From 76cb841cb886eef6b3bee341a2266c76578724ad Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Mon, 6 May 2024 03:02:30 +0200 Subject: Adding upstream version 4.19.249. Signed-off-by: Daniel Baumann --- Documentation/s390/00-INDEX | 28 + Documentation/s390/3270.ChangeLog | 44 + Documentation/s390/3270.txt | 271 +++++ Documentation/s390/CommonIO | 125 ++ Documentation/s390/DASD | 73 ++ Documentation/s390/Debugging390.txt | 2142 +++++++++++++++++++++++++++++++++++ Documentation/s390/cds.txt | 472 ++++++++ Documentation/s390/config3270.sh | 76 ++ Documentation/s390/driver-model.txt | 287 +++++ Documentation/s390/monreader.txt | 197 ++++ Documentation/s390/qeth.txt | 50 + Documentation/s390/s390dbf.txt | 667 +++++++++++ Documentation/s390/vfio-ccw.txt | 300 +++++ Documentation/s390/zfcpdump.txt | 48 + 14 files changed, 4780 insertions(+) create mode 100644 Documentation/s390/00-INDEX create mode 100644 Documentation/s390/3270.ChangeLog create mode 100644 Documentation/s390/3270.txt create mode 100644 Documentation/s390/CommonIO create mode 100644 Documentation/s390/DASD create mode 100644 Documentation/s390/Debugging390.txt create mode 100644 Documentation/s390/cds.txt create mode 100644 Documentation/s390/config3270.sh create mode 100644 Documentation/s390/driver-model.txt create mode 100644 Documentation/s390/monreader.txt create mode 100644 Documentation/s390/qeth.txt create mode 100644 Documentation/s390/s390dbf.txt create mode 100644 Documentation/s390/vfio-ccw.txt create mode 100644 Documentation/s390/zfcpdump.txt (limited to 'Documentation/s390') diff --git a/Documentation/s390/00-INDEX b/Documentation/s390/00-INDEX new file mode 100644 index 000000000..317f0378a --- /dev/null +++ b/Documentation/s390/00-INDEX @@ -0,0 +1,28 @@ +00-INDEX + - this file. +3270.ChangeLog + - ChangeLog for the UTS Global 3270-support patch (outdated). +3270.txt + - how to use the IBM 3270 display system support. +cds.txt + - s390 common device support (common I/O layer). +CommonIO + - common I/O layer command line parameters, procfs and debugfs entries +config3270.sh + - example configuration for 3270 devices. +DASD + - information on the DASD disk device driver. +Debugging390.txt + - hints for debugging on s390 systems. +driver-model.txt + - information on s390 devices and the driver model. +monreader.txt + - information on accessing the z/VM monitor stream from Linux. +qeth.txt + - HiperSockets Bridge Port Support. +s390dbf.txt + - information on using the s390 debug feature. +vfio-ccw.txt + information on the vfio-ccw I/O subchannel driver. +zfcpdump.txt + - information on the s390 SCSI dump tool. diff --git a/Documentation/s390/3270.ChangeLog b/Documentation/s390/3270.ChangeLog new file mode 100644 index 000000000..031c36081 --- /dev/null +++ b/Documentation/s390/3270.ChangeLog @@ -0,0 +1,44 @@ +ChangeLog for the UTS Global 3270-support patch + +Sep 2002: Get bootup colors right on 3270 console + * In tubttybld.c, substantially revise ESC processing so that + ESC sequences (especially coloring ones) and the strings + they affect work as right as 3270 can get them. Also, set + screen height to omit the two rows used for input area, in + tty3270_open() in tubtty.c. + +Sep 2002: Dynamically get 3270 input buffer + * Oversize 3270 screen widths may exceed GEOM_MAXINPLEN columns, + so get input-area buffer dynamically when sizing the device in + tubmakemin() in tuball.c (if it's the console) or tty3270_open() + in tubtty.c (if needed). Change tubp->tty_input to be a + pointer rather than an array, in tubio.h. + +Sep 2002: Fix tubfs kmalloc()s + * Do read and write lengths correctly in fs3270_read() + and fs3270_write(), whilst never asking kmalloc() + for more than 0x800 bytes. Affects tubfs.c and tubio.h. + +Sep 2002: Recognize 3270 control unit type 3174 + * Recognize control-unit type 0x3174 as well as 0x327?. + The IBM 2047 device emulates a 3174 control unit. + Modularize control-unit recognition in tuball.c by + adding and invoking new tub3270_is_ours(). + +Apr 2002: Fix 3270 console reboot loop + * (Belated log entry) Fixed reboot loop if 3270 console, + in tubtty.c:ttu3270_bh(). + +Feb 6, 2001: + * This changelog is new + * tub3270 now supports 3270 console: + Specify y for CONFIG_3270 and y for CONFIG_3270_CONSOLE. + Support for 3215 will not appear if 3270 console support + is chosen. + NOTE: The default is 3270 console support, NOT 3215. + * the components are remodularized: added source modules are + tubttybld.c and tubttyscl.c, for screen-building code and + scroll-timeout code. + * tub3270 source for this (2.4.0) version is #ifdeffed to + build with both 2.4.0 and 2.2.16.2. + * color support and minimal other ESC-sequence support is added. diff --git a/Documentation/s390/3270.txt b/Documentation/s390/3270.txt new file mode 100644 index 000000000..7c715de99 --- /dev/null +++ b/Documentation/s390/3270.txt @@ -0,0 +1,271 @@ +IBM 3270 Display System support + +This file describes the driver that supports local channel attachment +of IBM 3270 devices. It consists of three sections: + * Introduction + * Installation + * Operation + + +INTRODUCTION. + +This paper describes installing and operating 3270 devices under +Linux/390. A 3270 device is a block-mode rows-and-columns terminal of +which I'm sure hundreds of millions were sold by IBM and clonemakers +twenty and thirty years ago. + +You may have 3270s in-house and not know it. If you're using the +VM-ESA operating system, define a 3270 to your virtual machine by using +the command "DEF GRAF " This paper presumes you will be +defining four 3270s with the CP/CMS commands + + DEF GRAF 620 + DEF GRAF 621 + DEF GRAF 622 + DEF GRAF 623 + +Your network connection from VM-ESA allows you to use x3270, tn3270, or +another 3270 emulator, started from an xterm window on your PC or +workstation. With the DEF GRAF command, an application such as xterm, +and this Linux-390 3270 driver, you have another way of talking to your +Linux box. + +This paper covers installation of the driver and operation of a +dialed-in x3270. + + +INSTALLATION. + +You install the driver by installing a patch, doing a kernel build, and +running the configuration script (config3270.sh, in this directory). + +WARNING: If you are using 3270 console support, you must rerun the +configuration script every time you change the console's address (perhaps +by using the condev= parameter in silo's /boot/parmfile). More precisely, +you should rerun the configuration script every time your set of 3270s, +including the console 3270, changes subchannel identifier relative to +one another. ReIPL as soon as possible after running the configuration +script and the resulting /tmp/mkdev3270. + +If you have chosen to make tub3270 a module, you add a line to a +configuration file under /etc/modprobe.d/. If you are working on a VM +virtual machine, you can use DEF GRAF to define virtual 3270 devices. + +You may generate both 3270 and 3215 console support, or one or the +other, or neither. If you generate both, the console type under VM is +not changed. Use #CP Q TERM to see what the current console type is. +Use #CP TERM CONMODE 3270 to change it to 3270. If you generate only +3270 console support, then the driver automatically converts your console +at boot time to a 3270 if it is a 3215. + +In brief, these are the steps: + 1. Install the tub3270 patch + 2. (If a module) add a line to a file in /etc/modprobe.d/*.conf + 3. (If VM) define devices with DEF GRAF + 4. Reboot + 5. Configure + +To test that everything works, assuming VM and x3270, + 1. Bring up an x3270 window. + 2. Use the DIAL command in that window. + 3. You should immediately see a Linux login screen. + +Here are the installation steps in detail: + + 1. The 3270 driver is a part of the official Linux kernel + source. Build a tree with the kernel source and any necessary + patches. Then do + make oldconfig + (If you wish to disable 3215 console support, edit + .config; change CONFIG_TN3215's value to "n"; + and rerun "make oldconfig".) + make image + make modules + make modules_install + + 2. (Perform this step only if you have configured tub3270 as a + module.) Add a line to a file /etc/modprobe.d/*.conf to automatically + load the driver when it's needed. With this line added, you will see + login prompts appear on your 3270s as soon as boot is complete (or + with emulated 3270s, as soon as you dial into your vm guest using the + command "DIAL "). Since the line-mode major number is + 227, the line to add should be: + alias char-major-227 tub3270 + + 3. Define graphic devices to your vm guest machine, if you + haven't already. Define them before you reboot (reipl): + DEFINE GRAF 620 + DEFINE GRAF 621 + DEFINE GRAF 622 + DEFINE GRAF 623 + + 4. Reboot. The reboot process scans hardware devices, including + 3270s, and this enables the tub3270 driver once loaded to respond + correctly to the configuration requests of the next step. If + you have chosen 3270 console support, your console now behaves + as a 3270, not a 3215. + + 5. Run the 3270 configuration script config3270. It is + distributed in this same directory, Documentation/s390, as + config3270.sh. Inspect the output script it produces, + /tmp/mkdev3270, and then run that script. This will create the + necessary character special device files and make the necessary + changes to /etc/inittab. + + Then notify /sbin/init that /etc/inittab has changed, by issuing + the telinit command with the q operand: + cd Documentation/s390 + sh config3270.sh + sh /tmp/mkdev3270 + telinit q + + This should be sufficient for your first time. If your 3270 + configuration has changed and you're reusing config3270, you + should follow these steps: + Change 3270 configuration + Reboot + Run config3270 and /tmp/mkdev3270 + Reboot + +Here are the testing steps in detail: + + 1. Bring up an x3270 window, or use an actual hardware 3278 or + 3279, or use the 3270 emulator of your choice. You would be + running the emulator on your PC or workstation. You would use + the command, for example, + x3270 vm-esa-domain-name & + if you wanted a 3278 Model 4 with 43 rows of 80 columns, the + default model number. The driver does not take advantage of + extended attributes. + + The screen you should now see contains a VM logo with input + lines near the bottom. Use TAB to move to the bottom line, + probably labeled "COMMAND ===>". + + 2. Use the DIAL command instead of the LOGIN command to connect + to one of the virtual 3270s you defined with the DEF GRAF + commands: + dial my-vm-guest-name + + 3. You should immediately see a login prompt from your + Linux-390 operating system. If that does not happen, you would + see instead the line "DIALED TO my-vm-guest-name 0620". + + To troubleshoot: do these things. + + A. Is the driver loaded? Use the lsmod command (no operands) + to find out. Probably it isn't. Try loading it manually, with + the command "insmod tub3270". Does that command give error + messages? Ha! There's your problem. + + B. Is the /etc/inittab file modified as in installation step 3 + above? Use the grep command to find out; for instance, issue + "grep 3270 /etc/inittab". Nothing found? There's your + problem! + + C. Are the device special files created, as in installation + step 2 above? Use the ls -l command to find out; for instance, + issue "ls -l /dev/3270/tty620". The output should start with the + letter "c" meaning character device and should contain "227, 1" + just to the left of the device name. No such file? no "c"? + Wrong major number? Wrong minor number? There's your + problem! + + D. Do you get the message + "HCPDIA047E my-vm-guest-name 0620 does not exist"? + If so, you must issue the command "DEF GRAF 620" from your VM + 3215 console and then reboot the system. + + + +OPERATION. + +The driver defines three areas on the 3270 screen: the log area, the +input area, and the status area. + +The log area takes up all but the bottom two lines of the screen. The +driver writes terminal output to it, starting at the top line and going +down. When it fills, the status area changes from "Linux Running" to +"Linux More...". After a scrolling timeout of (default) 5 sec, the +screen clears and more output is written, from the top down. + +The input area extends from the beginning of the second-to-last screen +line to the start of the status area. You type commands in this area +and hit ENTER to execute them. + +The status area initializes to "Linux Running" to give you a warm +fuzzy feeling. When the log area fills up and output awaits, it +changes to "Linux More...". At this time you can do several things or +nothing. If you do nothing, the screen will clear in (default) 5 sec +and more output will appear. You may hit ENTER with nothing typed in +the input area to toggle between "Linux More..." and "Linux Holding", +which indicates no scrolling will occur. (If you hit ENTER with "Linux +Running" and nothing typed, the application receives a newline.) + +You may change the scrolling timeout value. For example, the following +command line: + echo scrolltime=60 > /proc/tty/driver/tty3270 +changes the scrolling timeout value to 60 sec. Set scrolltime to 0 if +you wish to prevent scrolling entirely. + +Other things you may do when the log area fills up are: hit PA2 to +clear the log area and write more output to it, or hit CLEAR to clear +the log area and the input area and write more output to the log area. + +Some of the Program Function (PF) and Program Attention (PA) keys are +preassigned special functions. The ones that are not yield an alarm +when pressed. + +PA1 causes a SIGINT to the currently running application. You may do +the same thing from the input area, by typing "^C" and hitting ENTER. + +PA2 causes the log area to be cleared. If output awaits, it is then +written to the log area. + +PF3 causes an EOF to be received as input by the application. You may +cause an EOF also by typing "^D" and hitting ENTER. + +No PF key is preassigned to cause a job suspension, but you may cause a +job suspension by typing "^Z" and hitting ENTER. You may wish to +assign this function to a PF key. To make PF7 cause job suspension, +execute the command: + echo pf7=^z > /proc/tty/driver/tty3270 + +If the input you type does not end with the two characters "^n", the +driver appends a newline character and sends it to the tty driver; +otherwise the driver strips the "^n" and does not append a newline. +The IBM 3215 driver behaves similarly. + +Pf10 causes the most recent command to be retrieved from the tube's +command stack (default depth 20) and displayed in the input area. You +may hit PF10 again for the next-most-recent command, and so on. A +command is entered into the stack only when the input area is not made +invisible (such as for password entry) and it is not identical to the +current top entry. PF10 rotates backward through the command stack; +PF11 rotates forward. You may assign the backward function to any PF +key (or PA key, for that matter), say, PA3, with the command: + echo -e pa3=\\033k > /proc/tty/driver/tty3270 +This assigns the string ESC-k to PA3. Similarly, the string ESC-j +performs the forward function. (Rationale: In bash with vi-mode line +editing, ESC-k and ESC-j retrieve backward and forward history. +Suggestions welcome.) + +Is a stack size of twenty commands not to your liking? Change it on +the fly. To change to saving the last 100 commands, execute the +command: + echo recallsize=100 > /proc/tty/driver/tty3270 + +Have a command you issue frequently? Assign it to a PF or PA key! Use +the command + echo pf24="mkdir foobar; cd foobar" > /proc/tty/driver/tty3270 +to execute the commands mkdir foobar and cd foobar immediately when you +hit PF24. Want to see the command line first, before you execute it? +Use the -n option of the echo command: + echo -n pf24="mkdir foo; cd foo" > /proc/tty/driver/tty3270 + + + +Happy testing! I welcome any and all comments about this document, the +driver, etc etc. + +Dick Hitt diff --git a/Documentation/s390/CommonIO b/Documentation/s390/CommonIO new file mode 100644 index 000000000..6e0f63f34 --- /dev/null +++ b/Documentation/s390/CommonIO @@ -0,0 +1,125 @@ +S/390 common I/O-Layer - command line parameters, procfs and debugfs entries +============================================================================ + +Command line parameters +----------------------- + +* ccw_timeout_log + + Enable logging of debug information in case of ccw device timeouts. + +* cio_ignore = device[,device[,..]] + + device := {all | [!]ipldev | [!]condev | [!] | [!]-} + + The given devices will be ignored by the common I/O-layer; no detection + and device sensing will be done on any of those devices. The subchannel to + which the device in question is attached will be treated as if no device was + attached. + + An ignored device can be un-ignored later; see the "/proc entries"-section for + details. + + The devices must be given either as bus ids (0.x.abcd) or as hexadecimal + device numbers (0xabcd or abcd, for 2.4 backward compatibility). If you + give a device number 0xabcd, it will be interpreted as 0.0.abcd. + + You can use the 'all' keyword to ignore all devices. The 'ipldev' and 'condev' + keywords can be used to refer to the CCW based boot device and CCW console + device respectively (these are probably useful only when combined with the '!' + operator). The '!' operator will cause the I/O-layer to _not_ ignore a device. + The command line is parsed from left to right. + + For example, + cio_ignore=0.0.0023-0.0.0042,0.0.4711 + will ignore all devices ranging from 0.0.0023 to 0.0.0042 and the device + 0.0.4711, if detected. + As another example, + cio_ignore=all,!0.0.4711,!0.0.fd00-0.0.fd02 + will ignore all devices but 0.0.4711, 0.0.fd00, 0.0.fd01, 0.0.fd02. + + By default, no devices are ignored. + + +/proc entries +------------- + +* /proc/cio_ignore + + Lists the ranges of devices (by bus id) which are ignored by common I/O. + + You can un-ignore certain or all devices by piping to /proc/cio_ignore. + "free all" will un-ignore all ignored devices, + "free , , ..." will un-ignore the specified + devices. + + For example, if devices 0.0.0023 to 0.0.0042 and 0.0.4711 are ignored, + - echo free 0.0.0030-0.0.0032 > /proc/cio_ignore + will un-ignore devices 0.0.0030 to 0.0.0032 and will leave devices 0.0.0023 + to 0.0.002f, 0.0.0033 to 0.0.0042 and 0.0.4711 ignored; + - echo free 0.0.0041 > /proc/cio_ignore will furthermore un-ignore device + 0.0.0041; + - echo free all > /proc/cio_ignore will un-ignore all remaining ignored + devices. + + When a device is un-ignored, device recognition and sensing is performed and + the device driver will be notified if possible, so the device will become + available to the system. Note that un-ignoring is performed asynchronously. + + You can also add ranges of devices to be ignored by piping to + /proc/cio_ignore; "add , , ..." will ignore the + specified devices. + + Note: While already known devices can be added to the list of devices to be + ignored, there will be no effect on then. However, if such a device + disappears and then reappears, it will then be ignored. To make + known devices go away, you need the "purge" command (see below). + + For example, + "echo add 0.0.a000-0.0.accc, 0.0.af00-0.0.afff > /proc/cio_ignore" + will add 0.0.a000-0.0.accc and 0.0.af00-0.0.afff to the list of ignored + devices. + + You can remove already known but now ignored devices via + "echo purge > /proc/cio_ignore" + All devices ignored but still registered and not online (= not in use) + will be deregistered and thus removed from the system. + + The devices can be specified either by bus id (0.x.abcd) or, for 2.4 backward + compatibility, by the device number in hexadecimal (0xabcd or abcd). Device + numbers given as 0xabcd will be interpreted as 0.0.abcd. + +* /proc/cio_settle + + A write request to this file is blocked until all queued cio actions are + handled. This will allow userspace to wait for pending work affecting + device availability after changing cio_ignore or the hardware configuration. + +* For some of the information present in the /proc filesystem in 2.4 (namely, + /proc/subchannels and /proc/chpids), see driver-model.txt. + Information formerly in /proc/irq_count is now in /proc/interrupts. + + +debugfs entries +--------------- + +* /sys/kernel/debug/s390dbf/cio_*/ (S/390 debug feature) + + Some views generated by the debug feature to hold various debug outputs. + + - /sys/kernel/debug/s390dbf/cio_crw/sprintf + Messages from the processing of pending channel report words (machine check + handling). + + - /sys/kernel/debug/s390dbf/cio_msg/sprintf + Various debug messages from the common I/O-layer. + + - /sys/kernel/debug/s390dbf/cio_trace/hex_ascii + Logs the calling of functions in the common I/O-layer and, if applicable, + which subchannel they were called for, as well as dumps of some data + structures (like irb in an error case). + + The level of logging can be changed to be more or less verbose by piping to + /sys/kernel/debug/s390dbf/cio_*/level a number between 0 and 6; see the + documentation on the S/390 debug feature (Documentation/s390/s390dbf.txt) + for details. diff --git a/Documentation/s390/DASD b/Documentation/s390/DASD new file mode 100644 index 000000000..9963f1e9c --- /dev/null +++ b/Documentation/s390/DASD @@ -0,0 +1,73 @@ +DASD device driver + +S/390's disk devices (DASDs) are managed by Linux via the DASD device +driver. It is valid for all types of DASDs and represents them to +Linux as block devices, namely "dd". Currently the DASD driver uses a +single major number (254) and 4 minor numbers per volume (1 for the +physical volume and 3 for partitions). With respect to partitions see +below. Thus you may have up to 64 DASD devices in your system. + +The kernel parameter 'dasd=from-to,...' may be issued arbitrary times +in the kernel's parameter line or not at all. The 'from' and 'to' +parameters are to be given in hexadecimal notation without a leading +0x. +If you supply kernel parameters the different instances are processed +in order of appearance and a minor number is reserved for any device +covered by the supplied range up to 64 volumes. Additional DASDs are +ignored. If you do not supply the 'dasd=' kernel parameter at all, the +DASD driver registers all supported DASDs of your system to a minor +number in ascending order of the subchannel number. + +The driver currently supports ECKD-devices and there are stubs for +support of the FBA and CKD architectures. For the FBA architecture +only some smart data structures are missing to make the support +complete. +We performed our testing on 3380 and 3390 type disks of different +sizes, under VM and on the bare hardware (LPAR), using internal disks +of the multiprise as well as a RAMAC virtual array. Disks exported by +an Enterprise Storage Server (Seascape) should work fine as well. + +We currently implement one partition per volume, which is the whole +volume, skipping the first blocks up to the volume label. These are +reserved for IPL records and IBM's volume label to assure +accessibility of the DASD from other OSs. In a later stage we will +provide support of partitions, maybe VTOC oriented or using a kind of +partition table in the label record. + +USAGE + +-Low-level format (?CKD only) +For using an ECKD-DASD as a Linux harddisk you have to low-level +format the tracks by issuing the BLKDASDFORMAT-ioctl on that +device. This will erase any data on that volume including IBM volume +labels, VTOCs etc. The ioctl may take a 'struct format_data *' or +'NULL' as an argument. +typedef struct { + int start_unit; + int stop_unit; + int blksize; +} format_data_t; +When a NULL argument is passed to the BLKDASDFORMAT ioctl the whole +disk is formatted to a blocksize of 1024 bytes. Otherwise start_unit +and stop_unit are the first and last track to be formatted. If +stop_unit is -1 it implies that the DASD is formatted from start_unit +up to the last track. blksize can be any power of two between 512 and +4096. We recommend no blksize lower than 1024 because the ext2fs uses +1kB blocks anyway and you gain approx. 50% of capacity increasing your +blksize from 512 byte to 1kB. + +-Make a filesystem +Then you can mk??fs the filesystem of your choice on that volume or +partition. For reasons of sanity you should build your filesystem on +the partition /dev/dd?1 instead of the whole volume. You only lose 3kB +but may be sure that you can reuse your data after introduction of a +real partition table. + +BUGS: +- Performance sometimes is rather low because we don't fully exploit clustering + +TODO-List: +- Add IBM'S Disk layout to genhd +- Enhance driver to use more than one major number +- Enable usage as a module +- Support Cache fast write and DASD fast write (ECKD) diff --git a/Documentation/s390/Debugging390.txt b/Documentation/s390/Debugging390.txt new file mode 100644 index 000000000..5ae7f868a --- /dev/null +++ b/Documentation/s390/Debugging390.txt @@ -0,0 +1,2142 @@ + + Debugging on Linux for s/390 & z/Architecture + by + Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + Copyright (C) 2000-2001 IBM Deutschland Entwicklung GmbH, IBM Corporation + Best viewed with fixed width fonts + +Overview of Document: +===================== +This document is intended to give a good overview of how to debug Linux for +s/390 and z/Architecture. It is not intended as a complete reference and not a +tutorial on the fundamentals of C & assembly. It doesn't go into +390 IO in any detail. It is intended to complement the documents in the +reference section below & any other worthwhile references you get. + +It is intended like the Enterprise Systems Architecture/390 Reference Summary +to be printed out & used as a quick cheat sheet self help style reference when +problems occur. + +Contents +======== +Register Set +Address Spaces on Intel Linux +Address Spaces on Linux for s/390 & z/Architecture +The Linux for s/390 & z/Architecture Kernel Task Structure +Register Usage & Stackframes on Linux for s/390 & z/Architecture +A sample program with comments +Compiling programs for debugging on Linux for s/390 & z/Architecture +Debugging under VM +s/390 & z/Architecture IO Overview +Debugging IO on s/390 & z/Architecture under VM +GDB on s/390 & z/Architecture +Stack chaining in gdb by hand +Examining core dumps +ldd +Debugging modules +The proc file system +SysRq +References +Special Thanks + +Register Set +============ +The current architectures have the following registers. + +16 General propose registers, 32 bit on s/390 and 64 bit on z/Architecture, +r0-r15 (or gpr0-gpr15), used for arithmetic and addressing. + +16 Control registers, 32 bit on s/390 and 64 bit on z/Architecture, cr0-cr15, +kernel usage only, used for memory management, interrupt control, debugging +control etc. + +16 Access registers (ar0-ar15), 32 bit on both s/390 and z/Architecture, +normally not used by normal programs but potentially could be used as +temporary storage. These registers have a 1:1 association with general +purpose registers and are designed to be used in the so-called access +register mode to select different address spaces. +Access register 0 (and access register 1 on z/Architecture, which needs a +64 bit pointer) is currently used by the pthread library as a pointer to +the current running threads private area. + +16 64 bit floating point registers (fp0-fp15 ) IEEE & HFP floating +point format compliant on G5 upwards & a Floating point control reg (FPC) +4 64 bit registers (fp0,fp2,fp4 & fp6) HFP only on older machines. +Note: +Linux (currently) always uses IEEE & emulates G5 IEEE format on older machines, +( provided the kernel is configured for this ). + + +The PSW is the most important register on the machine it +is 64 bit on s/390 & 128 bit on z/Architecture & serves the roles of +a program counter (pc), condition code register,memory space designator. +In IBM standard notation I am counting bit 0 as the MSB. +It has several advantages over a normal program counter +in that you can change address translation & program counter +in a single instruction. To change address translation, +e.g. switching address translation off requires that you +have a logical=physical mapping for the address you are +currently running at. + + Bit Value +s/390 z/Architecture +0 0 Reserved ( must be 0 ) otherwise specification exception occurs. + +1 1 Program Event Recording 1 PER enabled, + PER is used to facilitate debugging e.g. single stepping. + +2-4 2-4 Reserved ( must be 0 ). + +5 5 Dynamic address translation 1=DAT on. + +6 6 Input/Output interrupt Mask + +7 7 External interrupt Mask used primarily for interprocessor + signalling and clock interrupts. + +8-11 8-11 PSW Key used for complex memory protection mechanism + (not used under linux) + +12 12 1 on s/390 0 on z/Architecture + +13 13 Machine Check Mask 1=enable machine check interrupts + +14 14 Wait State. Set this to 1 to stop the processor except for + interrupts and give time to other LPARS. Used in CPU idle in + the kernel to increase overall usage of processor resources. + +15 15 Problem state ( if set to 1 certain instructions are disabled ) + all linux user programs run with this bit 1 + ( useful info for debugging under VM ). + +16-17 16-17 Address Space Control + + 00 Primary Space Mode: + The register CR1 contains the primary address-space control ele- + ment (PASCE), which points to the primary space region/segment + table origin. + + 01 Access register mode + + 10 Secondary Space Mode: + The register CR7 contains the secondary address-space control + element (SASCE), which points to the secondary space region or + segment table origin. + + 11 Home Space Mode: + The register CR13 contains the home space address-space control + element (HASCE), which points to the home space region/segment + table origin. + + See "Address Spaces on Linux for s/390 & z/Architecture" below + for more information about address space usage in Linux. + +18-19 18-19 Condition codes (CC) + +20 20 Fixed point overflow mask if 1=FPU exceptions for this event + occur ( normally 0 ) + +21 21 Decimal overflow mask if 1=FPU exceptions for this event occur + ( normally 0 ) + +22 22 Exponent underflow mask if 1=FPU exceptions for this event occur + ( normally 0 ) + +23 23 Significance Mask if 1=FPU exceptions for this event occur + ( normally 0 ) + +24-31 24-30 Reserved Must be 0. + + 31 Extended Addressing Mode + 32 Basic Addressing Mode + Used to set addressing mode + PSW 31 PSW 32 + 0 0 24 bit + 0 1 31 bit + 1 1 64 bit + +32 1=31 bit addressing mode 0=24 bit addressing mode (for backward + compatibility), linux always runs with this bit set to 1 + +33-64 Instruction address. + 33-63 Reserved must be 0 + 64-127 Address + In 24 bits mode bits 64-103=0 bits 104-127 Address + In 31 bits mode bits 64-96=0 bits 97-127 Address + Note: unlike 31 bit mode on s/390 bit 96 must be zero + when loading the address with LPSWE otherwise a + specification exception occurs, LPSW is fully backward + compatible. + + +Prefix Page(s) +-------------- +This per cpu memory area is too intimately tied to the processor not to mention. +It exists between the real addresses 0-4096 on s/390 and between 0-8192 on +z/Architecture and is exchanged with one page on s/390 or two pages on +z/Architecture in absolute storage by the set prefix instruction during Linux +startup. +This page is mapped to a different prefix for each processor in an SMP +configuration (assuming the OS designer is sane of course). +Bytes 0-512 (200 hex) on s/390 and 0-512, 4096-4544, 4604-5119 currently on +z/Architecture are used by the processor itself for holding such information +as exception indications and entry points for exceptions. +Bytes after 0xc00 hex are used by linux for per processor globals on s/390 and +z/Architecture (there is a gap on z/Architecture currently between 0xc00 and +0x1000, too, which is used by Linux). +The closest thing to this on traditional architectures is the interrupt +vector table. This is a good thing & does simplify some of the kernel coding +however it means that we now cannot catch stray NULL pointers in the +kernel without hard coded checks. + + + +Address Spaces on Intel Linux +============================= + +The traditional Intel Linux is approximately mapped as follows forgive +the ascii art. +0xFFFFFFFF 4GB Himem ***************** + * * + * Kernel Space * + * * + ***************** **************** +User Space Himem * User Stack * * * +(typically 0xC0000000 3GB ) ***************** * * + * Shared Libs * * Next Process * + ***************** * to * + * * <== * Run * <== + * User Program * * * + * Data BSS * * * + * Text * * * + * Sections * * * +0x00000000 ***************** **************** + +Now it is easy to see that on Intel it is quite easy to recognise a kernel +address as being one greater than user space himem (in this case 0xC0000000), +and addresses of less than this are the ones in the current running program on +this processor (if an smp box). +If using the virtual machine ( VM ) as a debugger it is quite difficult to +know which user process is running as the address space you are looking at +could be from any process in the run queue. + +The limitation of Intels addressing technique is that the linux +kernel uses a very simple real address to virtual addressing technique +of Real Address=Virtual Address-User Space Himem. +This means that on Intel the kernel linux can typically only address +Himem=0xFFFFFFFF-0xC0000000=1GB & this is all the RAM these machines +can typically use. +They can lower User Himem to 2GB or lower & thus be +able to use 2GB of RAM however this shrinks the maximum size +of User Space from 3GB to 2GB they have a no win limit of 4GB unless +they go to 64 Bit. + + +On 390 our limitations & strengths make us slightly different. +For backward compatibility we are only allowed use 31 bits (2GB) +of our 32 bit addresses, however, we use entirely separate address +spaces for the user & kernel. + +This means we can support 2GB of non Extended RAM on s/390, & more +with the Extended memory management swap device & +currently 4TB of physical memory currently on z/Architecture. + + +Address Spaces on Linux for s/390 & z/Architecture +================================================== + +Our addressing scheme is basically as follows: + + Primary Space Home Space +Himem 0x7fffffff 2GB on s/390 ***************** **************** +currently 0x3ffffffffff (2^42)-1 * User Stack * * * +on z/Architecture. ***************** * * + * Shared Libs * * * + ***************** * * + * * * Kernel * + * User Program * * * + * Data BSS * * * + * Text * * * + * Sections * * * +0x00000000 ***************** **************** + +This also means that we need to look at the PSW problem state bit and the +addressing mode to decide whether we are looking at user or kernel space. + +User space runs in primary address mode (or access register mode within +the vdso code). + +The kernel usually also runs in home space mode, however when accessing +user space the kernel switches to primary or secondary address mode if +the mvcos instruction is not available or if a compare-and-swap (futex) +instruction on a user space address is performed. + +When also looking at the ASCE control registers, this means: + +User space: +- runs in primary or access register mode +- cr1 contains the user asce +- cr7 contains the user asce +- cr13 contains the kernel asce + +Kernel space: +- runs in home space mode +- cr1 contains the user or kernel asce + -> the kernel asce is loaded when a uaccess requires primary or + secondary address mode +- cr7 contains the user or kernel asce, (changed with set_fs()) +- cr13 contains the kernel asce + +In case of uaccess the kernel changes to: +- primary space mode in case of a uaccess (copy_to_user) and uses + e.g. the mvcp instruction to access user space. However the kernel + will stay in home space mode if the mvcos instruction is available +- secondary space mode in case of futex atomic operations, so that the + instructions come from primary address space and data from secondary + space + +In case of KVM, the kernel runs in home space mode, but cr1 gets switched +to contain the gmap asce before the SIE instruction gets executed. When +the SIE instruction is finished, cr1 will be switched back to contain the +user asce. + + +Virtual Addresses on s/390 & z/Architecture +=========================================== + +A virtual address on s/390 is made up of 3 parts +The SX (segment index, roughly corresponding to the PGD & PMD in Linux +terminology) being bits 1-11. +The PX (page index, corresponding to the page table entry (pte) in Linux +terminology) being bits 12-19. +The remaining bits BX (the byte index are the offset in the page ) +i.e. bits 20 to 31. + +On z/Architecture in linux we currently make up an address from 4 parts. +The region index bits (RX) 0-32 we currently use bits 22-32 +The segment index (SX) being bits 33-43 +The page index (PX) being bits 44-51 +The byte index (BX) being bits 52-63 + +Notes: +1) s/390 has no PMD so the PMD is really the PGD also. +A lot of this stuff is defined in pgtable.h. + +2) Also seeing as s/390's page indexes are only 1k in size +(bits 12-19 x 4 bytes per pte ) we use 1 ( page 4k ) +to make the best use of memory by updating 4 segment indices +entries each time we mess with a PMD & use offsets +0,1024,2048 & 3072 in this page as for our segment indexes. +On z/Architecture our page indexes are now 2k in size +( bits 12-19 x 8 bytes per pte ) we do a similar trick +but only mess with 2 segment indices each time we mess with +a PMD. + +3) As z/Architecture supports up to a massive 5-level page table lookup we +can only use 3 currently on Linux ( as this is all the generic kernel +currently supports ) however this may change in future +this allows us to access ( according to my sums ) +4TB of virtual storage per process i.e. +4096*512(PTES)*1024(PMDS)*2048(PGD) = 4398046511104 bytes, +enough for another 2 or 3 of years I think :-). +to do this we use a region-third-table designation type in +our address space control registers. + + +The Linux for s/390 & z/Architecture Kernel Task Structure +========================================================== +Each process/thread under Linux for S390 has its own kernel task_struct +defined in linux/include/linux/sched.h +The S390 on initialisation & resuming of a process on a cpu sets +the __LC_KERNEL_STACK variable in the spare prefix area for this cpu +(which we use for per-processor globals). + +The kernel stack pointer is intimately tied with the task structure for +each processor as follows. + + s/390 + ************************ + * 1 page kernel stack * + * ( 4K ) * + ************************ + * 1 page task_struct * + * ( 4K ) * +8K aligned ************************ + + z/Architecture + ************************ + * 2 page kernel stack * + * ( 8K ) * + ************************ + * 2 page task_struct * + * ( 8K ) * +16K aligned ************************ + +What this means is that we don't need to dedicate any register or global +variable to point to the current running process & can retrieve it with the +following very simple construct for s/390 & one very similar for z/Architecture. + +static inline struct task_struct * get_current(void) +{ + struct task_struct *current; + __asm__("lhi %0,-8192\n\t" + "nr %0,15" + : "=r" (current) ); + return current; +} + +i.e. just anding the current kernel stack pointer with the mask -8192. +Thankfully because Linux doesn't have support for nested IO interrupts +& our devices have large buffers can survive interrupts being shut for +short amounts of time we don't need a separate stack for interrupts. + + + + +Register Usage & Stackframes on Linux for s/390 & z/Architecture +================================================================= +Overview: +--------- +This is the code that gcc produces at the top & the bottom of +each function. It usually is fairly consistent & similar from +function to function & if you know its layout you can probably +make some headway in finding the ultimate cause of a problem +after a crash without a source level debugger. + +Note: To follow stackframes requires a knowledge of C or Pascal & +limited knowledge of one assembly language. + +It should be noted that there are some differences between the +s/390 and z/Architecture stack layouts as the z/Architecture stack layout +didn't have to maintain compatibility with older linkage formats. + +Glossary: +--------- +alloca: +This is a built in compiler function for runtime allocation +of extra space on the callers stack which is obviously freed +up on function exit ( e.g. the caller may choose to allocate nothing +of a buffer of 4k if required for temporary purposes ), it generates +very efficient code ( a few cycles ) when compared to alternatives +like malloc. + +automatics: These are local variables on the stack, +i.e they aren't in registers & they aren't static. + +back-chain: +This is a pointer to the stack pointer before entering a +framed functions ( see frameless function ) prologue got by +dereferencing the address of the current stack pointer, + i.e. got by accessing the 32 bit value at the stack pointers +current location. + +base-pointer: +This is a pointer to the back of the literal pool which +is an area just behind each procedure used to store constants +in each function. + +call-clobbered: The caller probably needs to save these registers if there +is something of value in them, on the stack or elsewhere before making a +call to another procedure so that it can restore it later. + +epilogue: +The code generated by the compiler to return to the caller. + +frameless-function +A frameless function in Linux for s390 & z/Architecture is one which doesn't +need more than the register save area (96 bytes on s/390, 160 on z/Architecture) +given to it by the caller. +A frameless function never: +1) Sets up a back chain. +2) Calls alloca. +3) Calls other normal functions +4) Has automatics. + +GOT-pointer: +This is a pointer to the global-offset-table in ELF +( Executable Linkable Format, Linux'es most common executable format ), +all globals & shared library objects are found using this pointer. + +lazy-binding +ELF shared libraries are typically only loaded when routines in the shared +library are actually first called at runtime. This is lazy binding. + +procedure-linkage-table +This is a table found from the GOT which contains pointers to routines +in other shared libraries which can't be called to by easier means. + +prologue: +The code generated by the compiler to set up the stack frame. + +outgoing-args: +This is extra area allocated on the stack of the calling function if the +parameters for the callee's cannot all be put in registers, the same +area can be reused by each function the caller calls. + +routine-descriptor: +A COFF executable format based concept of a procedure reference +actually being 8 bytes or more as opposed to a simple pointer to the routine. +This is typically defined as follows +Routine Descriptor offset 0=Pointer to Function +Routine Descriptor offset 4=Pointer to Table of Contents +The table of contents/TOC is roughly equivalent to a GOT pointer. +& it means that shared libraries etc. can be shared between several +environments each with their own TOC. + + +static-chain: This is used in nested functions a concept adopted from pascal +by gcc not used in ansi C or C++ ( although quite useful ), basically it +is a pointer used to reference local variables of enclosing functions. +You might come across this stuff once or twice in your lifetime. + +e.g. +The function below should return 11 though gcc may get upset & toss warnings +about unused variables. +int FunctionA(int a) +{ + int b; + FunctionC(int c) + { + b=c+1; + } + FunctionC(10); + return(b); +} + + +s/390 & z/Architecture Register usage +===================================== +r0 used by syscalls/assembly call-clobbered +r1 used by syscalls/assembly call-clobbered +r2 argument 0 / return value 0 call-clobbered +r3 argument 1 / return value 1 (if long long) call-clobbered +r4 argument 2 call-clobbered +r5 argument 3 call-clobbered +r6 argument 4 saved +r7 pointer-to arguments 5 to ... saved +r8 this & that saved +r9 this & that saved +r10 static-chain ( if nested function ) saved +r11 frame-pointer ( if function used alloca ) saved +r12 got-pointer saved +r13 base-pointer saved +r14 return-address saved +r15 stack-pointer saved + +f0 argument 0 / return value ( float/double ) call-clobbered +f2 argument 1 call-clobbered +f4 z/Architecture argument 2 saved +f6 z/Architecture argument 3 saved +The remaining floating points +f1,f3,f5 f7-f15 are call-clobbered. + +Notes: +------ +1) The only requirement is that registers which are used +by the callee are saved, e.g. the compiler is perfectly +capable of using r11 for purposes other than a frame a +frame pointer if a frame pointer is not needed. +2) In functions with variable arguments e.g. printf the calling procedure +is identical to one without variable arguments & the same number of +parameters. However, the prologue of this function is somewhat more +hairy owing to it having to move these parameters to the stack to +get va_start, va_arg & va_end to work. +3) Access registers are currently unused by gcc but are used in +the kernel. Possibilities exist to use them at the moment for +temporary storage but it isn't recommended. +4) Only 4 of the floating point registers are used for +parameter passing as older machines such as G3 only have only 4 +& it keeps the stack frame compatible with other compilers. +However with IEEE floating point emulation under linux on the +older machines you are free to use the other 12. +5) A long long or double parameter cannot be have the +first 4 bytes in a register & the second four bytes in the +outgoing args area. It must be purely in the outgoing args +area if crossing this boundary. +6) Floating point parameters are mixed with outgoing args +on the outgoing args area in the order the are passed in as parameters. +7) Floating point arguments 2 & 3 are saved in the outgoing args area for +z/Architecture + + +Stack Frame Layout +------------------ +s/390 z/Architecture +0 0 back chain ( a 0 here signifies end of back chain ) +4 8 eos ( end of stack, not used on Linux for S390 used in other linkage formats ) +8 16 glue used in other s/390 linkage formats for saved routine descriptors etc. +12 24 glue used in other s/390 linkage formats for saved routine descriptors etc. +16 32 scratch area +20 40 scratch area +24 48 saved r6 of caller function +28 56 saved r7 of caller function +32 64 saved r8 of caller function +36 72 saved r9 of caller function +40 80 saved r10 of caller function +44 88 saved r11 of caller function +48 96 saved r12 of caller function +52 104 saved r13 of caller function +56 112 saved r14 of caller function +60 120 saved r15 of caller function +64 128 saved f4 of caller function +72 132 saved f6 of caller function +80 undefined +96 160 outgoing args passed from caller to callee +96+x 160+x possible stack alignment ( 8 bytes desirable ) +96+x+y 160+x+y alloca space of caller ( if used ) +96+x+y+z 160+x+y+z automatics of caller ( if used ) +0 back-chain + +A sample program with comments. +=============================== + +Comments on the function test +----------------------------- +1) It didn't need to set up a pointer to the constant pool gpr13 as it is not +used ( :-( ). +2) This is a frameless function & no stack is bought. +3) The compiler was clever enough to recognise that it could return the +value in r2 as well as use it for the passed in parameter ( :-) ). +4) The basr ( branch relative & save ) trick works as follows the instruction +has a special case with r0,r0 with some instruction operands is understood as +the literal value 0, some risc architectures also do this ). So now +we are branching to the next address & the address new program counter is +in r13,so now we subtract the size of the function prologue we have executed ++ the size of the literal pool to get to the top of the literal pool +0040037c int test(int b) +{ # Function prologue below + 40037c: 90 de f0 34 stm %r13,%r14,52(%r15) # Save registers r13 & r14 + 400380: 0d d0 basr %r13,%r0 # Set up pointer to constant pool using + 400382: a7 da ff fa ahi %r13,-6 # basr trick + return(5+b); + # Huge main program + 400386: a7 2a 00 05 ahi %r2,5 # add 5 to r2 + + # Function epilogue below + 40038a: 98 de f0 34 lm %r13,%r14,52(%r15) # restore registers r13 & 14 + 40038e: 07 fe br %r14 # return +} + +Comments on the function main +----------------------------- +1) The compiler did this function optimally ( 8-) ) + +Literal pool for main. +400390: ff ff ff ec .long 0xffffffec +main(int argc,char *argv[]) +{ # Function prologue below + 400394: 90 bf f0 2c stm %r11,%r15,44(%r15) # Save necessary registers + 400398: 18 0f lr %r0,%r15 # copy stack pointer to r0 + 40039a: a7 fa ff a0 ahi %r15,-96 # Make area for callee saving + 40039e: 0d d0 basr %r13,%r0 # Set up r13 to point to + 4003a0: a7 da ff f0 ahi %r13,-16 # literal pool + 4003a4: 50 00 f0 00 st %r0,0(%r15) # Save backchain + + return(test(5)); # Main Program Below + 4003a8: 58 e0 d0 00 l %r14,0(%r13) # load relative address of test from + # literal pool + 4003ac: a7 28 00 05 lhi %r2,5 # Set first parameter to 5 + 4003b0: 4d ee d0 00 bas %r14,0(%r14,%r13) # jump to test setting r14 as return + # address using branch & save instruction. + + # Function Epilogue below + 4003b4: 98 bf f0 8c lm %r11,%r15,140(%r15)# Restore necessary registers. + 4003b8: 07 fe br %r14 # return to do program exit +} + + +Compiler updates +---------------- + +main(int argc,char *argv[]) +{ + 4004fc: 90 7f f0 1c stm %r7,%r15,28(%r15) + 400500: a7 d5 00 04 bras %r13,400508 + 400504: 00 40 04 f4 .long 0x004004f4 + # compiler now puts constant pool in code to so it saves an instruction + 400508: 18 0f lr %r0,%r15 + 40050a: a7 fa ff a0 ahi %r15,-96 + 40050e: 50 00 f0 00 st %r0,0(%r15) + return(test(5)); + 400512: 58 10 d0 00 l %r1,0(%r13) + 400516: a7 28 00 05 lhi %r2,5 + 40051a: 0d e1 basr %r14,%r1 + # compiler adds 1 extra instruction to epilogue this is done to + # avoid processor pipeline stalls owing to data dependencies on g5 & + # above as register 14 in the old code was needed directly after being loaded + # by the lm %r11,%r15,140(%r15) for the br %14. + 40051c: 58 40 f0 98 l %r4,152(%r15) + 400520: 98 7f f0 7c lm %r7,%r15,124(%r15) + 400524: 07 f4 br %r4 +} + + +Hartmut ( our compiler developer ) also has been threatening to take out the +stack backchain in optimised code as this also causes pipeline stalls, you +have been warned. + +64 bit z/Architecture code disassembly +-------------------------------------- + +If you understand the stuff above you'll understand the stuff +below too so I'll avoid repeating myself & just say that +some of the instructions have g's on the end of them to indicate +they are 64 bit & the stack offsets are a bigger, +the only other difference you'll find between 32 & 64 bit is that +we now use f4 & f6 for floating point arguments on 64 bit. +00000000800005b0 : +int test(int b) +{ + return(5+b); + 800005b0: a7 2a 00 05 ahi %r2,5 + 800005b4: b9 14 00 22 lgfr %r2,%r2 # downcast to integer + 800005b8: 07 fe br %r14 + 800005ba: 07 07 bcr 0,%r7 + + +} + +00000000800005bc
: +main(int argc,char *argv[]) +{ + 800005bc: eb bf f0 58 00 24 stmg %r11,%r15,88(%r15) + 800005c2: b9 04 00 1f lgr %r1,%r15 + 800005c6: a7 fb ff 60 aghi %r15,-160 + 800005ca: e3 10 f0 00 00 24 stg %r1,0(%r15) + return(test(5)); + 800005d0: a7 29 00 05 lghi %r2,5 + # brasl allows jumps > 64k & is overkill here bras would do fune + 800005d4: c0 e5 ff ff ff ee brasl %r14,800005b0 + 800005da: e3 40 f1 10 00 04 lg %r4,272(%r15) + 800005e0: eb bf f0 f8 00 04 lmg %r11,%r15,248(%r15) + 800005e6: 07 f4 br %r4 +} + + + +Compiling programs for debugging on Linux for s/390 & z/Architecture +==================================================================== +-gdwarf-2 now works it should be considered the default debugging +format for s/390 & z/Architecture as it is more reliable for debugging +shared libraries, normal -g debugging works much better now +Thanks to the IBM java compiler developers bug reports. + +This is typically done adding/appending the flags -g or -gdwarf-2 to the +CFLAGS & LDFLAGS variables Makefile of the program concerned. + +If using gdb & you would like accurate displays of registers & + stack traces compile without optimisation i.e make sure +that there is no -O2 or similar on the CFLAGS line of the Makefile & +the emitted gcc commands, obviously this will produce worse code +( not advisable for shipment ) but it is an aid to the debugging process. + +This aids debugging because the compiler will copy parameters passed in +in registers onto the stack so backtracing & looking at passed in +parameters will work, however some larger programs which use inline functions +will not compile without optimisation. + +Debugging with optimisation has since much improved after fixing +some bugs, please make sure you are using gdb-5.0 or later developed +after Nov'2000. + + + +Debugging under VM +================== + +Notes +----- +Addresses & values in the VM debugger are always hex never decimal +Address ranges are of the format - or +. +For example, the address range 0x2000 to 0x3000 can be described as 2000-3000 +or 2000.1000 + +The VM Debugger is case insensitive. + +VM's strengths are usually other debuggers weaknesses you can get at any +resource no matter how sensitive e.g. memory management resources, change +address translation in the PSW. For kernel hacking you will reap dividends if +you get good at it. + +The VM Debugger displays operators but not operands, and also the debugger +displays useful information on the same line as the author of the code probably +felt that it was a good idea not to go over the 80 columns on the screen. +This isn't as unintuitive as it may seem as the s/390 instructions are easy to +decode mentally and you can make a good guess at a lot of them as all the +operands are nibble (half byte aligned). +So if you have an objdump listing by hand, it is quite easy to follow, and if +you don't have an objdump listing keep a copy of the s/390 Reference Summary +or alternatively the s/390 principles of operation next to you. +e.g. even I can guess that +0001AFF8' LR 180F CC 0 +is a ( load register ) lr r0,r15 + +Also it is very easy to tell the length of a 390 instruction from the 2 most +significant bits in the instruction (not that this info is really useful except +if you are trying to make sense of a hexdump of code). +Here is a table +Bits Instruction Length +------------------------------------------ +00 2 Bytes +01 4 Bytes +10 4 Bytes +11 6 Bytes + +The debugger also displays other useful info on the same line such as the +addresses being operated on destination addresses of branches & condition codes. +e.g. +00019736' AHI A7DAFF0E CC 1 +000198BA' BRC A7840004 -> 000198C2' CC 0 +000198CE' STM 900EF068 >> 0FA95E78 CC 2 + + + +Useful VM debugger commands +--------------------------- + +I suppose I'd better mention this before I start +to list the current active traces do +Q TR +there can be a maximum of 255 of these per set +( more about trace sets later ). +To stop traces issue a +TR END. +To delete a particular breakpoint issue +TR DEL + +The PA1 key drops to CP mode so you can issue debugger commands, +Doing alt c (on my 3270 console at least ) clears the screen. +hitting b comes back to the running operating system +from cp mode ( in our case linux ). +It is typically useful to add shortcuts to your profile.exec file +if you have one ( this is roughly equivalent to autoexec.bat in DOS ). +file here are a few from mine. +/* this gives me command history on issuing f12 */ +set pf12 retrieve +/* this continues */ +set pf8 imm b +/* goes to trace set a */ +set pf1 imm tr goto a +/* goes to trace set b */ +set pf2 imm tr goto b +/* goes to trace set c */ +set pf3 imm tr goto c + + + +Instruction Tracing +------------------- +Setting a simple breakpoint +TR I PSWA
+To debug a particular function try +TR I R +TR I on its own will single step. +TR I DATA will trace for particular mnemonics +e.g. +TR I DATA 4D R 0197BC.4000 +will trace for BAS'es ( opcode 4D ) in the range 0197BC.4000 +if you were inclined you could add traces for all branch instructions & +suffix them with the run prefix so you would have a backtrace on screen +when a program crashes. +TR BR will trace branches into or out of an address. +e.g. +TR BR INTO 0 is often quite useful if a program is getting awkward & deciding +to branch to 0 & crashing as this will stop at the address before in jumps to 0. +TR I R
RUN cmd d g +single steps a range of addresses but stays running & +displays the gprs on each step. + + + +Displaying & modifying Registers +-------------------------------- +D G will display all the gprs +Adding a extra G to all the commands is necessary to access the full 64 bit +content in VM on z/Architecture. Obviously this isn't required for access +registers as these are still 32 bit. +e.g. DGG instead of DG +D X will display all the control registers +D AR will display all the access registers +D AR4-7 will display access registers 4 to 7 +CPU ALL D G will display the GRPS of all CPUS in the configuration +D PSW will display the current PSW +st PSW 2000 will put the value 2000 into the PSW & +cause crash your machine. +D PREFIX displays the prefix offset + + +Displaying Memory +----------------- +To display memory mapped using the current PSW's mapping try +D +To make VM display a message each time it hits a particular address and +continue try +D I will disassemble/display a range of instructions. +ST addr 32 bit word will store a 32 bit aligned address +D T will display the EBCDIC in an address (if you are that way inclined) +D R will display real addresses ( without DAT ) but with prefixing. +There are other complex options to display if you need to get at say home space +but are in primary space the easiest thing to do is to temporarily +modify the PSW to the other addressing mode, display the stuff & then +restore it. + + + +Hints +----- +If you want to issue a debugger command without halting your virtual machine +with the PA1 key try prefixing the command with #CP e.g. +#cp tr i pswa 2000 +also suffixing most debugger commands with RUN will cause them not +to stop just display the mnemonic at the current instruction on the console. +If you have several breakpoints you want to put into your program & +you get fed up of cross referencing with System.map +you can do the following trick for several symbols. +grep do_signal System.map +which emits the following among other things +0001f4e0 T do_signal +now you can do + +TR I PSWA 0001f4e0 cmd msg * do_signal +This sends a message to your own console each time do_signal is entered. +( As an aside I wrote a perl script once which automatically generated a REXX +script with breakpoints on every kernel procedure, this isn't a good idea +because there are thousands of these routines & VM can only set 255 breakpoints +at a time so you nearly had to spend as long pruning the file down as you would +entering the msgs by hand), however, the trick might be useful for a single +object file. In the 3270 terminal emulator x3270 there is a very useful option +in the file menu called "Save Screen In File" - this is very good for keeping a +copy of traces. + +From CMS help will give you online help on a particular command. +e.g. +HELP DISPLAY + +Also CP has a file called profile.exec which automatically gets called +on startup of CMS ( like autoexec.bat ), keeping on a DOS analogy session +CP has a feature similar to doskey, it may be useful for you to +use profile.exec to define some keystrokes. +e.g. +SET PF9 IMM B +This does a single step in VM on pressing F8. +SET PF10 ^ +This sets up the ^ key. +which can be used for ^c (ctrl-c),^z (ctrl-z) which can't be typed directly +into some 3270 consoles. +SET PF11 ^- +This types the starting keystrokes for a sysrq see SysRq below. +SET PF12 RETRIEVE +This retrieves command history on pressing F12. + + +Sometimes in VM the display is set up to scroll automatically this +can be very annoying if there are messages you wish to look at +to stop this do +TERM MORE 255 255 +This will nearly stop automatic screen updates, however it will +cause a denial of service if lots of messages go to the 3270 console, +so it would be foolish to use this as the default on a production machine. + + +Tracing particular processes +---------------------------- +The kernel's text segment is intentionally at an address in memory that it will +very seldom collide with text segments of user programs ( thanks Martin ), +this simplifies debugging the kernel. +However it is quite common for user processes to have addresses which collide +this can make debugging a particular process under VM painful under normal +circumstances as the process may change when doing a +TR I R
. +Thankfully after reading VM's online help I figured out how to debug +I particular process. + +Your first problem is to find the STD ( segment table designation ) +of the program you wish to debug. +There are several ways you can do this here are a few +1) objdump --syms | grep main +To get the address of main in the program. +tr i pswa
+Start the program, if VM drops to CP on what looks like the entry +point of the main function this is most likely the process you wish to debug. +Now do a D X13 or D XG13 on z/Architecture. +On 31 bit the STD is bits 1-19 ( the STO segment table origin ) +& 25-31 ( the STL segment table length ) of CR13. +now type +TR I R STD 0.7fffffff +e.g. +TR I R STD 8F32E1FF 0.7fffffff +Another very useful variation is +TR STORE INTO STD
+for finding out when a particular variable changes. + +An alternative way of finding the STD of a currently running process +is to do the following, ( this method is more complex but +could be quite convenient if you aren't updating the kernel much & +so your kernel structures will stay constant for a reasonable period of +time ). + +grep task /proc//status +from this you should see something like +task: 0f160000 ksp: 0f161de8 pt_regs: 0f161f68 +This now gives you a pointer to the task structure. +Now make CC:="s390-gcc -g" kernel/sched.s +To get the task_struct stabinfo. +( task_struct is defined in include/linux/sched.h ). +Now we want to look at +task->active_mm->pgd +on my machine the active_mm in the task structure stab is +active_mm:(4,12),672,32 +its offset is 672/8=84=0x54 +the pgd member in the mm_struct stab is +pgd:(4,6)=*(29,5),96,32 +so its offset is 96/8=12=0xc + +so we'll +hexdump -s 0xf160054 /dev/mem | more +i.e. task_struct+active_mm offset +to look at the active_mm member +f160054 0fee cc60 0019 e334 0000 0000 0000 0011 +hexdump -s 0x0feecc6c /dev/mem | more +i.e. active_mm+pgd offset +feecc6c 0f2c 0000 0000 0001 0000 0001 0000 0010 +we get something like +now do +TR I R STD 0.7fffffff +i.e. the 0x7f is added because the pgd only +gives the page table origin & we need to set the low bits +to the maximum possible segment table length. +TR I R STD 0f2c007f 0.7fffffff +on z/Architecture you'll probably need to do +TR I R STD 0.ffffffffffffffff +to set the TableType to 0x1 & the Table length to 3. + + + +Tracing Program Exceptions +-------------------------- +If you get a crash which says something like +illegal operation or specification exception followed by a register dump +You can restart linux & trace these using the tr prog trace +option. + + +The most common ones you will normally be tracing for is +1=operation exception +2=privileged operation exception +4=protection exception +5=addressing exception +6=specification exception +10=segment translation exception +11=page translation exception + +The full list of these is on page 22 of the current s/390 Reference Summary. +e.g. +tr prog 10 will trace segment translation exceptions. +tr prog on its own will trace all program interruption codes. + +Trace Sets +---------- +On starting VM you are initially in the INITIAL trace set. +You can do a Q TR to verify this. +If you have a complex tracing situation where you wish to wait for instance +till a driver is open before you start tracing IO, but know in your +heart that you are going to have to make several runs through the code till you +have a clue whats going on. + +What you can do is +TR I PSWA +hit b to continue till breakpoint +reach the breakpoint +now do your +TR GOTO B +TR IO 7c08-7c09 inst int run +or whatever the IO channels you wish to trace are & hit b + +To got back to the initial trace set do +TR GOTO INITIAL +& the TR I PSWA will be the only active breakpoint again. + + +Tracing linux syscalls under VM +------------------------------- +Syscalls are implemented on Linux for S390 by the Supervisor call instruction +(SVC). There 256 possibilities of these as the instruction is made up of a 0xA +opcode and the second byte being the syscall number. They are traced using the +simple command: +TR SVC +the syscalls are defined in linux/arch/s390/include/asm/unistd.h +e.g. to trace all file opens just do +TR SVC 5 ( as this is the syscall number of open ) + + +SMP Specific commands +--------------------- +To find out how many cpus you have +Q CPUS displays all the CPU's available to your virtual machine +To find the cpu that the current cpu VM debugger commands are being directed at +do Q CPU to change the current cpu VM debugger commands are being directed at do +CPU + +On a SMP guest issue a command to all CPUs try prefixing the command with cpu +all. To issue a command to a particular cpu try cpu e.g. +CPU 01 TR I R 2000.3000 +If you are running on a guest with several cpus & you have a IO related problem +& cannot follow the flow of code but you know it isn't smp related. +from the bash prompt issue +shutdown -h now or halt. +do a Q CPUS to find out how many cpus you have +detach each one of them from cp except cpu 0 +by issuing a +DETACH CPU 01-(number of cpus in configuration) +& boot linux again. +TR SIGP will trace inter processor signal processor instructions. +DEFINE CPU 01-(number in configuration) +will get your guests cpus back. + + +Help for displaying ascii textstrings +------------------------------------- +On the very latest VM Nucleus'es VM can now display ascii +( thanks Neale for the hint ) by doing +D TX. +e.g. +D TX0.100 + +Alternatively +============= +Under older VM debuggers (I love EBDIC too) you can use following little +program which converts a command line of hex digits to ascii text. It can be +compiled under linux and you can copy the hex digits from your x3270 terminal +to your xterm if you are debugging from a linuxbox. + +This is quite useful when looking at a parameter passed in as a text string +under VM ( unless you are good at decoding ASCII in your head ). + +e.g. consider tracing an open syscall +TR SVC 5 +We have stopped at a breakpoint +000151B0' SVC 0A05 -> 0001909A' CC 0 + +D 20.8 to check the SVC old psw in the prefix area and see was it from userspace +(for the layout of the prefix area consult the "Fixed Storage Locations" +chapter of the s/390 Reference Summary if you have it available). +V00000020 070C2000 800151B2 +The problem state bit wasn't set & it's also too early in the boot sequence +for it to be a userspace SVC if it was we would have to temporarily switch the +psw to user space addressing so we could get at the first parameter of the open +in gpr2. +Next do a +D G2 +GPR 2 = 00014CB4 +Now display what gpr2 is pointing to +D 00014CB4.20 +V00014CB4 2F646576 2F636F6E 736F6C65 00001BF5 +V00014CC4 FC00014C B4001001 E0001000 B8070707 +Now copy the text till the first 00 hex ( which is the end of the string +to an xterm & do hex2ascii on it. +hex2ascii 2F646576 2F636F6E 736F6C65 00 +outputs +Decoded Hex:=/ d e v / c o n s o l e 0x00 +We were opening the console device, + +You can compile the code below yourself for practice :-), +/* + * hex2ascii.c + * a useful little tool for converting a hexadecimal command line to ascii + * + * Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com) + * (C) 2000 IBM Deutschland Entwicklung GmbH, IBM Corporation. + */ +#include + +int main(int argc,char *argv[]) +{ + int cnt1,cnt2,len,toggle=0; + int startcnt=1; + unsigned char c,hex; + + if(argc>1&&(strcmp(argv[1],"-a")==0)) + startcnt=2; + printf("Decoded Hex:="); + for(cnt1=startcnt;cnt1='0'&&c<='9') + c=c-'0'; + if(c>='A'&&c<='F') + c=c-'A'+10; + if(c>='a'&&c<='f') + c=c-'a'+10; + switch(toggle) + { + case 0: + hex=c<<4; + toggle=1; + break; + case 1: + hex+=c; + if(hex<32||hex>127) + { + if(startcnt==1) + printf("0x%02X ",(int)hex); + else + printf("."); + } + else + { + printf("%c",hex); + if(startcnt==1) + printf(" "); + } + toggle=0; + break; + } + } + } + printf("\n"); +} + + + + +Stack tracing under VM +---------------------- +A basic backtrace +----------------- + +Here are the tricks I use 9 out of 10 times it works pretty well, + +When your backchain reaches a dead end +-------------------------------------- +This can happen when an exception happens in the kernel and the kernel is +entered twice. If you reach the NULL pointer at the end of the back chain you +should be able to sniff further back if you follow the following tricks. +1) A kernel address should be easy to recognise since it is in +primary space & the problem state bit isn't set & also +The Hi bit of the address is set. +2) Another backchain should also be easy to recognise since it is an +address pointing to another address approximately 100 bytes or 0x70 hex +behind the current stackpointer. + + +Here is some practice. +boot the kernel & hit PA1 at some random time +d g to display the gprs, this should display something like +GPR 0 = 00000001 00156018 0014359C 00000000 +GPR 4 = 00000001 001B8888 000003E0 00000000 +GPR 8 = 00100080 00100084 00000000 000FE000 +GPR 12 = 00010400 8001B2DC 8001B36A 000FFED8 +Note that GPR14 is a return address but as we are real men we are going to +trace the stack. +display 0x40 bytes after the stack pointer. + +V000FFED8 000FFF38 8001B838 80014C8E 000FFF38 +V000FFEE8 00000000 00000000 000003E0 00000000 +V000FFEF8 00100080 00100084 00000000 000FE000 +V000FFF08 00010400 8001B2DC 8001B36A 000FFED8 + + +Ah now look at whats in sp+56 (sp+0x38) this is 8001B36A our saved r14 if +you look above at our stackframe & also agrees with GPR14. + +now backchain +d 000FFF38.40 +we now are taking the contents of SP to get our first backchain. + +V000FFF38 000FFFA0 00000000 00014995 00147094 +V000FFF48 00147090 001470A0 000003E0 00000000 +V000FFF58 00100080 00100084 00000000 001BF1D0 +V000FFF68 00010400 800149BA 80014CA6 000FFF38 + +This displays a 2nd return address of 80014CA6 + +now do d 000FFFA0.40 for our 3rd backchain + +V000FFFA0 04B52002 0001107F 00000000 00000000 +V000FFFB0 00000000 00000000 FF000000 0001107F +V000FFFC0 00000000 00000000 00000000 00000000 +V000FFFD0 00010400 80010802 8001085A 000FFFA0 + + +our 3rd return address is 8001085A + +as the 04B52002 looks suspiciously like rubbish it is fair to assume that the +kernel entry routines for the sake of optimisation don't set up a backchain. + +now look at System.map to see if the addresses make any sense. + +grep -i 0001b3 System.map +outputs among other things +0001b304 T cpu_idle +so 8001B36A +is cpu_idle+0x66 ( quiet the cpu is asleep, don't wake it ) + + +grep -i 00014 System.map +produces among other things +00014a78 T start_kernel +so 0014CA6 is start_kernel+some hex number I can't add in my head. + +grep -i 00108 System.map +this produces +00010800 T _stext +so 8001085A is _stext+0x5a + +Congrats you've done your first backchain. + + + +s/390 & z/Architecture IO Overview +================================== + +I am not going to give a course in 390 IO architecture as this would take me +quite a while and I'm no expert. Instead I'll give a 390 IO architecture +summary for Dummies. If you have the s/390 principles of operation available +read this instead. If nothing else you may find a few useful keywords in here +and be able to use them on a web search engine to find more useful information. + +Unlike other bus architectures modern 390 systems do their IO using mostly +fibre optics and devices such as tapes and disks can be shared between several +mainframes. Also S390 can support up to 65536 devices while a high end PC based +system might be choking with around 64. + +Here is some of the common IO terminology: + +Subchannel: +This is the logical number most IO commands use to talk to an IO device. There +can be up to 0x10000 (65536) of these in a configuration, typically there are a +few hundred. Under VM for simplicity they are allocated contiguously, however +on the native hardware they are not. They typically stay consistent between +boots provided no new hardware is inserted or removed. +Under Linux for s390 we use these as IRQ's and also when issuing an IO command +(CLEAR SUBCHANNEL, HALT SUBCHANNEL, MODIFY SUBCHANNEL, RESUME SUBCHANNEL, +START SUBCHANNEL, STORE SUBCHANNEL and TEST SUBCHANNEL). We use this as the ID +of the device we wish to talk to. The most important of these instructions are +START SUBCHANNEL (to start IO), TEST SUBCHANNEL (to check whether the IO +completed successfully) and HALT SUBCHANNEL (to kill IO). A subchannel can have +up to 8 channel paths to a device, this offers redundancy if one is not +available. + +Device Number: +This number remains static and is closely tied to the hardware. There are 65536 +of these, made up of a CHPID (Channel Path ID, the most significant 8 bits) and +another lsb 8 bits. These remain static even if more devices are inserted or +removed from the hardware. There is a 1 to 1 mapping between subchannels and +device numbers, provided devices aren't inserted or removed. + +Channel Control Words: +CCWs are linked lists of instructions initially pointed to by an operation +request block (ORB), which is initially given to Start Subchannel (SSCH) +command along with the subchannel number for the IO subsystem to process +while the CPU continues executing normal code. +CCWs come in two flavours, Format 0 (24 bit for backward compatibility) and +Format 1 (31 bit). These are typically used to issue read and write (and many +other) instructions. They consist of a length field and an absolute address +field. +Each IO typically gets 1 or 2 interrupts, one for channel end (primary status) +when the channel is idle, and the second for device end (secondary status). +Sometimes you get both concurrently. You check how the IO went on by issuing a +TEST SUBCHANNEL at each interrupt, from which you receive an Interruption +response block (IRB). If you get channel and device end status in the IRB +without channel checks etc. your IO probably went okay. If you didn't you +probably need to examine the IRB, extended status word etc. +If an error occurs, more sophisticated control units have a facility known as +concurrent sense. This means that if an error occurs Extended sense information +will be presented in the Extended status word in the IRB. If not you have to +issue a subsequent SENSE CCW command after the test subchannel. + + +TPI (Test pending interrupt) can also be used for polled IO, but in +multitasking multiprocessor systems it isn't recommended except for +checking special cases (i.e. non looping checks for pending IO etc.). + +Store Subchannel and Modify Subchannel can be used to examine and modify +operating characteristics of a subchannel (e.g. channel paths). + +Other IO related Terms: +Sysplex: S390's Clustering Technology +QDIO: S390's new high speed IO architecture to support devices such as gigabit +ethernet, this architecture is also designed to be forward compatible with +upcoming 64 bit machines. + + +General Concepts + +Input Output Processors (IOP's) are responsible for communicating between +the mainframe CPU's & the channel & relieve the mainframe CPU's from the +burden of communicating with IO devices directly, this allows the CPU's to +concentrate on data processing. + +IOP's can use one or more links ( known as channel paths ) to talk to each +IO device. It first checks for path availability & chooses an available one, +then starts ( & sometimes terminates IO ). +There are two types of channel path: ESCON & the Parallel IO interface. + +IO devices are attached to control units, control units provide the +logic to interface the channel paths & channel path IO protocols to +the IO devices, they can be integrated with the devices or housed separately +& often talk to several similar devices ( typical examples would be raid +controllers or a control unit which connects to 1000 3270 terminals ). + + + +---------------------------------------------------------------+ + | +-----+ +-----+ +-----+ +-----+ +----------+ +----------+ | + | | CPU | | CPU | | CPU | | CPU | | Main | | Expanded | | + | | | | | | | | | | Memory | | Storage | | + | +-----+ +-----+ +-----+ +-----+ +----------+ +----------+ | + |---------------------------------------------------------------+ + | IOP | IOP | IOP | + |--------------------------------------------------------------- + | C | C | C | C | C | C | C | C | C | C | C | C | C | C | C | C | + ---------------------------------------------------------------- + || || + || Bus & Tag Channel Path || ESCON + || ====================== || Channel + || || || || Path + +----------+ +----------+ +----------+ + | | | | | | + | CU | | CU | | CU | + | | | | | | + +----------+ +----------+ +----------+ + | | | | | ++----------+ +----------+ +----------+ +----------+ +----------+ +|I/O Device| |I/O Device| |I/O Device| |I/O Device| |I/O Device| ++----------+ +----------+ +----------+ +----------+ +----------+ + CPU = Central Processing Unit + C = Channel + IOP = IP Processor + CU = Control Unit + +The 390 IO systems come in 2 flavours the current 390 machines support both + +The Older 360 & 370 Interface,sometimes called the Parallel I/O interface, +sometimes called Bus-and Tag & sometimes Original Equipment Manufacturers +Interface (OEMI). + +This byte wide Parallel channel path/bus has parity & data on the "Bus" cable +and control lines on the "Tag" cable. These can operate in byte multiplex mode +for sharing between several slow devices or burst mode and monopolize the +channel for the whole burst. Up to 256 devices can be addressed on one of these +cables. These cables are about one inch in diameter. The maximum unextended +length supported by these cables is 125 Meters but this can be extended up to +2km with a fibre optic channel extended such as a 3044. The maximum burst speed +supported is 4.5 megabytes per second. However, some really old processors +support only transfer rates of 3.0, 2.0 & 1.0 MB/sec. +One of these paths can be daisy chained to up to 8 control units. + + +ESCON if fibre optic it is also called FICON +Was introduced by IBM in 1990. Has 2 fibre optic cables and uses either leds or +lasers for communication at a signaling rate of up to 200 megabits/sec. As +10bits are transferred for every 8 bits info this drops to 160 megabits/sec +and to 18.6 Megabytes/sec once control info and CRC are added. ESCON only +operates in burst mode. + +ESCONs typical max cable length is 3km for the led version and 20km for the +laser version known as XDF (extended distance facility). This can be further +extended by using an ESCON director which triples the above mentioned ranges. +Unlike Bus & Tag as ESCON is serial it uses a packet switching architecture, +the standard Bus & Tag control protocol is however present within the packets. +Up to 256 devices can be attached to each control unit that uses one of these +interfaces. + +Common 390 Devices include: +Network adapters typically OSA2,3172's,2116's & OSA-E gigabit ethernet adapters, +Consoles 3270 & 3215 (a teletype emulated under linux for a line mode console). +DASD's direct access storage devices ( otherwise known as hard disks ). +Tape Drives. +CTC ( Channel to Channel Adapters ), +ESCON or Parallel Cables used as a very high speed serial link +between 2 machines. + + +Debugging IO on s/390 & z/Architecture under VM +=============================================== + +Now we are ready to go on with IO tracing commands under VM + +A few self explanatory queries: +Q OSA +Q CTC +Q DISK ( This command is CMS specific ) +Q DASD + + + + + + +Q OSA on my machine returns +OSA 7C08 ON OSA 7C08 SUBCHANNEL = 0000 +OSA 7C09 ON OSA 7C09 SUBCHANNEL = 0001 +OSA 7C14 ON OSA 7C14 SUBCHANNEL = 0002 +OSA 7C15 ON OSA 7C15 SUBCHANNEL = 0003 + +If you have a guest with certain privileges you may be able to see devices +which don't belong to you. To avoid this, add the option V. +e.g. +Q V OSA + +Now using the device numbers returned by this command we will +Trace the io starting up on the first device 7c08 & 7c09 +In our simplest case we can trace the +start subchannels +like TR SSCH 7C08-7C09 +or the halt subchannels +or TR HSCH 7C08-7C09 +MSCH's ,STSCH's I think you can guess the rest + +A good trick is tracing all the IO's and CCWS and spooling them into the reader +of another VM guest so he can ftp the logfile back to his own machine. I'll do +a small bit of this and give you a look at the output. + +1) Spool stdout to VM reader +SP PRT TO (another vm guest ) or * for the local vm guest +2) Fill the reader with the trace +TR IO 7c08-7c09 INST INT CCW PRT RUN +3) Start up linux +i 00c +4) Finish the trace +TR END +5) close the reader +C PRT +6) list reader contents +RDRLIST +7) copy it to linux4's minidisk +RECEIVE / LOG TXT A1 ( replace +8) +filel & press F11 to look at it +You should see something like: + +00020942' SSCH B2334000 0048813C CC 0 SCH 0000 DEV 7C08 + CPA 000FFDF0 PARM 00E2C9C4 KEY 0 FPI C0 LPM 80 + CCW 000FFDF0 E4200100 00487FE8 0000 E4240100 ........ + IDAL 43D8AFE8 + IDAL 0FB76000 +00020B0A' I/O DEV 7C08 -> 000197BC' SCH 0000 PARM 00E2C9C4 +00021628' TSCH B2354000 >> 00488164 CC 0 SCH 0000 DEV 7C08 + CCWA 000FFDF8 DEV STS 0C SCH STS 00 CNT 00EC + KEY 0 FPI C0 CC 0 CTLS 4007 +00022238' STSCH B2344000 >> 00488108 CC 0 SCH 0000 DEV 7C08 + +If you don't like messing up your readed ( because you possibly booted from it ) +you can alternatively spool it to another readers guest. + + +Other common VM device related commands +--------------------------------------------- +These commands are listed only because they have +been of use to me in the past & may be of use to +you too. For more complete info on each of the commands +use type HELP from CMS. +detaching devices +DET +ATT +attach a device to guest * for your own guest +READY cause VM to issue a fake interrupt. + +The VARY command is normally only available to VM administrators. +VARY ON PATH TO +VARY OFF PATH FROM +This is used to switch on or off channel paths to devices. + +Q CHPID +This displays state of devices using this channel path +D SCHIB +This displays the subchannel information SCHIB block for the device. +this I believe is also only available to administrators. +DEFINE CTC +defines a virtual CTC channel to channel connection +2 need to be defined on each guest for the CTC driver to use. +COUPLE devno userid remote devno +Joins a local virtual device to a remote virtual device +( commonly used for the CTC driver ). + +Building a VM ramdisk under CMS which linux can use +def vfb- +blocksize is commonly 4096 for linux. +Formatting it +format (blksize + +Sharing a disk between multiple guests +LINK userid devno1 devno2 mode password + + + +GDB on S390 +=========== +N.B. if compiling for debugging gdb works better without optimisation +( see Compiling programs for debugging ) + +invocation +---------- +gdb + +Online help +----------- +help: gives help on commands +e.g. +help +help display +Note gdb's online help is very good use it. + + +Assembly +-------- +info registers: displays registers other than floating point. +info all-registers: displays floating points as well. +disassemble: disassembles +e.g. +disassemble without parameters will disassemble the current function +disassemble $pc $pc+10 + +Viewing & modifying variables +----------------------------- +print or p: displays variable or register +e.g. p/x $sp will display the stack pointer + +display: prints variable or register each time program stops +e.g. +display/x $pc will display the program counter +display argc + +undisplay : undo's display's + +info breakpoints: shows all current breakpoints + +info stack: shows stack back trace (if this doesn't work too well, I'll show +you the stacktrace by hand below). + +info locals: displays local variables. + +info args: display current procedure arguments. + +set args: will set argc & argv each time the victim program is invoked. + +set =value +set argc=100 +set $pc=0 + + + +Modifying execution +------------------- +step: steps n lines of sourcecode +step steps 1 line. +step 100 steps 100 lines of code. + +next: like step except this will not step into subroutines + +stepi: steps a single machine code instruction. +e.g. stepi 100 + +nexti: steps a single machine code instruction but will not step into +subroutines. + +finish: will run until exit of the current routine + +run: (re)starts a program + +cont: continues a program + +quit: exits gdb. + + +breakpoints +------------ + +break +sets a breakpoint +e.g. + +break main + +break *$pc + +break *0x400618 + +Here's a really useful one for large programs +rbr +Set a breakpoint for all functions matching REGEXP +e.g. +rbr 390 +will set a breakpoint with all functions with 390 in their name. + +info breakpoints +lists all breakpoints + +delete: delete breakpoint by number or delete them all +e.g. +delete 1 will delete the first breakpoint +delete will delete them all + +watch: This will set a watchpoint ( usually hardware assisted ), +This will watch a variable till it changes +e.g. +watch cnt, will watch the variable cnt till it changes. +As an aside unfortunately gdb's, architecture independent watchpoint code +is inconsistent & not very good, watchpoints usually work but not always. + +info watchpoints: Display currently active watchpoints + +condition: ( another useful one ) +Specify breakpoint number N to break only if COND is true. +Usage is `condition N COND', where N is an integer and COND is an +expression to be evaluated whenever breakpoint N is reached. + + + +User defined functions/macros +----------------------------- +define: ( Note this is very very useful,simple & powerful ) +usage define end + +examples which you should consider putting into .gdbinit in your home directory +define d +stepi +disassemble $pc $pc+10 +end + +define e +nexti +disassemble $pc $pc+10 +end + + +Other hard to classify stuff +---------------------------- +signal n: +sends the victim program a signal. +e.g. signal 3 will send a SIGQUIT. + +info signals: +what gdb does when the victim receives certain signals. + +list: +e.g. +list lists current function source +list 1,10 list first 10 lines of current file. +list test.c:1,10 + + +directory: +Adds directories to be searched for source if gdb cannot find the source. +(note it is a bit sensitive about slashes) +e.g. To add the root of the filesystem to the searchpath do +directory // + + +call +This calls a function in the victim program, this is pretty powerful +e.g. +(gdb) call printf("hello world") +outputs: +$1 = 11 + +You might now be thinking that the line above didn't work, something extra had +to be done. +(gdb) call fflush(stdout) +hello world$2 = 0 +As an aside the debugger also calls malloc & free under the hood +to make space for the "hello world" string. + + + +hints +----- +1) command completion works just like bash +( if you are a bad typist like me this really helps ) +e.g. hit br & cursor up & down :-). + +2) if you have a debugging problem that takes a few steps to recreate +put the steps into a file called .gdbinit in your current working directory +if you have defined a few extra useful user defined commands put these in +your home directory & they will be read each time gdb is launched. + +A typical .gdbinit file might be. +break main +run +break runtime_exception +cont + + +stack chaining in gdb by hand +----------------------------- +This is done using a the same trick described for VM +p/x (*($sp+56))&0x7fffffff get the first backchain. + +For z/Architecture +Replace 56 with 112 & ignore the &0x7fffffff +in the macros below & do nasty casts to longs like the following +as gdb unfortunately deals with printed arguments as ints which +messes up everything. +i.e. here is a 3rd backchain dereference +p/x *(long *)(***(long ***)$sp+112) + + +this outputs +$5 = 0x528f18 +on my machine. +Now you can use +info symbol (*($sp+56))&0x7fffffff +you might see something like. +rl_getc + 36 in section .text telling you what is located at address 0x528f18 +Now do. +p/x (*(*$sp+56))&0x7fffffff +This outputs +$6 = 0x528ed0 +Now do. +info symbol (*(*$sp+56))&0x7fffffff +rl_read_key + 180 in section .text +now do +p/x (*(**$sp+56))&0x7fffffff +& so on. + +Disassembling instructions without debug info +--------------------------------------------- +gdb typically complains if there is a lack of debugging +symbols in the disassemble command with +"No function contains specified address." To get around +this do +x/xi
+e.g. +x/20xi 0x400730 + + + +Note: Remember gdb has history just like bash you don't need to retype the +whole line just use the up & down arrows. + + + +For more info +------------- +From your linuxbox do +man gdb or info gdb. + +core dumps +---------- +What a core dump ?, +A core dump is a file generated by the kernel (if allowed) which contains the +registers and all active pages of the program which has crashed. +From this file gdb will allow you to look at the registers, stack trace and +memory of the program as if it just crashed on your system. It is usually +called core and created in the current working directory. +This is very useful in that a customer can mail a core dump to a technical +support department and the technical support department can reconstruct what +happened. Provided they have an identical copy of this program with debugging +symbols compiled in and the source base of this build is available. +In short it is far more useful than something like a crash log could ever hope +to be. + +Why have I never seen one ?. +Probably because you haven't used the command +ulimit -c unlimited in bash +to allow core dumps, now do +ulimit -a +to verify that the limit was accepted. + +A sample core dump +To create this I'm going to do +ulimit -c unlimited +gdb +to launch gdb (my victim app. ) now be bad & do the following from another +telnet/xterm session to the same machine +ps -aux | grep gdb +kill -SIGSEGV +or alternatively use killall -SIGSEGV gdb if you have the killall command. +Now look at the core dump. +./gdb core +Displays the following +GNU gdb 4.18 +Copyright 1998 Free Software Foundation, Inc. +GDB is free software, covered by the GNU General Public License, and you are +welcome to change it and/or distribute copies of it under certain conditions. +Type "show copying" to see the conditions. +There is absolutely no warranty for GDB. Type "show warranty" for details. +This GDB was configured as "s390-ibm-linux"... +Core was generated by `./gdb'. +Program terminated with signal 11, Segmentation fault. +Reading symbols from /usr/lib/libncurses.so.4...done. +Reading symbols from /lib/libm.so.6...done. +Reading symbols from /lib/libc.so.6...done. +Reading symbols from /lib/ld-linux.so.2...done. +#0 0x40126d1a in read () from /lib/libc.so.6 +Setting up the environment for debugging gdb. +Breakpoint 1 at 0x4dc6f8: file utils.c, line 471. +Breakpoint 2 at 0x4d87a4: file top.c, line 2609. +(top-gdb) info stack +#0 0x40126d1a in read () from /lib/libc.so.6 +#1 0x528f26 in rl_getc (stream=0x7ffffde8) at input.c:402 +#2 0x528ed0 in rl_read_key () at input.c:381 +#3 0x5167e6 in readline_internal_char () at readline.c:454 +#4 0x5168ee in readline_internal_charloop () at readline.c:507 +#5 0x51692c in readline_internal () at readline.c:521 +#6 0x5164fe in readline (prompt=0x7ffff810) + at readline.c:349 +#7 0x4d7a8a in command_line_input (prompt=0x564420 "(gdb) ", repeat=1, + annotation_suffix=0x4d6b44 "prompt") at top.c:2091 +#8 0x4d6cf0 in command_loop () at top.c:1345 +#9 0x4e25bc in main (argc=1, argv=0x7ffffdf4) at main.c:635 + + +LDD +=== +This is a program which lists the shared libraries which a library needs, +Note you also get the relocations of the shared library text segments which +help when using objdump --source. +e.g. + ldd ./gdb +outputs +libncurses.so.4 => /usr/lib/libncurses.so.4 (0x40018000) +libm.so.6 => /lib/libm.so.6 (0x4005e000) +libc.so.6 => /lib/libc.so.6 (0x40084000) +/lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000) + + +Debugging shared libraries +========================== +Most programs use shared libraries, however it can be very painful +when you single step instruction into a function like printf for the +first time & you end up in functions like _dl_runtime_resolve this is +the ld.so doing lazy binding, lazy binding is a concept in ELF where +shared library functions are not loaded into memory unless they are +actually used, great for saving memory but a pain to debug. +To get around this either relink the program -static or exit gdb type +export LD_BIND_NOW=true this will stop lazy binding & restart the gdb'ing +the program in question. + + + +Debugging modules +================= +As modules are dynamically loaded into the kernel their address can be +anywhere to get around this use the -m option with insmod to emit a load +map which can be piped into a file if required. + +The proc file system +==================== +What is it ?. +It is a filesystem created by the kernel with files which are created on demand +by the kernel if read, or can be used to modify kernel parameters, +it is a powerful concept. + +e.g. + +cat /proc/sys/net/ipv4/ip_forward +On my machine outputs +0 +telling me ip_forwarding is not on to switch it on I can do +echo 1 > /proc/sys/net/ipv4/ip_forward +cat it again +cat /proc/sys/net/ipv4/ip_forward +On my machine now outputs +1 +IP forwarding is on. +There is a lot of useful info in here best found by going in and having a look +around, so I'll take you through some entries I consider important. + +All the processes running on the machine have their own entry defined by +/proc/ +So lets have a look at the init process +cd /proc/1 + +cat cmdline +emits +init [2] + +cd /proc/1/fd +This contains numerical entries of all the open files, +some of these you can cat e.g. stdout (2) + +cat /proc/29/maps +on my machine emits + +00400000-00478000 r-xp 00000000 5f:00 4103 /bin/bash +00478000-0047e000 rw-p 00077000 5f:00 4103 /bin/bash +0047e000-00492000 rwxp 00000000 00:00 0 +40000000-40015000 r-xp 00000000 5f:00 14382 /lib/ld-2.1.2.so +40015000-40016000 rw-p 00014000 5f:00 14382 /lib/ld-2.1.2.so +40016000-40017000 rwxp 00000000 00:00 0 +40017000-40018000 rw-p 00000000 00:00 0 +40018000-4001b000 r-xp 00000000 5f:00 14435 /lib/libtermcap.so.2.0.8 +4001b000-4001c000 rw-p 00002000 5f:00 14435 /lib/libtermcap.so.2.0.8 +4001c000-4010d000 r-xp 00000000 5f:00 14387 /lib/libc-2.1.2.so +4010d000-40111000 rw-p 000f0000 5f:00 14387 /lib/libc-2.1.2.so +40111000-40114000 rw-p 00000000 00:00 0 +40114000-4011e000 r-xp 00000000 5f:00 14408 /lib/libnss_files-2.1.2.so +4011e000-4011f000 rw-p 00009000 5f:00 14408 /lib/libnss_files-2.1.2.so +7fffd000-80000000 rwxp ffffe000 00:00 0 + + +Showing us the shared libraries init uses where they are in memory +& memory access permissions for each virtual memory area. + +/proc/1/cwd is a softlink to the current working directory. +/proc/1/root is the root of the filesystem for this process. + +/proc/1/mem is the current running processes memory which you +can read & write to like a file. +strace uses this sometimes as it is a bit faster than the +rather inefficient ptrace interface for peeking at DATA. + + +cat status + +Name: init +State: S (sleeping) +Pid: 1 +PPid: 0 +Uid: 0 0 0 0 +Gid: 0 0 0 0 +Groups: +VmSize: 408 kB +VmLck: 0 kB +VmRSS: 208 kB +VmData: 24 kB +VmStk: 8 kB +VmExe: 368 kB +VmLib: 0 kB +SigPnd: 0000000000000000 +SigBlk: 0000000000000000 +SigIgn: 7fffffffd7f0d8fc +SigCgt: 00000000280b2603 +CapInh: 00000000fffffeff +CapPrm: 00000000ffffffff +CapEff: 00000000fffffeff + +User PSW: 070de000 80414146 +task: 004b6000 tss: 004b62d8 ksp: 004b7ca8 pt_regs: 004b7f68 +User GPRS: +00000400 00000000 0000000b 7ffffa90 +00000000 00000000 00000000 0045d9f4 +0045cafc 7ffffa90 7fffff18 0045cb08 +00010400 804039e8 80403af8 7ffff8b0 +User ACRS: +00000000 00000000 00000000 00000000 +00000001 00000000 00000000 00000000 +00000000 00000000 00000000 00000000 +00000000 00000000 00000000 00000000 +Kernel BackChain CallChain BackChain CallChain + 004b7ca8 8002bd0c 004b7d18 8002b92c + 004b7db8 8005cd50 004b7e38 8005d12a + 004b7f08 80019114 +Showing among other things memory usage & status of some signals & +the processes'es registers from the kernel task_structure +as well as a backchain which may be useful if a process crashes +in the kernel for some unknown reason. + +Some driver debugging techniques +================================ +debug feature +------------- +Some of our drivers now support a "debug feature" in +/proc/s390dbf see s390dbf.txt in the linux/Documentation directory +for more info. +e.g. +to switch on the lcs "debug feature" +echo 5 > /proc/s390dbf/lcs/level +& then after the error occurred. +cat /proc/s390dbf/lcs/sprintf >/logfile +the logfile now contains some information which may help +tech support resolve a problem in the field. + + + +high level debugging network drivers +------------------------------------ +ifconfig is a quite useful command +it gives the current state of network drivers. + +If you suspect your network device driver is dead +one way to check is type +ifconfig +e.g. tr0 +You should see something like +tr0 Link encap:16/4 Mbps Token Ring (New) HWaddr 00:04:AC:20:8E:48 + inet addr:9.164.185.132 Bcast:9.164.191.255 Mask:255.255.224.0 + UP BROADCAST RUNNING MULTICAST MTU:2000 Metric:1 + RX packets:246134 errors:0 dropped:0 overruns:0 frame:0 + TX packets:5 errors:0 dropped:0 overruns:0 carrier:0 + collisions:0 txqueuelen:100 + +if the device doesn't say up +try +/etc/rc.d/init.d/network start +( this starts the network stack & hopefully calls ifconfig tr0 up ). +ifconfig looks at the output of /proc/net/dev and presents it in a more +presentable form. +Now ping the device from a machine in the same subnet. +if the RX packets count & TX packets counts don't increment you probably +have problems. +next +cat /proc/net/arp +Do you see any hardware addresses in the cache if not you may have problems. +Next try +ping -c 5 i.e. the Bcast field above in the output of +ifconfig. Do you see any replies from machines other than the local machine +if not you may have problems. also if the TX packets count in ifconfig +hasn't incremented either you have serious problems in your driver +(e.g. the txbusy field of the network device being stuck on ) +or you may have multiple network devices connected. + + +chandev +------- +There is a new device layer for channel devices, some +drivers e.g. lcs are registered with this layer. +If the device uses the channel device layer you'll be +able to find what interrupts it uses & the current state +of the device. +See the manpage chandev.8 &type cat /proc/chandev for more info. + + +SysRq +===== +This is now supported by linux for s/390 & z/Architecture. +To enable it do compile the kernel with +Kernel Hacking -> Magic SysRq Key Enabled +echo "1" > /proc/sys/kernel/sysrq +also type +echo "8" >/proc/sys/kernel/printk +To make printk output go to console. +On 390 all commands are prefixed with +^- +e.g. +^-t will show tasks. +^-? or some unknown command will display help. +The sysrq key reading is very picky ( I have to type the keys in an + xterm session & paste them into the x3270 console ) +& it may be wise to predefine the keys as described in the VM hints above + +This is particularly useful for syncing disks unmounting & rebooting +if the machine gets partially hung. + +Read Documentation/admin-guide/sysrq.rst for more info + +References: +=========== +Enterprise Systems Architecture Reference Summary +Enterprise Systems Architecture Principles of Operation +Hartmut Penners s390 stack frame sheet. +IBM Mainframe Channel Attachment a technology brief from a CISCO webpage +Various bits of man & info pages of Linux. +Linux & GDB source. +Various info & man pages. +CMS Help on tracing commands. +Linux for s/390 Elf Application Binary Interface +Linux for z/Series Elf Application Binary Interface ( Both Highly Recommended ) +z/Architecture Principles of Operation SA22-7832-00 +Enterprise Systems Architecture/390 Reference Summary SA22-7209-01 & the +Enterprise Systems Architecture/390 Principles of Operation SA22-7201-05 + +Special Thanks +============== +Special thanks to Neale Ferguson who maintains a much +prettier HTML version of this page at +http://linuxvm.org/penguinvm/ +Bob Grainger Stefan Bader & others for reporting bugs diff --git a/Documentation/s390/cds.txt b/Documentation/s390/cds.txt new file mode 100644 index 000000000..480a78ef5 --- /dev/null +++ b/Documentation/s390/cds.txt @@ -0,0 +1,472 @@ +Linux for S/390 and zSeries + +Common Device Support (CDS) +Device Driver I/O Support Routines + +Authors : Ingo Adlung + Cornelia Huck + +Copyright, IBM Corp. 1999-2002 + +Introduction + +This document describes the common device support routines for Linux/390. +Different than other hardware architectures, ESA/390 has defined a unified +I/O access method. This gives relief to the device drivers as they don't +have to deal with different bus types, polling versus interrupt +processing, shared versus non-shared interrupt processing, DMA versus port +I/O (PIO), and other hardware features more. However, this implies that +either every single device driver needs to implement the hardware I/O +attachment functionality itself, or the operating system provides for a +unified method to access the hardware, providing all the functionality that +every single device driver would have to provide itself. + +The document does not intend to explain the ESA/390 hardware architecture in +every detail.This information can be obtained from the ESA/390 Principles of +Operation manual (IBM Form. No. SA22-7201). + +In order to build common device support for ESA/390 I/O interfaces, a +functional layer was introduced that provides generic I/O access methods to +the hardware. + +The common device support layer comprises the I/O support routines defined +below. Some of them implement common Linux device driver interfaces, while +some of them are ESA/390 platform specific. + +Note: +In order to write a driver for S/390, you also need to look into the interface +described in Documentation/s390/driver-model.txt. + +Note for porting drivers from 2.4: +The major changes are: +* The functions use a ccw_device instead of an irq (subchannel). +* All drivers must define a ccw_driver (see driver-model.txt) and the associated + functions. +* request_irq() and free_irq() are no longer done by the driver. +* The oper_handler is (kindof) replaced by the probe() and set_online() functions + of the ccw_driver. +* The not_oper_handler is (kindof) replaced by the remove() and set_offline() + functions of the ccw_driver. +* The channel device layer is gone. +* The interrupt handlers must be adapted to use a ccw_device as argument. + Moreover, they don't return a devstat, but an irb. +* Before initiating an io, the options must be set via ccw_device_set_options(). +* Instead of calling read_dev_chars()/read_conf_data(), the driver issues + the channel program and handles the interrupt itself. + +ccw_device_get_ciw() + get commands from extended sense data. + +ccw_device_start() +ccw_device_start_timeout() +ccw_device_start_key() +ccw_device_start_key_timeout() + initiate an I/O request. + +ccw_device_resume() + resume channel program execution. + +ccw_device_halt() + terminate the current I/O request processed on the device. + +do_IRQ() + generic interrupt routine. This function is called by the interrupt entry + routine whenever an I/O interrupt is presented to the system. The do_IRQ() + routine determines the interrupt status and calls the device specific + interrupt handler according to the rules (flags) defined during I/O request + initiation with do_IO(). + +The next chapters describe the functions other than do_IRQ() in more details. +The do_IRQ() interface is not described, as it is called from the Linux/390 +first level interrupt handler only and does not comprise a device driver +callable interface. Instead, the functional description of do_IO() also +describes the input to the device specific interrupt handler. + +Note: All explanations apply also to the 64 bit architecture s390x. + + +Common Device Support (CDS) for Linux/390 Device Drivers + +General Information + +The following chapters describe the I/O related interface routines the +Linux/390 common device support (CDS) provides to allow for device specific +driver implementations on the IBM ESA/390 hardware platform. Those interfaces +intend to provide the functionality required by every device driver +implementation to allow to drive a specific hardware device on the ESA/390 +platform. Some of the interface routines are specific to Linux/390 and some +of them can be found on other Linux platforms implementations too. +Miscellaneous function prototypes, data declarations, and macro definitions +can be found in the architecture specific C header file +linux/arch/s390/include/asm/irq.h. + +Overview of CDS interface concepts + +Different to other hardware platforms, the ESA/390 architecture doesn't define +interrupt lines managed by a specific interrupt controller and bus systems +that may or may not allow for shared interrupts, DMA processing, etc.. Instead, +the ESA/390 architecture has implemented a so called channel subsystem, that +provides a unified view of the devices physically attached to the systems. +Though the ESA/390 hardware platform knows about a huge variety of different +peripheral attachments like disk devices (aka. DASDs), tapes, communication +controllers, etc. they can all be accessed by a well defined access method and +they are presenting I/O completion a unified way : I/O interruptions. Every +single device is uniquely identified to the system by a so called subchannel, +where the ESA/390 architecture allows for 64k devices be attached. + +Linux, however, was first built on the Intel PC architecture, with its two +cascaded 8259 programmable interrupt controllers (PICs), that allow for a +maximum of 15 different interrupt lines. All devices attached to such a system +share those 15 interrupt levels. Devices attached to the ISA bus system must +not share interrupt levels (aka. IRQs), as the ISA bus bases on edge triggered +interrupts. MCA, EISA, PCI and other bus systems base on level triggered +interrupts, and therewith allow for shared IRQs. However, if multiple devices +present their hardware status by the same (shared) IRQ, the operating system +has to call every single device driver registered on this IRQ in order to +determine the device driver owning the device that raised the interrupt. + +Up to kernel 2.4, Linux/390 used to provide interfaces via the IRQ (subchannel). +For internal use of the common I/O layer, these are still there. However, +device drivers should use the new calling interface via the ccw_device only. + +During its startup the Linux/390 system checks for peripheral devices. Each +of those devices is uniquely defined by a so called subchannel by the ESA/390 +channel subsystem. While the subchannel numbers are system generated, each +subchannel also takes a user defined attribute, the so called device number. +Both subchannel number and device number cannot exceed 65535. During sysfs +initialisation, the information about control unit type and device types that +imply specific I/O commands (channel command words - CCWs) in order to operate +the device are gathered. Device drivers can retrieve this set of hardware +information during their initialization step to recognize the devices they +support using the information saved in the struct ccw_device given to them. +This methods implies that Linux/390 doesn't require to probe for free (not +armed) interrupt request lines (IRQs) to drive its devices with. Where +applicable, the device drivers can use issue the READ DEVICE CHARACTERISTICS +ccw to retrieve device characteristics in its online routine. + +In order to allow for easy I/O initiation the CDS layer provides a +ccw_device_start() interface that takes a device specific channel program (one +or more CCWs) as input sets up the required architecture specific control blocks +and initiates an I/O request on behalf of the device driver. The +ccw_device_start() routine allows to specify whether it expects the CDS layer +to notify the device driver for every interrupt it observes, or with final status +only. See ccw_device_start() for more details. A device driver must never issue +ESA/390 I/O commands itself, but must use the Linux/390 CDS interfaces instead. + +For long running I/O request to be canceled, the CDS layer provides the +ccw_device_halt() function. Some devices require to initially issue a HALT +SUBCHANNEL (HSCH) command without having pending I/O requests. This function is +also covered by ccw_device_halt(). + + +get_ciw() - get command information word + +This call enables a device driver to get information about supported commands +from the extended SenseID data. + +struct ciw * +ccw_device_get_ciw(struct ccw_device *cdev, __u32 cmd); + +cdev - The ccw_device for which the command is to be retrieved. +cmd - The command type to be retrieved. + +ccw_device_get_ciw() returns: +NULL - No extended data available, invalid device or command not found. +!NULL - The command requested. + + +ccw_device_start() - Initiate I/O Request + +The ccw_device_start() routines is the I/O request front-end processor. All +device driver I/O requests must be issued using this routine. A device driver +must not issue ESA/390 I/O commands itself. Instead the ccw_device_start() +routine provides all interfaces required to drive arbitrary devices. + +This description also covers the status information passed to the device +driver's interrupt handler as this is related to the rules (flags) defined +with the associated I/O request when calling ccw_device_start(). + +int ccw_device_start(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + unsigned long flags); +int ccw_device_start_timeout(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + unsigned long flags, + int expires); +int ccw_device_start_key(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + __u8 key, + unsigned long flags); +int ccw_device_start_key_timeout(struct ccw_device *cdev, + struct ccw1 *cpa, + unsigned long intparm, + __u8 lpm, + __u8 key, + unsigned long flags, + int expires); + +cdev : ccw_device the I/O is destined for +cpa : logical start address of channel program +user_intparm : user specific interrupt information; will be presented + back to the device driver's interrupt handler. Allows a + device driver to associate the interrupt with a + particular I/O request. +lpm : defines the channel path to be used for a specific I/O + request. A value of 0 will make cio use the opm. +key : the storage key to use for the I/O (useful for operating on a + storage with a storage key != default key) +flag : defines the action to be performed for I/O processing +expires : timeout value in jiffies. The common I/O layer will terminate + the running program after this and call the interrupt handler + with ERR_PTR(-ETIMEDOUT) as irb. + +Possible flag values are : + +DOIO_ALLOW_SUSPEND - channel program may become suspended +DOIO_DENY_PREFETCH - don't allow for CCW prefetch; usually + this implies the channel program might + become modified +DOIO_SUPPRESS_INTER - don't call the handler on intermediate status + +The cpa parameter points to the first format 1 CCW of a channel program : + +struct ccw1 { + __u8 cmd_code;/* command code */ + __u8 flags; /* flags, like IDA addressing, etc. */ + __u16 count; /* byte count */ + __u32 cda; /* data address */ +} __attribute__ ((packed,aligned(8))); + +with the following CCW flags values defined : + +CCW_FLAG_DC - data chaining +CCW_FLAG_CC - command chaining +CCW_FLAG_SLI - suppress incorrect length +CCW_FLAG_SKIP - skip +CCW_FLAG_PCI - PCI +CCW_FLAG_IDA - indirect addressing +CCW_FLAG_SUSPEND - suspend + + +Via ccw_device_set_options(), the device driver may specify the following +options for the device: + +DOIO_EARLY_NOTIFICATION - allow for early interrupt notification +DOIO_REPORT_ALL - report all interrupt conditions + + +The ccw_device_start() function returns : + + 0 - successful completion or request successfully initiated +-EBUSY - The device is currently processing a previous I/O request, or there is + a status pending at the device. +-ENODEV - cdev is invalid, the device is not operational or the ccw_device is + not online. + +When the I/O request completes, the CDS first level interrupt handler will +accumulate the status in a struct irb and then call the device interrupt handler. +The intparm field will contain the value the device driver has associated with a +particular I/O request. If a pending device status was recognized, +intparm will be set to 0 (zero). This may happen during I/O initiation or delayed +by an alert status notification. In any case this status is not related to the +current (last) I/O request. In case of a delayed status notification no special +interrupt will be presented to indicate I/O completion as the I/O request was +never started, even though ccw_device_start() returned with successful completion. + +The irb may contain an error value, and the device driver should check for this +first: + +-ETIMEDOUT: the common I/O layer terminated the request after the specified + timeout value +-EIO: the common I/O layer terminated the request due to an error state + +If the concurrent sense flag in the extended status word (esw) in the irb is +set, the field erw.scnt in the esw describes the number of device specific +sense bytes available in the extended control word irb->scsw.ecw[]. No device +sensing by the device driver itself is required. + +The device interrupt handler can use the following definitions to investigate +the primary unit check source coded in sense byte 0 : + +SNS0_CMD_REJECT 0x80 +SNS0_INTERVENTION_REQ 0x40 +SNS0_BUS_OUT_CHECK 0x20 +SNS0_EQUIPMENT_CHECK 0x10 +SNS0_DATA_CHECK 0x08 +SNS0_OVERRUN 0x04 +SNS0_INCOMPL_DOMAIN 0x01 + +Depending on the device status, multiple of those values may be set together. +Please refer to the device specific documentation for details. + +The irb->scsw.cstat field provides the (accumulated) subchannel status : + +SCHN_STAT_PCI - program controlled interrupt +SCHN_STAT_INCORR_LEN - incorrect length +SCHN_STAT_PROG_CHECK - program check +SCHN_STAT_PROT_CHECK - protection check +SCHN_STAT_CHN_DATA_CHK - channel data check +SCHN_STAT_CHN_CTRL_CHK - channel control check +SCHN_STAT_INTF_CTRL_CHK - interface control check +SCHN_STAT_CHAIN_CHECK - chaining check + +The irb->scsw.dstat field provides the (accumulated) device status : + +DEV_STAT_ATTENTION - attention +DEV_STAT_STAT_MOD - status modifier +DEV_STAT_CU_END - control unit end +DEV_STAT_BUSY - busy +DEV_STAT_CHN_END - channel end +DEV_STAT_DEV_END - device end +DEV_STAT_UNIT_CHECK - unit check +DEV_STAT_UNIT_EXCEP - unit exception + +Please see the ESA/390 Principles of Operation manual for details on the +individual flag meanings. + +Usage Notes : + +ccw_device_start() must be called disabled and with the ccw device lock held. + +The device driver is allowed to issue the next ccw_device_start() call from +within its interrupt handler already. It is not required to schedule a +bottom-half, unless a non deterministically long running error recovery procedure +or similar needs to be scheduled. During I/O processing the Linux/390 generic +I/O device driver support has already obtained the IRQ lock, i.e. the handler +must not try to obtain it again when calling ccw_device_start() or we end in a +deadlock situation! + +If a device driver relies on an I/O request to be completed prior to start the +next it can reduce I/O processing overhead by chaining a NoOp I/O command +CCW_CMD_NOOP to the end of the submitted CCW chain. This will force Channel-End +and Device-End status to be presented together, with a single interrupt. +However, this should be used with care as it implies the channel will remain +busy, not being able to process I/O requests for other devices on the same +channel. Therefore e.g. read commands should never use this technique, as the +result will be presented by a single interrupt anyway. + +In order to minimize I/O overhead, a device driver should use the +DOIO_REPORT_ALL only if the device can report intermediate interrupt +information prior to device-end the device driver urgently relies on. In this +case all I/O interruptions are presented to the device driver until final +status is recognized. + +If a device is able to recover from asynchronously presented I/O errors, it can +perform overlapping I/O using the DOIO_EARLY_NOTIFICATION flag. While some +devices always report channel-end and device-end together, with a single +interrupt, others present primary status (channel-end) when the channel is +ready for the next I/O request and secondary status (device-end) when the data +transmission has been completed at the device. + +Above flag allows to exploit this feature, e.g. for communication devices that +can handle lost data on the network to allow for enhanced I/O processing. + +Unless the channel subsystem at any time presents a secondary status interrupt, +exploiting this feature will cause only primary status interrupts to be +presented to the device driver while overlapping I/O is performed. When a +secondary status without error (alert status) is presented, this indicates +successful completion for all overlapping ccw_device_start() requests that have +been issued since the last secondary (final) status. + +Channel programs that intend to set the suspend flag on a channel command word +(CCW) must start the I/O operation with the DOIO_ALLOW_SUSPEND option or the +suspend flag will cause a channel program check. At the time the channel program +becomes suspended an intermediate interrupt will be generated by the channel +subsystem. + +ccw_device_resume() - Resume Channel Program Execution + +If a device driver chooses to suspend the current channel program execution by +setting the CCW suspend flag on a particular CCW, the channel program execution +is suspended. In order to resume channel program execution the CIO layer +provides the ccw_device_resume() routine. + +int ccw_device_resume(struct ccw_device *cdev); + +cdev - ccw_device the resume operation is requested for + +The ccw_device_resume() function returns: + + 0 - suspended channel program is resumed +-EBUSY - status pending +-ENODEV - cdev invalid or not-operational subchannel +-EINVAL - resume function not applicable +-ENOTCONN - there is no I/O request pending for completion + +Usage Notes: +Please have a look at the ccw_device_start() usage notes for more details on +suspended channel programs. + +ccw_device_halt() - Halt I/O Request Processing + +Sometimes a device driver might need a possibility to stop the processing of +a long-running channel program or the device might require to initially issue +a halt subchannel (HSCH) I/O command. For those purposes the ccw_device_halt() +command is provided. + +ccw_device_halt() must be called disabled and with the ccw device lock held. + +int ccw_device_halt(struct ccw_device *cdev, + unsigned long intparm); + +cdev : ccw_device the halt operation is requested for +intparm : interruption parameter; value is only used if no I/O + is outstanding, otherwise the intparm associated with + the I/O request is returned + +The ccw_device_halt() function returns : + + 0 - request successfully initiated +-EBUSY - the device is currently busy, or status pending. +-ENODEV - cdev invalid. +-EINVAL - The device is not operational or the ccw device is not online. + +Usage Notes : + +A device driver may write a never-ending channel program by writing a channel +program that at its end loops back to its beginning by means of a transfer in +channel (TIC) command (CCW_CMD_TIC). Usually this is performed by network +device drivers by setting the PCI CCW flag (CCW_FLAG_PCI). Once this CCW is +executed a program controlled interrupt (PCI) is generated. The device driver +can then perform an appropriate action. Prior to interrupt of an outstanding +read to a network device (with or without PCI flag) a ccw_device_halt() +is required to end the pending operation. + +ccw_device_clear() - Terminage I/O Request Processing + +In order to terminate all I/O processing at the subchannel, the clear subchannel +(CSCH) command is used. It can be issued via ccw_device_clear(). + +ccw_device_clear() must be called disabled and with the ccw device lock held. + +int ccw_device_clear(struct ccw_device *cdev, unsigned long intparm); + +cdev: ccw_device the clear operation is requested for +intparm: interruption parameter (see ccw_device_halt()) + +The ccw_device_clear() function returns: + + 0 - request successfully initiated +-ENODEV - cdev invalid +-EINVAL - The device is not operational or the ccw device is not online. + +Miscellaneous Support Routines + +This chapter describes various routines to be used in a Linux/390 device +driver programming environment. + +get_ccwdev_lock() + +Get the address of the device specific lock. This is then used in +spin_lock() / spin_unlock() calls. + + +__u8 ccw_device_get_path_mask(struct ccw_device *cdev); + +Get the mask of the path currently available for cdev. diff --git a/Documentation/s390/config3270.sh b/Documentation/s390/config3270.sh new file mode 100644 index 000000000..515e2f431 --- /dev/null +++ b/Documentation/s390/config3270.sh @@ -0,0 +1,76 @@ +#!/bin/sh +# +# config3270 -- Autoconfigure /dev/3270/* and /etc/inittab +# +# Usage: +# config3270 +# +# Output: +# /tmp/mkdev3270 +# +# Operation: +# 1. Run this script +# 2. Run the script it produces: /tmp/mkdev3270 +# 3. Issue "telinit q" or reboot, as appropriate. +# +P=/proc/tty/driver/tty3270 +ROOT= +D=$ROOT/dev +SUBD=3270 +TTY=$SUBD/tty +TUB=$SUBD/tub +SCR=$ROOT/tmp/mkdev3270 +SCRTMP=$SCR.a +GETTYLINE=:2345:respawn:/sbin/mingetty +INITTAB=$ROOT/etc/inittab +NINITTAB=$ROOT/etc/NEWinittab +OINITTAB=$ROOT/etc/OLDinittab +ADDNOTE=\\"# Additional mingettys for the 3270/tty* driver, tub3270 ---\\" + +if ! ls $P > /dev/null 2>&1; then + modprobe tub3270 > /dev/null 2>&1 +fi +ls $P > /dev/null 2>&1 || exit 1 + +# Initialize two files, one for /dev/3270 commands and one +# to replace the /etc/inittab file (old one saved in OLDinittab) +echo "#!/bin/sh" > $SCR || exit 1 +echo " " >> $SCR +echo "# Script built by /sbin/config3270" >> $SCR +if [ ! -d /dev/dasd ]; then + echo rm -rf "$D/$SUBD/*" >> $SCR +fi +echo "grep -v $TTY $INITTAB > $NINITTAB" > $SCRTMP || exit 1 +echo "echo $ADDNOTE >> $NINITTAB" >> $SCRTMP +if [ ! -d /dev/dasd ]; then + echo mkdir -p $D/$SUBD >> $SCR +fi + +# Now query the tub3270 driver for 3270 device information +# and add appropriate mknod and mingetty lines to our files +echo what=config > $P +while read devno maj min;do + if [ $min = 0 ]; then + fsmaj=$maj + if [ ! -d /dev/dasd ]; then + echo mknod $D/$TUB c $fsmaj 0 >> $SCR + echo chmod 666 $D/$TUB >> $SCR + fi + elif [ $maj = CONSOLE ]; then + if [ ! -d /dev/dasd ]; then + echo mknod $D/$TUB$devno c $fsmaj $min >> $SCR + fi + else + if [ ! -d /dev/dasd ]; then + echo mknod $D/$TTY$devno c $maj $min >>$SCR + echo mknod $D/$TUB$devno c $fsmaj $min >> $SCR + fi + echo "echo t$min$GETTYLINE $TTY$devno >> $NINITTAB" >> $SCRTMP + fi +done < $P + +echo mv $INITTAB $OINITTAB >> $SCRTMP || exit 1 +echo mv $NINITTAB $INITTAB >> $SCRTMP +cat $SCRTMP >> $SCR +rm $SCRTMP +exit 0 diff --git a/Documentation/s390/driver-model.txt b/Documentation/s390/driver-model.txt new file mode 100644 index 000000000..ed265cf54 --- /dev/null +++ b/Documentation/s390/driver-model.txt @@ -0,0 +1,287 @@ +S/390 driver model interfaces +----------------------------- + +1. CCW devices +-------------- + +All devices which can be addressed by means of ccws are called 'CCW devices' - +even if they aren't actually driven by ccws. + +All ccw devices are accessed via a subchannel, this is reflected in the +structures under devices/: + +devices/ + - system/ + - css0/ + - 0.0.0000/0.0.0815/ + - 0.0.0001/0.0.4711/ + - 0.0.0002/ + - 0.1.0000/0.1.1234/ + ... + - defunct/ + +In this example, device 0815 is accessed via subchannel 0 in subchannel set 0, +device 4711 via subchannel 1 in subchannel set 0, and subchannel 2 is a non-I/O +subchannel. Device 1234 is accessed via subchannel 0 in subchannel set 1. + +The subchannel named 'defunct' does not represent any real subchannel on the +system; it is a pseudo subchannel where disconnected ccw devices are moved to +if they are displaced by another ccw device becoming operational on their +former subchannel. The ccw devices will be moved again to a proper subchannel +if they become operational again on that subchannel. + +You should address a ccw device via its bus id (e.g. 0.0.4711); the device can +be found under bus/ccw/devices/. + +All ccw devices export some data via sysfs. + +cutype: The control unit type / model. + +devtype: The device type / model, if applicable. + +availability: Can be 'good' or 'boxed'; 'no path' or 'no device' for + disconnected devices. + +online: An interface to set the device online and offline. + In the special case of the device being disconnected (see the + notify function under 1.2), piping 0 to online will forcibly delete + the device. + +The device drivers can add entries to export per-device data and interfaces. + +There is also some data exported on a per-subchannel basis (see under +bus/css/devices/): + +chpids: Via which chpids the device is connected. + +pimpampom: The path installed, path available and path operational masks. + +There also might be additional data, for example for block devices. + + +1.1 Bringing up a ccw device +---------------------------- + +This is done in several steps. + +a. Each driver can provide one or more parameter interfaces where parameters can + be specified. These interfaces are also in the driver's responsibility. +b. After a. has been performed, if necessary, the device is finally brought up + via the 'online' interface. + + +1.2 Writing a driver for ccw devices +------------------------------------ + +The basic struct ccw_device and struct ccw_driver data structures can be found +under include/asm/ccwdev.h. + +struct ccw_device { + spinlock_t *ccwlock; + struct ccw_device_private *private; + struct ccw_device_id id; + + struct ccw_driver *drv; + struct device dev; + int online; + + void (*handler) (struct ccw_device *dev, unsigned long intparm, + struct irb *irb); +}; + +struct ccw_driver { + struct module *owner; + struct ccw_device_id *ids; + int (*probe) (struct ccw_device *); + int (*remove) (struct ccw_device *); + int (*set_online) (struct ccw_device *); + int (*set_offline) (struct ccw_device *); + int (*notify) (struct ccw_device *, int); + struct device_driver driver; + char *name; +}; + +The 'private' field contains data needed for internal i/o operation only, and +is not available to the device driver. + +Each driver should declare in a MODULE_DEVICE_TABLE into which CU types/models +and/or device types/models it is interested. This information can later be found +in the struct ccw_device_id fields: + +struct ccw_device_id { + __u16 match_flags; + + __u16 cu_type; + __u16 dev_type; + __u8 cu_model; + __u8 dev_model; + + unsigned long driver_info; +}; + +The functions in ccw_driver should be used in the following way: +probe: This function is called by the device layer for each device the driver + is interested in. The driver should only allocate private structures + to put in dev->driver_data and create attributes (if needed). Also, + the interrupt handler (see below) should be set here. + +int (*probe) (struct ccw_device *cdev); + +Parameters: cdev - the device to be probed. + + +remove: This function is called by the device layer upon removal of the driver, + the device or the module. The driver should perform cleanups here. + +int (*remove) (struct ccw_device *cdev); + +Parameters: cdev - the device to be removed. + + +set_online: This function is called by the common I/O layer when the device is + activated via the 'online' attribute. The driver should finally + setup and activate the device here. + +int (*set_online) (struct ccw_device *); + +Parameters: cdev - the device to be activated. The common layer has + verified that the device is not already online. + + +set_offline: This function is called by the common I/O layer when the device is + de-activated via the 'online' attribute. The driver should shut + down the device, but not de-allocate its private data. + +int (*set_offline) (struct ccw_device *); + +Parameters: cdev - the device to be deactivated. The common layer has + verified that the device is online. + + +notify: This function is called by the common I/O layer for some state changes + of the device. + Signalled to the driver are: + * In online state, device detached (CIO_GONE) or last path gone + (CIO_NO_PATH). The driver must return !0 to keep the device; for + return code 0, the device will be deleted as usual (also when no + notify function is registered). If the driver wants to keep the + device, it is moved into disconnected state. + * In disconnected state, device operational again (CIO_OPER). The + common I/O layer performs some sanity checks on device number and + Device / CU to be reasonably sure if it is still the same device. + If not, the old device is removed and a new one registered. By the + return code of the notify function the device driver signals if it + wants the device back: !0 for keeping, 0 to make the device being + removed and re-registered. + +int (*notify) (struct ccw_device *, int); + +Parameters: cdev - the device whose state changed. + event - the event that happened. This can be one of CIO_GONE, + CIO_NO_PATH or CIO_OPER. + +The handler field of the struct ccw_device is meant to be set to the interrupt +handler for the device. In order to accommodate drivers which use several +distinct handlers (e.g. multi subchannel devices), this is a member of ccw_device +instead of ccw_driver. +The handler is registered with the common layer during set_online() processing +before the driver is called, and is deregistered during set_offline() after the +driver has been called. Also, after registering / before deregistering, path +grouping resp. disbanding of the path group (if applicable) are performed. + +void (*handler) (struct ccw_device *dev, unsigned long intparm, struct irb *irb); + +Parameters: dev - the device the handler is called for + intparm - the intparm which allows the device driver to identify + the i/o the interrupt is associated with, or to recognize + the interrupt as unsolicited. + irb - interruption response block which contains the accumulated + status. + +The device driver is called from the common ccw_device layer and can retrieve +information about the interrupt from the irb parameter. + + +1.3 ccwgroup devices +-------------------- + +The ccwgroup mechanism is designed to handle devices consisting of multiple ccw +devices, like lcs or ctc. + +The ccw driver provides a 'group' attribute. Piping bus ids of ccw devices to +this attributes creates a ccwgroup device consisting of these ccw devices (if +possible). This ccwgroup device can be set online or offline just like a normal +ccw device. + +Each ccwgroup device also provides an 'ungroup' attribute to destroy the device +again (only when offline). This is a generic ccwgroup mechanism (the driver does +not need to implement anything beyond normal removal routines). + +A ccw device which is a member of a ccwgroup device carries a pointer to the +ccwgroup device in the driver_data of its device struct. This field must not be +touched by the driver - it should use the ccwgroup device's driver_data for its +private data. + +To implement a ccwgroup driver, please refer to include/asm/ccwgroup.h. Keep in +mind that most drivers will need to implement both a ccwgroup and a ccw +driver. + + +2. Channel paths +----------------- + +Channel paths show up, like subchannels, under the channel subsystem root (css0) +and are called 'chp0.'. They have no driver and do not belong to any bus. +Please note, that unlike /proc/chpids in 2.4, the channel path objects reflect +only the logical state and not the physical state, since we cannot track the +latter consistently due to lacking machine support (we don't need to be aware +of it anyway). + +status - Can be 'online' or 'offline'. + Piping 'on' or 'off' sets the chpid logically online/offline. + Piping 'on' to an online chpid triggers path reprobing for all devices + the chpid connects to. This can be used to force the kernel to re-use + a channel path the user knows to be online, but the machine hasn't + created a machine check for. + +type - The physical type of the channel path. + +shared - Whether the channel path is shared. + +cmg - The channel measurement group. + +3. System devices +----------------- + +3.1 xpram +--------- + +xpram shows up under devices/system/ as 'xpram'. + +3.2 cpus +-------- + +For each cpu, a directory is created under devices/system/cpu/. Each cpu has an +attribute 'online' which can be 0 or 1. + + +4. Other devices +---------------- + +4.1 Netiucv +----------- + +The netiucv driver creates an attribute 'connection' under +bus/iucv/drivers/netiucv. Piping to this attribute creates a new netiucv +connection to the specified host. + +Netiucv connections show up under devices/iucv/ as "netiucv". The interface +number is assigned sequentially to the connections defined via the 'connection' +attribute. + +user - shows the connection partner. + +buffer - maximum buffer size. + Pipe to it to change buffer size. + + diff --git a/Documentation/s390/monreader.txt b/Documentation/s390/monreader.txt new file mode 100644 index 000000000..d3729585f --- /dev/null +++ b/Documentation/s390/monreader.txt @@ -0,0 +1,197 @@ + +Date : 2004-Nov-26 +Author: Gerald Schaefer (geraldsc@de.ibm.com) + + + Linux API for read access to z/VM Monitor Records + ================================================= + + +Description +=========== +This item delivers a new Linux API in the form of a misc char device that is +usable from user space and allows read access to the z/VM Monitor Records +collected by the *MONITOR System Service of z/VM. + + +User Requirements +================= +The z/VM guest on which you want to access this API needs to be configured in +order to allow IUCV connections to the *MONITOR service, i.e. it needs the +IUCV *MONITOR statement in its user entry. If the monitor DCSS to be used is +restricted (likely), you also need the NAMESAVE statement. +This item will use the IUCV device driver to access the z/VM services, so you +need a kernel with IUCV support. You also need z/VM version 4.4 or 5.1. + +There are two options for being able to load the monitor DCSS (examples assume +that the monitor DCSS begins at 144 MB and ends at 152 MB). You can query the +location of the monitor DCSS with the Class E privileged CP command Q NSS MAP +(the values BEGPAG and ENDPAG are given in units of 4K pages). + +See also "CP Command and Utility Reference" (SC24-6081-00) for more information +on the DEF STOR and Q NSS MAP commands, as well as "Saved Segments Planning +and Administration" (SC24-6116-00) for more information on DCSSes. + +1st option: +----------- +You can use the CP command DEF STOR CONFIG to define a "memory hole" in your +guest virtual storage around the address range of the DCSS. + +Example: DEF STOR CONFIG 0.140M 200M.200M + +This defines two blocks of storage, the first is 140MB in size an begins at +address 0MB, the second is 200MB in size and begins at address 200MB, +resulting in a total storage of 340MB. Note that the first block should +always start at 0 and be at least 64MB in size. + +2nd option: +----------- +Your guest virtual storage has to end below the starting address of the DCSS +and you have to specify the "mem=" kernel parameter in your parmfile with a +value greater than the ending address of the DCSS. + +Example: DEF STOR 140M + +This defines 140MB storage size for your guest, the parameter "mem=160M" is +added to the parmfile. + + +User Interface +============== +The char device is implemented as a kernel module named "monreader", +which can be loaded via the modprobe command, or it can be compiled into the +kernel instead. There is one optional module (or kernel) parameter, "mondcss", +to specify the name of the monitor DCSS. If the module is compiled into the +kernel, the kernel parameter "monreader.mondcss=" can be specified +in the parmfile. + +The default name for the DCSS is "MONDCSS" if none is specified. In case that +there are other users already connected to the *MONITOR service (e.g. +Performance Toolkit), the monitor DCSS is already defined and you have to use +the same DCSS. The CP command Q MONITOR (Class E privileged) shows the name +of the monitor DCSS, if already defined, and the users connected to the +*MONITOR service. +Refer to the "z/VM Performance" book (SC24-6109-00) on how to create a monitor +DCSS if your z/VM doesn't have one already, you need Class E privileges to +define and save a DCSS. + +Example: +-------- +modprobe monreader mondcss=MYDCSS + +This loads the module and sets the DCSS name to "MYDCSS". + +NOTE: +----- +This API provides no interface to control the *MONITOR service, e.g. specify +which data should be collected. This can be done by the CP command MONITOR +(Class E privileged), see "CP Command and Utility Reference". + +Device nodes with udev: +----------------------- +After loading the module, a char device will be created along with the device +node //monreader. + +Device nodes without udev: +-------------------------- +If your distribution does not support udev, a device node will not be created +automatically and you have to create it manually after loading the module. +Therefore you need to know the major and minor numbers of the device. These +numbers can be found in /sys/class/misc/monreader/dev. +Typing cat /sys/class/misc/monreader/dev will give an output of the form +:. The device node can be created via the mknod command, enter +mknod c , where is the name of the device node +to be created. + +Example: +-------- +# modprobe monreader +# cat /sys/class/misc/monreader/dev +10:63 +# mknod /dev/monreader c 10 63 + +This loads the module with the default monitor DCSS (MONDCSS) and creates a +device node. + +File operations: +---------------- +The following file operations are supported: open, release, read, poll. +There are two alternative methods for reading: either non-blocking read in +conjunction with polling, or blocking read without polling. IOCTLs are not +supported. + +Read: +----- +Reading from the device provides a 12 Byte monitor control element (MCE), +followed by a set of one or more contiguous monitor records (similar to the +output of the CMS utility MONWRITE without the 4K control blocks). The MCE +contains information on the type of the following record set (sample/event +data), the monitor domains contained within it and the start and end address +of the record set in the monitor DCSS. The start and end address can be used +to determine the size of the record set, the end address is the address of the +last byte of data. The start address is needed to handle "end-of-frame" records +correctly (domain 1, record 13), i.e. it can be used to determine the record +start offset relative to a 4K page (frame) boundary. + +See "Appendix A: *MONITOR" in the "z/VM Performance" document for a description +of the monitor control element layout. The layout of the monitor records can +be found here (z/VM 5.1): http://www.vm.ibm.com/pubs/mon510/index.html + +The layout of the data stream provided by the monreader device is as follows: +... +<0 byte read> + \ + | +... |- data set + | + / +<0 byte read> +... + +There may be more than one combination of MCE and corresponding record set +within one data set and the end of each data set is indicated by a successful +read with a return value of 0 (0 byte read). +Any received data must be considered invalid until a complete set was +read successfully, including the closing 0 byte read. Therefore you should +always read the complete set into a buffer before processing the data. + +The maximum size of a data set can be as large as the size of the +monitor DCSS, so design the buffer adequately or use dynamic memory allocation. +The size of the monitor DCSS will be printed into syslog after loading the +module. You can also use the (Class E privileged) CP command Q NSS MAP to +list all available segments and information about them. + +As with most char devices, error conditions are indicated by returning a +negative value for the number of bytes read. In this case, the errno variable +indicates the error condition: + +EIO: reply failed, read data is invalid and the application + should discard the data read since the last successful read with 0 size. +EFAULT: copy_to_user failed, read data is invalid and the application should + discard the data read since the last successful read with 0 size. +EAGAIN: occurs on a non-blocking read if there is no data available at the + moment. There is no data missing or corrupted, just try again or rather + use polling for non-blocking reads. +EOVERFLOW: message limit reached, the data read since the last successful + read with 0 size is valid but subsequent records may be missing. + +In the last case (EOVERFLOW) there may be missing data, in the first two cases +(EIO, EFAULT) there will be missing data. It's up to the application if it will +continue reading subsequent data or rather exit. + +Open: +----- +Only one user is allowed to open the char device. If it is already in use, the +open function will fail (return a negative value) and set errno to EBUSY. +The open function may also fail if an IUCV connection to the *MONITOR service +cannot be established. In this case errno will be set to EIO and an error +message with an IPUSER SEVER code will be printed into syslog. The IPUSER SEVER +codes are described in the "z/VM Performance" book, Appendix A. + +NOTE: +----- +As soon as the device is opened, incoming messages will be accepted and they +will account for the message limit, i.e. opening the device without reading +from it will provoke the "message limit reached" error (EOVERFLOW error code) +eventually. + diff --git a/Documentation/s390/qeth.txt b/Documentation/s390/qeth.txt new file mode 100644 index 000000000..aa06fcf5f --- /dev/null +++ b/Documentation/s390/qeth.txt @@ -0,0 +1,50 @@ +IBM s390 QDIO Ethernet Driver + +OSA and HiperSockets Bridge Port Support + +Uevents + +To generate the events the device must be assigned a role of either +a primary or a secondary Bridge Port. For more information, see +"z/VM Connectivity, SC24-6174". + +When run on an OSA or HiperSockets Bridge Capable Port hardware, and the state +of some configured Bridge Port device on the channel changes, a udev +event with ACTION=CHANGE is emitted on behalf of the corresponding +ccwgroup device. The event has the following attributes: + +BRIDGEPORT=statechange - indicates that the Bridge Port device changed + its state. + +ROLE={primary|secondary|none} - the role assigned to the port. + +STATE={active|standby|inactive} - the newly assumed state of the port. + +When run on HiperSockets Bridge Capable Port hardware with host address +notifications enabled, a udev event with ACTION=CHANGE is emitted. +It is emitted on behalf of the corresponding ccwgroup device when a host +or a VLAN is registered or unregistered on the network served by the device. +The event has the following attributes: + +BRIDGEDHOST={reset|register|deregister|abort} - host address + notifications are started afresh, a new host or VLAN is registered or + deregistered on the Bridge Port HiperSockets channel, or address + notifications are aborted. + +VLAN=numeric-vlan-id - VLAN ID on which the event occurred. Not included + if no VLAN is involved in the event. + +MAC=xx:xx:xx:xx:xx:xx - MAC address of the host that is being registered + or deregistered from the HiperSockets channel. Not reported if the + event reports the creation or destruction of a VLAN. + +NTOK_BUSID=x.y.zzzz - device bus ID (CSSID, SSID and device number). + +NTOK_IID=xx - device IID. + +NTOK_CHPID=xx - device CHPID. + +NTOK_CHID=xxxx - device channel ID. + +Note that the NTOK_* attributes refer to devices other than the one +connected to the system on which the OS is running. diff --git a/Documentation/s390/s390dbf.txt b/Documentation/s390/s390dbf.txt new file mode 100644 index 000000000..61329fd62 --- /dev/null +++ b/Documentation/s390/s390dbf.txt @@ -0,0 +1,667 @@ +S390 Debug Feature +================== + +files: arch/s390/kernel/debug.c + arch/s390/include/asm/debug.h + +Description: +------------ +The goal of this feature is to provide a kernel debug logging API +where log records can be stored efficiently in memory, where each component +(e.g. device drivers) can have one separate debug log. +One purpose of this is to inspect the debug logs after a production system crash +in order to analyze the reason for the crash. +If the system still runs but only a subcomponent which uses dbf fails, +it is possible to look at the debug logs on a live system via the Linux +debugfs filesystem. +The debug feature may also very useful for kernel and driver development. + +Design: +------- +Kernel components (e.g. device drivers) can register themselves at the debug +feature with the function call debug_register(). This function initializes a +debug log for the caller. For each debug log exists a number of debug areas +where exactly one is active at one time. Each debug area consists of contiguous +pages in memory. In the debug areas there are stored debug entries (log records) +which are written by event- and exception-calls. + +An event-call writes the specified debug entry to the active debug +area and updates the log pointer for the active area. If the end +of the active debug area is reached, a wrap around is done (ring buffer) +and the next debug entry will be written at the beginning of the active +debug area. + +An exception-call writes the specified debug entry to the log and +switches to the next debug area. This is done in order to be sure +that the records which describe the origin of the exception are not +overwritten when a wrap around for the current area occurs. + +The debug areas themselves are also ordered in form of a ring buffer. +When an exception is thrown in the last debug area, the following debug +entries are then written again in the very first area. + +There are three versions for the event- and exception-calls: One for +logging raw data, one for text and one for numbers. + +Each debug entry contains the following data: + +- Timestamp +- Cpu-Number of calling task +- Level of debug entry (0...6) +- Return Address to caller +- Flag, if entry is an exception or not + +The debug logs can be inspected in a live system through entries in +the debugfs-filesystem. Under the toplevel directory "s390dbf" there is +a directory for each registered component, which is named like the +corresponding component. The debugfs normally should be mounted to +/sys/kernel/debug therefore the debug feature can be accessed under +/sys/kernel/debug/s390dbf. + +The content of the directories are files which represent different views +to the debug log. Each component can decide which views should be +used through registering them with the function debug_register_view(). +Predefined views for hex/ascii, sprintf and raw binary data are provided. +It is also possible to define other views. The content of +a view can be inspected simply by reading the corresponding debugfs file. + +All debug logs have an actual debug level (range from 0 to 6). +The default level is 3. Event and Exception functions have a 'level' +parameter. Only debug entries with a level that is lower or equal +than the actual level are written to the log. This means, when +writing events, high priority log entries should have a low level +value whereas low priority entries should have a high one. +The actual debug level can be changed with the help of the debugfs-filesystem +through writing a number string "x" to the 'level' debugfs file which is +provided for every debug log. Debugging can be switched off completely +by using "-" on the 'level' debugfs file. + +Example: + +> echo "-" > /sys/kernel/debug/s390dbf/dasd/level + +It is also possible to deactivate the debug feature globally for every +debug log. You can change the behavior using 2 sysctl parameters in +/proc/sys/s390dbf: +There are currently 2 possible triggers, which stop the debug feature +globally. The first possibility is to use the "debug_active" sysctl. If +set to 1 the debug feature is running. If "debug_active" is set to 0 the +debug feature is turned off. +The second trigger which stops the debug feature is a kernel oops. +That prevents the debug feature from overwriting debug information that +happened before the oops. After an oops you can reactivate the debug feature +by piping 1 to /proc/sys/s390dbf/debug_active. Nevertheless, its not +suggested to use an oopsed kernel in a production environment. +If you want to disallow the deactivation of the debug feature, you can use +the "debug_stoppable" sysctl. If you set "debug_stoppable" to 0 the debug +feature cannot be stopped. If the debug feature is already stopped, it +will stay deactivated. + +Kernel Interfaces: +------------------ + +---------------------------------------------------------------------------- +debug_info_t *debug_register(char *name, int pages, int nr_areas, + int buf_size); + +Parameter: name: Name of debug log (e.g. used for debugfs entry) + pages: number of pages, which will be allocated per area + nr_areas: number of debug areas + buf_size: size of data area in each debug entry + +Return Value: Handle for generated debug area + NULL if register failed + +Description: Allocates memory for a debug log + Must not be called within an interrupt handler + +---------------------------------------------------------------------------- +debug_info_t *debug_register_mode(char *name, int pages, int nr_areas, + int buf_size, mode_t mode, uid_t uid, + gid_t gid); + +Parameter: name: Name of debug log (e.g. used for debugfs entry) + pages: Number of pages, which will be allocated per area + nr_areas: Number of debug areas + buf_size: Size of data area in each debug entry + mode: File mode for debugfs files. E.g. S_IRWXUGO + uid: User ID for debugfs files. Currently only 0 is + supported. + gid: Group ID for debugfs files. Currently only 0 is + supported. + +Return Value: Handle for generated debug area + NULL if register failed + +Description: Allocates memory for a debug log + Must not be called within an interrupt handler + +--------------------------------------------------------------------------- +void debug_unregister (debug_info_t * id); + +Parameter: id: handle for debug log + +Return Value: none + +Description: frees memory for a debug log and removes all registered debug + views. + Must not be called within an interrupt handler + +--------------------------------------------------------------------------- +void debug_set_level (debug_info_t * id, int new_level); + +Parameter: id: handle for debug log + new_level: new debug level + +Return Value: none + +Description: Sets new actual debug level if new_level is valid. + +--------------------------------------------------------------------------- +bool debug_level_enabled (debug_info_t * id, int level); + +Parameter: id: handle for debug log + level: debug level + +Return Value: True if level is less or equal to the current debug level. + +Description: Returns true if debug events for the specified level would be + logged. Otherwise returns false. +--------------------------------------------------------------------------- +void debug_stop_all(void); + +Parameter: none + +Return Value: none + +Description: stops the debug feature if stopping is allowed. Currently + used in case of a kernel oops. + +--------------------------------------------------------------------------- +debug_entry_t* debug_event (debug_info_t* id, int level, void* data, + int length); + +Parameter: id: handle for debug log + level: debug level + data: pointer to data for debug entry + length: length of data in bytes + +Return Value: Address of written debug entry + +Description: writes debug entry to active debug area (if level <= actual + debug level) + +--------------------------------------------------------------------------- +debug_entry_t* debug_int_event (debug_info_t * id, int level, + unsigned int data); +debug_entry_t* debug_long_event(debug_info_t * id, int level, + unsigned long data); + +Parameter: id: handle for debug log + level: debug level + data: integer value for debug entry + +Return Value: Address of written debug entry + +Description: writes debug entry to active debug area (if level <= actual + debug level) + +--------------------------------------------------------------------------- +debug_entry_t* debug_text_event (debug_info_t * id, int level, + const char* data); + +Parameter: id: handle for debug log + level: debug level + data: string for debug entry + +Return Value: Address of written debug entry + +Description: writes debug entry in ascii format to active debug area + (if level <= actual debug level) + +--------------------------------------------------------------------------- +debug_entry_t* debug_sprintf_event (debug_info_t * id, int level, + char* string,...); + +Parameter: id: handle for debug log + level: debug level + string: format string for debug entry + ...: varargs used as in sprintf() + +Return Value: Address of written debug entry + +Description: writes debug entry with format string and varargs (longs) to + active debug area (if level $<=$ actual debug level). + floats and long long datatypes cannot be used as varargs. + +--------------------------------------------------------------------------- + +debug_entry_t* debug_exception (debug_info_t* id, int level, void* data, + int length); + +Parameter: id: handle for debug log + level: debug level + data: pointer to data for debug entry + length: length of data in bytes + +Return Value: Address of written debug entry + +Description: writes debug entry to active debug area (if level <= actual + debug level) and switches to next debug area + +--------------------------------------------------------------------------- +debug_entry_t* debug_int_exception (debug_info_t * id, int level, + unsigned int data); +debug_entry_t* debug_long_exception(debug_info_t * id, int level, + unsigned long data); + +Parameter: id: handle for debug log + level: debug level + data: integer value for debug entry + +Return Value: Address of written debug entry + +Description: writes debug entry to active debug area (if level <= actual + debug level) and switches to next debug area + +--------------------------------------------------------------------------- +debug_entry_t* debug_text_exception (debug_info_t * id, int level, + const char* data); + +Parameter: id: handle for debug log + level: debug level + data: string for debug entry + +Return Value: Address of written debug entry + +Description: writes debug entry in ascii format to active debug area + (if level <= actual debug level) and switches to next debug + area + +--------------------------------------------------------------------------- +debug_entry_t* debug_sprintf_exception (debug_info_t * id, int level, + char* string,...); + +Parameter: id: handle for debug log + level: debug level + string: format string for debug entry + ...: varargs used as in sprintf() + +Return Value: Address of written debug entry + +Description: writes debug entry with format string and varargs (longs) to + active debug area (if level $<=$ actual debug level) and + switches to next debug area. + floats and long long datatypes cannot be used as varargs. + +--------------------------------------------------------------------------- + +int debug_register_view (debug_info_t * id, struct debug_view *view); + +Parameter: id: handle for debug log + view: pointer to debug view struct + +Return Value: 0 : ok + < 0: Error + +Description: registers new debug view and creates debugfs dir entry + +--------------------------------------------------------------------------- +int debug_unregister_view (debug_info_t * id, struct debug_view *view); + +Parameter: id: handle for debug log + view: pointer to debug view struct + +Return Value: 0 : ok + < 0: Error + +Description: unregisters debug view and removes debugfs dir entry + + + +Predefined views: +----------------- + +extern struct debug_view debug_hex_ascii_view; +extern struct debug_view debug_raw_view; +extern struct debug_view debug_sprintf_view; + +Examples +-------- + +/* + * hex_ascii- + raw-view Example + */ + +#include +#include + +static debug_info_t* debug_info; + +static int init(void) +{ + /* register 4 debug areas with one page each and 4 byte data field */ + + debug_info = debug_register ("test", 1, 4, 4 ); + debug_register_view(debug_info,&debug_hex_ascii_view); + debug_register_view(debug_info,&debug_raw_view); + + debug_text_event(debug_info, 4 , "one "); + debug_int_exception(debug_info, 4, 4711); + debug_event(debug_info, 3, &debug_info, 4); + + return 0; +} + +static void cleanup(void) +{ + debug_unregister (debug_info); +} + +module_init(init); +module_exit(cleanup); + +--------------------------------------------------------------------------- + +/* + * sprintf-view Example + */ + +#include +#include + +static debug_info_t* debug_info; + +static int init(void) +{ + /* register 4 debug areas with one page each and data field for */ + /* format string pointer + 2 varargs (= 3 * sizeof(long)) */ + + debug_info = debug_register ("test", 1, 4, sizeof(long) * 3); + debug_register_view(debug_info,&debug_sprintf_view); + + debug_sprintf_event(debug_info, 2 , "first event in %s:%i\n",__FILE__,__LINE__); + debug_sprintf_exception(debug_info, 1, "pointer to debug info: %p\n",&debug_info); + + return 0; +} + +static void cleanup(void) +{ + debug_unregister (debug_info); +} + +module_init(init); +module_exit(cleanup); + + + +Debugfs Interface +---------------- +Views to the debug logs can be investigated through reading the corresponding +debugfs-files: + +Example: + +> ls /sys/kernel/debug/s390dbf/dasd +flush hex_ascii level pages raw +> cat /sys/kernel/debug/s390dbf/dasd/hex_ascii | sort -k2,2 -s +00 00974733272:680099 2 - 02 0006ad7e 07 ea 4a 90 | .... +00 00974733272:682210 2 - 02 0006ade6 46 52 45 45 | FREE +00 00974733272:682213 2 - 02 0006adf6 07 ea 4a 90 | .... +00 00974733272:682281 1 * 02 0006ab08 41 4c 4c 43 | EXCP +01 00974733272:682284 2 - 02 0006ab16 45 43 4b 44 | ECKD +01 00974733272:682287 2 - 02 0006ab28 00 00 00 04 | .... +01 00974733272:682289 2 - 02 0006ab3e 00 00 00 20 | ... +01 00974733272:682297 2 - 02 0006ad7e 07 ea 4a 90 | .... +01 00974733272:684384 2 - 00 0006ade6 46 52 45 45 | FREE +01 00974733272:684388 2 - 00 0006adf6 07 ea 4a 90 | .... + +See section about predefined views for explanation of the above output! + +Changing the debug level +------------------------ + +Example: + + +> cat /sys/kernel/debug/s390dbf/dasd/level +3 +> echo "5" > /sys/kernel/debug/s390dbf/dasd/level +> cat /sys/kernel/debug/s390dbf/dasd/level +5 + +Flushing debug areas +-------------------- +Debug areas can be flushed with piping the number of the desired +area (0...n) to the debugfs file "flush". When using "-" all debug areas +are flushed. + +Examples: + +1. Flush debug area 0: +> echo "0" > /sys/kernel/debug/s390dbf/dasd/flush + +2. Flush all debug areas: +> echo "-" > /sys/kernel/debug/s390dbf/dasd/flush + +Changing the size of debug areas +------------------------------------ +It is possible the change the size of debug areas through piping +the number of pages to the debugfs file "pages". The resize request will +also flush the debug areas. + +Example: + +Define 4 pages for the debug areas of debug feature "dasd": +> echo "4" > /sys/kernel/debug/s390dbf/dasd/pages + +Stooping the debug feature +-------------------------- +Example: + +1. Check if stopping is allowed +> cat /proc/sys/s390dbf/debug_stoppable +2. Stop debug feature +> echo 0 > /proc/sys/s390dbf/debug_active + +lcrash Interface +---------------- +It is planned that the dump analysis tool lcrash gets an additional command +'s390dbf' to display all the debug logs. With this tool it will be possible +to investigate the debug logs on a live system and with a memory dump after +a system crash. + +Investigating raw memory +------------------------ +One last possibility to investigate the debug logs at a live +system and after a system crash is to look at the raw memory +under VM or at the Service Element. +It is possible to find the anker of the debug-logs through +the 'debug_area_first' symbol in the System map. Then one has +to follow the correct pointers of the data-structures defined +in debug.h and find the debug-areas in memory. +Normally modules which use the debug feature will also have +a global variable with the pointer to the debug-logs. Following +this pointer it will also be possible to find the debug logs in +memory. + +For this method it is recommended to use '16 * x + 4' byte (x = 0..n) +for the length of the data field in debug_register() in +order to see the debug entries well formatted. + + +Predefined Views +---------------- + +There are three predefined views: hex_ascii, raw and sprintf. +The hex_ascii view shows the data field in hex and ascii representation +(e.g. '45 43 4b 44 | ECKD'). +The raw view returns a bytestream as the debug areas are stored in memory. + +The sprintf view formats the debug entries in the same way as the sprintf +function would do. The sprintf event/exception functions write to the +debug entry a pointer to the format string (size = sizeof(long)) +and for each vararg a long value. So e.g. for a debug entry with a format +string plus two varargs one would need to allocate a (3 * sizeof(long)) +byte data area in the debug_register() function. + +IMPORTANT: Using "%s" in sprintf event functions is dangerous. You can only +use "%s" in the sprintf event functions, if the memory for the passed string is +available as long as the debug feature exists. The reason behind this is that +due to performance considerations only a pointer to the string is stored in +the debug feature. If you log a string that is freed afterwards, you will get +an OOPS when inspecting the debug feature, because then the debug feature will +access the already freed memory. + +NOTE: If using the sprintf view do NOT use other event/exception functions +than the sprintf-event and -exception functions. + +The format of the hex_ascii and sprintf view is as follows: +- Number of area +- Timestamp (formatted as seconds and microseconds since 00:00:00 Coordinated + Universal Time (UTC), January 1, 1970) +- level of debug entry +- Exception flag (* = Exception) +- Cpu-Number of calling task +- Return Address to caller +- data field + +The format of the raw view is: +- Header as described in debug.h +- datafield + +A typical line of the hex_ascii view will look like the following (first line +is only for explanation and will not be displayed when 'cating' the view): + +area time level exception cpu caller data (hex + ascii) +-------------------------------------------------------------------------- +00 00964419409:440690 1 - 00 88023fe + + +Defining views +-------------- + +Views are specified with the 'debug_view' structure. There are defined +callback functions which are used for reading and writing the debugfs files: + +struct debug_view { + char name[DEBUG_MAX_PROCF_LEN]; + debug_prolog_proc_t* prolog_proc; + debug_header_proc_t* header_proc; + debug_format_proc_t* format_proc; + debug_input_proc_t* input_proc; + void* private_data; +}; + +where + +typedef int (debug_header_proc_t) (debug_info_t* id, + struct debug_view* view, + int area, + debug_entry_t* entry, + char* out_buf); + +typedef int (debug_format_proc_t) (debug_info_t* id, + struct debug_view* view, char* out_buf, + const char* in_buf); +typedef int (debug_prolog_proc_t) (debug_info_t* id, + struct debug_view* view, + char* out_buf); +typedef int (debug_input_proc_t) (debug_info_t* id, + struct debug_view* view, + struct file* file, const char* user_buf, + size_t in_buf_size, loff_t* offset); + + +The "private_data" member can be used as pointer to view specific data. +It is not used by the debug feature itself. + +The output when reading a debugfs file is structured like this: + +"prolog_proc output" + +"header_proc output 1" "format_proc output 1" +"header_proc output 2" "format_proc output 2" +"header_proc output 3" "format_proc output 3" +... + +When a view is read from the debugfs, the Debug Feature calls the +'prolog_proc' once for writing the prolog. +Then 'header_proc' and 'format_proc' are called for each +existing debug entry. + +The input_proc can be used to implement functionality when it is written to +the view (e.g. like with 'echo "0" > /sys/kernel/debug/s390dbf/dasd/level). + +For header_proc there can be used the default function +debug_dflt_header_fn() which is defined in debug.h. +and which produces the same header output as the predefined views. +E.g: +00 00964419409:440761 2 - 00 88023ec + +In order to see how to use the callback functions check the implementation +of the default views! + +Example + +#include + +#define UNKNOWNSTR "data: %08x" + +const char* messages[] = +{"This error...........\n", + "That error...........\n", + "Problem..............\n", + "Something went wrong.\n", + "Everything ok........\n", + NULL +}; + +static int debug_test_format_fn( + debug_info_t * id, struct debug_view *view, + char *out_buf, const char *in_buf +) +{ + int i, rc = 0; + + if(id->buf_size >= 4) { + int msg_nr = *((int*)in_buf); + if(msg_nr < sizeof(messages)/sizeof(char*) - 1) + rc += sprintf(out_buf, "%s", messages[msg_nr]); + else + rc += sprintf(out_buf, UNKNOWNSTR, msg_nr); + } + out: + return rc; +} + +struct debug_view debug_test_view = { + "myview", /* name of view */ + NULL, /* no prolog */ + &debug_dflt_header_fn, /* default header for each entry */ + &debug_test_format_fn, /* our own format function */ + NULL, /* no input function */ + NULL /* no private data */ +}; + +===== +test: +===== +debug_info_t *debug_info; +... +debug_info = debug_register ("test", 0, 4, 4 )); +debug_register_view(debug_info, &debug_test_view); +for(i = 0; i < 10; i ++) debug_int_event(debug_info, 1, i); + +> cat /sys/kernel/debug/s390dbf/test/myview +00 00964419734:611402 1 - 00 88042ca This error........... +00 00964419734:611405 1 - 00 88042ca That error........... +00 00964419734:611408 1 - 00 88042ca Problem.............. +00 00964419734:611411 1 - 00 88042ca Something went wrong. +00 00964419734:611414 1 - 00 88042ca Everything ok........ +00 00964419734:611417 1 - 00 88042ca data: 00000005 +00 00964419734:611419 1 - 00 88042ca data: 00000006 +00 00964419734:611422 1 - 00 88042ca data: 00000007 +00 00964419734:611425 1 - 00 88042ca data: 00000008 +00 00964419734:611428 1 - 00 88042ca data: 00000009 diff --git a/Documentation/s390/vfio-ccw.txt b/Documentation/s390/vfio-ccw.txt new file mode 100644 index 000000000..2be11ad86 --- /dev/null +++ b/Documentation/s390/vfio-ccw.txt @@ -0,0 +1,300 @@ +vfio-ccw: the basic infrastructure +================================== + +Introduction +------------ + +Here we describe the vfio support for I/O subchannel devices for +Linux/s390. Motivation for vfio-ccw is to passthrough subchannels to a +virtual machine, while vfio is the means. + +Different than other hardware architectures, s390 has defined a unified +I/O access method, which is so called Channel I/O. It has its own access +patterns: +- Channel programs run asynchronously on a separate (co)processor. +- The channel subsystem will access any memory designated by the caller + in the channel program directly, i.e. there is no iommu involved. +Thus when we introduce vfio support for these devices, we realize it +with a mediated device (mdev) implementation. The vfio mdev will be +added to an iommu group, so as to make itself able to be managed by the +vfio framework. And we add read/write callbacks for special vfio I/O +regions to pass the channel programs from the mdev to its parent device +(the real I/O subchannel device) to do further address translation and +to perform I/O instructions. + +This document does not intend to explain the s390 I/O architecture in +every detail. More information/reference could be found here: +- A good start to know Channel I/O in general: + https://en.wikipedia.org/wiki/Channel_I/O +- s390 architecture: + s390 Principles of Operation manual (IBM Form. No. SA22-7832) +- The existing QEMU code which implements a simple emulated channel + subsystem could also be a good reference. It makes it easier to follow + the flow. + qemu/hw/s390x/css.c + +For vfio mediated device framework: +- Documentation/vfio-mediated-device.txt + +Motivation of vfio-ccw +---------------------- + +Typically, a guest virtualized via QEMU/KVM on s390 only sees +paravirtualized virtio devices via the "Virtio Over Channel I/O +(virtio-ccw)" transport. This makes virtio devices discoverable via +standard operating system algorithms for handling channel devices. + +However this is not enough. On s390 for the majority of devices, which +use the standard Channel I/O based mechanism, we also need to provide +the functionality of passing through them to a QEMU virtual machine. +This includes devices that don't have a virtio counterpart (e.g. tape +drives) or that have specific characteristics which guests want to +exploit. + +For passing a device to a guest, we want to use the same interface as +everybody else, namely vfio. We implement this vfio support for channel +devices via the vfio mediated device framework and the subchannel device +driver "vfio_ccw". + +Access patterns of CCW devices +------------------------------ + +s390 architecture has implemented a so called channel subsystem, that +provides a unified view of the devices physically attached to the +systems. Though the s390 hardware platform knows about a huge variety of +different peripheral attachments like disk devices (aka. DASDs), tapes, +communication controllers, etc. They can all be accessed by a well +defined access method and they are presenting I/O completion a unified +way: I/O interruptions. + +All I/O requires the use of channel command words (CCWs). A CCW is an +instruction to a specialized I/O channel processor. A channel program is +a sequence of CCWs which are executed by the I/O channel subsystem. To +issue a channel program to the channel subsystem, it is required to +build an operation request block (ORB), which can be used to point out +the format of the CCW and other control information to the system. The +operating system signals the I/O channel subsystem to begin executing +the channel program with a SSCH (start sub-channel) instruction. The +central processor is then free to proceed with non-I/O instructions +until interrupted. The I/O completion result is received by the +interrupt handler in the form of interrupt response block (IRB). + +Back to vfio-ccw, in short: +- ORBs and channel programs are built in guest kernel (with guest + physical addresses). +- ORBs and channel programs are passed to the host kernel. +- Host kernel translates the guest physical addresses to real addresses + and starts the I/O with issuing a privileged Channel I/O instruction + (e.g SSCH). +- channel programs run asynchronously on a separate processor. +- I/O completion will be signaled to the host with I/O interruptions. + And it will be copied as IRB to user space to pass it back to the + guest. + +Physical vfio ccw device and its child mdev +------------------------------------------- + +As mentioned above, we realize vfio-ccw with a mdev implementation. + +Channel I/O does not have IOMMU hardware support, so the physical +vfio-ccw device does not have an IOMMU level translation or isolation. + +Subchannel I/O instructions are all privileged instructions. When +handling the I/O instruction interception, vfio-ccw has the software +policing and translation how the channel program is programmed before +it gets sent to hardware. + +Within this implementation, we have two drivers for two types of +devices: +- The vfio_ccw driver for the physical subchannel device. + This is an I/O subchannel driver for the real subchannel device. It + realizes a group of callbacks and registers to the mdev framework as a + parent (physical) device. As a consequence, mdev provides vfio_ccw a + generic interface (sysfs) to create mdev devices. A vfio mdev could be + created by vfio_ccw then and added to the mediated bus. It is the vfio + device that added to an IOMMU group and a vfio group. + vfio_ccw also provides an I/O region to accept channel program + request from user space and store I/O interrupt result for user + space to retrieve. To notify user space an I/O completion, it offers + an interface to setup an eventfd fd for asynchronous signaling. + +- The vfio_mdev driver for the mediated vfio ccw device. + This is provided by the mdev framework. It is a vfio device driver for + the mdev that created by vfio_ccw. + It realizes a group of vfio device driver callbacks, adds itself to a + vfio group, and registers itself to the mdev framework as a mdev + driver. + It uses a vfio iommu backend that uses the existing map and unmap + ioctls, but rather than programming them into an IOMMU for a device, + it simply stores the translations for use by later requests. This + means that a device programmed in a VM with guest physical addresses + can have the vfio kernel convert that address to process virtual + address, pin the page and program the hardware with the host physical + address in one step. + For a mdev, the vfio iommu backend will not pin the pages during the + VFIO_IOMMU_MAP_DMA ioctl. Mdev framework will only maintain a database + of the iova<->vaddr mappings in this operation. And they export a + vfio_pin_pages and a vfio_unpin_pages interfaces from the vfio iommu + backend for the physical devices to pin and unpin pages by demand. + +Below is a high Level block diagram. + + +-------------+ + | | + | +---------+ | mdev_register_driver() +--------------+ + | | Mdev | +<-----------------------+ | + | | bus | | | vfio_mdev.ko | + | | driver | +----------------------->+ |<-> VFIO user + | +---------+ | probe()/remove() +--------------+ APIs + | | + | MDEV CORE | + | MODULE | + | mdev.ko | + | +---------+ | mdev_register_device() +--------------+ + | |Physical | +<-----------------------+ | + | | device | | | vfio_ccw.ko |<-> subchannel + | |interface| +----------------------->+ | device + | +---------+ | callback +--------------+ + +-------------+ + +The process of how these work together. +1. vfio_ccw.ko drives the physical I/O subchannel, and registers the + physical device (with callbacks) to mdev framework. + When vfio_ccw probing the subchannel device, it registers device + pointer and callbacks to the mdev framework. Mdev related file nodes + under the device node in sysfs would be created for the subchannel + device, namely 'mdev_create', 'mdev_destroy' and + 'mdev_supported_types'. +2. Create a mediated vfio ccw device. + Use the 'mdev_create' sysfs file, we need to manually create one (and + only one for our case) mediated device. +3. vfio_mdev.ko drives the mediated ccw device. + vfio_mdev is also the vfio device drvier. It will probe the mdev and + add it to an iommu_group and a vfio_group. Then we could pass through + the mdev to a guest. + +vfio-ccw I/O region +------------------- + +An I/O region is used to accept channel program request from user +space and store I/O interrupt result for user space to retrieve. The +definition of the region is: + +struct ccw_io_region { +#define ORB_AREA_SIZE 12 + __u8 orb_area[ORB_AREA_SIZE]; +#define SCSW_AREA_SIZE 12 + __u8 scsw_area[SCSW_AREA_SIZE]; +#define IRB_AREA_SIZE 96 + __u8 irb_area[IRB_AREA_SIZE]; + __u32 ret_code; +} __packed; + +While starting an I/O request, orb_area should be filled with the +guest ORB, and scsw_area should be filled with the SCSW of the Virtual +Subchannel. + +irb_area stores the I/O result. + +ret_code stores a return code for each access of the region. + +vfio-ccw operation details +-------------------------- + +vfio-ccw follows what vfio-pci did on the s390 platform and uses +vfio-iommu-type1 as the vfio iommu backend. + +* CCW translation APIs + A group of APIs (start with 'cp_') to do CCW translation. The CCWs + passed in by a user space program are organized with their guest + physical memory addresses. These APIs will copy the CCWs into kernel + space, and assemble a runnable kernel channel program by updating the + guest physical addresses with their corresponding host physical addresses. + Note that we have to use IDALs even for direct-access CCWs, as the + referenced memory can be located anywhere, including above 2G. + +* vfio_ccw device driver + This driver utilizes the CCW translation APIs and introduces + vfio_ccw, which is the driver for the I/O subchannel devices you want + to pass through. + vfio_ccw implements the following vfio ioctls: + VFIO_DEVICE_GET_INFO + VFIO_DEVICE_GET_IRQ_INFO + VFIO_DEVICE_GET_REGION_INFO + VFIO_DEVICE_RESET + VFIO_DEVICE_SET_IRQS + This provides an I/O region, so that the user space program can pass a + channel program to the kernel, to do further CCW translation before + issuing them to a real device. + This also provides the SET_IRQ ioctl to setup an event notifier to + notify the user space program the I/O completion in an asynchronous + way. + +The use of vfio-ccw is not limited to QEMU, while QEMU is definitely a +good example to get understand how these patches work. Here is a little +bit more detail how an I/O request triggered by the QEMU guest will be +handled (without error handling). + +Explanation: +Q1-Q7: QEMU side process. +K1-K5: Kernel side process. + +Q1. Get I/O region info during initialization. +Q2. Setup event notifier and handler to handle I/O completion. + +... ... + +Q3. Intercept a ssch instruction. +Q4. Write the guest channel program and ORB to the I/O region. + K1. Copy from guest to kernel. + K2. Translate the guest channel program to a host kernel space + channel program, which becomes runnable for a real device. + K3. With the necessary information contained in the orb passed in + by QEMU, issue the ccwchain to the device. + K4. Return the ssch CC code. +Q5. Return the CC code to the guest. + +... ... + + K5. Interrupt handler gets the I/O result and write the result to + the I/O region. + K6. Signal QEMU to retrieve the result. +Q6. Get the signal and event handler reads out the result from the I/O + region. +Q7. Update the irb for the guest. + +Limitations +----------- + +The current vfio-ccw implementation focuses on supporting basic commands +needed to implement block device functionality (read/write) of DASD/ECKD +device only. Some commands may need special handling in the future, for +example, anything related to path grouping. + +DASD is a kind of storage device. While ECKD is a data recording format. +More information for DASD and ECKD could be found here: +https://en.wikipedia.org/wiki/Direct-access_storage_device +https://en.wikipedia.org/wiki/Count_key_data + +Together with the corresponding work in QEMU, we can bring the passed +through DASD/ECKD device online in a guest now and use it as a block +device. + +While the current code allows the guest to start channel programs via +START SUBCHANNEL, support for HALT SUBCHANNEL or CLEAR SUBCHANNEL is +not yet implemented. + +vfio-ccw supports classic (command mode) channel I/O only. Transport +mode (HPF) is not supported. + +QDIO subchannels are currently not supported. Classic devices other than +DASD/ECKD might work, but have not been tested. + +Reference +--------- +1. ESA/s390 Principles of Operation manual (IBM Form. No. SA22-7832) +2. ESA/390 Common I/O Device Commands manual (IBM Form. No. SA22-7204) +3. https://en.wikipedia.org/wiki/Channel_I/O +4. Documentation/s390/cds.txt +5. Documentation/vfio.txt +6. Documentation/vfio-mediated-device.txt diff --git a/Documentation/s390/zfcpdump.txt b/Documentation/s390/zfcpdump.txt new file mode 100644 index 000000000..b064aa597 --- /dev/null +++ b/Documentation/s390/zfcpdump.txt @@ -0,0 +1,48 @@ +The s390 SCSI dump tool (zfcpdump) + +System z machines (z900 or higher) provide hardware support for creating system +dumps on SCSI disks. The dump process is initiated by booting a dump tool, which +has to create a dump of the current (probably crashed) Linux image. In order to +not overwrite memory of the crashed Linux with data of the dump tool, the +hardware saves some memory plus the register sets of the boot CPU before the +dump tool is loaded. There exists an SCLP hardware interface to obtain the saved +memory afterwards. Currently 32 MB are saved. + +This zfcpdump implementation consists of a Linux dump kernel together with +a user space dump tool, which are loaded together into the saved memory region +below 32 MB. zfcpdump is installed on a SCSI disk using zipl (as contained in +the s390-tools package) to make the device bootable. The operator of a Linux +system can then trigger a SCSI dump by booting the SCSI disk, where zfcpdump +resides on. + +The user space dump tool accesses the memory of the crashed system by means +of the /proc/vmcore interface. This interface exports the crashed system's +memory and registers in ELF core dump format. To access the memory which has +been saved by the hardware SCLP requests will be created at the time the data +is needed by /proc/vmcore. The tail part of the crashed systems memory which +has not been stashed by hardware can just be copied from real memory. + +To build a dump enabled kernel the kernel config option CONFIG_CRASH_DUMP +has to be set. + +To get a valid zfcpdump kernel configuration use "make zfcpdump_defconfig". + +The s390 zipl tool looks for the zfcpdump kernel and optional initrd/initramfs +under the following locations: + +* kernel: /zfcpdump.image +* ramdisk: /zfcpdump.rd + +The zfcpdump directory is defined in the s390-tools package. + +The user space application of zfcpdump can reside in an intitramfs or an +initrd. It can also be included in a built-in kernel initramfs. The application +reads from /proc/vmcore or zcore/mem and writes the system dump to a SCSI disk. + +The s390-tools package version 1.24.0 and above builds an external zfcpdump +initramfs with a user space application that writes the dump to a SCSI +partition. + +For more information on how to use zfcpdump refer to the s390 'Using the Dump +Tools book', which is available from +http://www.ibm.com/developerworks/linux/linux390. -- cgit v1.2.3