced->cpumask = cpumask_of(smp_processor_id());
+ clockevents_register_device(ced);
+}
+
+static __init int uv_rtc_setup_clock(void)
+{
+ int rc;
+
+ if (!is_uv_system())
+ return -ENODEV;
+
+ rc = clocksource_register_hz(&clocksource_uv, sn_rtc_cycles_per_second);
+ if (rc)
+ printk(KERN_INFO "UV RTC clocksource failed rc %d\n", rc);
+ else
+ printk(KERN_INFO "UV RTC clocksource registered freq %lu MHz\n",
+ sn_rtc_cycles_per_second/(unsigned long)1E6);
+
+ if (rc || !uv_rtc_evt_enable || x86_platform_ipi_callback)
+ return rc;
+
+ /* Setup and register clockevents */
+ rc = uv_rtc_allocate_timers();
+ if (rc)
+ goto error;
+
+ x86_platform_ipi_callback = uv_rtc_interrupt;
+
+ clock_event_device_uv.mult = div_sc(sn_rtc_cycles_per_second,
+ NSEC_PER_SEC, clock_event_device_uv.shift);
+
+ clock_event_device_uv.min_delta_ns = NSEC_PER_SEC /
+ sn_rtc_cycles_per_second;
+ clock_event_device_uv.min_delta_ticks = 1;
+
+ clock_event_device_uv.max_delta_ns = clocksource_uv.mask *
+ (NSEC_PER_SEC / sn_rtc_cycles_per_second);
+ clock_event_device_uv.max_delta_ticks = clocksource_uv.mask;
+
+ rc = schedule_on_each_cpu(uv_rtc_register_clockevents);
+ if (rc) {
+ x86_platform_ipi_callback = NULL;
+ uv_rtc_deallocate_timers();
+ goto error;
+ }
+
+ printk(KERN_INFO "UV RTC clockevents registered\n");
+
+ return 0;
+
+error:
+ clocksource_unregister(&clocksource_uv);
+ printk(KERN_INFO "UV RTC clockevents failed rc %d\n", rc);
+
+ return rc;
+}
+arch_initcall(uv_rtc_setup_clock);