Page MenuHomePhabricator

ezhulenev (Eugene Zhulenev)
User

Projects

User does not belong to any projects.

User Details

User Since
Mar 6 2020, 7:42 PM (121 w, 3 d)

Recent Activity

Sat, Jul 2

ezhulenev added inline comments to D129035: Add MLIR interpreter framework.
Sat, Jul 2, 12:26 PM · Restricted Project, Restricted Project
ezhulenev added inline comments to D129035: Add MLIR interpreter framework.
Sat, Jul 2, 8:23 AM · Restricted Project, Restricted Project
ezhulenev added inline comments to D129035: Add MLIR interpreter framework.
Sat, Jul 2, 5:37 AM · Restricted Project, Restricted Project

Fri, Jun 10

ezhulenev accepted D127471: [Coroutines] Convert coroutine.presplit to enum attr.

Yes, that's the correct MLIR way of passing attributes to LLVM.

Fri, Jun 10, 12:05 PM · Restricted Project, Restricted Project, Restricted Project, Restricted Project

May 20 2022

ezhulenev committed rG705f048cbbfd: [mlir] MemRefToLLVM: convert memref.view operations for empty memrefs (authored by ezhulenev).
[mlir] MemRefToLLVM: convert memref.view operations for empty memrefs
May 20 2022, 4:44 PM · Restricted Project, Restricted Project
ezhulenev closed D126094: [mlir] MemRefToLLVM: convert memref.view operations for empty memrefs.
May 20 2022, 4:44 PM · Restricted Project, Restricted Project
ezhulenev updated the diff for D126094: [mlir] MemRefToLLVM: convert memref.view operations for empty memrefs.

Fix empty strides assertion error

May 20 2022, 2:59 PM · Restricted Project, Restricted Project
ezhulenev added a reviewer for D126094: [mlir] MemRefToLLVM: convert memref.view operations for empty memrefs: mehdi_amini.
May 20 2022, 2:46 PM · Restricted Project, Restricted Project
ezhulenev requested review of D126094: [mlir] MemRefToLLVM: convert memref.view operations for empty memrefs.
May 20 2022, 2:45 PM · Restricted Project, Restricted Project

May 2 2022

ezhulenev committed rG38d0df557706: [mlir] CRunnerUtils: qualify UnrankedMemRefType to avoid collisions with mlir… (authored by ezhulenev).
[mlir] CRunnerUtils: qualify UnrankedMemRefType to avoid collisions with mlir…
May 2 2022, 10:19 AM · Restricted Project, Restricted Project
ezhulenev closed D124744: [mlir] CRunnerUtils: qualify UnrankedMemRefType to avoid collisions with mlir::UnrankedMemRefType.
May 2 2022, 10:19 AM · Restricted Project, Restricted Project

May 1 2022

ezhulenev updated the summary of D124744: [mlir] CRunnerUtils: qualify UnrankedMemRefType to avoid collisions with mlir::UnrankedMemRefType.
May 1 2022, 1:30 PM · Restricted Project, Restricted Project
ezhulenev requested review of D124744: [mlir] CRunnerUtils: qualify UnrankedMemRefType to avoid collisions with mlir::UnrankedMemRefType.
May 1 2022, 1:29 PM · Restricted Project, Restricted Project

Apr 11 2022

ezhulenev committed rGb35b9e307fdd: [mlir] Add msan memory unpoisoning macros to mlir ExecutionEngine (authored by ezhulenev).
[mlir] Add msan memory unpoisoning macros to mlir ExecutionEngine
Apr 11 2022, 6:59 PM · Restricted Project, Restricted Project
ezhulenev closed D123557: [mlir] Add msan memory unpoisoning macros to mlir ExecutionEngine.
Apr 11 2022, 6:58 PM · Restricted Project, Restricted Project
ezhulenev updated the summary of D123557: [mlir] Add msan memory unpoisoning macros to mlir ExecutionEngine.
Apr 11 2022, 5:30 PM · Restricted Project, Restricted Project
ezhulenev requested review of D123557: [mlir] Add msan memory unpoisoning macros to mlir ExecutionEngine.
Apr 11 2022, 5:29 PM · Restricted Project, Restricted Project

Feb 23 2022

ezhulenev accepted D120415: [mlir] Add sectionMemoryMapper to ExecutionEngineOptions.
Feb 23 2022, 11:20 AM · Restricted Project
ezhulenev accepted D120415: [mlir] Add sectionMemoryMapper to ExecutionEngineOptions.
Feb 23 2022, 10:45 AM · Restricted Project
ezhulenev committed rGbeff16f7bd63: [mlir] Async: update condition for dispatching block-aligned compute function (authored by ezhulenev).
[mlir] Async: update condition for dispatching block-aligned compute function
Feb 23 2022, 10:30 AM
ezhulenev closed D120075: [mlir] Async: update condition for dispatching block-aligned compute function.
Feb 23 2022, 10:30 AM · Restricted Project

Feb 17 2022

ezhulenev abandoned D116440: [mlir] Async: update condition for dispatching block-aligned compute function.
Feb 17 2022, 11:19 AM · Restricted Project
ezhulenev updated the summary of D120075: [mlir] Async: update condition for dispatching block-aligned compute function.
Feb 17 2022, 10:24 AM · Restricted Project
ezhulenev requested review of D120075: [mlir] Async: update condition for dispatching block-aligned compute function.
Feb 17 2022, 10:22 AM · Restricted Project
ezhulenev removed reviewers for D114496: Implement a cost model to drive the lowering of scf.parallel.: ezhulenev, nicolasvasilache.
Feb 17 2022, 10:08 AM · Restricted Project
ezhulenev abandoned D119976: [mlir] Async: add alwaysinline/noinline attributes to parallel-for functions.
Feb 17 2022, 10:07 AM · Restricted Project

Feb 16 2022

ezhulenev committed rGabe2dee5ebb9: [mlir] NFC Async: always use 'b' for the current builder (authored by ezhulenev).
[mlir] NFC Async: always use 'b' for the current builder
Feb 16 2022, 9:21 PM
ezhulenev closed D120003: [mlir] NFC Async: always use 'b' for the current builder.
Feb 16 2022, 9:21 PM · Restricted Project
ezhulenev retitled D120003: [mlir] NFC Async: always use 'b' for the current builder from [mlir] Async: always use 'b' for the current builder to [mlir] NFC Async: always use 'b' for the current builder.
Feb 16 2022, 8:37 PM · Restricted Project
ezhulenev updated the summary of D120003: [mlir] NFC Async: always use 'b' for the current builder.
Feb 16 2022, 8:37 PM · Restricted Project
ezhulenev requested review of D120003: [mlir] NFC Async: always use 'b' for the current builder.
Feb 16 2022, 8:35 PM · Restricted Project
ezhulenev committed rGb171583ae7d1: [mlir] Async: create async.group inside the scf.if branch (authored by ezhulenev).
[mlir] Async: create async.group inside the scf.if branch
Feb 16 2022, 2:47 PM
ezhulenev closed D119959: [mlir] Async: create async.group inside the scf.if branch.
Feb 16 2022, 2:47 PM · Restricted Project
ezhulenev retitled D119976: [mlir] Async: add alwaysinline/noinline attributes to parallel-for functions from [mlir] Async: add alwaysinline/noinlile attributes to parallel-for functions to [mlir] Async: add alwaysinline/noinline attributes to parallel-for functions.
Feb 16 2022, 1:49 PM · Restricted Project
ezhulenev requested review of D119976: [mlir] Async: add alwaysinline/noinline attributes to parallel-for functions.
Feb 16 2022, 1:48 PM · Restricted Project
ezhulenev added a reviewer for D119959: [mlir] Async: create async.group inside the scf.if branch: cota.
Feb 16 2022, 10:40 AM · Restricted Project
ezhulenev requested review of D119959: [mlir] Async: create async.group inside the scf.if branch.
Feb 16 2022, 10:40 AM · Restricted Project

Feb 8 2022

ezhulenev accepted D118968: [mlir][math] Expand coverage of atan2 expansion.
Feb 8 2022, 8:56 AM · Restricted Project

Feb 6 2022

ezhulenev committed rGedca177cbeb6: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops (authored by ezhulenev).
[mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops
Feb 6 2022, 2:47 PM
ezhulenev closed D119043: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops.
Feb 6 2022, 2:46 PM · Restricted Project
ezhulenev updated the diff for D119043: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops.

Address comments

Feb 6 2022, 2:38 PM · Restricted Project
ezhulenev added inline comments to D119043: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops.
Feb 6 2022, 2:38 PM · Restricted Project

Feb 4 2022

ezhulenev updated the diff for D119043: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops.

Rebased

Feb 4 2022, 3:32 PM · Restricted Project
ezhulenev committed rG981f0a14f1de: [mlir] Add canonicalizer to merge shape.assuming_all ops (authored by ezhulenev).
[mlir] Add canonicalizer to merge shape.assuming_all ops
Feb 4 2022, 3:28 PM
ezhulenev closed D119025: [mlir] Add canonicalizer to merge shape.assuming_all ops.
Feb 4 2022, 3:27 PM · Restricted Project
ezhulenev added a reviewer for D119043: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops: frgossen.
Feb 4 2022, 3:18 PM · Restricted Project
ezhulenev requested review of D119043: [mlir] Add canonicalizer to remove redundant shape.cstr_broadcastable ops.
Feb 4 2022, 3:18 PM · Restricted Project
ezhulenev updated the diff for D119025: [mlir] Add canonicalizer to merge shape.assuming_all ops.

Rebased

Feb 4 2022, 11:35 AM · Restricted Project
ezhulenev committed rG2cff9ee46bff: [mlir] Use SetVector to deduplicate shape ops operands (authored by ezhulenev).
[mlir] Use SetVector to deduplicate shape ops operands
Feb 4 2022, 11:34 AM
ezhulenev closed D119021: [mlir] Use SetVector to deduplicate shape ops operands.
Feb 4 2022, 11:34 AM · Restricted Project
ezhulenev added a reviewer for D119025: [mlir] Add canonicalizer to merge shape.assuming_all ops: frgossen.
Feb 4 2022, 11:32 AM · Restricted Project
ezhulenev requested review of D119025: [mlir] Add canonicalizer to merge shape.assuming_all ops.
Feb 4 2022, 11:32 AM · Restricted Project
ezhulenev added a reviewer for D119021: [mlir] Use SetVector to deduplicate shape ops operands: frgossen.
Feb 4 2022, 10:21 AM · Restricted Project
ezhulenev requested review of D119021: [mlir] Use SetVector to deduplicate shape ops operands.
Feb 4 2022, 10:21 AM · Restricted Project
ezhulenev added a comment to D118968: [mlir][math] Expand coverage of atan2 expansion.

I think it's ok as a fallback for f16/f32, but maybe @mehdi_amini has a different opinion.

Feb 4 2022, 9:56 AM · Restricted Project

Feb 3 2022

ezhulenev committed rG0557c6a7970d: [mlir] Keep sorted vector of registered operation names for efficient lookup (authored by ezhulenev).
[mlir] Keep sorted vector of registered operation names for efficient lookup
Feb 3 2022, 2:20 PM
ezhulenev closed D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.
Feb 3 2022, 2:19 PM · Restricted Project
ezhulenev updated the diff for D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.

Upd

Feb 3 2022, 1:16 PM · Restricted Project
ezhulenev added a comment to D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.

It will lead to slightly slower startup times, but we can always switch to deferred sorting if that is a problem. (It isn't clear how much it matters in practice anyways).

Feb 3 2022, 1:16 PM · Restricted Project
ezhulenev added inline comments to D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.
Feb 3 2022, 1:03 PM · Restricted Project
ezhulenev added a comment to D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.

Do you have an idea on the cost of this vs. say computing on demand?

Feb 3 2022, 12:59 PM · Restricted Project
ezhulenev updated the summary of D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.
Feb 3 2022, 12:13 PM · Restricted Project
ezhulenev requested review of D118937: [mlir] Keep sorted vector of registered operation names for efficient lookup.
Feb 3 2022, 12:11 PM · Restricted Project

Jan 31 2022

ezhulenev committed rG149311b4055a: [async] Get the number of worker threads from the runtime. (authored by bakhtiyar <bakhtiyar@x.team>).
[async] Get the number of worker threads from the runtime.
Jan 31 2022, 12:06 PM
ezhulenev closed D117751: [async] Get the number of worker threads from the runtime..
Jan 31 2022, 12:06 PM · Restricted Project
ezhulenev accepted D117751: [async] Get the number of worker threads from the runtime..
Jan 31 2022, 12:05 PM · Restricted Project

Jan 26 2022

ezhulenev added a comment to D117751: [async] Get the number of worker threads from the runtime..

I get this error locally:

Jan 26 2022, 12:18 PM · Restricted Project
ezhulenev accepted D117751: [async] Get the number of worker threads from the runtime..
Jan 26 2022, 11:54 AM · Restricted Project

Jan 21 2022

ezhulenev added a comment to D117751: [async] Get the number of worker threads from the runtime..

Can you please also add tests cases to tests (benchmarks?) in this folder: https://github.com/llvm/llvm-project/tree/main/mlir/test/Integration/Dialect/Async/CPU

Jan 21 2022, 7:52 AM · Restricted Project
ezhulenev added inline comments to D117751: [async] Get the number of worker threads from the runtime..
Jan 21 2022, 6:24 AM · Restricted Project

Jan 20 2022

ezhulenev added inline comments to D117751: [async] Get the number of worker threads from the runtime..
Jan 20 2022, 12:37 PM · Restricted Project
ezhulenev added inline comments to D117751: [async] Get the number of worker threads from the runtime..
Jan 20 2022, 12:20 PM · Restricted Project
ezhulenev added inline comments to D117751: [async] Get the number of worker threads from the runtime..
Jan 20 2022, 12:16 PM · Restricted Project
ezhulenev added inline comments to D117751: [async] Get the number of worker threads from the runtime..
Jan 20 2022, 12:15 PM · Restricted Project

Jan 18 2022

ezhulenev accepted D117542: [Coroutines] Offering llvm.coro.align intrinsic.

thank you! For our MLIR use case this is exactly what we need.

Jan 18 2022, 3:49 AM · Restricted Project

Jan 17 2022

ezhulenev accepted D117501: [AsyncToLLVM] Align frames to 64 bytes.
Jan 17 2022, 9:47 AM · Restricted Project

Jan 14 2022

ezhulenev committed rG69bc334be569: [mlir] Remove getNumberOfExecutions from RegionBranchOpInterface (authored by ezhulenev).
[mlir] Remove getNumberOfExecutions from RegionBranchOpInterface
Jan 14 2022, 1:15 PM
ezhulenev closed D117347: [mlir] Remove getNumberOfExecutions from RegionBranchOpInterface.
Jan 14 2022, 1:15 PM · Restricted Project
ezhulenev updated the summary of D117347: [mlir] Remove getNumberOfExecutions from RegionBranchOpInterface.
Jan 14 2022, 11:11 AM · Restricted Project
ezhulenev requested review of D117347: [mlir] Remove getNumberOfExecutions from RegionBranchOpInterface.
Jan 14 2022, 11:08 AM · Restricted Project

Jan 13 2022

ezhulenev committed rG764e52f0d4b6: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map… (authored by ezhulenev).
[DebugInfo][InstrRef] Short-circuit unnecessary preferred location map…
Jan 13 2022, 6:25 AM
ezhulenev closed D117162: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map construction.
Jan 13 2022, 6:25 AM · Restricted Project
ezhulenev updated the diff for D117162: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map construction.

Rebase

Jan 13 2022, 3:35 AM · Restricted Project
ezhulenev added a reviewer for D117162: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map construction: cota.
Jan 13 2022, 2:44 AM · Restricted Project
ezhulenev updated the diff for D117162: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map construction.

Fix bugs

Jan 13 2022, 2:20 AM · Restricted Project

Jan 12 2022

ezhulenev added a reviewer for D117162: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map construction: mehdi_amini.
Jan 12 2022, 2:45 PM · Restricted Project
ezhulenev requested review of D117162: [DebugInfo][InstrRef] Short-circuit unnecessary preferred location map construction.
Jan 12 2022, 2:44 PM · Restricted Project

Dec 31 2021

ezhulenev updated the summary of D116440: [mlir] Async: update condition for dispatching block-aligned compute function.
Dec 31 2021, 5:12 AM · Restricted Project
ezhulenev requested review of D116440: [mlir] Async: update condition for dispatching block-aligned compute function.
Dec 31 2021, 5:10 AM · Restricted Project

Dec 24 2021

ezhulenev accepted D115790: [Coroutines] Set presplit attribute in Clang and mlir.
Dec 24 2021, 8:32 AM · Restricted Project, Restricted Project, Restricted Project

Dec 19 2021

ezhulenev committed rGec0e4545caa1: Make AsyncParallelForRewrite parameterizable with a cost model which drives… (authored by bakhtiyar <bakhtiyar@x.team>).
Make AsyncParallelForRewrite parameterizable with a cost model which drives…
Dec 19 2021, 8:41 AM
ezhulenev closed D115423: Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity..
Dec 19 2021, 8:41 AM · Restricted Project

Dec 17 2021

ezhulenev updated subscribers of D115790: [Coroutines] Set presplit attribute in Clang and mlir.

There are two places where in MLIR you can put an attribute to coroutine functions:

Dec 17 2021, 7:25 AM · Restricted Project, Restricted Project, Restricted Project

Dec 15 2021

ezhulenev added inline comments to D115423: Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity..
Dec 15 2021, 2:14 AM · Restricted Project
ezhulenev added a comment to D115423: Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity..

Build is failing: https://buildkite.com/llvm-project/premerge-checks/builds/70123#b97ba1d4-10ba-4287-899a-0bcf5260aefe

Dec 15 2021, 2:10 AM · Restricted Project
ezhulenev accepted D115423: Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity..
Dec 15 2021, 2:05 AM · Restricted Project

Dec 14 2021

ezhulenev added inline comments to D115423: Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity..
Dec 14 2021, 4:28 AM · Restricted Project

Dec 9 2021

ezhulenev accepted D115423: Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity..
Dec 9 2021, 3:19 PM · Restricted Project
ezhulenev committed rG49ce40e9ab25: [mlir] AsyncParallelFor: align block size to be a multiple of inner loops… (authored by ezhulenev).
[mlir] AsyncParallelFor: align block size to be a multiple of inner loops…
Dec 9 2021, 6:51 AM
ezhulenev closed D115436: [mlir] AsyncParallelFor: align block size to be a multiple of inner loops iterations.
Dec 9 2021, 6:51 AM · Restricted Project