Page MenuHomePhabricator

mravishankar (Mahesh Ravishankar)
User

Projects

User does not belong to any projects.

User Details

User Since
Jan 6 2015, 1:11 PM (268 w, 23 h)

Recent Activity

Thu, Feb 20

mravishankar accepted D74874: [mlir][spirv] Add lowering for load/store zero-rank memref from std to SPIR-V..

This looks fine to me. Thanks for adding this.
But please add serialization tests as well

Thu, Feb 20, 11:47 PM · Restricted Project
mravishankar requested changes to D74874: [mlir][spirv] Add lowering for load/store zero-rank memref from std to SPIR-V..

Might be worth checking that the serialization/deserialization works as well

Thu, Feb 20, 12:58 PM · Restricted Project

Wed, Feb 19

mravishankar accepted D74852: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.indexed_generic ops..
Wed, Feb 19, 2:09 PM · Restricted Project

Tue, Feb 18

mravishankar committed rGa8355b5c0f67: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg. (authored by mravishankar).
[mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.
Tue, Feb 18, 1:31 PM
mravishankar closed D74638: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.generic ops..
Tue, Feb 18, 1:31 PM · Restricted Project

Fri, Feb 14

mravishankar updated the diff for D74638: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.generic ops..

Change to test to check for scalar load

Fri, Feb 14, 1:43 PM · Restricted Project
mravishankar updated the diff for D74638: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.generic ops..

Addressing comments and fixing failing test

Fri, Feb 14, 1:34 PM · Restricted Project
mravishankar updated the diff for D74638: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.generic ops..

Removing errant ;

Fri, Feb 14, 1:07 PM · Restricted Project
mravishankar created D74638: [mlir][Linalg] Allow specifiying zero-rank shaped type operands to linalg.generic ops..
Fri, Feb 14, 12:57 PM · Restricted Project

Wed, Feb 12

mravishankar accepted D74491: [mlir][Linalg] NFC - Refactor in preparation for automatic Linalg "named" ops..

Overall looks fine to me.

Wed, Feb 12, 10:15 AM · Restricted Project

Tue, Feb 11

mravishankar added inline comments to D74365: [mlir][Linalg] Update semantics for Linalg generic ops with tensors..
Tue, Feb 11, 1:18 PM · Restricted Project
mravishankar resigned from D73893: [MLIR][GPU] Implement initial mapping from loop.parallel to gpu.launch..

Thanks Stephan! I see overall where this is headed. This is nice to have indeed.

Tue, Feb 11, 1:10 PM · Restricted Project

Mon, Feb 10

mravishankar accepted D74012: [mlir][spirv] Use spv.entry_point_abi in GPU to SPIR-V conversions.

THanks Lei! More I think about it, it is better to use this approach.

Mon, Feb 10, 10:13 AM · Restricted Project
mravishankar added a comment to D74012: [mlir][spirv] Use spv.entry_point_abi in GPU to SPIR-V conversions.

Maybe I am missing something here, but from the GPU dialect, the sizes are passed to the gpu.launch, so you can take them from there. If you want to specialize a kernel for specific sizes, you need to ensure compatible call sites, like in other function specialization. Is this more about driving upper layers of code generation so that you end up with a gpu.launch that has sizes you want? Or do you want to make gpu.func usable independent of the gpu.launch?

Mon, Feb 10, 10:13 AM · Restricted Project
mravishankar accepted D74226: [mlir][spirv] Introduce spv.function.

Thanks Lei! Accepting conditioned on addressing other comments.

Mon, Feb 10, 10:04 AM · Restricted Project

Sat, Feb 8

mravishankar committed rGaaddca1efd1d: [mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with… (authored by mravishankar).
[mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with…
Sat, Feb 8, 10:34 PM
mravishankar added inline comments to D74270: [mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with Vulkan SPEC.
Sat, Feb 8, 6:24 PM · Restricted Project
mravishankar updated the diff for D74270: [mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with Vulkan SPEC.

Addressing missed comment.

Sat, Feb 8, 6:24 PM · Restricted Project
mravishankar updated the diff for D74270: [mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with Vulkan SPEC.

Addressing comments

Sat, Feb 8, 6:20 PM · Restricted Project
mravishankar requested changes to D74226: [mlir][spirv] Introduce spv.function.

THanks for taking this one now. Its already touching so much code, if we left it for longer it would be more effort to use spv.function

Sat, Feb 8, 2:29 PM · Restricted Project
mravishankar requested changes to D73893: [MLIR][GPU] Implement initial mapping from loop.parallel to gpu.launch..

(Sorry for jumping on this late)

Sat, Feb 8, 10:55 AM · Restricted Project

Fri, Feb 7

mravishankar created D74270: [mlir][GPUToSPIRV] Modify the lowering of gpu.block_dim to be consistent with Vulkan SPEC.
Fri, Feb 7, 5:26 PM · Restricted Project
mravishankar committed rGd06dd29e09b6: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors. (authored by mravishankar).
[mlir][Linalg] Implement fusion of linalg.generic operation on tensors.
Fri, Feb 7, 10:45 AM
mravishankar closed D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..
Fri, Feb 7, 10:45 AM · Restricted Project
mravishankar added inline comments to D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..
Fri, Feb 7, 10:08 AM · Restricted Project
mravishankar updated the diff for D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..

Updating after addressing comments

Fri, Feb 7, 10:08 AM · Restricted Project

Thu, Feb 6

mravishankar updated the diff for D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..

Addressing some clang format issues

Thu, Feb 6, 11:09 PM · Restricted Project
mravishankar updated the diff for D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..

Actually updating the file to fix build errors.

Thu, Feb 6, 11:01 PM · Restricted Project
mravishankar accepted D74151: Adding sin op in the GLSL extension.

THanks for adding the cosine tests too...

Thu, Feb 6, 10:52 PM · Restricted Project
mravishankar updated the diff for D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..

Updating with changes to fix build failures

Thu, Feb 6, 5:18 PM · Restricted Project
mravishankar created D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..
Thu, Feb 6, 2:52 PM · Restricted Project
mravishankar updated subscribers of D74172: [mlir][Linalg] Implement fusion of linalg.generic operation on tensors..
Thu, Feb 6, 2:52 PM · Restricted Project
mravishankar requested changes to D74151: Adding sin op in the GLSL extension.

Thanks Natasha! This needs tests though.
See mlir/test/Dialect/SPIRV/glslops.mlir for similar ops. These are round-trip tests. It checks that mlir can parse the operation and print it out correctly
You also need "Serialization" tests, i.e. MLIR can generate valid SPIR-V binary for this operation. See mlir/test/Dialect/SPIRV/Serialization/glsl-ops.mlir

Thu, Feb 6, 2:26 PM · Restricted Project

Wed, Feb 5

mravishankar added a comment to D74012: [mlir][spirv] Use spv.entry_point_abi in GPU to SPIR-V conversions.

Lets go with this for now. We can clean this up when the attribute story gets fixed up.

Wed, Feb 5, 11:01 PM · Restricted Project

Tue, Feb 4

mravishankar requested changes to D74012: [mlir][spirv] Use spv.entry_point_abi in GPU to SPIR-V conversions.
Tue, Feb 4, 11:02 PM · Restricted Project
mravishankar added a comment to D74012: [mlir][spirv] Use spv.entry_point_abi in GPU to SPIR-V conversions.

I understand what the intent is here, but the input already has an attribute that belongs to the SPIR-V dialect before lowering. That makes things a bit non-composable. In cases where someone lowers to the GPU dialect and then conditionally decides to lower to SPIR-V dialect or the NVVM dialect, with this change on the SPIR-V side a separate pass will be needed to add this attribute. Ideally the input should be only in GPU dialect, whereas here it isnt.
Is it possible instead to add an attribute to GPU dialect itself which contains information about the workgroup size. Then while lowering we can convert one attribute to another.

Tue, Feb 4, 11:02 PM · Restricted Project

Wed, Jan 29

mravishankar accepted D73437: [mlir][spirv] Convert linalg.generic for reduction to SPIR-V ops.
Wed, Jan 29, 12:07 PM · Restricted Project, Restricted Project, Restricted Project
mravishankar accepted D73563: [mlir][spirv] Add GroupNonUniform min and max operations..

Thanks!

Wed, Jan 29, 9:49 AM · Restricted Project

Tue, Jan 28

mravishankar added inline comments to D73437: [mlir][spirv] Convert linalg.generic for reduction to SPIR-V ops.
Tue, Jan 28, 4:44 PM · Restricted Project, Restricted Project, Restricted Project

Mon, Jan 27

mravishankar added inline comments to D73491: [mlir][spirv] Add GroupNonUniform arithmetic operations..
Mon, Jan 27, 3:02 PM · Restricted Project

Jan 27 2020

mravishankar requested changes to D73491: [mlir][spirv] Add GroupNonUniform arithmetic operations..

The (de)serialization needs to be implemented. Will that come later?

Jan 27 2020, 11:07 AM · Restricted Project
mravishankar requested changes to D73437: [mlir][spirv] Convert linalg.generic for reduction to SPIR-V ops.
Jan 27 2020, 10:48 AM · Restricted Project, Restricted Project, Restricted Project

Jan 20 2020

mravishankar added inline comments to D73022: [mlir][spirv] Add lowering for composite constant..
Jan 20 2020, 11:40 AM · Restricted Project
mravishankar requested changes to D73022: [mlir][spirv] Add lowering for composite constant..
Jan 20 2020, 11:40 AM · Restricted Project

Jan 6 2020

mravishankar accepted D72296: [mlir][spirv] Add lowering for std cmp ops..
Jan 6 2020, 2:54 PM · Restricted Project
mravishankar added a comment to D72296: [mlir][spirv] Add lowering for std cmp ops..

THanks Denis. Looks great!

Jan 6 2020, 2:54 PM · Restricted Project
mravishankar added a comment to D72205: [mlir][spirv] Add lowering for std bit ops..

Thanks Denis for taking these up and driving them

Jan 6 2020, 2:54 PM · Restricted Project
mravishankar accepted D72205: [mlir][spirv] Add lowering for std bit ops..
Jan 6 2020, 2:54 PM · Restricted Project
mravishankar requested changes to D72205: [mlir][spirv] Add lowering for std bit ops..

Could you add some serialization tests as well. Thanks

Jan 6 2020, 12:14 PM · Restricted Project
mravishankar requested changes to D72296: [mlir][spirv] Add lowering for std cmp ops..

Could you add some serialization tests as well?

Jan 6 2020, 12:14 PM · Restricted Project
mravishankar accepted D72205: [mlir][spirv] Add lowering for std bit ops..
Jan 6 2020, 12:05 PM · Restricted Project
mravishankar requested changes to D72205: [mlir][spirv] Add lowering for std bit ops..
Jan 6 2020, 12:05 PM · Restricted Project

Jan 5 2020

mravishankar committed rG8aae6455c074: [mlir][spirv] Update SPIR-V documentation with information about lowering to… (authored by mravishankar).
[mlir][spirv] Update SPIR-V documentation with information about lowering to…
Jan 5 2020, 11:12 PM

Jan 3 2020

mravishankar updated the diff for D72116: [mlir][spirv] Update SPIR-V docs with information about utilities to convert to SPIR-V.

Adding link to chunked Vulkan Spec pages.

Jan 3 2020, 5:07 PM · Restricted Project
mravishankar added inline comments to D72116: [mlir][spirv] Update SPIR-V docs with information about utilities to convert to SPIR-V.
Jan 3 2020, 5:07 PM · Restricted Project
mravishankar updated the diff for D72116: [mlir][spirv] Update SPIR-V docs with information about utilities to convert to SPIR-V.

Addressing comments from antiagainst

Jan 3 2020, 4:57 PM · Restricted Project
mravishankar accepted D72137: [mlir][spirv] Add lowering for std.fpext, std.fptrunc, std.sitofp..

Thanks!

Jan 3 2020, 11:40 AM · Restricted Project
mravishankar added inline comments to D72094: [mlir][linalg] Lower linalg to affine loops.
Jan 3 2020, 12:00 AM · Restricted Project

Jan 2 2020

mravishankar accepted D72062: [mlir][spirv] Fix shader ABI attribute prefix and add verification.
Jan 2 2020, 4:59 PM · Restricted Project
mravishankar created D72116: [mlir][spirv] Update SPIR-V docs with information about utilities to convert to SPIR-V.
Jan 2 2020, 3:54 PM · Restricted Project
mravishankar added inline comments to D72062: [mlir][spirv] Fix shader ABI attribute prefix and add verification.
Jan 2 2020, 2:49 PM · Restricted Project
mravishankar added a comment to D72022: [mlir][Linalg] Extend generic ops to allow tensors.

@nicolasvasilache : Totally agree with all the points you make. I understand the advantage of performing fusion at the tensor level. I was under the impression that there would be a separate dialect for "Linalg operations on tensors", and the buffer allocation pass would "convert" from that dialect into the existing Linalg dialect. But, I guess inter-mixing of operations from different dialects to having different "incarnations" of the operation (one operating on tensors and other on buffers) is equivalent. I see that this makes sense for generic ops, but how would this work for other operations like linalg.matmul and linalg.matvec, etc. where I am not sure you can have a tensor argument. So it seems like this is making generic ops different than other linalg ops.
Another question is about buffer allocation. Is there some buffer allocation already implemented in Linalg?
In any case, since the contract between Linalg and other "backend" dialects is that the lowering is only supported if Linalg ops work on memrefs, I dont really have any major concerns and happy to see how things work out.

Jan 2 2020, 10:58 AM · Restricted Project
mravishankar accepted D72095: [mlir][spirv] Update SPIR-V enums and ops with availability spec.

This is awesome! Fantastic work!

Jan 2 2020, 10:46 AM · Restricted Project
mravishankar requested changes to D72062: [mlir][spirv] Fix shader ABI attribute prefix and add verification.

Thanks for fixing the naming!

Jan 2 2020, 10:17 AM · Restricted Project
mravishankar added a comment to D72022: [mlir][Linalg] Extend generic ops to allow tensors.

Without commenting on the specifics of the change itself, was wondering what advantage there is to allow generic ops to take tensor values as well. Until now lowering from Linalg to (say) GPU Dialect was fairly straight-forward cause Linalg operated on memref types. I am not sure what would happen if the generic op with tensors would be lowered to GPU dialect (presumably should be an error), and lowering it correctly would need some buffer allocation to kick in. So it seems like this is adding complexity. Could you please provide some context behind the need for this change?

Jan 2 2020, 10:08 AM · Restricted Project
mravishankar accepted D71947: [MLIR][spirv] Allow specifying availability on enum attribute cases.

SPIR-V parts look good to me. Would still prefer the version/capability related methods into a separate file. They seem to have disjoint functionality (but its a personal preference).

Jan 2 2020, 9:58 AM · Restricted Project

Jan 1 2020

mravishankar accepted D72051: [mlir][spirv] Fix links in docs and update dialect docs.
Jan 1 2020, 3:25 PM · Restricted Project
mravishankar accepted D72054: [mlir][spirv] NFC: Move shader ABI attributes to a new file.

Just wondering why the refactoring?

Jan 1 2020, 3:16 PM · Restricted Project

Dec 27 2019

mravishankar accepted D71930: [mlir][spirv] Add basic definitions for supporting availability.
Dec 27 2019, 1:15 PM · Restricted Project
mravishankar requested changes to D71930: [mlir][spirv] Add basic definitions for supporting availability.

The SPIR-V part looks fine to me. I have no major comments on the tblgen part. Maybe get some one with more knowledge to review it.

Dec 27 2019, 11:58 AM · Restricted Project
mravishankar committed rGc3d3569d4ca6: [mlir] Convert std.and/std.or ops to spv.LogicalAnd/spv.LogicalOr (authored by mravishankar).
[mlir] Convert std.and/std.or ops to spv.LogicalAnd/spv.LogicalOr
Dec 27 2019, 11:39 AM
mravishankar closed D71881: [mlir] Convert std.and/std.or ops to spv.LogicalAnd/spv.LogicalOr.
Dec 27 2019, 11:39 AM · Restricted Project
mravishankar accepted D71928: [mlir][spirv] Update docs regarding how to define new ops and types.

Thanks. Looks good.

Dec 27 2019, 11:39 AM · Restricted Project

Dec 25 2019

mravishankar created D71881: [mlir] Convert std.and/std.or ops to spv.LogicalAnd/spv.LogicalOr.
Dec 25 2019, 4:51 PM · Restricted Project

Dec 23 2019

A. Unique TensorFlower <gardener@tensorflow.org> committed rG319cca3bbe69: Add missing virtual inliner interface method in SPIR-V dialect. (authored by mravishankar).
Add missing virtual inliner interface method in SPIR-V dialect.
Dec 23 2019, 9:24 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG80ec474a65a2: Add atomic operations to SPIR-V dialect. (authored by mravishankar).
Add atomic operations to SPIR-V dialect.
Dec 23 2019, 9:24 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rGa0557ea9d654: Fix (de)serialization generation for SPV_ScopeAttr, SPV_MemorySemanticsAttr. (authored by mravishankar).
Fix (de)serialization generation for SPV_ScopeAttr, SPV_MemorySemanticsAttr.
Dec 23 2019, 9:24 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rGb909299d205d: Add missing CMake dependency for MLIRTestIR. (authored by mravishankar).
Add missing CMake dependency for MLIRTestIR.
Dec 23 2019, 9:22 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG652fc261d7f2: Expose a convenience function to add interface attributes to a function. (authored by mravishankar).
Expose a convenience function to add interface attributes to a function.
Dec 23 2019, 9:22 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG04fdd33daf5e: More convenience build methods for SPIR-V ops. (authored by mravishankar).
More convenience build methods for SPIR-V ops.
Dec 23 2019, 9:21 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG4a62019eb88f: Add lowering for module with gpu.kernel_module attribute. (authored by mravishankar).
Add lowering for module with gpu.kernel_module attribute.
Dec 23 2019, 9:21 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG6500b7e0c085: NFC: Separate implementation and definition in ConvertStandardToSPIRV.cpp… (authored by mravishankar).
NFC: Separate implementation and definition in ConvertStandardToSPIRV.cpp…
Dec 23 2019, 9:20 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG883f5557263e: During serialization do a walk of ops in module to find spv.module. (authored by mravishankar).
During serialization do a walk of ops in module to find spv.module.
Dec 23 2019, 9:20 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG4d61a79db46e: Allow specification of the workgroup size for GPUToSPIRV lowering. (authored by mravishankar).
Allow specification of the workgroup size for GPUToSPIRV lowering.
Dec 23 2019, 9:19 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rGc5ba37b6ae00: Add a pass to legalize operations before lowering to SPIR-V. (authored by mravishankar).
Add a pass to legalize operations before lowering to SPIR-V.
Dec 23 2019, 9:18 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG353fb2bd3878: Convert MemRefType to a linearized array in SPIR-V lowering. (authored by mravishankar).
Convert MemRefType to a linearized array in SPIR-V lowering.
Dec 23 2019, 9:18 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG03620fa70ad2: Misc changes to lowering to SPIR-V. (authored by mravishankar).
Misc changes to lowering to SPIR-V.
Dec 23 2019, 9:17 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rGf87b2fd41bcc: NFC: Actually expose the implementation of createGPUToSPIRVLoweringPass. (authored by mravishankar).
NFC: Actually expose the implementation of createGPUToSPIRVLoweringPass.
Dec 23 2019, 9:17 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG7fd46bf258e8: Add missing rule to generate SPIR-V ABI Attribute using tblgen to CMake. (authored by mravishankar).
Add missing rule to generate SPIR-V ABI Attribute using tblgen to CMake.
Dec 23 2019, 9:17 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rGbd485afda031: Introduce attributes that specify the final ABI for a spirv::ModuleOp. (authored by mravishankar).
Introduce attributes that specify the final ABI for a spirv::ModuleOp.
Dec 23 2019, 9:17 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG1ea231bd39ea: Allow memref_cast from static strides to dynamic strides. (authored by mravishankar).
Allow memref_cast from static strides to dynamic strides.
Dec 23 2019, 9:17 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG6db8530c26c8: Add more canonicalizations for SubViewOp. (authored by mravishankar).
Add more canonicalizations for SubViewOp.
Dec 23 2019, 9:16 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG1145cebdab37: Verify subview op result has dynamic shape, when sizes are specified. (authored by mravishankar).
Verify subview op result has dynamic shape, when sizes are specified.
Dec 23 2019, 9:16 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG19212105dd09: Changes to SubViewOp to make it more amenable to canonicalization. (authored by mravishankar).
Changes to SubViewOp to make it more amenable to canonicalization.
Dec 23 2019, 9:16 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rGa78bd84cf84c: NFC: Refactor Dialect Conversion targeting SPIR-V. (authored by mravishankar).
NFC: Refactor Dialect Conversion targeting SPIR-V.
Dec 23 2019, 9:14 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG2be53603e929: Add operations needed to support lowering of AffineExpr to SPIR-V. (authored by mravishankar).
Add operations needed to support lowering of AffineExpr to SPIR-V.
Dec 23 2019, 9:14 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG9d985141ef84: Make legality check in GPU->SPIR-V lowering of FuncOp kernel specific. (authored by mravishankar).
Make legality check in GPU->SPIR-V lowering of FuncOp kernel specific.
Dec 23 2019, 9:14 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG104af84f4c87: Add Conversion to lower loop::ForOp to spirv::LoopOp. (authored by mravishankar).
Add Conversion to lower loop::ForOp to spirv::LoopOp.
Dec 23 2019, 9:14 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG9cbbd8f4dfa4: Support lowering of imperfectly nested loops into GPU dialect. (authored by mravishankar).
Support lowering of imperfectly nested loops into GPU dialect.
Dec 23 2019, 9:12 PM
A. Unique TensorFlower <gardener@tensorflow.org> committed rG61225d678e5d: Add a convenient operation build method for spirv::SelectOp (authored by mravishankar).
Add a convenient operation build method for spirv::SelectOp
Dec 23 2019, 9:11 PM