[SYCL][Offload] Add SYCLBIN format and dump tool #16873

steffenlarsen · 2025-02-03T16:03:09Z

This commit adds a new option --syclbin to the clang-linker-wrapper tool to produce SYCLBIN files from packaged device code binaries. The SYCLBIN format parsing added with these changes follow the format defined in the SYCLBIN design document.

To help both users and the testing framework, a tool (syclbin-dump) for printing information about a SYCLBIN file is added.

This commit adds a new option `--syclbin` to the clang-linker-wrapper tool to produce SYCLBIN files from packaged device code binaries. The SYCLBIN format parsing added with these changes follow the format defined in the SYCLBIN design document. To help both users and the testing framework, a tool (syclbin-dump) for printing information about a SYCLBIN file is added. Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen · 2025-02-03T16:03:20Z

Design document: #16872

clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp

sarnex

just a quick first pass

clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp

sarnex · 2025-02-03T16:13:55Z

llvm/include/llvm/SYCLLowerIR/ModuleSplitter.h

@@ -18,6 +18,7 @@
 #include "llvm/ADT/SetVector.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/IR/Function.h"
+#include "llvm/IR/Module.h"


nit: do we need this instead of forward declaring?

We do. Unlike shared_ptr, unique_ptr insists on the contained type not being incomplete.

But why do we need this in a header file which is otherwise unchanged? It seems like either this #include should be moved straight into .cpp file, or into another .h file

It seems to be a dependency bug that just wasn't noticed because Module.h was included before ModuleSplitter.h in all other cases.

sycl/test/syclbin/simple_kernel.cpp

sycl/tools/syclbin-dump/syclbin-dump.cpp

clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp

llvm/lib/Object/SYCLBIN.cpp

llvm/include/llvm/Object/SYCLBIN.h

Signed-off-by: Larsen, Steffen <[email protected]>

AlexeySachkov

Haven't yet looked at clang-linker-wrapper and SYCLBIN::read/write implementations

llvm/include/llvm/Object/SYCLBIN.h

AlexeySachkov · 2025-02-04T12:41:44Z

llvm/include/llvm/Object/SYCLBIN.h

+  SmallVector<AbstractModule, 4> AbstractModules;
+
+private:
+  SYCLBIN(const SYCLBIN &Other) = delete;


Why do you need it to be declared as private to delete it? If we explicitly deleting it, then I think it is better to put it next to other constructors so that they are all next to each other.

llvm/include/llvm/Object/SYCLBIN.h

llvm/include/llvm/Object/Binary.h

llvm/include/llvm/Object/SYCLBIN.h

llvm/lib/Object/SYCLBIN.cpp

AlexeySachkov · 2025-02-04T13:20:20Z

llvm/lib/Object/SYCLBIN.cpp

+    for (const ModuleDesc &Desc : ModuleDescs) {
+      for (const module_split::SplitModule &SM : Desc.SplitModules) {


This is a very long for loop, I suggest that we outline some elements of it into smaller helper functions so that it is easier to grasp high-level logic and keep track of indentation levels

I've separated it a bit.

AlexeySachkov · 2025-02-04T13:26:58Z

llvm/include/llvm/Object/SYCLBIN.h

+  static Expected<SmallString<0>> write(const SmallVector<ModuleDesc> &);
+
+  static Expected<std::unique_ptr<SYCLBIN>> read(MemoryBufferRef Source);


Implementation of read/write is quite complex, it has to be exhaustively tested. I'm not saying that all the testing should be here from day 1, but we at least need to lay some groundwork for it. For example what you can do is to have a unit-test which does write, then read and checks that result is the same as the initial input. See e471ba3 for inspiration.

Unittest has been added. Definitely could use some expanding on, but there are also a handful of TODOs in the current version, so it will be expanded upon over time.

sycl/tools/syclbin-dump/syclbin-dump.cpp

sycl/tools/syclbin-dump/CMakeLists.txt

Signed-off-by: Larsen, Steffen <[email protected]>

…_wrapper_impl

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen · 2025-04-23T10:13:46Z

@maksimsab & @AlexeySachkov & @asudarsa - I've refactored the implementation based on discussions with @maksimsab. It should reduce the memory relocation now by pre-calculating the size needed for the serialized SYCLBIN in memory. Let me know what you think!

llvm/lib/Object/SYCLBIN.cpp

maksimsab · 2025-04-24T12:04:08Z

llvm/lib/Object/SYCLBIN.cpp

+    if (!BlobOrError)
+      return BlobOrError.takeError();
+
+    std::string PropStr{BlobOrError->str()};


nit.
We could create here a MemoryBuffer right from the StringRef.

Sadly it won't work. PropertySetRegistry::read expects null-termination and the binary blob doesn't have that. It gets it when we copy it to a string though.

maksimsab · 2025-04-24T12:14:42Z

llvm/lib/Object/SYCLBIN.cpp

+
+namespace {
+
+class SYCLBINBlockReader {


nit.
These 3 classes could be replaced with simple functions making the maintenance easier for developers.

I think the benefit of the class is that we can implicitly track how much memory is left. This lets us avoid reading outside the buffer, which is good for security. We could of course track that through functions as well, but it would require some additional variables we pass around. Personally I think this is cleaner, but if you think having free-functions for it is better, I won't die on this hill. 😉

…_wrapper_impl

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen · 2025-04-28T14:06:32Z

Friendly ping @KseniyaTikhomirova .

KseniyaTikhomirova

LGTM

asudarsa

LGTM.

This commit avoids a warning produced by intel#16873 when building syclbin-dump using compilers other than gcc and clang, where the -fno-rtti option isn't available. Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen · 2025-04-29T06:43:49Z

Windows post-commit failure should be addressed by #18233.

This commit avoids a warning produced by #16873 when building syclbin-dump using compilers other than gcc and clang, where the -fno-rtti option isn't available. Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen requested a review from asudarsa February 3, 2025 16:03

steffenlarsen requested review from a team as code owners February 3, 2025 16:03

steffenlarsen requested a review from KseniyaTikhomirova February 3, 2025 16:03

steffenlarsen had a problem deploying to WindowsCILock February 3, 2025 16:04 — with GitHub Actions Failure

sarnex reviewed Feb 3, 2025

View reviewed changes

clang/tools/clang-linker-wrapper/ClangLinkerWrapper.cpp Show resolved Hide resolved

sarnex reviewed Feb 3, 2025

View reviewed changes

AlexeySachkov reviewed Feb 3, 2025

View reviewed changes

llvm/include/llvm/Object/SYCLBIN.h Outdated Show resolved Hide resolved

steffenlarsen added 3 commits February 4, 2025 00:46

Fix include mistake

43c6227

Signed-off-by: Larsen, Steffen <[email protected]>

Address comments

86e3128

Signed-off-by: Larsen, Steffen <[email protected]>

SPIRV -> SPIR-V

1685f0f

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen had a problem deploying to WindowsCILock February 4, 2025 08:56 — with GitHub Actions Failure

steffenlarsen temporarily deployed to WindowsCILock February 4, 2025 09:51 — with GitHub Actions Inactive

Ignore unused variable in syclbin-dump

a024f11

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen had a problem deploying to WindowsCILock February 4, 2025 10:34 — with GitHub Actions Error

Disable on Windows until driver option is in place

716bdab

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen had a problem deploying to WindowsCILock February 4, 2025 11:13 — with GitHub Actions Error

Fix fault revert

d07b06c

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen temporarily deployed to WindowsCILock February 4, 2025 11:53 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock February 4, 2025 12:40 — with GitHub Actions Inactive

AlexeySachkov reviewed Feb 4, 2025

View reviewed changes

AlexeySachkov mentioned this pull request Feb 4, 2025

[SYCL][Docs] Add SYCLBIN feature and format design document #16872

Merged

KseniyaTikhomirova reviewed Feb 5, 2025

View reviewed changes

sycl/tools/syclbin-dump/CMakeLists.txt Outdated Show resolved Hide resolved

steffenlarsen added 2 commits February 6, 2025 02:32

Address comments and improve code

b598b66

Signed-off-by: Larsen, Steffen <[email protected]>

Merge remote-tracking branch 'intel/sycl' into steffen/syclbin_linker…

a4324ab

…_wrapper_impl

steffenlarsen temporarily deployed to WindowsCILock February 6, 2025 10:37 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock February 6, 2025 11:27 — with GitHub Actions Inactive

steffenlarsen marked this pull request as draft February 10, 2025 08:29

Refactor to write to ostream

799dcc6

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen temporarily deployed to WindowsCILock April 23, 2025 10:09 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock April 23, 2025 10:38 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock April 23, 2025 10:50 — with GitHub Actions Inactive

steffenlarsen had a problem deploying to WindowsCILock April 23, 2025 10:50 — with GitHub Actions Failure

steffenlarsen requested a review from maksimsab April 24, 2025 04:38

steffenlarsen temporarily deployed to WindowsCILock April 24, 2025 04:39 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock April 24, 2025 05:37 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock April 24, 2025 05:48 — with GitHub Actions Inactive

steffenlarsen had a problem deploying to WindowsCILock April 24, 2025 05:48 — with GitHub Actions Failure

maksimsab reviewed Apr 24, 2025

View reviewed changes

llvm/lib/Object/SYCLBIN.cpp Outdated Show resolved Hide resolved

maksimsab reviewed Apr 24, 2025

View reviewed changes

maksimsab approved these changes Apr 24, 2025

View reviewed changes

steffenlarsen added 2 commits April 25, 2025 01:22

Merge remote-tracking branch 'intel/sycl' into steffen/syclbin_linker…

84b619f

…_wrapper_impl

Remove noexcept

498779e

Signed-off-by: Larsen, Steffen <[email protected]>

steffenlarsen temporarily deployed to WindowsCILock April 25, 2025 08:41 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock April 25, 2025 09:40 — with GitHub Actions Inactive

steffenlarsen temporarily deployed to WindowsCILock April 25, 2025 09:49 — with GitHub Actions Inactive

KseniyaTikhomirova approved these changes Apr 28, 2025

View reviewed changes

asudarsa approved these changes Apr 28, 2025

View reviewed changes

steffenlarsen requested a review from mdtoguchi April 28, 2025 15:04

steffenlarsen merged commit c997f77 into intel:sycl Apr 29, 2025
38 of 39 checks passed

steffenlarsen mentioned this pull request Apr 29, 2025

[SYCL] Fix syclbin-dump build on non clang/gcc compilers #18233

Merged

		for (const ModuleDesc &Desc : ModuleDescs) {
		for (const module_split::SplitModule &SM : Desc.SplitModules) {

		static Expected<SmallString<0>> write(const SmallVector<ModuleDesc> &);

		static Expected<std::unique_ptr<SYCLBIN>> read(MemoryBufferRef Source);

[SYCL][Offload] Add SYCLBIN format and dump tool #16873

[SYCL][Offload] Add SYCLBIN format and dump tool #16873

Uh oh!

Conversation

steffenlarsen commented Feb 3, 2025

Uh oh!

steffenlarsen commented Feb 3, 2025

Uh oh!

Uh oh!

sarnex left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

AlexeySachkov left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

steffenlarsen commented Apr 23, 2025

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

steffenlarsen commented Apr 28, 2025

Uh oh!

KseniyaTikhomirova left a comment

Choose a reason for hiding this comment

Uh oh!

asudarsa left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

steffenlarsen commented Apr 29, 2025

Uh oh!

Uh oh!