Page MenuHomePhabricator

kzhuravl (Konstantin Zhuravlyov)
User

Projects

User does not belong to any projects.

User Details

User Since
Feb 16 2016, 12:13 PM (333 w, 1 d)

Recent Activity

Wed, Jun 29

kzhuravl accepted D128344: [AMDGPU] Add the uses_dynamic_stack field to the kernel descriptor and the kernel metadata map.
Wed, Jun 29, 10:20 AM · Restricted Project, Restricted Project

Tue, Jun 28

kzhuravl accepted D115562: AMDGPU: Remove amdgpu-waves-per-eu support in old attribute pass.

LGTM, thanks

Tue, Jun 28, 6:50 AM · Restricted Project, Restricted Project

Fri, Jun 24

kzhuravl committed rG7736ce1c56c7: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence (authored by kzhuravl).
AMDGPU: Clear kill flags when optimizing vcmp save exec sequence
Fri, Jun 24, 8:33 AM · Restricted Project, Restricted Project
kzhuravl closed D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.
Fri, Jun 24, 8:33 AM · Restricted Project, Restricted Project

Wed, Jun 22

kzhuravl requested changes to D128344: [AMDGPU] Add the uses_dynamic_stack field to the kernel descriptor and the kernel metadata map.

Do we want to tie this bit to a particular code object version? Code object v4 and up? Why did you pick code object v3 and up?

Wed, Jun 22, 8:48 AM · Restricted Project, Restricted Project
kzhuravl added a comment to D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.

@arsenm, any additional feedback?

Wed, Jun 22, 8:11 AM · Restricted Project, Restricted Project

Mon, Jun 20

kzhuravl added inline comments to D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.
Mon, Jun 20, 10:01 AM · Restricted Project, Restricted Project
kzhuravl updated the diff for D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.

Clear kill flags for Src0 if its a reg.

Mon, Jun 20, 10:01 AM · Restricted Project, Restricted Project

Tue, Jun 14

kzhuravl updated the diff for D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.

Handle non register case and add a test for it.

Tue, Jun 14, 12:40 PM · Restricted Project, Restricted Project
kzhuravl added a comment to D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.

Why not to remove original v_cmp instead? It should have no other uses.

Original v_cmp_eq_u32 got removed in favor of v_cmpx_eq_u32. v_cmp_ne_u32 is left as is.

OK, I see. That's another instruction in between kills it..

What if src1 is not register?

Tue, Jun 14, 12:39 PM · Restricted Project, Restricted Project
kzhuravl added a comment to D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.

Why not to remove original v_cmp instead? It should have no other uses.

Tue, Jun 14, 11:47 AM · Restricted Project, Restricted Project
kzhuravl updated the summary of D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.
Tue, Jun 14, 10:36 AM · Restricted Project, Restricted Project
kzhuravl requested review of D127768: AMDGPU: Clear kill flags when optimizing vcmp save exec sequence.
Tue, Jun 14, 10:35 AM · Restricted Project, Restricted Project

Thu, Jun 9

kzhuravl accepted D127241: [AMDGPU] gfx11 add bits to COMPUTE_PGM_RSRC3.

LGTM, can you wait and see if @rampitec has any additional comments?

Thu, Jun 9, 7:54 AM · Restricted Project, Restricted Project
kzhuravl added a comment to D127241: [AMDGPU] gfx11 add bits to COMPUTE_PGM_RSRC3.

Can you add documentation to AMDGPUUsage in this patch as well?

That's in D127402 which is in review right now.

Thu, Jun 9, 7:53 AM · Restricted Project, Restricted Project
kzhuravl added a comment to D127241: [AMDGPU] gfx11 add bits to COMPUTE_PGM_RSRC3.

This looks like it affects gfx10 as well, while probably should not?

@kzhuravl Do you know about this?

GFX10 is only using COMPUTE_PGM_RSRC3_GFX10_PLUS.SHARED_VGPR_COUNT, the rest of the bits for GFX10 are reserved and must be 0.

GFX11 is using COMPUTE_PGM_RSRC3_GFX10_PLUS.INST_PREF_SIZE and others.

This is similar to how we do other compute program resource registers. E.g.:

...
  COMPUTE_PGM_RSRC1(BULKY, 24, 1),
  COMPUTE_PGM_RSRC1(CDBG_USER, 25, 1),
  COMPUTE_PGM_RSRC1(FP16_OVFL, 26, 1),    // GFX9+
  COMPUTE_PGM_RSRC1(RESERVED0, 27, 2),
  COMPUTE_PGM_RSRC1(WGP_MODE, 29, 1),     // GFX10+
  COMPUTE_PGM_RSRC1(MEM_ORDERED, 30, 1),  // GFX10+
  COMPUTE_PGM_RSRC1(FWD_PROGRESS, 31, 1), // GFX10+
...

The reason we have 2 different compute program resource register 3 is because GFX90A and GFX10 have completely different layout for those. So I think this header file should be fine.

Can you add documentation to AMDGPUUsage in this patch as well?

Thanks

Thu, Jun 9, 7:47 AM · Restricted Project, Restricted Project
kzhuravl added a comment to D127241: [AMDGPU] gfx11 add bits to COMPUTE_PGM_RSRC3.

This looks like it affects gfx10 as well, while probably should not?

@kzhuravl Do you know about this?

Thu, Jun 9, 7:45 AM · Restricted Project, Restricted Project

Apr 28 2022

kzhuravl accepted D124536: [AMDGPU] Add gfx11 subtarget ELF definition.

LGTM, thanks

Apr 28 2022, 9:18 AM · Restricted Project, Restricted Project, Restricted Project
kzhuravl requested changes to D124536: [AMDGPU] Add gfx11 subtarget ELF definition.

Need to update processor table: https://llvm.org/docs/AMDGPUUsage.html#processors
Need to update EF_AMDGPU_MACH_AMDGCN table: https://llvm.org/docs/AMDGPUUsage.html#amdgpu-ef-amdgpu-mach-table

Apr 28 2022, 8:30 AM · Restricted Project, Restricted Project, Restricted Project
kzhuravl added a comment to D124536: [AMDGPU] Add gfx11 subtarget ELF definition.

Looks like this change is missing doc updates?

Apr 28 2022, 8:28 AM · Restricted Project, Restricted Project, Restricted Project
kzhuravl accepted D124537: [AMDGPU][clang] Definition of gfx11 subtarget.
Apr 28 2022, 8:25 AM · Restricted Project, Restricted Project

Apr 13 2022

kzhuravl accepted D123687: [AMDGPU] Increate hazard for store dwordx3/4 to 2 waitstates on gfx940.

LGTM, thanks!

Apr 13 2022, 2:05 PM · Restricted Project, Restricted Project

Mar 29 2022

kzhuravl accepted D122339: [AMDGPU] gfx940 VALU hazard recognizer.

LGTM with clang formatting fixed

Mar 29 2022, 9:36 AM · Restricted Project, Restricted Project

Mar 24 2022

kzhuravl accepted D122044: [AMDGPU] New gfx940 mfma instructions.

LGTM, unless @foad has any additional questions

Mar 24 2022, 11:50 AM · Restricted Project, Restricted Project, Restricted Project

Mar 23 2022

kzhuravl accepted D122191: [AMDGPU] Support gfx940 smfmac instructions.

LGTM with a minor nit

Mar 23 2022, 12:43 PM · Restricted Project, Restricted Project, Restricted Project

Mar 22 2022

kzhuravl added a comment to D122190: [AMDGPU][LLD] Adding support for ABI version 5 option.

Do we need to add the test for it?

Mar 22 2022, 1:02 PM · Restricted Project, Restricted Project

Mar 18 2022

kzhuravl accepted D121966: [AMDGPU] gfx940 basic speed model.
Mar 18 2022, 12:36 PM · Restricted Project, Restricted Project
kzhuravl accepted D121745: [AMDGPU] reuse blgp as neg in 2 mfma operations on gfx940.

LGTM

Mar 18 2022, 12:32 PM · Restricted Project, Restricted Project

Mar 17 2022

kzhuravl accepted D121741: [AMDGPU] New MFMA names for existing instructions.

LGTM

Mar 17 2022, 1:18 PM · Restricted Project, Restricted Project
kzhuravl accepted D121811: [AMDGPU][DOC][NFC] Add links to MI200 documentation.

LGTM, thanks

Mar 17 2022, 12:41 PM · Restricted Project, Restricted Project
kzhuravl added inline comments to D121745: [AMDGPU] reuse blgp as neg in 2 mfma operations on gfx940.
Mar 17 2022, 12:38 PM · Restricted Project, Restricted Project
kzhuravl accepted D120849: [AMDGPU] Add gfx90a and gfx940 to get_elf_mach_gfx_name.cpp.

lgtm

Mar 17 2022, 12:31 PM · Restricted Project, Restricted Project

Mar 14 2022

kzhuravl accepted D121634: [AMDGPU] gfx940: disable OP_SEL on V_DOT instructions.
Mar 14 2022, 4:43 PM · Restricted Project, Restricted Project

Mar 11 2022

kzhuravl accepted D121234: [AMDGPU] gfx940 MUBUF format changes.

LGTM with the nit fixed

Mar 11 2022, 10:51 AM · Restricted Project, Restricted Project

Mar 9 2022

kzhuravl added a reviewer for D121242: [AMDGPU] gfx940 memory model: t-tye.

Adding Tony

Mar 9 2022, 10:18 AM · Restricted Project, Restricted Project

Mar 7 2022

kzhuravl accepted D105507: [AMDGPU] Add gfx10 assembler directive to specify shared VGPR count.

LGTM, thanks for taking this over!

Mar 7 2022, 1:43 PM · Restricted Project, Restricted Project
kzhuravl accepted D120769: [AMDGPU] gfx940: support V_FMAMK_F32 and V_FMAAK_F32.

LGTM, thanks

Mar 7 2022, 11:09 AM · Restricted Project, Restricted Project
kzhuravl accepted D120855: [AMDGPU] gfx940 uses new names for coherency bits.

Other than a TODO comment, LGTM, thanks

Mar 7 2022, 10:23 AM · Restricted Project, Restricted Project
kzhuravl accepted D121028: [AMDGPU] new gfx940 fp atomics.

lgtm

Mar 7 2022, 8:52 AM · Restricted Project, Restricted Project, Restricted Project
kzhuravl added inline comments to D121028: [AMDGPU] new gfx940 fp atomics.
Mar 7 2022, 8:40 AM · Restricted Project, Restricted Project, Restricted Project

Mar 3 2022

Herald added a project to D117797: Revert "[amdgpu] Improve the from f32 to i64.": Restricted Project.

ping

Mar 3 2022, 9:30 AM · Restricted Project, Restricted Project

Mar 2 2022

kzhuravl accepted D120763: [AMDGPU] gfx940 removes all image instructions.

lgtm

Mar 2 2022, 11:10 AM · Restricted Project, Restricted Project

Mar 1 2022

kzhuravl accepted D120688: [AMDGPU] Add gfx940 target.

LGTM, maybe give a day or so before merging this, so other folks have a chance to take a look

Mar 1 2022, 5:01 PM · Restricted Project, Restricted Project, Restricted Project, Restricted Project

Feb 15 2022

kzhuravl accepted D119869: [AMDGPU] Do not define GET_INSTRINFO_SCHED_ENUM.

lgtm, thanks

Feb 15 2022, 12:50 PM · Restricted Project

Feb 10 2022

kzhuravl added a comment to D117797: Revert "[amdgpu] Improve the from f32 to i64.".

ping

Feb 10 2022, 9:21 AM · Restricted Project, Restricted Project

Feb 3 2022

kzhuravl added a comment to D117797: Revert "[amdgpu] Improve the from f32 to i64.".

ping

Feb 3 2022, 9:14 AM · Restricted Project, Restricted Project

Feb 2 2022

kzhuravl updated subscribers of D118229: [AMDGPUHSAMetadataStreamer] Do not assume ABI alignment for pointers.

+Mark,Ron

Feb 2 2022, 8:27 AM · Restricted Project
kzhuravl added a comment to D118229: [AMDGPUHSAMetadataStreamer] Do not assume ABI alignment for pointers.

This change introduces the regression in OpenCL conformance test: basic - kernel_memory_alignment_local. Does it require any corresponding runtime changes?

Feb 2 2022, 8:23 AM · Restricted Project

Jan 24 2022

kzhuravl accepted D117845: [AMDGPU][NFC] Update to AMDGPUUsage for default Code Object Version.

lgtm

Jan 24 2022, 1:07 PM · Restricted Project
kzhuravl added a comment to D117845: [AMDGPU][NFC] Update to AMDGPUUsage for default Code Object Version.

Can you also update summary to drop AMDGPULLVMExtensionsForHeterogeneousDebugging?

Jan 24 2022, 11:16 AM · Restricted Project
kzhuravl added inline comments to D117845: [AMDGPU][NFC] Update to AMDGPUUsage for default Code Object Version.
Jan 24 2022, 11:09 AM · Restricted Project
kzhuravl added a comment to D117845: [AMDGPU][NFC] Update to AMDGPUUsage for default Code Object Version.

Any reason you are bringing in heterogeneous debugging bits in this change?

Jan 24 2022, 11:09 AM · Restricted Project
kzhuravl added a comment to D117845: [AMDGPU][NFC] Update to AMDGPUUsage for default Code Object Version.

Can you upload a diff with the context?

Jan 24 2022, 10:58 AM · Restricted Project

Jan 20 2022

kzhuravl added a comment to D117797: Revert "[amdgpu] Improve the from f32 to i64.".

Breaks how / on what values?

Jan 20 2022, 9:13 AM · Restricted Project, Restricted Project
kzhuravl added a reverting change for rG940efa4f6981: [amdgpu] Improve the from f32 to i64.: D117797: Revert "[amdgpu] Improve the from f32 to i64.".
Jan 20 2022, 8:58 AM
kzhuravl added a reverting change for D104427: [amdgpu] Improve the from f32 to i64.: D117797: Revert "[amdgpu] Improve the from f32 to i64.".
Jan 20 2022, 8:58 AM · Restricted Project
kzhuravl requested review of D117797: Revert "[amdgpu] Improve the from f32 to i64.".
Jan 20 2022, 8:58 AM · Restricted Project, Restricted Project

Jan 11 2022

kzhuravl accepted D116686: Revert D109159 : Revert "[amdgpu] Enable selection of `s_cselect_b64`.".

LGTM, I have looked through the implementation, and it looks like a correct revert of https://reviews.llvm.org/rG640beb38e7710b939b3cfb3f4c54accc694b1d30. I have looked through most tests, but not all. Can you make sure to run lit tests before pushing your change? Thank you

Jan 11 2022, 12:15 PM · Restricted Project

Dec 28 2021

kzhuravl accepted D116275: [AMDGPU][NFC] Update to DWARF extension for heterogeneous debugging.
Dec 28 2021, 8:50 AM · Restricted Project
kzhuravl accepted D116265: [AMDGPU][NFC] Update DWARF extension allowing locations on stack.
Dec 28 2021, 8:27 AM · Restricted Project

Dec 22 2021

kzhuravl added a comment to D116140: [AMDGPU] Add agpr_count to metadata and AsmParser.

Add @tpr to look at pal metadata changes

Dec 22 2021, 8:11 AM · Restricted Project
kzhuravl added a reviewer for D116140: [AMDGPU] Add agpr_count to metadata and AsmParser: tpr.
Dec 22 2021, 8:11 AM · Restricted Project

Dec 20 2021

kzhuravl accepted D116077: [AMDGPU][NFC] Update DWARF extension for locations on the stack.
Dec 20 2021, 9:10 PM · Restricted Project
kzhuravl accepted D115960: Revert D109159 "[amdgpu] Enable selection of `s_cselect_b64`.".
Dec 20 2021, 9:33 AM · Restricted Project, Restricted Project, Restricted Project, Restricted Project, Restricted Project, Restricted Project, Restricted Project

Nov 23 2021

kzhuravl accepted D114289: [AMDGPU] Remove a no-op check in the gfx90a hazard recognizer.
Nov 23 2021, 3:37 PM · Restricted Project

Nov 10 2021

kzhuravl added inline comments to D113203: [AMDGPU] Small correction in SITargetLowering::performOrCombine().
Nov 10 2021, 1:19 PM · Restricted Project, Restricted Project
kzhuravl accepted D113203: [AMDGPU] Small correction in SITargetLowering::performOrCombine().

LGTM with a nit below. Thanks!

Nov 10 2021, 11:52 AM · Restricted Project, Restricted Project

Nov 9 2021

kzhuravl accepted D112716: AMDGPU: Account for implicit argument alignment for kernarg segment.

The metadata treats the minimum as 16?

Nov 9 2021, 12:44 PM · Restricted Project

Nov 4 2021

kzhuravl added a comment to D113203: [AMDGPU] Small correction in SITargetLowering::performOrCombine().

Agreed on needing a lit test

Nov 4 2021, 10:19 AM · Restricted Project, Restricted Project

Oct 5 2021

kzhuravl added a comment to D108322: AMDGPU: Add options to override SGPR and VGPR blocks.

ping

Oct 5 2021, 10:55 AM · Restricted Project

Sep 24 2021

kzhuravl accepted D110376: [AMDGPU] Always reserve flat scratch SGPR for architected flat scratch.

LGTM, thanks!

Sep 24 2021, 5:35 AM · Restricted Project

Sep 23 2021

kzhuravl added a comment to D110337: [AMDGPU] Use "hostcall" module flag instead of searching for ockl_hostcall_internal() declaration..

Probably need docs https://llvm.org/docs/AMDGPUUsage.html#llvm-ir-attributes

Sep 23 2021, 8:35 AM · Restricted Project, Restricted Project

Aug 27 2021

kzhuravl added a comment to D108322: AMDGPU: Add options to override SGPR and VGPR blocks.

ping

Aug 27 2021, 9:08 AM · Restricted Project

Aug 20 2021

kzhuravl updated the diff for D108322: AMDGPU: Add options to override SGPR and VGPR blocks.

Address review feedback.

Aug 20 2021, 11:34 AM · Restricted Project

Aug 19 2021

kzhuravl added inline comments to D108322: AMDGPU: Add options to override SGPR and VGPR blocks.
Aug 19 2021, 8:46 AM · Restricted Project

Aug 18 2021

kzhuravl requested review of D108322: AMDGPU: Add options to override SGPR and VGPR blocks.
Aug 18 2021, 12:39 PM · Restricted Project

Jul 7 2021

kzhuravl added a comment to D105507: [AMDGPU] Add gfx10 assembler directive to specify shared VGPR count.

Also needs disassembler change.

Jul 7 2021, 8:03 AM · Restricted Project, Restricted Project

Jul 6 2021

kzhuravl requested review of D105507: [AMDGPU] Add gfx10 assembler directive to specify shared VGPR count.
Jul 6 2021, 12:30 PM · Restricted Project, Restricted Project
kzhuravl accepted D105377: [NFC][AMDGPU] Add link to AMD GPU gfx906 instruction set architecture.
Jul 6 2021, 7:58 AM · Restricted Project

Jun 24 2021

kzhuravl accepted D104872: [AMDGPU] Reserve AMDGPU ELF e_flags machine 0x43.

lgtm

Jun 24 2021, 1:58 PM · Restricted Project

Jun 4 2021

kzhuravl committed rG4d9f8527dbfb: CUDA/HIP: Change device-use-host-var.cu's NOT "external" check to include… (authored by kzhuravl).
CUDA/HIP: Change device-use-host-var.cu's NOT "external" check to include…
Jun 4 2021, 10:10 AM
kzhuravl closed D103658: CUDA/HIP: Change device-use-host-var.cu's NOT "external" check to include "addrspace".
Jun 4 2021, 10:10 AM · Restricted Project

Jun 3 2021

kzhuravl updated the diff for D103658: CUDA/HIP: Change device-use-host-var.cu's NOT "external" check to include "addrspace".

Address review feedback.

Jun 3 2021, 5:24 PM · Restricted Project
kzhuravl requested review of D103658: CUDA/HIP: Change device-use-host-var.cu's NOT "external" check to include "addrspace".
Jun 3 2021, 4:31 PM · Restricted Project

May 20 2021

kzhuravl accepted D102859: [NFC][AMDGPU] Add documentation for AMD Instinct MI100 accelerator.
May 20 2021, 11:55 AM · Restricted Project

May 19 2021

kzhuravl closed D102708: AMDGPU/NFC: Replace EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3E with EF_AMDGPU_MACH_AMDGCN_GFX1034.
May 19 2021, 8:32 AM · Restricted Project

May 18 2021

kzhuravl added inline comments to D102306: Add gfx1034.
May 18 2021, 2:06 PM · Restricted Project, Restricted Project, Restricted Project
kzhuravl committed rG4e297dcd18da: AMDGPU/Docs: Remove reserved MACH 0x3E (it is no longer reserved), sort MACHs… (authored by kzhuravl).
AMDGPU/Docs: Remove reserved MACH 0x3E (it is no longer reserved), sort MACHs…
May 18 2021, 1:58 PM
kzhuravl committed rG0f544be244e3: AMDGPU/NFC: Replace EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3E with… (authored by kzhuravl).
AMDGPU/NFC: Replace EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3E with…
May 18 2021, 12:12 PM
kzhuravl closed D102708: AMDGPU/NFC: Replace EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3E with EF_AMDGPU_MACH_AMDGCN_GFX1034.
May 18 2021, 12:12 PM · Restricted Project
kzhuravl requested review of D102708: AMDGPU/NFC: Replace EF_AMDGPU_MACH_AMDGCN_RESERVED_0X3E with EF_AMDGPU_MACH_AMDGCN_GFX1034.
May 18 2021, 11:32 AM · Restricted Project

May 17 2021

kzhuravl updated subscribers of D102306: Add gfx1034.

Is there documentation for the mapping from product names to gfx numbers?

See the Processors section of https://llvm.org/docs/AMDGPUUsage.html

It appears to be missing some GPUs. None of WX series are listed, for example.

May 17 2021, 8:49 AM · Restricted Project, Restricted Project, Restricted Project

May 13 2021

kzhuravl accepted D102432: [AMDGPU] Add support for architected flat scratch.

LGTM, but please run clang-format, thanks!

May 13 2021, 6:10 PM · Restricted Project

May 3 2021

kzhuravl added a comment to D101304: AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying.

Hi, llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v3.s is failing for me.
/android0/llvm-project/llvm/build/bin/llvm-mc: error: : error: unable to get target for 'amdgcn-amd-amdhsa', see --version and --triple.
/android0/llvm-project/llvm/build/bin/llvm-readobj: error: '<stdin>': The file was not recognized as a valid object file
FileCheck error: '<stdin>' is empty.

May 3 2021, 6:48 AM · Restricted Project
kzhuravl added a comment to D101304: AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying.

Hello,

The reland version of this patch for which I cannot find the differential of has broken LLVM::note-amd-valid-v2.test on PowerPC:

/home/buildbots/ppc64be-clang-test/clang-ppc64be/llvm/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test:17:14: error: LLVM-NEXT: expected string not found in input
# LLVM-NEXT: AMD HSA Code Object Version: [Major: 2, Minor: 1]
             ^
<stdin>:15:68: note: scanning from here
 Type: NT_AMD_HSA_CODE_OBJECT_VERSION (AMD HSA Code Object Version)
                                                                   ^
<stdin>:16:2: note: possible intended match here
 AMD HSA Code Object Version: [Major: 33554432, Minor: 16777216]
 ^
Input file: <stdin>
Check file: /home/buildbots/ppc64be-clang-test/clang-ppc64be/llvm/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test
-dump-input=help explains the following input dump.
Input was:
<<<<<<
           .
           .
           .
          10:  Offset: 0x40 
          11:  Size: 0x18 
          12:  Note { 
          13:  Owner: AMD 
          14:  Data size: 0x8 
          15:  Type: NT_AMD_HSA_CODE_OBJECT_VERSION (AMD HSA Code Object Version) 
next:17'0                                                                        X error: no match found
          16:  AMD HSA Code Object Version: [Major: 33554432, Minor: 16777216] 
next:17'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
next:17'1      ?                                                                possible intended match

Seems like a small fix but if you could have a push for it soon that'd be great. Thanks! FYI @rksharma

Hi,

Reland patch fixed an issue if AMDGPU backend was not built, which was a one liner (I reverted original patch to not create the noise).

Seems like it is an endianness issue.

Unfortunately I don't have PowerPC, will adding these lines in note-amd-valid-v2.test mark it as XFAIL for you:

diff --git a/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test b/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test
index 688bbd50994c..f9cac212ce9f 100644
--- a/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test
+++ b/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test
@@ -1,6 +1,9 @@
 ## This test is checking the handling of valid note entries for AMDGPU code
 ## object v2.

+## Big endian not supported.
+# XFAIL: host-byteorder-big-endian
+
 # RUN: yaml2obj %s -o %t.o
 # RUN: llvm-readobj --notes %t.o | FileCheck %s --match-full-lines --check-prefix=LLVM
 # RUN: llvm-readelf --notes %t.o | FileCheck %s --match-full-lines --check-prefix=GNU

?

That would fix it yes, thank you!
Another possible solution would be to add a BE specific runline, but I'm not really familiar with the AMD test files so it's just a suggestion.

May 3 2021, 6:46 AM · Restricted Project
kzhuravl committed rG2055cc8ef4b9: AMDGPU: XFAIL LLVM::note-amd-valid-v2.test for big endian (authored by kzhuravl).
AMDGPU: XFAIL LLVM::note-amd-valid-v2.test for big endian
May 3 2021, 6:45 AM
kzhuravl added a comment to D101304: AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying.

Hello,

The reland version of this patch for which I cannot find the differential of has broken LLVM::note-amd-valid-v2.test on PowerPC:

/home/buildbots/ppc64be-clang-test/clang-ppc64be/llvm/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test:17:14: error: LLVM-NEXT: expected string not found in input
# LLVM-NEXT: AMD HSA Code Object Version: [Major: 2, Minor: 1]
             ^
<stdin>:15:68: note: scanning from here
 Type: NT_AMD_HSA_CODE_OBJECT_VERSION (AMD HSA Code Object Version)
                                                                   ^
<stdin>:16:2: note: possible intended match here
 AMD HSA Code Object Version: [Major: 33554432, Minor: 16777216]
 ^
Input file: <stdin>
Check file: /home/buildbots/ppc64be-clang-test/clang-ppc64be/llvm/llvm/test/tools/llvm-readobj/ELF/note-amd-valid-v2.test
-dump-input=help explains the following input dump.
Input was:
<<<<<<
           .
           .
           .
          10:  Offset: 0x40 
          11:  Size: 0x18 
          12:  Note { 
          13:  Owner: AMD 
          14:  Data size: 0x8 
          15:  Type: NT_AMD_HSA_CODE_OBJECT_VERSION (AMD HSA Code Object Version) 
next:17'0                                                                        X error: no match found
          16:  AMD HSA Code Object Version: [Major: 33554432, Minor: 16777216] 
next:17'0     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
next:17'1      ?                                                                possible intended match

Seems like a small fix but if you could have a push for it soon that'd be great. Thanks! FYI @rksharma

May 3 2021, 6:11 AM · Restricted Project

May 2 2021

kzhuravl added a reverting change for rG54aad6365951: Revert "AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying": rG94aaf3ddd96f: Reland "AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying".
May 2 2021, 7:56 PM
kzhuravl committed rG94aaf3ddd96f: Reland "AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying" (authored by kzhuravl).
Reland "AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying"
May 2 2021, 7:56 PM

Apr 30 2021

kzhuravl added a reverting change for D101304: AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying: rG54aad6365951: Revert "AMDGPU/llvm-readobj: Add missing tests for note parsing/displaying".
Apr 30 2021, 11:46 AM · Restricted Project