Page MenuHomePhabricator

Please use GitHub pull requests for new patches. Phabricator shutdown timeline

krzysz00 (Krzysztof Drewniak)
User

Projects

User does not belong to any projects.

User Details

User Since
May 3 2021, 10:02 AM (124 w, 6 d)

Recent Activity

Today

krzysz00 added a comment to D157228: Add Lowerings for GPU WMMA F16/F32 ops to ROCDL dialect.

I think this is headed in a good direction, I just have some thoughts, mostly about how the pass options were set up

Sun, Sep 24, 9:49 AM · Restricted Project, Restricted Project

Fri, Sep 22

krzysz00 updated the diff for D152457: [mlir][AMDGPU] Add packed 8-bit float conversion ops and lowering.

Address naming comment from review

Fri, Sep 22, 1:54 PM · Restricted Project, Restricted Project
krzysz00 committed rGbfa501b8927a: [mlir][AMDGPU] Move to new buffer resource intrinsics (authored by krzysz00).
[mlir][AMDGPU] Move to new buffer resource intrinsics
Fri, Sep 22, 12:48 PM · Restricted Project, Restricted Project
krzysz00 closed D157053: [mlir][AMDGPU] Move to new buffer resource intrinsics.
Fri, Sep 22, 12:48 PM · Restricted Project, Restricted Project

Tue, Sep 19

krzysz00 added inline comments to D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.
Tue, Sep 19, 2:39 PM · Restricted Project, Restricted Project
krzysz00 updated the diff for D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.

Move to i32 for offsets, fix function handling

Tue, Sep 19, 2:38 PM · Restricted Project, Restricted Project

Wed, Sep 13

krzysz00 committed rGdf852599f3a1: [mlir] Split up VectorToLLVM pass (authored by krzysz00).
[mlir] Split up VectorToLLVM pass
Wed, Sep 13, 9:10 AM · Restricted Project, Restricted Project
krzysz00 closed D158287: [mlir] Split up VectorToLLVM pass.
Wed, Sep 13, 9:10 AM · Restricted Project, Restricted Project
krzysz00 committed rGba6d7a0f25f3: [mlir][AMDGPU] Add gfx941 to buffer atomics emulation (authored by krzysz00).
[mlir][AMDGPU] Add gfx941 to buffer atomics emulation
Wed, Sep 13, 9:07 AM · Restricted Project, Restricted Project
krzysz00 closed D152299: [mlir][AMDGPU] Add gfx941 to buffer atomics emulation.
Wed, Sep 13, 9:07 AM · Restricted Project, Restricted Project

Tue, Sep 12

krzysz00 updated the diff for D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.

Rebase and update tests

Tue, Sep 12, 10:34 AM · Restricted Project, Restricted Project

Mon, Sep 11

krzysz00 updated the diff for D158287: [mlir] Split up VectorToLLVM pass.

Fix NVVM tests

Mon, Sep 11, 8:26 AM · Restricted Project, Restricted Project
krzysz00 committed rGfbf67bfaf0e0: [mlir][GPU] Handle LLVM pointer attributes on memref arguments. (authored by stefankoncarevic).
[mlir][GPU] Handle LLVM pointer attributes on memref arguments.
Mon, Sep 11, 8:11 AM · Restricted Project, Restricted Project
krzysz00 closed D157082: [mlir][GPU] Handle LLVM pointer attributes on memref arguments..
Mon, Sep 11, 8:11 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D157228: Add Lowerings for GPU WMMA F16/F32 ops to ROCDL dialect.

Just wanted to check on what you're planning here

Mon, Sep 11, 8:00 AM · Restricted Project, Restricted Project
krzysz00 accepted D157082: [mlir][GPU] Handle LLVM pointer attributes on memref arguments..

Approving to move this along

Mon, Sep 11, 7:53 AM · Restricted Project, Restricted Project

Fri, Sep 8

krzysz00 added a reviewer for D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7: piotr.
Fri, Sep 8, 3:15 PM · Restricted Project, Restricted Project
krzysz00 updated the diff for D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.

Move after codegenprepare, rebase, fix lack of an or constant

Fri, Sep 8, 3:14 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D152299: [mlir][AMDGPU] Add gfx941 to buffer atomics emulation.

Ping

Fri, Sep 8, 12:56 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D157053: [mlir][AMDGPU] Move to new buffer resource intrinsics.

Ping

Fri, Sep 8, 12:42 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D158287: [mlir] Split up VectorToLLVM pass.

Ping on what needs to be done here

Fri, Sep 8, 12:42 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D156362: [mlir][Arith] Change F32 to BF16 truncation to match __truncsfbf2.

Does anyone know where that code in ExecutionEngine came from, then? Because perhaps that needs to be corrected as well

Fri, Sep 8, 12:41 PM · Restricted Project, Restricted Project

Tue, Aug 29

krzysz00 committed rG1c5fd1534cfd: [mlir][Tosa] Allow non-fp32 tosa.cast to integers (authored by krzysz00).
[mlir][Tosa] Allow non-fp32 tosa.cast to integers
Tue, Aug 29, 7:11 AM · Restricted Project, Restricted Project
krzysz00 closed D158738: [mlir][Tosa] Allow non-fp32 tosa.cast to integers.
Tue, Aug 29, 7:11 AM · Restricted Project, Restricted Project

Aug 25 2023

krzysz00 added inline comments to D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.
Aug 25 2023, 7:22 AM · Restricted Project, Restricted Project

Aug 24 2023

krzysz00 updated the diff for D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.

Use data layout widths instead of hardcoded pointer widths

Aug 24 2023, 10:09 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D157183: Add a generic "convert-to-llvm" pass delegating to an interface.
  1. I'm asking in relation to https://discourse.llvm.org/t/rfc-almost-all-uses-of-typeconverter-should-be-const/72689/9 and the existence of the need to create target-specific overrides of the LLVM type conversion (for instance, AMDGPU lowers bf16 to i16 to match the expectations of the backend). I'm not sure the new pass has much to do with this
Aug 24 2023, 8:24 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D156362: [mlir][Arith] Change F32 to BF16 truncation to match __truncsfbf2.
  1. I couldn't find this discussion. @rsuderman, do you have pointers?
Aug 24 2023, 8:01 AM · Restricted Project, Restricted Project
krzysz00 added reviewers for D158738: [mlir][Tosa] Allow non-fp32 tosa.cast to integers: rafaelubalmw, eric-k256, jpienaar.
Aug 24 2023, 7:44 AM · Restricted Project, Restricted Project
krzysz00 requested review of D158738: [mlir][Tosa] Allow non-fp32 tosa.cast to integers.
Aug 24 2023, 7:34 AM · Restricted Project, Restricted Project

Aug 22 2023

krzysz00 added a comment to D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.

@nlopes

  • Two of the UndefValue uses are in splitting up an existing undef
  • One is respecting "When an Instruction is deleted, its debug uses change to undef. " from the documentation on how to update debug info for pass authors
  • The last has been changed to poison
Aug 22 2023, 7:54 AM · Restricted Project, Restricted Project
krzysz00 updated the diff for D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.

Swap one of the undefs for poison

Aug 22 2023, 7:49 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D157183: Add a generic "convert-to-llvm" pass delegating to an interface.

So, I have some minor thoughts here:

  1. It sounds like this'll help with target-specific overrides of type conversion, am I right here?
  2. Similarly, it'd be nice if this general pass would take options like triple and chip so you could look up the LLVM data layout etc., since that's logic that'd be a pain to set up everywhere
  3. One interesting idea would be to generalize D158287 . I'd like, for example, the notion of "target-specific patterns". That way, for example, when we have these vector-to-llvm patterns, the avx patterns would live in a separate library and be pulled in only if I'm targetting x86. This would allow me to not even build the target-specific dialects (for instance, why does a library targetting AMDGPU need ArmSVE?) and to have that be a general pattern. (As another example, I'm having to pull in NVVM because of some memref patterns.)
Aug 22 2023, 7:46 AM · Restricted Project, Restricted Project

Aug 21 2023

krzysz00 requested review of D158463: [AMDGPU] Add IR-level pass to rewrite away address space 7.
Aug 21 2023, 2:59 PM · Restricted Project, Restricted Project
krzysz00 committed rG7db18533b99d: [mlir][MemRefToLLVM] Add fmin, fmax to AtomicRMW lowering (authored by krzysz00).
[mlir][MemRefToLLVM] Add fmin, fmax to AtomicRMW lowering
Aug 21 2023, 2:21 PM · Restricted Project, Restricted Project
krzysz00 closed D158283: [mlir][MemRefToLLVM] Add fmin, fmax to AtomicRMW lowering.
Aug 21 2023, 2:21 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D158287: [mlir] Split up VectorToLLVM pass.

@mehdi_amini What's this new ConvertToLLVM pass?

Aug 21 2023, 8:27 AM · Restricted Project, Restricted Project

Aug 18 2023

krzysz00 added a reviewer for D158287: [mlir] Split up VectorToLLVM pass: mehdi_amini.
Aug 18 2023, 8:38 AM · Restricted Project, Restricted Project
krzysz00 requested review of D158287: [mlir] Split up VectorToLLVM pass.
Aug 18 2023, 8:38 AM · Restricted Project, Restricted Project
krzysz00 requested review of D158283: [mlir][MemRefToLLVM] Add fmin, fmax to AtomicRMW lowering.
Aug 18 2023, 7:57 AM · Restricted Project, Restricted Project

Aug 17 2023

krzysz00 added a reviewer for D152457: [mlir][AMDGPU] Add packed 8-bit float conversion ops and lowering: rsuderman.
Aug 17 2023, 11:32 AM · Restricted Project, Restricted Project
krzysz00 added a reviewer for D152299: [mlir][AMDGPU] Add gfx941 to buffer atomics emulation: rsuderman.
Aug 17 2023, 11:32 AM · Restricted Project, Restricted Project
krzysz00 committed rG51b65d0895af: [mlir][AMDGPU] Improve BF16 handling through AMDGPU compilation (authored by krzysz00).
[mlir][AMDGPU] Improve BF16 handling through AMDGPU compilation
Aug 17 2023, 11:31 AM · Restricted Project, Restricted Project
krzysz00 closed D156361: [mlir][AMDGPU] Improve BF16 handling through AMDGPU compilation.
Aug 17 2023, 11:31 AM · Restricted Project, Restricted Project

Aug 14 2023

krzysz00 updated subscribers of D157228: Add Lowerings for GPU WMMA F16/F32 ops to ROCDL dialect.

The reason I want to lower the matrix ops from GPU to AMDGPU + memref/vector + ... instead of to ROCDL and LLVM directory is separation of concerns and progressive lowering.

Aug 14 2023, 7:18 AM · Restricted Project, Restricted Project

Aug 11 2023

krzysz00 accepted D154129: [mlir][ROCDL] Adds the ROCDL target attribute..

Works for me

Aug 11 2023, 10:56 AM · Restricted Project, Restricted Project

Aug 10 2023

Herald added a reviewer for D126444: [mlir]Implement SoftwareBF16 to handle the bf16 type: nicolasvasilache.

Since we got rid of this pass downstream, we should abandon this

Aug 10 2023, 7:43 AM · Restricted Project, Restricted Project
krzysz00 added a reverting change for rG7c349c369847: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates: rG3fe8ec7906a8: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates".
Aug 10 2023, 7:13 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 committed rG3fe8ec7906a8: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates" (authored by krzysz00).
Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates"
Aug 10 2023, 7:13 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added a reverting change for D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates: rG3fe8ec7906a8: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates".
Aug 10 2023, 7:13 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 closed D157543: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates".
Aug 10 2023, 7:13 AM · Restricted Project, Restricted Project, Restricted Project

Aug 9 2023

krzysz00 added a comment to D154129: [mlir][ROCDL] Adds the ROCDL target attribute..

So, this code looks reasonable to me (and, as a bonus, I can tell how to invoke this when I'm generating binaries as a library), but I don't want to hand out a green check without @arsenm or someone else who's more involved in the intricacies of how LLVM gets called for GPU compilations signing off, unless we can't get a hold of anyone.

Aug 9 2023, 12:43 PM · Restricted Project, Restricted Project
krzysz00 added a reverting change for rG7c349c369847: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates: D157543: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates".
Aug 9 2023, 12:34 PM · Restricted Project, Restricted Project, Restricted Project
krzysz00 requested review of D157543: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates".
Aug 9 2023, 12:34 PM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added a reverting change for D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates: D157543: Revert "[mlir][Arith] Add arith.is_nan and arith.is_inf predicates".
Aug 9 2023, 12:34 PM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added a comment to rG7c349c369847: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.

@arsenm I'd be fine lowering is_nan and is_inf to those canonical forms when lowering to LLVM.

Aug 9 2023, 12:07 PM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added a comment to D157228: Add Lowerings for GPU WMMA F16/F32 ops to ROCDL dialect.

Re 1: The motivation is that, while the rocdl dialect provides direct access to the LLVM intrinsics, the amdgpu dialect provides a wrapper around those intrinsics that is more MLIR-flavored. This means that there is only one place in the code that needs to know how the exact details of WMMA, any relevant type mangling (ex. the fact that i8 inputs should be turned into i64s).

Aug 9 2023, 8:22 AM · Restricted Project, Restricted Project

Aug 6 2023

krzysz00 added a comment to D157228: Add Lowerings for GPU WMMA F16/F32 ops to ROCDL dialect.

(amdgpu.wmma is at https://github.com/llvm/llvm-project/blob/main/mlir/include/mlir/Dialect/AMDGPU/IR/AMDGPU.td#L445 )

Aug 6 2023, 7:54 AM · Restricted Project, Restricted Project
krzysz00 requested changes to D157228: Add Lowerings for GPU WMMA F16/F32 ops to ROCDL dialect.
  1. In order to maintain good abstractions, could you refactor this into a GPUToAMDGPU pass instead, targetting the amdgpu.wmma operation, which will then be lowered to rocdl? This should make the code less fragile and mean that we only have one place to change if we need to adjust the compiler intrinsics in the future.
  2. https://reviews.llvm.org/D154666 defines a lowering for lane_id, please use it
  3. Minor comments on test organization
Aug 6 2023, 7:52 AM · Restricted Project, Restricted Project

Aug 3 2023

krzysz00 added reviewers for D157053: [mlir][AMDGPU] Move to new buffer resource intrinsics: arsenm, foad.
Aug 3 2023, 4:05 PM · Restricted Project, Restricted Project
krzysz00 requested review of D157053: [mlir][AMDGPU] Move to new buffer resource intrinsics.
Aug 3 2023, 4:04 PM · Restricted Project, Restricted Project

Aug 2 2023

krzysz00 committed rG2f8d8c78a89e: Generalize finding the rocm_agent_enumberator program (authored by krzysz00).
Generalize finding the rocm_agent_enumberator program
Aug 2 2023, 1:36 PM · Restricted Project, Restricted Project
krzysz00 closed D156826: Generalize finding the rocm_agent_enumberator program.
Aug 2 2023, 1:36 PM · Restricted Project, Restricted Project
krzysz00 committed rG7c349c369847: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates (authored by krzysz00).
[mlir][Arith] Add arith.is_nan and arith.is_inf predicates
Aug 2 2023, 7:43 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 closed D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.
Aug 2 2023, 7:42 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 retitled D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates from [mlir][Arith] Add arith.isnan and arith.isinf predicates to [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.
Aug 2 2023, 7:41 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 accepted D156826: Generalize finding the rocm_agent_enumberator program.

Do you need me to land this?

Aug 2 2023, 7:04 AM · Restricted Project, Restricted Project

Aug 1 2023

krzysz00 added reviewers for D154129: [mlir][ROCDL] Adds the ROCDL target attribute.: krzysz00, lamb-j, kzhuravl, arsenm.

Adding some comgr folks in the interests of making sure this revision isn't subtly wrong

Aug 1 2023, 10:46 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.

Address comments

Aug 1 2023, 10:43 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 updated the diff for D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.

Address review comments

Aug 1 2023, 10:42 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added a comment to D154100: [mlir][Target][LLVM] Adds an utility class for serializing operations to binary strings..

Having gone back and re-read, yeah, the only thing here is turning the test off when native codegen isn't available

Aug 1 2023, 10:37 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D154147: [mlir][gpu] Add the Select Object compilation attribute..

More nitpicking than actual issues - this translation seems reasonable

Aug 1 2023, 10:35 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D154132: [mlir][gpu] Add `gpu.binary` op and `#gpu.object` attribute..

This all seems reasonable to me

Aug 1 2023, 10:30 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D155960: [NaryReassociate] Use new access type aware getGEPCost.

This seems fine, but I don't know the code well enough to approve it

Aug 1 2023, 10:20 AM · Restricted Project, Restricted Project

Jul 31 2023

krzysz00 added a comment to D154100: [mlir][Target][LLVM] Adds an utility class for serializing operations to binary strings..

I'd like to see this working with just AMDGPU target enabled (and with X86 or the native target) off so I can use this in a library without needing to pull in extra targets

Jul 31 2023, 1:51 AM · Restricted Project, Restricted Project

Jul 26 2023

krzysz00 requested review of D156362: [mlir][Arith] Change F32 to BF16 truncation to match __truncsfbf2.
Jul 26 2023, 1:17 PM · Restricted Project, Restricted Project
krzysz00 requested review of D156361: [mlir][AMDGPU] Improve BF16 handling through AMDGPU compilation.
Jul 26 2023, 1:16 PM · Restricted Project, Restricted Project
krzysz00 committed rGcdf7ca6db76b: [MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL (authored by SJW <swaters@amd.com>).
[MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL
Jul 26 2023, 8:13 AM · Restricted Project, Restricted Project
krzysz00 closed D154666: [MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL.
Jul 26 2023, 8:13 AM · Restricted Project, Restricted Project

Jul 25 2023

krzysz00 accepted D154666: [MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL.

Overall, looks good

Jul 25 2023, 10:57 AM · Restricted Project, Restricted Project
krzysz00 updated the diff for D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.

Update flang tests

Jul 25 2023, 8:08 AM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added inline comments to D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.
Jul 25 2023, 7:38 AM · Restricted Project, Restricted Project, Restricted Project

Jul 24 2023

krzysz00 added inline comments to D154666: [MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL.
Jul 24 2023, 2:30 PM · Restricted Project, Restricted Project
krzysz00 requested review of D156169: [mlir][Arith] Add arith.is_nan and arith.is_inf predicates.
Jul 24 2023, 2:24 PM · Restricted Project, Restricted Project, Restricted Project
krzysz00 added a reviewer for D154666: [MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL: krzysz00.
Jul 24 2023, 7:11 AM · Restricted Project, Restricted Project
krzysz00 added a comment to D154666: [MLIR][ROCDL] Add conversion for gpu.lane_id to ROCDL.

This overall design works, but I've got minor nitpicks

Jul 24 2023, 7:11 AM · Restricted Project, Restricted Project
krzysz00 accepted D154104: [mlir][gpu] Add GPU target attribute interface..

Hold removed since my concerns are addressed

Jul 24 2023, 7:07 AM · Restricted Project, Restricted Project

Jul 21 2023

krzysz00 accepted D156009: [mlir][rocdl] Change the translation of `GridDim*Op` to `__ockl_get_num_groups`.
Jul 21 2023, 7:12 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D156009: [mlir][rocdl] Change the translation of `GridDim*Op` to `__ockl_get_num_groups`.

Looks good to me, thanks for catching this!

Jul 21 2023, 7:12 PM · Restricted Project, Restricted Project

Jul 20 2023

krzysz00 committed rG4b3eaee2701a: [mlir][AMDGPU] Define wrappers for WMMA matrix ops (authored by giuseros).
[mlir][AMDGPU] Define wrappers for WMMA matrix ops
Jul 20 2023, 11:39 AM · Restricted Project, Restricted Project
krzysz00 closed D152451: [mlir][AMDGPU] Define wrappers for WMMA matrix ops.
Jul 20 2023, 11:38 AM · Restricted Project, Restricted Project

Jul 14 2023

krzysz00 committed rGdb647f5bd8fd: [mlir][GPU] Initialize LLVM exactly once during GPU compiles (authored by krzysz00).
[mlir][GPU] Initialize LLVM exactly once during GPU compiles
Jul 14 2023, 12:11 PM · Restricted Project, Restricted Project
krzysz00 closed D155226: [mlir][GPU] Initialize LLVM exactly once during GPU compiles.
Jul 14 2023, 12:11 PM · Restricted Project, Restricted Project
krzysz00 updated the diff for D155226: [mlir][GPU] Initialize LLVM exactly once during GPU compiles.

Fixed clang-format issue caused by using old clang-format version

Jul 14 2023, 9:17 AM · Restricted Project, Restricted Project
krzysz00 updated the diff for D155226: [mlir][GPU] Initialize LLVM exactly once during GPU compiles.

Fix typos, migration issues

Jul 14 2023, 8:06 AM · Restricted Project, Restricted Project

Jul 13 2023

krzysz00 committed rG6dc2ab95663b: [mlir] Don't make the ROCm conversions depend on the execution engine (authored by krzysz00).
[mlir] Don't make the ROCm conversions depend on the execution engine
Jul 13 2023, 1:40 PM · Restricted Project, Restricted Project
krzysz00 closed D155227: [mlir] Don't make the ROCm conversions depend on the execution engine.
Jul 13 2023, 1:40 PM · Restricted Project, Restricted Project
krzysz00 added a reviewer for D155226: [mlir][GPU] Initialize LLVM exactly once during GPU compiles: fmorac.
Jul 13 2023, 1:38 PM · Restricted Project, Restricted Project
krzysz00 added a comment to D154152: [mlir][gpu] Add GPU target support to `gpu-to-llvm`..

Can we get a test for the bare pointer calling convention please?

Jul 13 2023, 12:34 PM · Restricted Project, Restricted Project
krzysz00 added inline comments to D154149: [mlir][gpu] Add the `gpu-module-to-binary` pass..
Jul 13 2023, 12:32 PM · Restricted Project, Restricted Project