diff options
author | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-11 08:27:49 +0000 |
---|---|---|
committer | Daniel Baumann <daniel.baumann@progress-linux.org> | 2024-04-11 08:27:49 +0000 |
commit | ace9429bb58fd418f0c81d4c2835699bddf6bde6 (patch) | |
tree | b2d64bc10158fdd5497876388cd68142ca374ed3 /drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm | |
parent | Initial commit. (diff) | |
download | linux-ace9429bb58fd418f0c81d4c2835699bddf6bde6.tar.xz linux-ace9429bb58fd418f0c81d4c2835699bddf6bde6.zip |
Adding upstream version 6.6.15.upstream/6.6.15
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm')
-rw-r--r-- | drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm | 1099 |
1 files changed, 1099 insertions, 0 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm new file mode 100644 index 0000000000..e506411ad2 --- /dev/null +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx9.asm @@ -0,0 +1,1099 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +/* To compile this assembly code: + * + * gfx9: + * cpp -DASIC_FAMILY=CHIP_VEGAM cwsr_trap_handler_gfx9.asm -P -o gfx9.sp3 + * sp3 gfx9.sp3 -hex gfx9.hex + * + * arcturus: + * cpp -DASIC_FAMILY=CHIP_ARCTURUS cwsr_trap_handler_gfx9.asm -P -o arcturus.sp3 + * sp3 arcturus.sp3 -hex arcturus.hex + * + * aldebaran: + * cpp -DASIC_FAMILY=CHIP_ALDEBARAN cwsr_trap_handler_gfx9.asm -P -o aldebaran.sp3 + * sp3 aldebaran.sp3 -hex aldebaran.hex + * + * gc_9_4_3: + * cpp -DASIC_FAMILY=GC_9_4_3 cwsr_trap_handler_gfx9.asm -P -o gc_9_4_3.sp3 + * sp3 gc_9_4_3.sp3 -hex gc_9_4_3.hex + */ + +#define CHIP_VEGAM 18 +#define CHIP_ARCTURUS 23 +#define CHIP_ALDEBARAN 25 +#define CHIP_GC_9_4_3 26 + +var ACK_SQC_STORE = 1 //workaround for suspected SQC store bug causing incorrect stores under concurrency +var SAVE_AFTER_XNACK_ERROR = 1 //workaround for TCP store failure after XNACK error when ALLOW_REPLAY=0, for debugger +var SINGLE_STEP_MISSED_WORKAROUND = (ASIC_FAMILY <= CHIP_ALDEBARAN) //workaround for lost MODE.DEBUG_EN exception when SAVECTX raised + +/**************************************************************************/ +/* variables */ +/**************************************************************************/ +var SQ_WAVE_STATUS_SPI_PRIO_SHIFT = 1 +var SQ_WAVE_STATUS_SPI_PRIO_MASK = 0x00000006 +var SQ_WAVE_STATUS_HALT_MASK = 0x2000 +var SQ_WAVE_STATUS_PRE_SPI_PRIO_SHIFT = 0 +var SQ_WAVE_STATUS_PRE_SPI_PRIO_SIZE = 1 +var SQ_WAVE_STATUS_POST_SPI_PRIO_SHIFT = 3 +var SQ_WAVE_STATUS_POST_SPI_PRIO_SIZE = 29 +var SQ_WAVE_STATUS_ALLOW_REPLAY_MASK = 0x400000 +var SQ_WAVE_STATUS_ECC_ERR_MASK = 0x20000 + +var SQ_WAVE_LDS_ALLOC_LDS_SIZE_SHIFT = 12 +var SQ_WAVE_LDS_ALLOC_LDS_SIZE_SIZE = 9 +var SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SIZE = 6 +var SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SIZE = 3 //FIXME sq.blk still has 4 bits at this time while SQ programming guide has 3 bits +var SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SHIFT = 24 + +#if ASIC_FAMILY >= CHIP_ALDEBARAN +var SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SHIFT = 6 +var SQ_WAVE_GPR_ALLOC_ACCV_OFFSET_SHIFT = 12 +var SQ_WAVE_GPR_ALLOC_ACCV_OFFSET_SIZE = 6 +#else +var SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SHIFT = 8 +#endif + +var SQ_WAVE_TRAPSTS_SAVECTX_MASK = 0x400 +var SQ_WAVE_TRAPSTS_EXCP_MASK = 0x1FF +var SQ_WAVE_TRAPSTS_SAVECTX_SHIFT = 10 +var SQ_WAVE_TRAPSTS_ADDR_WATCH_MASK = 0x80 +var SQ_WAVE_TRAPSTS_ADDR_WATCH_SHIFT = 7 +var SQ_WAVE_TRAPSTS_MEM_VIOL_MASK = 0x100 +var SQ_WAVE_TRAPSTS_MEM_VIOL_SHIFT = 8 +var SQ_WAVE_TRAPSTS_HOST_TRAP_MASK = 0x400000 +var SQ_WAVE_TRAPSTS_WAVE_BEGIN_MASK = 0x800000 +var SQ_WAVE_TRAPSTS_WAVE_END_MASK = 0x1000000 +var SQ_WAVE_TRAPSTS_TRAP_AFTER_INST_MASK = 0x2000000 +var SQ_WAVE_TRAPSTS_PRE_SAVECTX_MASK = 0x3FF +var SQ_WAVE_TRAPSTS_PRE_SAVECTX_SHIFT = 0x0 +var SQ_WAVE_TRAPSTS_PRE_SAVECTX_SIZE = 10 +var SQ_WAVE_TRAPSTS_POST_SAVECTX_MASK = 0xFFFFF800 +var SQ_WAVE_TRAPSTS_POST_SAVECTX_SHIFT = 11 +var SQ_WAVE_TRAPSTS_POST_SAVECTX_SIZE = 21 +var SQ_WAVE_TRAPSTS_ILLEGAL_INST_MASK = 0x800 +var SQ_WAVE_TRAPSTS_EXCP_HI_MASK = 0x7000 +var SQ_WAVE_TRAPSTS_XNACK_ERROR_MASK = 0x10000000 + +var SQ_WAVE_MODE_EXCP_EN_SHIFT = 12 +var SQ_WAVE_MODE_EXCP_EN_ADDR_WATCH_SHIFT = 19 + +var SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT = 15 //FIXME +var SQ_WAVE_IB_STS_RCNT_FIRST_REPLAY_MASK = 0x1F8000 + +var SQ_WAVE_MODE_DEBUG_EN_MASK = 0x800 + +var TTMP_SAVE_RCNT_FIRST_REPLAY_SHIFT = 26 // bits [31:26] unused by SPI debug data +var TTMP_SAVE_RCNT_FIRST_REPLAY_MASK = 0xFC000000 +var TTMP_DEBUG_TRAP_ENABLED_SHIFT = 23 +var TTMP_DEBUG_TRAP_ENABLED_MASK = 0x800000 + +/* Save */ +var S_SAVE_BUF_RSRC_WORD1_STRIDE = 0x00040000 //stride is 4 bytes +var S_SAVE_BUF_RSRC_WORD3_MISC = 0x00807FAC //SQ_SEL_X/Y/Z/W, BUF_NUM_FORMAT_FLOAT, (0 for MUBUF stride[17:14] when ADD_TID_ENABLE and BUF_DATA_FORMAT_32 for MTBUF), ADD_TID_ENABLE +var S_SAVE_PC_HI_TRAP_ID_MASK = 0x00FF0000 +var S_SAVE_PC_HI_HT_MASK = 0x01000000 +var S_SAVE_SPI_INIT_FIRST_WAVE_MASK = 0x04000000 //bit[26]: FirstWaveInTG +var S_SAVE_SPI_INIT_FIRST_WAVE_SHIFT = 26 + +var s_save_spi_init_lo = exec_lo +var s_save_spi_init_hi = exec_hi + +var s_save_pc_lo = ttmp0 //{TTMP1, TTMP0} = {3'h0,pc_rewind[3:0], HT[0],trapID[7:0], PC[47:0]} +var s_save_pc_hi = ttmp1 +var s_save_exec_lo = ttmp2 +var s_save_exec_hi = ttmp3 +var s_save_tmp = ttmp14 +var s_save_trapsts = ttmp15 //not really used until the end of the SAVE routine +var s_save_xnack_mask_lo = ttmp6 +var s_save_xnack_mask_hi = ttmp7 +var s_save_buf_rsrc0 = ttmp8 +var s_save_buf_rsrc1 = ttmp9 +var s_save_buf_rsrc2 = ttmp10 +var s_save_buf_rsrc3 = ttmp11 +var s_save_status = ttmp12 +var s_save_mem_offset = ttmp4 +var s_save_alloc_size = s_save_trapsts //conflict +var s_save_m0 = ttmp5 +var s_save_ttmps_lo = s_save_tmp //no conflict +var s_save_ttmps_hi = s_save_trapsts //no conflict +#if ASIC_FAMILY >= CHIP_GC_9_4_3 +var s_save_ib_sts = ttmp13 +#else +var s_save_ib_sts = ttmp11 +#endif + +/* Restore */ +var S_RESTORE_BUF_RSRC_WORD1_STRIDE = S_SAVE_BUF_RSRC_WORD1_STRIDE +var S_RESTORE_BUF_RSRC_WORD3_MISC = S_SAVE_BUF_RSRC_WORD3_MISC + +var S_RESTORE_SPI_INIT_FIRST_WAVE_MASK = 0x04000000 //bit[26]: FirstWaveInTG +var S_RESTORE_SPI_INIT_FIRST_WAVE_SHIFT = 26 + +var s_restore_spi_init_lo = exec_lo +var s_restore_spi_init_hi = exec_hi + +var s_restore_mem_offset = ttmp12 +var s_restore_tmp2 = ttmp13 +var s_restore_alloc_size = ttmp3 +var s_restore_tmp = ttmp2 +var s_restore_mem_offset_save = s_restore_tmp //no conflict +var s_restore_accvgpr_offset_save = ttmp7 + +var s_restore_m0 = s_restore_alloc_size //no conflict + +var s_restore_mode = s_restore_accvgpr_offset_save + +var s_restore_pc_lo = ttmp0 +var s_restore_pc_hi = ttmp1 +var s_restore_exec_lo = ttmp4 +var s_restore_exec_hi = ttmp5 +var s_restore_status = ttmp14 +var s_restore_trapsts = ttmp15 +var s_restore_xnack_mask_lo = xnack_mask_lo +var s_restore_xnack_mask_hi = xnack_mask_hi +var s_restore_buf_rsrc0 = ttmp8 +var s_restore_buf_rsrc1 = ttmp9 +var s_restore_buf_rsrc2 = ttmp10 +var s_restore_buf_rsrc3 = ttmp11 +var s_restore_ttmps_lo = s_restore_tmp //no conflict +var s_restore_ttmps_hi = s_restore_alloc_size //no conflict + +/**************************************************************************/ +/* trap handler entry points */ +/**************************************************************************/ +/* Shader Main*/ + +shader main + asic(DEFAULT) + type(CS) + + + s_branch L_SKIP_RESTORE //NOT restore. might be a regular trap or save + +L_JUMP_TO_RESTORE: + s_branch L_RESTORE //restore + +L_SKIP_RESTORE: + + s_getreg_b32 s_save_status, hwreg(HW_REG_STATUS) //save STATUS since we will change SCC + + // Clear SPI_PRIO: do not save with elevated priority. + // Clear ECC_ERR: prevents SQC store and triggers FATAL_HALT if setreg'd. + s_andn2_b32 s_save_status, s_save_status, SQ_WAVE_STATUS_SPI_PRIO_MASK|SQ_WAVE_STATUS_ECC_ERR_MASK + + s_getreg_b32 s_save_trapsts, hwreg(HW_REG_TRAPSTS) + + s_and_b32 ttmp2, s_save_status, SQ_WAVE_STATUS_HALT_MASK + s_cbranch_scc0 L_NOT_HALTED + +L_HALTED: + // Host trap may occur while wave is halted. + s_and_b32 ttmp2, s_save_pc_hi, S_SAVE_PC_HI_TRAP_ID_MASK + s_cbranch_scc1 L_FETCH_2ND_TRAP + +L_CHECK_SAVE: + s_and_b32 ttmp2, s_save_trapsts, SQ_WAVE_TRAPSTS_SAVECTX_MASK //check whether this is for save + s_cbranch_scc1 L_SAVE //this is the operation for save + + // Wave is halted but neither host trap nor SAVECTX is raised. + // Caused by instruction fetch memory violation. + // Spin wait until context saved to prevent interrupt storm. + s_sleep 0x10 + s_getreg_b32 s_save_trapsts, hwreg(HW_REG_TRAPSTS) + s_branch L_CHECK_SAVE + +L_NOT_HALTED: + // Let second-level handle non-SAVECTX exception or trap. + // Any concurrent SAVECTX will be handled upon re-entry once halted. + + // Check non-maskable exceptions. memory_violation, illegal_instruction + // and debugger (host trap, wave start/end, trap after instruction) + // exceptions always cause the wave to enter the trap handler. + s_and_b32 ttmp2, s_save_trapsts, \ + SQ_WAVE_TRAPSTS_MEM_VIOL_MASK | \ + SQ_WAVE_TRAPSTS_ILLEGAL_INST_MASK | \ + SQ_WAVE_TRAPSTS_HOST_TRAP_MASK | \ + SQ_WAVE_TRAPSTS_WAVE_BEGIN_MASK | \ + SQ_WAVE_TRAPSTS_WAVE_END_MASK | \ + SQ_WAVE_TRAPSTS_TRAP_AFTER_INST_MASK + s_cbranch_scc1 L_FETCH_2ND_TRAP + + // Check for maskable exceptions in trapsts.excp and trapsts.excp_hi. + // Maskable exceptions only cause the wave to enter the trap handler if + // their respective bit in mode.excp_en is set. + s_and_b32 ttmp2, s_save_trapsts, SQ_WAVE_TRAPSTS_EXCP_MASK|SQ_WAVE_TRAPSTS_EXCP_HI_MASK + s_cbranch_scc0 L_CHECK_TRAP_ID + + s_and_b32 ttmp3, s_save_trapsts, SQ_WAVE_TRAPSTS_ADDR_WATCH_MASK|SQ_WAVE_TRAPSTS_EXCP_HI_MASK + s_cbranch_scc0 L_NOT_ADDR_WATCH + s_bitset1_b32 ttmp2, SQ_WAVE_TRAPSTS_ADDR_WATCH_SHIFT // Check all addr_watch[123] exceptions against excp_en.addr_watch + +L_NOT_ADDR_WATCH: + s_getreg_b32 ttmp3, hwreg(HW_REG_MODE) + s_lshl_b32 ttmp2, ttmp2, SQ_WAVE_MODE_EXCP_EN_SHIFT + s_and_b32 ttmp2, ttmp2, ttmp3 + s_cbranch_scc1 L_FETCH_2ND_TRAP + +L_CHECK_TRAP_ID: + // Check trap_id != 0 + s_and_b32 ttmp2, s_save_pc_hi, S_SAVE_PC_HI_TRAP_ID_MASK + s_cbranch_scc1 L_FETCH_2ND_TRAP + +if SINGLE_STEP_MISSED_WORKAROUND + // Prioritize single step exception over context save. + // Second-level trap will halt wave and RFE, re-entering for SAVECTX. + s_getreg_b32 ttmp2, hwreg(HW_REG_MODE) + s_and_b32 ttmp2, ttmp2, SQ_WAVE_MODE_DEBUG_EN_MASK + s_cbranch_scc1 L_FETCH_2ND_TRAP +end + + s_and_b32 ttmp2, s_save_trapsts, SQ_WAVE_TRAPSTS_SAVECTX_MASK + s_cbranch_scc1 L_SAVE + +L_FETCH_2ND_TRAP: + // Preserve and clear scalar XNACK state before issuing scalar reads. + save_and_clear_ib_sts(ttmp14) + + // Read second-level TBA/TMA from first-level TMA and jump if available. + // ttmp[2:5] and ttmp12 can be used (others hold SPI-initialized debug data) + // ttmp12 holds SQ_WAVE_STATUS + s_getreg_b32 ttmp14, hwreg(HW_REG_SQ_SHADER_TMA_LO) + s_getreg_b32 ttmp15, hwreg(HW_REG_SQ_SHADER_TMA_HI) + s_lshl_b64 [ttmp14, ttmp15], [ttmp14, ttmp15], 0x8 + + s_bitcmp1_b32 ttmp15, 0xF + s_cbranch_scc0 L_NO_SIGN_EXTEND_TMA + s_or_b32 ttmp15, ttmp15, 0xFFFF0000 +L_NO_SIGN_EXTEND_TMA: + + s_load_dword ttmp2, [ttmp14, ttmp15], 0x10 glc:1 // debug trap enabled flag + s_waitcnt lgkmcnt(0) + s_lshl_b32 ttmp2, ttmp2, TTMP_DEBUG_TRAP_ENABLED_SHIFT + s_andn2_b32 s_save_ib_sts, s_save_ib_sts, TTMP_DEBUG_TRAP_ENABLED_MASK + s_or_b32 s_save_ib_sts, s_save_ib_sts, ttmp2 + + s_load_dwordx2 [ttmp2, ttmp3], [ttmp14, ttmp15], 0x0 glc:1 // second-level TBA + s_waitcnt lgkmcnt(0) + s_load_dwordx2 [ttmp14, ttmp15], [ttmp14, ttmp15], 0x8 glc:1 // second-level TMA + s_waitcnt lgkmcnt(0) + + s_and_b64 [ttmp2, ttmp3], [ttmp2, ttmp3], [ttmp2, ttmp3] + s_cbranch_scc0 L_NO_NEXT_TRAP // second-level trap handler not been set + s_setpc_b64 [ttmp2, ttmp3] // jump to second-level trap handler + +L_NO_NEXT_TRAP: + // If not caused by trap then halt wave to prevent re-entry. + s_and_b32 ttmp2, s_save_pc_hi, (S_SAVE_PC_HI_TRAP_ID_MASK|S_SAVE_PC_HI_HT_MASK) + s_cbranch_scc1 L_TRAP_CASE + s_or_b32 s_save_status, s_save_status, SQ_WAVE_STATUS_HALT_MASK + + // If the PC points to S_ENDPGM then context save will fail if STATUS.HALT is set. + // Rewind the PC to prevent this from occurring. + s_sub_u32 ttmp0, ttmp0, 0x8 + s_subb_u32 ttmp1, ttmp1, 0x0 + + s_branch L_EXIT_TRAP + +L_TRAP_CASE: + // Host trap will not cause trap re-entry. + s_and_b32 ttmp2, s_save_pc_hi, S_SAVE_PC_HI_HT_MASK + s_cbranch_scc1 L_EXIT_TRAP + + // Advance past trap instruction to prevent re-entry. + s_add_u32 ttmp0, ttmp0, 0x4 + s_addc_u32 ttmp1, ttmp1, 0x0 + +L_EXIT_TRAP: + s_and_b32 ttmp1, ttmp1, 0xFFFF + + restore_ib_sts(ttmp14) + + // Restore SQ_WAVE_STATUS. + s_and_b64 exec, exec, exec // Restore STATUS.EXECZ, not writable by s_setreg_b32 + s_and_b64 vcc, vcc, vcc // Restore STATUS.VCCZ, not writable by s_setreg_b32 + set_status_without_spi_prio(s_save_status, ttmp2) + + s_rfe_b64 [ttmp0, ttmp1] + + // ********* End handling of non-CWSR traps ******************* + +/**************************************************************************/ +/* save routine */ +/**************************************************************************/ + +L_SAVE: + s_and_b32 s_save_pc_hi, s_save_pc_hi, 0x0000ffff //pc[47:32] + + s_mov_b32 s_save_tmp, 0 //clear saveCtx bit + s_setreg_b32 hwreg(HW_REG_TRAPSTS, SQ_WAVE_TRAPSTS_SAVECTX_SHIFT, 1), s_save_tmp //clear saveCtx bit + + save_and_clear_ib_sts(s_save_tmp) + + /* inform SPI the readiness and wait for SPI's go signal */ + s_mov_b32 s_save_exec_lo, exec_lo //save EXEC and use EXEC for the go signal from SPI + s_mov_b32 s_save_exec_hi, exec_hi + s_mov_b64 exec, 0x0 //clear EXEC to get ready to receive + + s_sendmsg sendmsg(MSG_SAVEWAVE) //send SPI a message and wait for SPI's write to EXEC + + // Set SPI_PRIO=2 to avoid starving instruction fetch in the waves we're waiting for. + s_or_b32 s_save_tmp, s_save_status, (2 << SQ_WAVE_STATUS_SPI_PRIO_SHIFT) + s_setreg_b32 hwreg(HW_REG_STATUS), s_save_tmp + + L_SLEEP: + s_sleep 0x2 // sleep 1 (64clk) is not enough for 8 waves per SIMD, which will cause SQ hang, since the 7,8th wave could not get arbit to exec inst, while other waves are stuck into the sleep-loop and waiting for wrexec!=0 + + s_cbranch_execz L_SLEEP + + // Save trap temporaries 4-11, 13 initialized by SPI debug dispatch logic + // ttmp SR memory offset : size(VGPR)+size(SGPR)+0x40 + get_vgpr_size_bytes(s_save_ttmps_lo) + get_sgpr_size_bytes(s_save_ttmps_hi) + s_add_u32 s_save_ttmps_lo, s_save_ttmps_lo, s_save_ttmps_hi + s_add_u32 s_save_ttmps_lo, s_save_ttmps_lo, s_save_spi_init_lo + s_addc_u32 s_save_ttmps_hi, s_save_spi_init_hi, 0x0 + s_and_b32 s_save_ttmps_hi, s_save_ttmps_hi, 0xFFFF + s_store_dwordx4 [ttmp4, ttmp5, ttmp6, ttmp7], [s_save_ttmps_lo, s_save_ttmps_hi], 0x50 glc:1 + ack_sqc_store_workaround() + s_store_dwordx4 [ttmp8, ttmp9, ttmp10, ttmp11], [s_save_ttmps_lo, s_save_ttmps_hi], 0x60 glc:1 + ack_sqc_store_workaround() + s_store_dword ttmp13, [s_save_ttmps_lo, s_save_ttmps_hi], 0x74 glc:1 + ack_sqc_store_workaround() + + /* setup Resource Contants */ + s_mov_b32 s_save_buf_rsrc0, s_save_spi_init_lo //base_addr_lo + s_and_b32 s_save_buf_rsrc1, s_save_spi_init_hi, 0x0000FFFF //base_addr_hi + s_or_b32 s_save_buf_rsrc1, s_save_buf_rsrc1, S_SAVE_BUF_RSRC_WORD1_STRIDE + s_mov_b32 s_save_buf_rsrc2, 0 //NUM_RECORDS initial value = 0 (in bytes) although not neccessarily inited + s_mov_b32 s_save_buf_rsrc3, S_SAVE_BUF_RSRC_WORD3_MISC + + //FIXME right now s_save_m0/s_save_mem_offset use tma_lo/tma_hi (might need to save them before using them?) + s_mov_b32 s_save_m0, m0 //save M0 + + /* global mem offset */ + s_mov_b32 s_save_mem_offset, 0x0 //mem offset initial value = 0 + + + + + /* save HW registers */ + ////////////////////////////// + + L_SAVE_HWREG: + // HWREG SR memory offset : size(VGPR)+size(SGPR) + get_vgpr_size_bytes(s_save_mem_offset) + get_sgpr_size_bytes(s_save_tmp) + s_add_u32 s_save_mem_offset, s_save_mem_offset, s_save_tmp + + + s_mov_b32 s_save_buf_rsrc2, 0x4 //NUM_RECORDS in bytes + s_mov_b32 s_save_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + + write_hwreg_to_mem(s_save_m0, s_save_buf_rsrc0, s_save_mem_offset) //M0 + write_hwreg_to_mem(s_save_pc_lo, s_save_buf_rsrc0, s_save_mem_offset) //PC + write_hwreg_to_mem(s_save_pc_hi, s_save_buf_rsrc0, s_save_mem_offset) + write_hwreg_to_mem(s_save_exec_lo, s_save_buf_rsrc0, s_save_mem_offset) //EXEC + write_hwreg_to_mem(s_save_exec_hi, s_save_buf_rsrc0, s_save_mem_offset) + write_hwreg_to_mem(s_save_status, s_save_buf_rsrc0, s_save_mem_offset) //STATUS + + //s_save_trapsts conflicts with s_save_alloc_size + s_getreg_b32 s_save_trapsts, hwreg(HW_REG_TRAPSTS) + write_hwreg_to_mem(s_save_trapsts, s_save_buf_rsrc0, s_save_mem_offset) //TRAPSTS + + write_hwreg_to_mem(xnack_mask_lo, s_save_buf_rsrc0, s_save_mem_offset) //XNACK_MASK_LO + write_hwreg_to_mem(xnack_mask_hi, s_save_buf_rsrc0, s_save_mem_offset) //XNACK_MASK_HI + + //use s_save_tmp would introduce conflict here between s_save_tmp and s_save_buf_rsrc2 + s_getreg_b32 s_save_m0, hwreg(HW_REG_MODE) //MODE + write_hwreg_to_mem(s_save_m0, s_save_buf_rsrc0, s_save_mem_offset) + + + + /* the first wave in the threadgroup */ + s_and_b32 s_save_tmp, s_save_spi_init_hi, S_SAVE_SPI_INIT_FIRST_WAVE_MASK // extract fisrt wave bit + s_mov_b32 s_save_exec_hi, 0x0 + s_or_b32 s_save_exec_hi, s_save_tmp, s_save_exec_hi // save first wave bit to s_save_exec_hi.bits[26] + + + /* save SGPRs */ + // Save SGPR before LDS save, then the s0 to s4 can be used during LDS save... + ////////////////////////////// + + // SGPR SR memory offset : size(VGPR) + get_vgpr_size_bytes(s_save_mem_offset) + // TODO, change RSRC word to rearrange memory layout for SGPRS + + s_getreg_b32 s_save_alloc_size, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SHIFT,SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SIZE) //spgr_size + s_add_u32 s_save_alloc_size, s_save_alloc_size, 1 + s_lshl_b32 s_save_alloc_size, s_save_alloc_size, 4 //Number of SGPRs = (sgpr_size + 1) * 16 (non-zero value) + + s_lshl_b32 s_save_buf_rsrc2, s_save_alloc_size, 2 //NUM_RECORDS in bytes + + s_mov_b32 s_save_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + + // backup s_save_buf_rsrc0,1 to s_save_pc_lo/hi, since write_16sgpr_to_mem function will change the rsrc0 + //s_mov_b64 s_save_pc_lo, s_save_buf_rsrc0 + s_mov_b64 s_save_xnack_mask_lo, s_save_buf_rsrc0 + s_add_u32 s_save_buf_rsrc0, s_save_buf_rsrc0, s_save_mem_offset + s_addc_u32 s_save_buf_rsrc1, s_save_buf_rsrc1, 0 + + s_mov_b32 m0, 0x0 //SGPR initial index value =0 + s_nop 0x0 //Manually inserted wait states + L_SAVE_SGPR_LOOP: + // SGPR is allocated in 16 SGPR granularity + s_movrels_b64 s0, s0 //s0 = s[0+m0], s1 = s[1+m0] + s_movrels_b64 s2, s2 //s2 = s[2+m0], s3 = s[3+m0] + s_movrels_b64 s4, s4 //s4 = s[4+m0], s5 = s[5+m0] + s_movrels_b64 s6, s6 //s6 = s[6+m0], s7 = s[7+m0] + s_movrels_b64 s8, s8 //s8 = s[8+m0], s9 = s[9+m0] + s_movrels_b64 s10, s10 //s10 = s[10+m0], s11 = s[11+m0] + s_movrels_b64 s12, s12 //s12 = s[12+m0], s13 = s[13+m0] + s_movrels_b64 s14, s14 //s14 = s[14+m0], s15 = s[15+m0] + + write_16sgpr_to_mem(s0, s_save_buf_rsrc0, s_save_mem_offset) //PV: the best performance should be using s_buffer_store_dwordx4 + s_add_u32 m0, m0, 16 //next sgpr index + s_cmp_lt_u32 m0, s_save_alloc_size //scc = (m0 < s_save_alloc_size) ? 1 : 0 + s_cbranch_scc1 L_SAVE_SGPR_LOOP //SGPR save is complete? + // restore s_save_buf_rsrc0,1 + //s_mov_b64 s_save_buf_rsrc0, s_save_pc_lo + s_mov_b64 s_save_buf_rsrc0, s_save_xnack_mask_lo + + + + + /* save first 4 VGPR, then LDS save could use */ + // each wave will alloc 4 vgprs at least... + ///////////////////////////////////////////////////////////////////////////////////// + + s_mov_b32 s_save_mem_offset, 0 + s_mov_b32 exec_lo, 0xFFFFFFFF //need every thread from now on + s_mov_b32 exec_hi, 0xFFFFFFFF + s_mov_b32 xnack_mask_lo, 0x0 + s_mov_b32 xnack_mask_hi, 0x0 + + s_mov_b32 s_save_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + + // VGPR Allocated in 4-GPR granularity + +if SAVE_AFTER_XNACK_ERROR + check_if_tcp_store_ok() + s_cbranch_scc1 L_SAVE_FIRST_VGPRS_WITH_TCP + + write_vgprs_to_mem_with_sqc(v0, 4, s_save_buf_rsrc0, s_save_mem_offset) + s_branch L_SAVE_LDS + +L_SAVE_FIRST_VGPRS_WITH_TCP: +end + + write_4vgprs_to_mem(s_save_buf_rsrc0, s_save_mem_offset) + + /* save LDS */ + ////////////////////////////// + + L_SAVE_LDS: + + // Change EXEC to all threads... + s_mov_b32 exec_lo, 0xFFFFFFFF //need every thread from now on + s_mov_b32 exec_hi, 0xFFFFFFFF + + s_getreg_b32 s_save_alloc_size, hwreg(HW_REG_LDS_ALLOC,SQ_WAVE_LDS_ALLOC_LDS_SIZE_SHIFT,SQ_WAVE_LDS_ALLOC_LDS_SIZE_SIZE) //lds_size + s_and_b32 s_save_alloc_size, s_save_alloc_size, 0xFFFFFFFF //lds_size is zero? + s_cbranch_scc0 L_SAVE_LDS_DONE //no lds used? jump to L_SAVE_DONE + + s_barrier //LDS is used? wait for other waves in the same TG + s_and_b32 s_save_tmp, s_save_exec_hi, S_SAVE_SPI_INIT_FIRST_WAVE_MASK //exec is still used here + s_cbranch_scc0 L_SAVE_LDS_DONE + + // first wave do LDS save; + + s_lshl_b32 s_save_alloc_size, s_save_alloc_size, 6 //LDS size in dwords = lds_size * 64dw + s_lshl_b32 s_save_alloc_size, s_save_alloc_size, 2 //LDS size in bytes + s_mov_b32 s_save_buf_rsrc2, s_save_alloc_size //NUM_RECORDS in bytes + + // LDS at offset: size(VGPR)+SIZE(SGPR)+SIZE(HWREG) + // + get_vgpr_size_bytes(s_save_mem_offset) + get_sgpr_size_bytes(s_save_tmp) + s_add_u32 s_save_mem_offset, s_save_mem_offset, s_save_tmp + s_add_u32 s_save_mem_offset, s_save_mem_offset, get_hwreg_size_bytes() + + + s_mov_b32 s_save_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + s_mov_b32 m0, 0x0 //lds_offset initial value = 0 + + + v_mbcnt_lo_u32_b32 v2, 0xffffffff, 0x0 + v_mbcnt_hi_u32_b32 v3, 0xffffffff, v2 // tid + +if SAVE_AFTER_XNACK_ERROR + check_if_tcp_store_ok() + s_cbranch_scc1 L_SAVE_LDS_WITH_TCP + + v_lshlrev_b32 v2, 2, v3 +L_SAVE_LDS_LOOP_SQC: + ds_read2_b32 v[0:1], v2 offset0:0 offset1:0x40 + s_waitcnt lgkmcnt(0) + + write_vgprs_to_mem_with_sqc(v0, 2, s_save_buf_rsrc0, s_save_mem_offset) + + v_add_u32 v2, 0x200, v2 + v_cmp_lt_u32 vcc[0:1], v2, s_save_alloc_size + s_cbranch_vccnz L_SAVE_LDS_LOOP_SQC + + s_branch L_SAVE_LDS_DONE + +L_SAVE_LDS_WITH_TCP: +end + + v_mul_i32_i24 v2, v3, 8 // tid*8 + v_mov_b32 v3, 256*2 + s_mov_b32 m0, 0x10000 + s_mov_b32 s0, s_save_buf_rsrc3 + s_and_b32 s_save_buf_rsrc3, s_save_buf_rsrc3, 0xFF7FFFFF // disable add_tid + s_or_b32 s_save_buf_rsrc3, s_save_buf_rsrc3, 0x58000 //DFMT + +L_SAVE_LDS_LOOP_VECTOR: + ds_read_b64 v[0:1], v2 //x =LDS[a], byte address + s_waitcnt lgkmcnt(0) + buffer_store_dwordx2 v[0:1], v2, s_save_buf_rsrc0, s_save_mem_offset offen:1 glc:1 slc:1 +// s_waitcnt vmcnt(0) +// v_add_u32 v2, vcc[0:1], v2, v3 + v_add_u32 v2, v2, v3 + v_cmp_lt_u32 vcc[0:1], v2, s_save_alloc_size + s_cbranch_vccnz L_SAVE_LDS_LOOP_VECTOR + + // restore rsrc3 + s_mov_b32 s_save_buf_rsrc3, s0 + +L_SAVE_LDS_DONE: + + + /* save VGPRs - set the Rest VGPRs */ + ////////////////////////////////////////////////////////////////////////////////////// + L_SAVE_VGPR: + // VGPR SR memory offset: 0 + // TODO rearrange the RSRC words to use swizzle for VGPR save... + + s_mov_b32 s_save_mem_offset, (0+256*4) // for the rest VGPRs + s_mov_b32 exec_lo, 0xFFFFFFFF //need every thread from now on + s_mov_b32 exec_hi, 0xFFFFFFFF + + get_num_arch_vgprs(s_save_alloc_size) + s_mov_b32 s_save_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + + // VGPR store using dw burst + s_mov_b32 m0, 0x4 //VGPR initial index value =0 + s_cmp_lt_u32 m0, s_save_alloc_size + s_cbranch_scc0 L_SAVE_VGPR_END + + + s_set_gpr_idx_on m0, 0x1 //M0[7:0] = M0[7:0] and M0[15:12] = 0x1 + s_add_u32 s_save_alloc_size, s_save_alloc_size, 0x1000 //add 0x1000 since we compare m0 against it later + +if SAVE_AFTER_XNACK_ERROR + check_if_tcp_store_ok() + s_cbranch_scc1 L_SAVE_VGPR_LOOP + +L_SAVE_VGPR_LOOP_SQC: + write_vgprs_to_mem_with_sqc(v0, 4, s_save_buf_rsrc0, s_save_mem_offset) + + s_add_u32 m0, m0, 4 + s_cmp_lt_u32 m0, s_save_alloc_size + s_cbranch_scc1 L_SAVE_VGPR_LOOP_SQC + + s_set_gpr_idx_off + s_branch L_SAVE_VGPR_END +end + + L_SAVE_VGPR_LOOP: + v_mov_b32 v0, v0 //v0 = v[0+m0] + v_mov_b32 v1, v1 //v0 = v[0+m0] + v_mov_b32 v2, v2 //v0 = v[0+m0] + v_mov_b32 v3, v3 //v0 = v[0+m0] + + write_4vgprs_to_mem(s_save_buf_rsrc0, s_save_mem_offset) + + s_add_u32 m0, m0, 4 //next vgpr index + s_add_u32 s_save_mem_offset, s_save_mem_offset, 256*4 //every buffer_store_dword does 256 bytes + s_cmp_lt_u32 m0, s_save_alloc_size //scc = (m0 < s_save_alloc_size) ? 1 : 0 + s_cbranch_scc1 L_SAVE_VGPR_LOOP //VGPR save is complete? + s_set_gpr_idx_off + +L_SAVE_VGPR_END: + +#if ASIC_FAMILY >= CHIP_ARCTURUS + // Save ACC VGPRs + +#if ASIC_FAMILY >= CHIP_ALDEBARAN + // ACC VGPR count may differ from ARCH VGPR count. + get_num_acc_vgprs(s_save_alloc_size, s_save_tmp) + s_and_b32 s_save_alloc_size, s_save_alloc_size, s_save_alloc_size + s_cbranch_scc0 L_SAVE_ACCVGPR_END + s_add_u32 s_save_alloc_size, s_save_alloc_size, 0x1000 //add 0x1000 since we compare m0 against it later +#endif + + s_mov_b32 m0, 0x0 //VGPR initial index value =0 + s_set_gpr_idx_on m0, 0x1 //M0[7:0] = M0[7:0] and M0[15:12] = 0x1 + +if SAVE_AFTER_XNACK_ERROR + check_if_tcp_store_ok() + s_cbranch_scc1 L_SAVE_ACCVGPR_LOOP + +L_SAVE_ACCVGPR_LOOP_SQC: + for var vgpr = 0; vgpr < 4; ++ vgpr + v_accvgpr_read v[vgpr], acc[vgpr] // v[N] = acc[N+m0] + end + + write_vgprs_to_mem_with_sqc(v0, 4, s_save_buf_rsrc0, s_save_mem_offset) + + s_add_u32 m0, m0, 4 + s_cmp_lt_u32 m0, s_save_alloc_size + s_cbranch_scc1 L_SAVE_ACCVGPR_LOOP_SQC + + s_set_gpr_idx_off + s_branch L_SAVE_ACCVGPR_END +end + +L_SAVE_ACCVGPR_LOOP: + for var vgpr = 0; vgpr < 4; ++ vgpr + v_accvgpr_read v[vgpr], acc[vgpr] // v[N] = acc[N+m0] + end + + write_4vgprs_to_mem(s_save_buf_rsrc0, s_save_mem_offset) + + s_add_u32 m0, m0, 4 + s_add_u32 s_save_mem_offset, s_save_mem_offset, 256*4 + s_cmp_lt_u32 m0, s_save_alloc_size + s_cbranch_scc1 L_SAVE_ACCVGPR_LOOP + s_set_gpr_idx_off + +L_SAVE_ACCVGPR_END: +#endif + + s_branch L_END_PGM + + + +/**************************************************************************/ +/* restore routine */ +/**************************************************************************/ + +L_RESTORE: + /* Setup Resource Contants */ + s_mov_b32 s_restore_buf_rsrc0, s_restore_spi_init_lo //base_addr_lo + s_and_b32 s_restore_buf_rsrc1, s_restore_spi_init_hi, 0x0000FFFF //base_addr_hi + s_or_b32 s_restore_buf_rsrc1, s_restore_buf_rsrc1, S_RESTORE_BUF_RSRC_WORD1_STRIDE + s_mov_b32 s_restore_buf_rsrc2, 0 //NUM_RECORDS initial value = 0 (in bytes) + s_mov_b32 s_restore_buf_rsrc3, S_RESTORE_BUF_RSRC_WORD3_MISC + + /* global mem offset */ +// s_mov_b32 s_restore_mem_offset, 0x0 //mem offset initial value = 0 + + /* the first wave in the threadgroup */ + s_and_b32 s_restore_tmp, s_restore_spi_init_hi, S_RESTORE_SPI_INIT_FIRST_WAVE_MASK + s_cbranch_scc0 L_RESTORE_VGPR + + /* restore LDS */ + ////////////////////////////// + L_RESTORE_LDS: + + s_mov_b32 exec_lo, 0xFFFFFFFF //need every thread from now on //be consistent with SAVE although can be moved ahead + s_mov_b32 exec_hi, 0xFFFFFFFF + + s_getreg_b32 s_restore_alloc_size, hwreg(HW_REG_LDS_ALLOC,SQ_WAVE_LDS_ALLOC_LDS_SIZE_SHIFT,SQ_WAVE_LDS_ALLOC_LDS_SIZE_SIZE) //lds_size + s_and_b32 s_restore_alloc_size, s_restore_alloc_size, 0xFFFFFFFF //lds_size is zero? + s_cbranch_scc0 L_RESTORE_VGPR //no lds used? jump to L_RESTORE_VGPR + s_lshl_b32 s_restore_alloc_size, s_restore_alloc_size, 6 //LDS size in dwords = lds_size * 64dw + s_lshl_b32 s_restore_alloc_size, s_restore_alloc_size, 2 //LDS size in bytes + s_mov_b32 s_restore_buf_rsrc2, s_restore_alloc_size //NUM_RECORDS in bytes + + // LDS at offset: size(VGPR)+SIZE(SGPR)+SIZE(HWREG) + // + get_vgpr_size_bytes(s_restore_mem_offset) + get_sgpr_size_bytes(s_restore_tmp) + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, s_restore_tmp + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, get_hwreg_size_bytes() //FIXME, Check if offset overflow??? + + + s_mov_b32 s_restore_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + s_mov_b32 m0, 0x0 //lds_offset initial value = 0 + + L_RESTORE_LDS_LOOP: + buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 // first 64DW + buffer_load_dword v0, v0, s_restore_buf_rsrc0, s_restore_mem_offset lds:1 offset:256 // second 64DW + s_add_u32 m0, m0, 256*2 // 128 DW + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*2 //mem offset increased by 128DW + s_cmp_lt_u32 m0, s_restore_alloc_size //scc=(m0 < s_restore_alloc_size) ? 1 : 0 + s_cbranch_scc1 L_RESTORE_LDS_LOOP //LDS restore is complete? + + + /* restore VGPRs */ + ////////////////////////////// + L_RESTORE_VGPR: + s_mov_b32 exec_lo, 0xFFFFFFFF //need every thread from now on //be consistent with SAVE although can be moved ahead + s_mov_b32 exec_hi, 0xFFFFFFFF + s_mov_b32 s_restore_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + // Save ARCH VGPRs 4-N, then all ACC VGPRs, then ARCH VGPRs 0-3. + get_num_arch_vgprs(s_restore_alloc_size) + s_add_u32 s_restore_alloc_size, s_restore_alloc_size, 0x8000 //add 0x8000 since we compare m0 against it later + + // ARCH VGPRs at offset: 0 + s_mov_b32 s_restore_mem_offset, 0x0 + s_mov_b32 s_restore_mem_offset_save, s_restore_mem_offset // restore start with v1, v0 will be the last + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*4 + s_mov_b32 m0, 4 //VGPR initial index value = 1 + s_set_gpr_idx_on m0, 0x8 //M0[7:0] = M0[7:0] and M0[15:12] = 0x8 + + L_RESTORE_VGPR_LOOP: + read_4vgprs_from_mem(s_restore_buf_rsrc0, s_restore_mem_offset) + v_mov_b32 v0, v0 //v[0+m0] = v0 + v_mov_b32 v1, v1 + v_mov_b32 v2, v2 + v_mov_b32 v3, v3 + s_add_u32 m0, m0, 4 //next vgpr index + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*4 //every buffer_load_dword does 256 bytes + s_cmp_lt_u32 m0, s_restore_alloc_size //scc = (m0 < s_restore_alloc_size) ? 1 : 0 + s_cbranch_scc1 L_RESTORE_VGPR_LOOP //VGPR restore (except v0) is complete? + +#if ASIC_FAMILY >= CHIP_ALDEBARAN + // ACC VGPR count may differ from ARCH VGPR count. + get_num_acc_vgprs(s_restore_alloc_size, s_restore_tmp2) + s_and_b32 s_restore_alloc_size, s_restore_alloc_size, s_restore_alloc_size + s_cbranch_scc0 L_RESTORE_ACCVGPR_END + s_add_u32 s_restore_alloc_size, s_restore_alloc_size, 0x8000 //add 0x8000 since we compare m0 against it later +#endif + +#if ASIC_FAMILY >= CHIP_ARCTURUS + // ACC VGPRs at offset: size(ARCH VGPRs) + s_mov_b32 m0, 0 + s_set_gpr_idx_on m0, 0x8 //M0[7:0] = M0[7:0] and M0[15:12] = 0x8 + + L_RESTORE_ACCVGPR_LOOP: + read_4vgprs_from_mem(s_restore_buf_rsrc0, s_restore_mem_offset) + + for var vgpr = 0; vgpr < 4; ++ vgpr + v_accvgpr_write acc[vgpr], v[vgpr] + end + + s_add_u32 m0, m0, 4 //next vgpr index + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, 256*4 //every buffer_load_dword does 256 bytes + s_cmp_lt_u32 m0, s_restore_alloc_size //scc = (m0 < s_restore_alloc_size) ? 1 : 0 + s_cbranch_scc1 L_RESTORE_ACCVGPR_LOOP //VGPR restore (except v0) is complete? + L_RESTORE_ACCVGPR_END: +#endif + + s_set_gpr_idx_off + + // Restore VGPRs 0-3 last, no longer needed. + read_4vgprs_from_mem(s_restore_buf_rsrc0, s_restore_mem_offset_save) + + /* restore SGPRs */ + ////////////////////////////// + + // SGPR SR memory offset : size(VGPR) + get_vgpr_size_bytes(s_restore_mem_offset) + get_sgpr_size_bytes(s_restore_tmp) + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, s_restore_tmp + s_sub_u32 s_restore_mem_offset, s_restore_mem_offset, 16*4 // restore SGPR from S[n] to S[0], by 16 sgprs group + // TODO, change RSRC word to rearrange memory layout for SGPRS + + s_getreg_b32 s_restore_alloc_size, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SHIFT,SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SIZE) //spgr_size + s_add_u32 s_restore_alloc_size, s_restore_alloc_size, 1 + s_lshl_b32 s_restore_alloc_size, s_restore_alloc_size, 4 //Number of SGPRs = (sgpr_size + 1) * 16 (non-zero value) + + s_lshl_b32 s_restore_buf_rsrc2, s_restore_alloc_size, 2 //NUM_RECORDS in bytes + s_mov_b32 s_restore_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + s_mov_b32 m0, s_restore_alloc_size + + L_RESTORE_SGPR_LOOP: + read_16sgpr_from_mem(s0, s_restore_buf_rsrc0, s_restore_mem_offset) //PV: further performance improvement can be made + s_waitcnt lgkmcnt(0) //ensure data ready + + s_sub_u32 m0, m0, 16 // Restore from S[n] to S[0] + s_nop 0 // hazard SALU M0=> S_MOVREL + + s_movreld_b64 s0, s0 //s[0+m0] = s0 + s_movreld_b64 s2, s2 + s_movreld_b64 s4, s4 + s_movreld_b64 s6, s6 + s_movreld_b64 s8, s8 + s_movreld_b64 s10, s10 + s_movreld_b64 s12, s12 + s_movreld_b64 s14, s14 + + s_cmp_eq_u32 m0, 0 //scc = (m0 < s_restore_alloc_size) ? 1 : 0 + s_cbranch_scc0 L_RESTORE_SGPR_LOOP //SGPR restore (except s0) is complete? + + /* restore HW registers */ + ////////////////////////////// + L_RESTORE_HWREG: + + + // HWREG SR memory offset : size(VGPR)+size(SGPR) + get_vgpr_size_bytes(s_restore_mem_offset) + get_sgpr_size_bytes(s_restore_tmp) + s_add_u32 s_restore_mem_offset, s_restore_mem_offset, s_restore_tmp + + + s_mov_b32 s_restore_buf_rsrc2, 0x4 //NUM_RECORDS in bytes + s_mov_b32 s_restore_buf_rsrc2, 0x1000000 //NUM_RECORDS in bytes + + read_hwreg_from_mem(s_restore_m0, s_restore_buf_rsrc0, s_restore_mem_offset) //M0 + read_hwreg_from_mem(s_restore_pc_lo, s_restore_buf_rsrc0, s_restore_mem_offset) //PC + read_hwreg_from_mem(s_restore_pc_hi, s_restore_buf_rsrc0, s_restore_mem_offset) + read_hwreg_from_mem(s_restore_exec_lo, s_restore_buf_rsrc0, s_restore_mem_offset) //EXEC + read_hwreg_from_mem(s_restore_exec_hi, s_restore_buf_rsrc0, s_restore_mem_offset) + read_hwreg_from_mem(s_restore_status, s_restore_buf_rsrc0, s_restore_mem_offset) //STATUS + read_hwreg_from_mem(s_restore_trapsts, s_restore_buf_rsrc0, s_restore_mem_offset) //TRAPSTS + read_hwreg_from_mem(xnack_mask_lo, s_restore_buf_rsrc0, s_restore_mem_offset) //XNACK_MASK_LO + read_hwreg_from_mem(xnack_mask_hi, s_restore_buf_rsrc0, s_restore_mem_offset) //XNACK_MASK_HI + read_hwreg_from_mem(s_restore_mode, s_restore_buf_rsrc0, s_restore_mem_offset) //MODE + + s_waitcnt lgkmcnt(0) //from now on, it is safe to restore STATUS and IB_STS + + s_mov_b32 m0, s_restore_m0 + s_mov_b32 exec_lo, s_restore_exec_lo + s_mov_b32 exec_hi, s_restore_exec_hi + + s_and_b32 s_restore_m0, SQ_WAVE_TRAPSTS_PRE_SAVECTX_MASK, s_restore_trapsts + s_setreg_b32 hwreg(HW_REG_TRAPSTS, SQ_WAVE_TRAPSTS_PRE_SAVECTX_SHIFT, SQ_WAVE_TRAPSTS_PRE_SAVECTX_SIZE), s_restore_m0 + s_and_b32 s_restore_m0, SQ_WAVE_TRAPSTS_POST_SAVECTX_MASK, s_restore_trapsts + s_lshr_b32 s_restore_m0, s_restore_m0, SQ_WAVE_TRAPSTS_POST_SAVECTX_SHIFT + s_setreg_b32 hwreg(HW_REG_TRAPSTS, SQ_WAVE_TRAPSTS_POST_SAVECTX_SHIFT, SQ_WAVE_TRAPSTS_POST_SAVECTX_SIZE), s_restore_m0 + //s_setreg_b32 hwreg(HW_REG_TRAPSTS), s_restore_trapsts //don't overwrite SAVECTX bit as it may be set through external SAVECTX during restore + s_setreg_b32 hwreg(HW_REG_MODE), s_restore_mode + + // Restore trap temporaries 4-11, 13 initialized by SPI debug dispatch logic + // ttmp SR memory offset : size(VGPR)+size(SGPR)+0x40 + get_vgpr_size_bytes(s_restore_ttmps_lo) + get_sgpr_size_bytes(s_restore_ttmps_hi) + s_add_u32 s_restore_ttmps_lo, s_restore_ttmps_lo, s_restore_ttmps_hi + s_add_u32 s_restore_ttmps_lo, s_restore_ttmps_lo, s_restore_buf_rsrc0 + s_addc_u32 s_restore_ttmps_hi, s_restore_buf_rsrc1, 0x0 + s_and_b32 s_restore_ttmps_hi, s_restore_ttmps_hi, 0xFFFF + s_load_dwordx4 [ttmp4, ttmp5, ttmp6, ttmp7], [s_restore_ttmps_lo, s_restore_ttmps_hi], 0x50 glc:1 + s_load_dwordx4 [ttmp8, ttmp9, ttmp10, ttmp11], [s_restore_ttmps_lo, s_restore_ttmps_hi], 0x60 glc:1 + s_load_dword ttmp13, [s_restore_ttmps_lo, s_restore_ttmps_hi], 0x74 glc:1 + s_waitcnt lgkmcnt(0) + + restore_ib_sts(s_restore_tmp) + + s_and_b32 s_restore_pc_hi, s_restore_pc_hi, 0x0000ffff //pc[47:32] //Do it here in order not to affect STATUS + s_and_b64 exec, exec, exec // Restore STATUS.EXECZ, not writable by s_setreg_b32 + s_and_b64 vcc, vcc, vcc // Restore STATUS.VCCZ, not writable by s_setreg_b32 + set_status_without_spi_prio(s_restore_status, s_restore_tmp) // SCC is included, which is changed by previous salu + + s_barrier //barrier to ensure the readiness of LDS before access attempts from any other wave in the same TG //FIXME not performance-optimal at this time + + s_rfe_b64 s_restore_pc_lo //Return to the main shader program and resume execution + + +/**************************************************************************/ +/* the END */ +/**************************************************************************/ +L_END_PGM: + s_endpgm + +end + + +/**************************************************************************/ +/* the helper functions */ +/**************************************************************************/ + +//Only for save hwreg to mem +function write_hwreg_to_mem(s, s_rsrc, s_mem_offset) + s_mov_b32 exec_lo, m0 //assuming exec_lo is not needed anymore from this point on + s_mov_b32 m0, s_mem_offset + s_buffer_store_dword s, s_rsrc, m0 glc:1 + ack_sqc_store_workaround() + s_add_u32 s_mem_offset, s_mem_offset, 4 + s_mov_b32 m0, exec_lo +end + + +// HWREG are saved before SGPRs, so all HWREG could be use. +function write_16sgpr_to_mem(s, s_rsrc, s_mem_offset) + + s_buffer_store_dwordx4 s[0], s_rsrc, 0 glc:1 + ack_sqc_store_workaround() + s_buffer_store_dwordx4 s[4], s_rsrc, 16 glc:1 + ack_sqc_store_workaround() + s_buffer_store_dwordx4 s[8], s_rsrc, 32 glc:1 + ack_sqc_store_workaround() + s_buffer_store_dwordx4 s[12], s_rsrc, 48 glc:1 + ack_sqc_store_workaround() + s_add_u32 s_rsrc[0], s_rsrc[0], 4*16 + s_addc_u32 s_rsrc[1], s_rsrc[1], 0x0 // +scc +end + + +function read_hwreg_from_mem(s, s_rsrc, s_mem_offset) + s_buffer_load_dword s, s_rsrc, s_mem_offset glc:1 + s_add_u32 s_mem_offset, s_mem_offset, 4 +end + +function read_16sgpr_from_mem(s, s_rsrc, s_mem_offset) + s_buffer_load_dwordx16 s, s_rsrc, s_mem_offset glc:1 + s_sub_u32 s_mem_offset, s_mem_offset, 4*16 +end + +function check_if_tcp_store_ok + // If STATUS.ALLOW_REPLAY=0 and TRAPSTS.XNACK_ERROR=1 then TCP stores will fail. + s_and_b32 s_save_tmp, s_save_status, SQ_WAVE_STATUS_ALLOW_REPLAY_MASK + s_cbranch_scc1 L_TCP_STORE_CHECK_DONE + + s_getreg_b32 s_save_tmp, hwreg(HW_REG_TRAPSTS) + s_andn2_b32 s_save_tmp, SQ_WAVE_TRAPSTS_XNACK_ERROR_MASK, s_save_tmp + +L_TCP_STORE_CHECK_DONE: +end + +function write_4vgprs_to_mem(s_rsrc, s_mem_offset) + buffer_store_dword v0, v0, s_rsrc, s_mem_offset slc:1 glc:1 + buffer_store_dword v1, v0, s_rsrc, s_mem_offset slc:1 glc:1 offset:256 + buffer_store_dword v2, v0, s_rsrc, s_mem_offset slc:1 glc:1 offset:256*2 + buffer_store_dword v3, v0, s_rsrc, s_mem_offset slc:1 glc:1 offset:256*3 +end + +function read_4vgprs_from_mem(s_rsrc, s_mem_offset) + buffer_load_dword v0, v0, s_rsrc, s_mem_offset slc:1 glc:1 + buffer_load_dword v1, v0, s_rsrc, s_mem_offset slc:1 glc:1 offset:256 + buffer_load_dword v2, v0, s_rsrc, s_mem_offset slc:1 glc:1 offset:256*2 + buffer_load_dword v3, v0, s_rsrc, s_mem_offset slc:1 glc:1 offset:256*3 + s_waitcnt vmcnt(0) +end + +function write_vgpr_to_mem_with_sqc(v, s_rsrc, s_mem_offset) + s_mov_b32 s4, 0 + +L_WRITE_VGPR_LANE_LOOP: + for var lane = 0; lane < 4; ++ lane + v_readlane_b32 s[lane], v, s4 + s_add_u32 s4, s4, 1 + end + + s_buffer_store_dwordx4 s[0:3], s_rsrc, s_mem_offset glc:1 + ack_sqc_store_workaround() + + s_add_u32 s_mem_offset, s_mem_offset, 0x10 + s_cmp_eq_u32 s4, 0x40 + s_cbranch_scc0 L_WRITE_VGPR_LANE_LOOP +end + +function write_vgprs_to_mem_with_sqc(v, n_vgprs, s_rsrc, s_mem_offset) + for var vgpr = 0; vgpr < n_vgprs; ++ vgpr + write_vgpr_to_mem_with_sqc(v[vgpr], s_rsrc, s_mem_offset) + end +end + +function get_lds_size_bytes(s_lds_size_byte) + // SQ LDS granularity is 64DW, while PGM_RSRC2.lds_size is in granularity 128DW + s_getreg_b32 s_lds_size_byte, hwreg(HW_REG_LDS_ALLOC, SQ_WAVE_LDS_ALLOC_LDS_SIZE_SHIFT, SQ_WAVE_LDS_ALLOC_LDS_SIZE_SIZE) // lds_size + s_lshl_b32 s_lds_size_byte, s_lds_size_byte, 8 //LDS size in dwords = lds_size * 64 *4Bytes // granularity 64DW +end + +function get_vgpr_size_bytes(s_vgpr_size_byte) + s_getreg_b32 s_vgpr_size_byte, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SHIFT,SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SIZE) //vpgr_size + s_add_u32 s_vgpr_size_byte, s_vgpr_size_byte, 1 + s_lshl_b32 s_vgpr_size_byte, s_vgpr_size_byte, (2+8) //Number of VGPRs = (vgpr_size + 1) * 4 * 64 * 4 (non-zero value) //FIXME for GFX, zero is possible + +#if ASIC_FAMILY >= CHIP_ARCTURUS + s_lshl_b32 s_vgpr_size_byte, s_vgpr_size_byte, 1 // Double size for ACC VGPRs +#endif +end + +function get_sgpr_size_bytes(s_sgpr_size_byte) + s_getreg_b32 s_sgpr_size_byte, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SHIFT,SQ_WAVE_GPR_ALLOC_SGPR_SIZE_SIZE) //spgr_size + s_add_u32 s_sgpr_size_byte, s_sgpr_size_byte, 1 + s_lshl_b32 s_sgpr_size_byte, s_sgpr_size_byte, 6 //Number of SGPRs = (sgpr_size + 1) * 16 *4 (non-zero value) +end + +function get_hwreg_size_bytes + return 128 //HWREG size 128 bytes +end + +function get_num_arch_vgprs(s_num_arch_vgprs) +#if ASIC_FAMILY >= CHIP_ALDEBARAN + // VGPR count includes ACC VGPRs, use ACC VGPR offset for ARCH VGPR count. + s_getreg_b32 s_num_arch_vgprs, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_ACCV_OFFSET_SHIFT,SQ_WAVE_GPR_ALLOC_ACCV_OFFSET_SIZE) +#else + s_getreg_b32 s_num_arch_vgprs, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SHIFT,SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SIZE) +#endif + + // Number of VGPRs = (vgpr_size + 1) * 4 + s_add_u32 s_num_arch_vgprs, s_num_arch_vgprs, 1 + s_lshl_b32 s_num_arch_vgprs, s_num_arch_vgprs, 2 +end + +#if ASIC_FAMILY >= CHIP_ALDEBARAN +function get_num_acc_vgprs(s_num_acc_vgprs, s_tmp) + // VGPR count = (GPR_ALLOC.VGPR_SIZE + 1) * 8 + s_getreg_b32 s_num_acc_vgprs, hwreg(HW_REG_GPR_ALLOC,SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SHIFT,SQ_WAVE_GPR_ALLOC_VGPR_SIZE_SIZE) + s_add_u32 s_num_acc_vgprs, s_num_acc_vgprs, 1 + s_lshl_b32 s_num_acc_vgprs, s_num_acc_vgprs, 3 + + // ACC VGPR count = VGPR count - ARCH VGPR count. + get_num_arch_vgprs(s_tmp) + s_sub_u32 s_num_acc_vgprs, s_num_acc_vgprs, s_tmp +end +#endif + +function ack_sqc_store_workaround + if ACK_SQC_STORE + s_waitcnt lgkmcnt(0) + end +end + +function set_status_without_spi_prio(status, tmp) + // Do not restore STATUS.SPI_PRIO since scheduler may have raised it. + s_lshr_b32 tmp, status, SQ_WAVE_STATUS_POST_SPI_PRIO_SHIFT + s_setreg_b32 hwreg(HW_REG_STATUS, SQ_WAVE_STATUS_POST_SPI_PRIO_SHIFT, SQ_WAVE_STATUS_POST_SPI_PRIO_SIZE), tmp + s_nop 0x2 // avoid S_SETREG => S_SETREG hazard + s_setreg_b32 hwreg(HW_REG_STATUS, SQ_WAVE_STATUS_PRE_SPI_PRIO_SHIFT, SQ_WAVE_STATUS_PRE_SPI_PRIO_SIZE), status +end + +function save_and_clear_ib_sts(tmp) + // Save IB_STS.FIRST_REPLAY[15] and IB_STS.RCNT[20:16] into unused space s_save_ib_sts[31:26]. + s_getreg_b32 tmp, hwreg(HW_REG_IB_STS) + s_and_b32 tmp, tmp, SQ_WAVE_IB_STS_RCNT_FIRST_REPLAY_MASK + s_lshl_b32 tmp, tmp, (TTMP_SAVE_RCNT_FIRST_REPLAY_SHIFT - SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT) + s_andn2_b32 s_save_ib_sts, s_save_ib_sts, TTMP_SAVE_RCNT_FIRST_REPLAY_MASK + s_or_b32 s_save_ib_sts, s_save_ib_sts, tmp + s_setreg_imm32_b32 hwreg(HW_REG_IB_STS), 0x0 +end + +function restore_ib_sts(tmp) + s_lshr_b32 tmp, s_save_ib_sts, (TTMP_SAVE_RCNT_FIRST_REPLAY_SHIFT - SQ_WAVE_IB_STS_FIRST_REPLAY_SHIFT) + s_and_b32 tmp, tmp, SQ_WAVE_IB_STS_RCNT_FIRST_REPLAY_MASK + s_setreg_b32 hwreg(HW_REG_IB_STS), tmp +end |