People kzhuravl

kzhuravl (Konstantin Zhuravlyov)
User

Projects

User does not belong to any projects.

User Details

User Since
Feb 16 2016, 12:13 PM (122 w, 1 d)

Recent Activity
View All

Yesterday

kzhuravl accepted D48385: AMDGPU: Fix scalar_to_vector for v4i16/v4f16.

LGTM

Wed, Jun 20, 12:12 PM

Fri, Jun 15

kzhuravl created D48234: AMDGPU: Remove ability to reserve VGPRs for debugger.
Fri, Jun 15, 1:25 PM
kzhuravl added inline comments to D47736: AMDHSA Code Object v3 assembler syntax update.
Fri, Jun 15, 7:31 AM

Thu, Jun 14

kzhuravl added inline comments to D47736: AMDHSA Code Object v3 assembler syntax update.
Thu, Jun 14, 2:29 PM
kzhuravl updated subscribers of D48191: AMDGPU/AMDHSA: Remove GridWorkGroupCountX/Y/Z.
Thu, Jun 14, 1:24 PM
kzhuravl created D48191: AMDGPU/AMDHSA: Remove GridWorkGroupCountX/Y/Z.
Thu, Jun 14, 1:11 PM
kzhuravl added inline comments to D47736: AMDHSA Code Object v3 assembler syntax update.
Thu, Jun 14, 10:03 AM

Wed, Jun 13

kzhuravl added inline comments to D47736: AMDHSA Code Object v3 assembler syntax update.
Wed, Jun 13, 6:50 PM
kzhuravl added a comment to D48103: [AMDGPU] Update code object metadata format documentation.

This seems like a duplicate of D47549?

Wed, Jun 13, 6:35 PM
kzhuravl accepted D47549: [AMDGPU] Update code object metadata format documentation.

looks good to me

Wed, Jun 13, 6:34 PM
kzhuravl accepted D48101: [AMDGPU] Document the AMDGPU LLVM attributes.

looks good to me

Wed, Jun 13, 6:33 PM

Tue, Jun 12

kzhuravl closed D42257: AMDGPU/MemoryModel: Fix monotonic atomic loads.

Done.

Tue, Jun 12, 11:43 AM
kzhuravl added a comment to D47736: AMDHSA Code Object v3 assembler syntax update.

Missing codegen tests.

Tue, Jun 12, 11:42 AM
kzhuravl closed D47566: AMDHSA: Code object v3 updates.

rL334519

Tue, Jun 12, 11:42 AM
kzhuravl committed rL334521: AMDHSA/NFC: Code object v3 updates (additional):.
AMDHSA/NFC: Code object v3 updates (additional):
Tue, Jun 12, 11:38 AM
kzhuravl committed rL334519: AMDHSA: Code object v3 updates.
AMDHSA: Code object v3 updates
Tue, Jun 12, 11:07 AM

Mon, Jun 11

kzhuravl committed rLLD334444: AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation.
AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation
Mon, Jun 11, 2:47 PM
kzhuravl committed rL334444: AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation.
AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation
Mon, Jun 11, 2:47 PM
kzhuravl closed D47734: AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation.
Mon, Jun 11, 2:47 PM
kzhuravl committed rL334443: AMDGPU: Add 64-bit relative variant kind.
AMDGPU: Add 64-bit relative variant kind
Mon, Jun 11, 2:42 PM
kzhuravl closed D47601: AMDGPU: Add 64-bit relative variant kind.
Mon, Jun 11, 2:42 PM

Thu, Jun 7

kzhuravl added inline comments to D47566: AMDHSA: Code object v3 updates.
Thu, Jun 7, 12:33 PM
kzhuravl accepted D47900: AMDGPU: Error on LDS global address in functions.

LGTM

Thu, Jun 7, 12:20 PM
kzhuravl accepted D47504: [AMDGPU] Simplify memory legalizer.

LGTM

Thu, Jun 7, 11:52 AM

Tue, Jun 5

kzhuravl added inline comments to D47566: AMDHSA: Code object v3 updates.
Tue, Jun 5, 5:42 PM
kzhuravl updated the diff for D47566: AMDHSA: Code object v3 updates.

Address review feedback.

Tue, Jun 5, 5:42 PM
kzhuravl added inline comments to D47736: AMDHSA Code Object v3 assembler syntax update.
Tue, Jun 5, 12:24 PM

Mon, Jun 4

kzhuravl added inline comments to D47566: AMDHSA: Code object v3 updates.
Mon, Jun 4, 3:24 PM
kzhuravl added a dependent revision for D47601: AMDGPU: Add 64-bit relative variant kind: D47734: AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation.
Mon, Jun 4, 1:06 PM
kzhuravl added a dependency for D47734: AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation: D47601: AMDGPU: Add 64-bit relative variant kind.
Mon, Jun 4, 1:06 PM
kzhuravl created D47734: AMDGPU/LLD: Handle R_AMDGPU_REL64 relocation.
Mon, Jun 4, 11:42 AM
kzhuravl updated the diff for D47566: AMDHSA: Code object v3 updates.

Fix symbol bindings and rebase.

Mon, Jun 4, 11:40 AM
kzhuravl updated the diff for D47601: AMDGPU: Add 64-bit relative variant kind.

Rebase.

Mon, Jun 4, 11:40 AM
kzhuravl added a comment to D47601: AMDGPU: Add 64-bit relative variant kind.
In D47601#1121159, @t-tye wrote:
In D47601#1120916, @nhaehnle wrote:
In D47601#1120903, @nhaehnle wrote:

What is this actually needed for? Having a relative relocation in a data segment doesn't seem that useful?

Never mind, I see it in the other commit. Could you please update the table of relocation types in AMDGPUUsage.rst? It has an R_AMDGPU_RELATIVE64, I don't know what that's about, but it doesn't mention R_AMDGPU_REL64.

The AMDGPU_REL64 is defined in AMDGPUUsage as:

'''
Relocation Type Kind Value Field Calculation
R_AMDGPU_REL64 Static 5 word64 S + A - P
'''

It is a static relocation used to create PC relative values between sections. For example, there is a field in the kernel descriptor that has to be the delta from the kernel descriptor to the kernel entry point. So if the kernel descriptor is put in a different section to the code, a static relocation record is need. It will be resolved when the executable is linked.

The R_AMDGPU_RELATIVE64 is a dynamic relocation record implemented by most targets to support having absolute addresses in data sections that need to be initialized to the address of a symbol in the same shared object. For example, a variable initialized to the address of another variable which is considered a compile time constant in languages such as C/C++.

Mon, Jun 4, 10:25 AM

Fri, Jun 1

kzhuravl accepted D47504: [AMDGPU] Simplify memory legalizer.

LGTM

Fri, Jun 1, 8:34 PM

Thu, May 31

kzhuravl added a dependency for D47566: AMDHSA: Code object v3 updates: D47601: AMDGPU: Add 64-bit relative variant kind.
Thu, May 31, 12:31 PM
kzhuravl added a dependent revision for D47601: AMDGPU: Add 64-bit relative variant kind: D47566: AMDHSA: Code object v3 updates.
Thu, May 31, 12:31 PM
kzhuravl created D47601: AMDGPU: Add 64-bit relative variant kind.
Thu, May 31, 12:31 PM
kzhuravl updated the diff for D47566: AMDHSA: Code object v3 updates.
  • This only applies to AMDHSA for now
  • Update tests
Thu, May 31, 12:30 PM

Wed, May 30

kzhuravl added a comment to D47566: AMDHSA: Code object v3 updates.

This change also requires a change in lld, which I plan to post tomorrow.

Wed, May 30, 5:15 PM
kzhuravl added a reviewer for D47566: AMDHSA: Code object v3 updates: nhaehnle.
Wed, May 30, 5:15 PM
kzhuravl created D47566: AMDHSA: Code object v3 updates.
Wed, May 30, 5:06 PM

Tue, May 29

kzhuravl committed rL333451: AMDGPU: Always set COMPUTE_PGM_RSRC2.ENABLE_TRAP_HANDLER to zero for AMDHSA as.
AMDGPU: Always set COMPUTE_PGM_RSRC2.ENABLE_TRAP_HANDLER to zero for AMDHSA as
Tue, May 29, 12:13 PM
kzhuravl closed D47392: AMDGPU: Always set COMPUTE_PGM_RSRC2.ENABLE_TRAP_HANDLER to zero for AMDHSA as it is set by CP.
Tue, May 29, 12:13 PM

Fri, May 25

kzhuravl added a comment to D47392: AMDGPU: Always set COMPUTE_PGM_RSRC2.ENABLE_TRAP_HANDLER to zero for AMDHSA as it is set by CP.

How does mesa handle it?

Fri, May 25, 1:56 PM
kzhuravl created D47392: AMDGPU: Always set COMPUTE_PGM_RSRC2.ENABLE_TRAP_HANDLER to zero for AMDHSA as it is set by CP.
Fri, May 25, 1:56 PM
kzhuravl accepted D47378: [AMDGPU][Waitcnt] Remove obsolete waitcnt option.

LGTM

Fri, May 25, 11:28 AM · Restricted Project
kzhuravl accepted D47361: AMDGPU: Pass function directly instead of MachineFunction.

LGTM

Fri, May 25, 6:13 AM

Tue, May 22

kzhuravl added a comment to D45968: StackSlotColoring: Decide colors per stack ID.

This patch fixes one of the issues with Luxmark I am looking into.

Tue, May 22, 1:24 PM

May 22 2018

kzhuravl accepted D47184: AMDGPU: Fix v2f16 fneg/fabs pattern.

LGTM

May 22 2018, 7:52 AM

May 18 2018

kzhuravl committed rL332753: AMDGPU/NFC: Set symbol's type that is coming from an argument in.
AMDGPU/NFC: Set symbol's type that is coming from an argument in
May 18 2018, 11:45 AM

May 16 2018

kzhuravl committed rL332523: AMDGPU : Recalculate SGPRs when trap handler is supported.
AMDGPU : Recalculate SGPRs when trap handler is supported
May 16 2018, 1:52 PM
kzhuravl closed D29911: AMDGPU : Recalculate SGPRs when trap handler is supported.
May 16 2018, 1:52 PM

May 15 2018

kzhuravl updated the diff for D29911: AMDGPU : Recalculate SGPRs when trap handler is supported.

Rebase + add test

May 15 2018, 2:30 PM
kzhuravl commandeered D29911: AMDGPU : Recalculate SGPRs when trap handler is supported.
May 15 2018, 2:29 PM
kzhuravl committed rL332387: AMDGPU: Fix v_dot{4, 8}* instruction encoding.
AMDGPU: Fix v_dot{4, 8}* instruction encoding
May 15 2018, 12:36 PM
kzhuravl closed D46848: AMDGPU: Fix v_dot{4, 8}* instruction encoding.
May 15 2018, 12:36 PM
kzhuravl added inline comments to D46853: AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm.
May 15 2018, 11:57 AM
kzhuravl committed rL332377: AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm.
AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm
May 15 2018, 10:45 AM
kzhuravl closed D46853: AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm.
May 15 2018, 10:45 AM

May 14 2018

kzhuravl added a dependent revision for D46853: AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm: D46848: AMDGPU: Fix v_dot{4, 8}* instruction encoding.
May 14 2018, 3:39 PM
kzhuravl added a dependency for D46848: AMDGPU: Fix v_dot{4, 8}* instruction encoding: D46853: AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm.
May 14 2018, 3:39 PM
kzhuravl updated the diff for D46848: AMDGPU: Fix v_dot{4, 8}* instruction encoding.

Added disasm tests

May 14 2018, 3:38 PM
kzhuravl created D46853: AMDGPU: Add disasm tests for deep learning instructions + fix v_fmac_f32 disasm.
May 14 2018, 3:38 PM
kzhuravl retitled D46848: AMDGPU: Fix v_dot{4, 8}* instruction encoding from AMDGPU: Fix v_dot_{4, 8}* instruction encoding to AMDGPU: Fix v_dot{4, 8}* instruction encoding.
May 14 2018, 1:46 PM
kzhuravl created D46848: AMDGPU: Fix v_dot{4, 8}* instruction encoding.
May 14 2018, 1:46 PM

May 11 2018

kzhuravl accepted D46781: [AMDGPU] Fix amdgpu-waves-per-eu accounting in scheduler.

LGTM

May 11 2018, 4:50 PM
kzhuravl accepted D46769: [AMDGPU] Change llvm.debugtrap to be a debug breakpoint that can resume execution..

LGTM

May 11 2018, 1:19 PM
kzhuravl accepted D46770: [AMDGPU] Fix compilation failure when IR contains comdat.

LGTM

May 11 2018, 1:19 PM
kzhuravl accepted D46744: AMDGPU: Make undef legal for v2i16/v2f16.

LGTM

May 11 2018, 1:39 AM

May 9 2018

kzhuravl accepted D46644: AMDGPU: Ignore any_extend in mul24 combine.

LGTM

May 9 2018, 9:32 AM

May 8 2018

kzhuravl accepted D46587: [AMDGPU] Provide machine -> name mapping.

LGTM

May 8 2018, 10:11 AM

May 7 2018

kzhuravl updated subscribers of D46366: AMDGPU: Add D16 instructions preserve unused bits feature.

Yes.

May 7 2018, 11:10 AM

May 4 2018

kzhuravl committed rL331564: AMDGPU/NFC: Update D16PreservesUnusedBits description based Tony Tye's comments.
AMDGPU/NFC: Update D16PreservesUnusedBits description based Tony Tye's comments
May 4 2018, 3:57 PM
kzhuravl committed rL331553: AMDGPU/NFC: Fix formatting for 900, 902 ISA Version features.
AMDGPU/NFC: Fix formatting for 900, 902 ISA Version features
May 4 2018, 1:25 PM
kzhuravl committed rL331551: AMDGPU: Add D16 instructions preserve unused bits feature.
AMDGPU: Add D16 instructions preserve unused bits feature
May 4 2018, 1:11 PM
kzhuravl closed D46366: AMDGPU: Add D16 instructions preserve unused bits feature.
May 4 2018, 1:11 PM

May 3 2018

kzhuravl added inline comments to D46366: AMDGPU: Add D16 instructions preserve unused bits feature.
May 3 2018, 8:39 PM
kzhuravl updated the diff for D46366: AMDGPU: Add D16 instructions preserve unused bits feature.

Address review feedback

May 3 2018, 8:38 PM

May 2 2018

kzhuravl created D46366: AMDGPU: Add D16 instructions preserve unused bits feature.
May 2 2018, 2:51 PM
kzhuravl added inline comments to D46365: AMDGPU: Separate R600 and GCN TableGen files.
May 2 2018, 1:23 PM

May 1 2018

kzhuravl committed rL331298: AMDGPU: Remove remnants of gfx901 (it was deprecated some time ago).
AMDGPU: Remove remnants of gfx901 (it was deprecated some time ago)
May 1 2018, 11:51 AM
kzhuravl committed rL331284: AMDGPU: Add missing gfx904 tests.
AMDGPU: Add missing gfx904 tests
May 1 2018, 10:09 AM
kzhuravl committed rL331278: AMDGPU/NFC: Use enum values for first/last machs instead of numbers.
AMDGPU/NFC: Use enum values for first/last machs instead of numbers
May 1 2018, 9:38 AM

Apr 25 2018

kzhuravl accepted D46067: [AMDGPU][Waitcnt] Take ISA target into account for s_waitcnt expcnt instr generation.

LGTM

Apr 25 2018, 9:09 AM · Restricted Project

Apr 24 2018

kzhuravl accepted D45987: [AMDGPU] Truncate packed inline constant.

LGTM.

Apr 24 2018, 10:36 AM

Apr 19 2018

kzhuravl accepted D45790: [AMDGPU] Use packed literals with zero either lower or hi part.

LGTM

Apr 19 2018, 1:56 PM

Apr 17 2018

kzhuravl accepted D45745: [AMDGPU] Enabled v2.16 literals for VOP3P.

LGTM, thanks.

Apr 17 2018, 4:11 PM
kzhuravl accepted D43831: [AMDGPU] Do not only rely on BB number when finding bottom loop.

LGTM.

Apr 17 2018, 12:51 PM

Apr 12 2018

kzhuravl accepted D45609: [AMDGPU] Add gfx902 product names.

LGTM

Apr 12 2018, 10:05 PM
kzhuravl accepted D45587: [AMDGPU] Update relocation record description..

LGTM

Apr 12 2018, 5:34 PM

Apr 9 2018

kzhuravl committed rL329620: AMDGPU: Remove max_scratch_backing_memory_byte_size from kernel header.
AMDGPU: Remove max_scratch_backing_memory_byte_size from kernel header
Apr 9 2018, 1:50 PM
kzhuravl closed D45452: AMDGPU: Remove max_scratch_backing_memory_byte_size from kernel header.
Apr 9 2018, 1:50 PM
kzhuravl added a comment to D45452: AMDGPU: Remove max_scratch_backing_memory_byte_size from kernel header.
In D45452#1062203, @artem.tamazov wrote:

Is it so that max_scratch_backing_memory_byte_size is not needed at all for Gfx7-9?

Apr 9 2018, 1:30 PM
kzhuravl created D45452: AMDGPU: Remove max_scratch_backing_memory_byte_size from kernel header.
Apr 9 2018, 12:29 PM

Apr 5 2018

kzhuravl committed rL329341: AMDGPU/Metadata: Always report a fixed number of hidden arguments.
AMDGPU/Metadata: Always report a fixed number of hidden arguments
Apr 5 2018, 1:50 PM
kzhuravl closed D45129: AMDGPU/Metadata: Always report a fixed number of hidden arguments.
Apr 5 2018, 1:49 PM
kzhuravl committed rL329325: llvm-exegesis: Fix unittests include dirs when llvm is a part of another project.
llvm-exegesis: Fix unittests include dirs when llvm is a part of another project
Apr 5 2018, 11:19 AM
kzhuravl closed D45328: llvm-exegesis: Fix unittests include dirs when llvm is a part of another project.
Apr 5 2018, 11:19 AM
kzhuravl added a comment to D44519: Add llvm-exegesis tool..

Hi, I am failing to build llvm-exegesis' unit tests when llvm is a part of another project. I have posted a fix here: D45328. Can you take a look? Thanks.

Apr 5 2018, 10:21 AM