[SPV-IR to OCL] Fix mutated builtin call attribute copying #2208
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
When SPIRVToOCL20Pass translates following SPV IR ``` %call = tail call spir_func noundef ptr @_Z42__spirv_GenericCastToPtrExplicit_ToPrivatePvi(ptr addrspace(4) noundef %Ptr, i32 noundef 0) declare spir_func ptr @_Z42__spirv_GenericCastToPtrExplicit_ToPrivatePvi(ptr addrspace(4), i32) ``` to OCL IR: ``` %call = tail call spir_func noundef ptr @__to_private(ptr addrspace(4) noundef %Ptr) declare spir_func ptr @__to_private(ptr addrspace(4)) ``` , there is error `Attribute after last parameter` at copying attribute to the new call because the old call has an additional AttributeSet for the last argument. The last argument is eliminated. Set tail call for new call if old call is tail call. Revert cafd7e0. Test isn't added since I can only reproduce using SPIRVToOCL20Pass.
MrSidims
reviewed
Nov 9, 2023
There was a problem hiding this comment.
%call = tail call spir_func noundef ptr @_Z42__spirv_GenericCastToPtrExplicit_ToPrivatePvi(ptr addrspace(4) noundef %Ptr, i32 noundef 0)
declare spir_func ptr @_Z42__spirv_GenericCastToPtrExplicit_ToPrivatePvi(ptr addrspace(4), i32)
to OCL IR:
%call = tail call spir_func noundef ptr @__to_private(ptr addrspace(4) noundef %Ptr)
declare spir_func ptr @__to_private(ptr addrspace(4))
Can this be used as a test? You can add .spt file with usage of OpGenericCastToPtrExplicit instruction, translate it back like llvm-spirv -r --spirv-target-env="CL2.0"
Can |
@MrSidims I've added above LIT and use opt to run. |
|
Restarting CI with a new HEAD |
MrSidims
approved these changes
Jan 29, 2024
MrSidims
reviewed
Jan 29, 2024
svenvh
reviewed
Jan 30, 2024
test/transcoding/call-attribute.ll
Outdated
There was a problem hiding this comment.
This isn't really a transcoding test (i.e., doesn't go LLVM IR -> SPIR-V -> LLVM IR), so I'd suggest moving it to a different place (outside of transcoding/ at least, top level would be fine with me for now).
There was a problem hiding this comment.
done, moved to top level
svenvh
approved these changes
Jan 31, 2024
|
@MrSidims could you please help to merge, thank you! |
MiloszSkobejko
added a commit
to MiloszSkobejko/SPIRV-LLVM-Translator
that referenced
this pull request
Mar 13, 2024
MiloszSkobejko
added a commit
to MiloszSkobejko/SPIRV-LLVM-Translator
that referenced
this pull request
Mar 19, 2024
Changes were cherry-picked from the following commit: KhronosGroup@c6fe12b Also added fixes from: KhronosGroup#2208 KhronosGroup#2192 This changes add SPIR-V translator support for the SPIR-V extension documented here: KhronosGroup/SPIRV-Registry#193. This extension adds one decoration to represent maximum error for FP operations and adds the related Capability. SPIRV Headers support for representing this in SPIR-V: KhronosGroup/SPIRV-Headers#363 intel/llvm#8134 added a new call-site attribute associated with FP builtin intrinsics. This attribute is named 'fpbuiltin-max-error'. Following example shows how this extension is supported in the translator. The input LLVM IR uses new LLVM builtin calls to represent FP operations. An attribute named 'fpbuiltin-max-error' is used to represent the max-error allowed in the FP operation. Example Input LLVM: %t6 = call float @llvm.fpbuiltin.sin.f32(float %f1) KhronosGroup#2 attributes KhronosGroup#2 = { "fpbuiltin-max-error"="2.5" } This is translated into a SPIR-V instruction (for add/sub/mul/div/rem) and OpenCl extended instruction for other instructions. A decoration to represent the max-error is attached to the SPIR-V instruction. SPIR-V code: 4 Decorate 97 FPMaxErrorDecorationINTEL 1075838976 6 ExtInst 2 97 1 sin 88 No new support is added to support translating this SPIR_V back to LLVM. Existing support is used. The decoration is translated back into named metadata associated with the LLVM instruction. This can be readily consumed by backends. Based on input from @andykaylor, we emit attributes when the FP operation is translated back to a call to a builtin function and emit metadata otherwise. Translated LLVM code for basic math functions (add/sub/mul/div/rem): %t6 = fmul float %f1, %f2, !fpbuiltin-max-error !7 !7 = !{!"2.500000"} Translated LLVM code for other math functions: %t6 = call spir_func float @_Z3sinf(float %f1) KhronosGroup#3 attributes KhronosGroup#3 = { "fpbuiltin-max-error"="4.000000" }
MiloszSkobejko
added a commit
to MiloszSkobejko/SPIRV-LLVM-Translator
that referenced
this pull request
Mar 19, 2024
Changes were cherry-picked from the following commit: KhronosGroup@c6fe12b Also cherry picked fixes from: KhronosGroup#2208 KhronosGroup#2192 This changes add SPIR-V translator support for the SPIR-V extension documented here: KhronosGroup/SPIRV-Registry#193. This extension adds one decoration to represent maximum error for FP operations and adds the related Capability. SPIRV Headers support for representing this in SPIR-V: KhronosGroup/SPIRV-Headers#363 intel/llvm#8134 added a new call-site attribute associated with FP builtin intrinsics. This attribute is named 'fpbuiltin-max-error'. Following example shows how this extension is supported in the translator. The input LLVM IR uses new LLVM builtin calls to represent FP operations. An attribute named 'fpbuiltin-max-error' is used to represent the max-error allowed in the FP operation. Example Input LLVM: %t6 = call float @llvm.fpbuiltin.sin.f32(float %f1) KhronosGroup#2 attributes KhronosGroup#2 = { "fpbuiltin-max-error"="2.5" } This is translated into a SPIR-V instruction (for add/sub/mul/div/rem) and OpenCl extended instruction for other instructions. A decoration to represent the max-error is attached to the SPIR-V instruction. SPIR-V code: 4 Decorate 97 FPMaxErrorDecorationINTEL 1075838976 6 ExtInst 2 97 1 sin 88 No new support is added to support translating this SPIR_V back to LLVM. Existing support is used. The decoration is translated back into named metadata associated with the LLVM instruction. This can be readily consumed by backends. Based on input from @andykaylor, we emit attributes when the FP operation is translated back to a call to a builtin function and emit metadata otherwise. Translated LLVM code for basic math functions (add/sub/mul/div/rem): %t6 = fmul float %f1, %f2, !fpbuiltin-max-error !7 !7 = !{!"2.500000"} Translated LLVM code for other math functions: %t6 = call spir_func float @_Z3sinf(float %f1) KhronosGroup#3 attributes KhronosGroup#3 = { "fpbuiltin-max-error"="4.000000" }
MrSidims
pushed a commit
that referenced
this pull request
Mar 21, 2024
Changes were cherry-picked from the following commit: c6fe12b Also cherry picked fixes from: #2208 #2192 This changes add SPIR-V translator support for the SPIR-V extension documented here: KhronosGroup/SPIRV-Registry#193. This extension adds one decoration to represent maximum error for FP operations and adds the related Capability. SPIRV Headers support for representing this in SPIR-V: KhronosGroup/SPIRV-Headers#363 intel/llvm#8134 added a new call-site attribute associated with FP builtin intrinsics. This attribute is named 'fpbuiltin-max-error'. Following example shows how this extension is supported in the translator. The input LLVM IR uses new LLVM builtin calls to represent FP operations. An attribute named 'fpbuiltin-max-error' is used to represent the max-error allowed in the FP operation. Example Input LLVM: %t6 = call float @llvm.fpbuiltin.sin.f32(float %f1) #2 attributes #2 = { "fpbuiltin-max-error"="2.5" } This is translated into a SPIR-V instruction (for add/sub/mul/div/rem) and OpenCl extended instruction for other instructions. A decoration to represent the max-error is attached to the SPIR-V instruction. SPIR-V code: 4 Decorate 97 FPMaxErrorDecorationINTEL 1075838976 6 ExtInst 2 97 1 sin 88 No new support is added to support translating this SPIR_V back to LLVM. Existing support is used. The decoration is translated back into named metadata associated with the LLVM instruction. This can be readily consumed by backends. Based on input from @andykaylor, we emit attributes when the FP operation is translated back to a call to a builtin function and emit metadata otherwise. Translated LLVM code for basic math functions (add/sub/mul/div/rem): %t6 = fmul float %f1, %f2, !fpbuiltin-max-error !7 !7 = !{!"2.500000"} Translated LLVM code for other math functions: %t6 = call spir_func float @_Z3sinf(float %f1) #3 attributes #3 = { "fpbuiltin-max-error"="4.000000" }
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
When SPIRVToOCL20Pass translates following SPV IR
to OCL IR:
, there is error
Attribute after last parameterat copying attributeto the new call because the old call has an additional AttributeSet for
the last argument. The last argument is eliminated in the new call.
Set tail call for new call if old call is tail call.
Revert cafd7e0. Partially revert c60327f.