[SYCL] Embed bfloat16 devicelib into executable if necessary #16729
Conversation
This file contains hidden or 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
Signed-off-by: jinge90 <[email protected]>
jinge90
had a problem deploying
to
WindowsCILock
GitHub Actions
Failure
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
jinge90
had a problem deploying
to
WindowsCILock
GitHub Actions
Failure
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
Co-authored-by: Steffen Larsen <[email protected]>
…mbed_bf16_devicelib_spv
steffenlarsen
approved these changes
Mar 7, 2025
|
Hi, @mdtoguchi |
|
Hi, @maksimsab |
Signed-off-by: jinge90 <[email protected]>
mdtoguchi
approved these changes
Mar 7, 2025
Hi, @maksimsab |
maksimsab
requested changes
Mar 10, 2025
| ; CHECK-BF16: [SYCL/imported symbols] | ||
| ; CHECK-BF16-NEXT: __devicelib_ConvertFToBF16INTEL | ||
|
|
||
| %"class.sycl::_V1::id" = type { %"class.sycl::_V1::detail::array" } |
There was a problem hiding this comment.
Please, remove all unnecessary instructions, attributes and metadata.
There was a problem hiding this comment.
Hi, @maksimsab
Done.
Thanks very much!
| @@ -751,7 +794,7 @@ bool isTargetCompatibleWithModule(const std::string &Target, | |||
| } | |||
|
|
|||
| std::vector<std::unique_ptr<util::SimpleTable>> | |||
| processInputModule(std::unique_ptr<Module> M) { | |||
| processInputModule(std::unique_ptr<Module> M, LLVMContext &Context) { | |||
There was a problem hiding this comment.
No need for LLVMContext here.
There was a problem hiding this comment.
Hi, @maksimsab
Done.
Thanks for the careful review.
| @@ -775,3 +752,44 @@ uint32_t llvm::getSYCLDeviceLibReqMask(const Module &M) { | |||
| } | |||
| return ReqMask; | |||
| } | |||
|
|
|||
| static llvm::SmallVector<const char *, 14> BF16DeviceLibFuncs = { | |||
There was a problem hiding this comment.
Please, use static llvm::SmallVector<StringRef, 14>.
There was a problem hiding this comment.
Hi, @maksimsab
Done.
Thanks very much!
Signed-off-by: jinge90 <[email protected]>
Signed-off-by: jinge90 <[email protected]>
|
Hi, @intel/llvm-gatekeepers |
sommerlukas
pushed a commit
that referenced
this pull request
Mar 14, 2025
Incorporates recent changes to `sycl-post-link` into the RTC-specific version: - #16729 - #16236 - #17211 --------- Signed-off-by: Julian Oppermann <[email protected]>
martygrant
pushed a commit
that referenced
this pull request
Apr 16, 2025
…ion (#18023) As agreement in #16729 , In general we discourage the use of the extension strings as it is a remnant of the OpenCL days. Instead we want to use urDeviceGetInfo when possible. This is a follow-up pr to replace checking OpenCL extension string with urGetDeviceInfo for native bfloat16 conversion extension. --------- Signed-off-by: jinge90 <[email protected]> Co-authored-by: Steffen Larsen <[email protected]>
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.
Currently, sycl bfloat16 conversion functions are implemented in 2 devicelib spvs(fallback version and native version).
The native version targets for any platform which supports "cl_intel_bfloat16_conversions" extension and fallback version is used for all other platforms.
SYCL runtime will select the bfloat16 spvs during execution time by checking bfloat16 extension.
The design requires us to ship 2 spv files together with sycl runtime which some users may dislike.
The PR uses sycl dynamic library mechanism to re-implement this behavior. These 2 bfloat16 lib files are regarded as dynamic library and embedded to final executable, so we don't need to ship any bfloat16 spv libs.
The PR consists following changes:
Fallback and native version of bfloat16 devicelib files have exactly same exported functions, we add a new metadata("SYCL_DEVICELIB_BF16_TYPE") to indicate the version in them. SYCL runtime will check cl_intel_bfloat16_conversions extension and this metadata to decide which version will be linked.