[SYCL][ABI] Subgroup Extension spec update #1600

garimagu · 2020-04-28T20:02:57Z

Implements the change as per the spec update at
https://github.com/intel/llvm/blob/sycl/sycl/doc/extensions/SubGroup/SYCL_INTEL_sub_group.asciidoc
and update from : #1565

ABI update: Symbols corresponding to the deleted APIs have been deleted.

Signed-off-by: Garima Gupta garima.gupta@intel.com

smaslov-intel · 2020-04-28T20:51:42Z

sycl/plugins/opencl/pi_opencl.cpp

+  if (ret_err != CL_SUCCESS)
+    return cast<pi_result>(ret_err);
+
+  *((uint32_t *)param_value) = (uint32_t)ret_val;


the OpenCL returns "size_t", why are you truncating this to 32-bit? If really needed (I don't think it is) we should document this for piKernelGetSubGroupInfo

The return type of the get_sub_group_info APIs have been changed to return uint32_t data instead of size_t, which is why this change was done. How should this be documented?

Add a comment in the pi.h for piKernelGetSubGroupInfo to indicated now supported parameters with their sizes.

But I suggest that we continue to use "size_t" in PI, and truncate at SYCL level.

Why is that?

I have made the call to PI_API check the returned size of the result value and truncate the value appropriately. Does this look better?
I'm essentially hesitant to make it size_t based on OpenCL spec, because if any new backend is developed based on the SYCL extension spec, it will have to change the Plugin to cast the uint32_t to size_t first, and then in the library will be casting the size_t to uint32_t. It will be weird that the PI APIs are not SYCL extension spec compliant.

I'm essentially hesitant to make it size_t based on OpenCL spec

I'd be fine making it uint32_t in PI, document this in the pi.h, and truncate in OpenCL plugin.
But we should stick to just one size that fits SYCL well.

@smaslov-intel : Do you mean the initial solution was ok with the document change?

Yes, I'd be fine with it.

Thanks. I have updated the review. Let me know if it looks good.

sycl/plugins/opencl/pi_opencl.cpp

alexbatashev · 2020-04-30T15:43:37Z

@garimagu your patch very likely conflicts with recently merged #1609. Please, update your branch.

garimagu · 2020-05-05T19:04:33Z

@smaslov-intel Ping.

alexbatashev

@garimagu since this PR removes two functions, you also need to uplift DEV_ABI version

llvm/sycl/CMakeLists.txt

Line 14 in 16d866b

set(SYCL_DEV_ABI_VERSION 1)

according to ABI Policy Guide.

garimagu · 2020-05-07T17:26:52Z

@garimagu since this PR removes two functions, you also need to uplift DEV_ABI version

llvm/sycl/CMakeLists.txt

Line 14 in 16d866b

set(SYCL_DEV_ABI_VERSION 1)

according to ABI Policy Guide.

What should the new SYCL_DEV_ABI_VERSION be?
Is there any document for ABI Policy Guide?

smaslov-intel

LGTM

alexbatashev · 2020-05-08T04:01:55Z

@garimagu since this PR removes two functions, you also need to uplift DEV_ABI version

llvm/sycl/CMakeLists.txt

Line 14 in 16d866b

set(SYCL_DEV_ABI_VERSION 1)

according to ABI Policy Guide.

What should the new SYCL_DEV_ABI_VERSION be?
Is there any document for ABI Policy Guide?

https://github.com/intel/llvm/blob/sycl/sycl/doc/ABIPolicyGuide.md
The current value is 1, so it must become 2.

garimagu · 2020-05-11T17:06:27Z

@garimagu since this PR removes two functions, you also need to uplift DEV_ABI version

llvm/sycl/CMakeLists.txt

Line 14 in 16d866b

set(SYCL_DEV_ABI_VERSION 1)

according to ABI Policy Guide.

What should the new SYCL_DEV_ABI_VERSION be?
Is there any document for ABI Policy Guide?

https://github.com/intel/llvm/blob/sycl/sycl/doc/ABIPolicyGuide.md
The current value is 1, so it must become 2.

Thank you for the link @alexbatashev . Let me know if I have missed something. I have updated the Major version, the ABI version and the PR commit message.

alexbatashev · 2020-05-11T17:08:59Z

sycl/CMakeLists.txt

@@ -8,10 +8,10 @@ set(CMAKE_CXX_EXTENSIONS OFF)
 option(SYCL_ENABLE_WERROR "Treat all warnings as errors in SYCL project" OFF)
 option(SYCL_ADD_DEV_VERSION_POSTFIX "Adds -V postfix to version string" ON)

-set(SYCL_MAJOR_VERSION 0)
+set(SYCL_MAJOR_VERSION 1)


@garimagu the major version is updated once per release. You only need to update DEV_ABI version.

Suggested change

set(SYCL_MAJOR_VERSION 1)

set(SYCL_MAJOR_VERSION 0)

uploaded the change. please review and approve. smaslov-intel has approved the changes other than the ABI version update.

I think we should (semi-)automate SYCL version update.
It's hard to manage multiple commits requiring version update.

@alexbatashev, could you create a post-commit job, which will create a PR with the version update if it's required, please?

@bader I'll take a look into it

Implements the change as per the spec update at https://github.com/intel/llvm/blob/sycl/sycl/doc/extensions/SubGroup/SYCL_INTEL_sub_group.asciidoc and update from : intel#1565 ABI update: Functions corresponding to the deleted APIs have been deleted. Edit info.cpp test to expect max sub-group size based on the spec update. Changed pi.h document and changed the plugin to always return a 32 bit value. ABI version update. Signed-off-by: Garima Gupta <garima.gupta@intel.com>

Signed-off-by: Garima Gupta <garima.gupta@intel.com>

garimagu · 2020-05-13T17:27:01Z

@bader Please approve and merge.
Sergey has approved it before the final small fix of pi_opencl_symbol_check.dump

bader · 2020-05-13T17:28:42Z

Waiting on code owner review from smaslov-intel and/or intel/llvm-reviewers-runtime.

bader

LGTM. A few minor suggestions.

sycl/test/sub_group/info.cpp

bader · 2020-05-13T17:40:58Z

sycl/source/detail/kernel_info.hpp

+template <info::kernel_sub_group Param> struct get_kernel_sub_group_info {
+  static uint32_t get(RT::PiKernel Kernel, RT::PiDevice Device,
+                      const plugin &Plugin) {
+    uint32_t Result;
    // TODO catch an exception and put it to list of asynchronous exceptions


Suggested change

// TODO catch an exception and put it to list of asynchronous exceptions

Should this comment be removed?

AFAIK, this comment is not valid anymore. Most of the errors now reported through synchronous exceptions.
Although, I'm not the owner of the runtime and let @intel/llvm-reviewers-runtime to decide.

Signed-off-by: Garima Gupta <garima.gupta@intel.com>

garimagu · 2020-05-13T18:02:19Z

@bader uploaded the changes. smaslov has also approved the changes just before the suggested changes were merged.

garimagu · 2020-05-15T16:13:06Z

@bader ping

…ctoring [L0] Refactoring of boolean event parameters

[L0] Refactoring of boolean event parameters

garimagu requested review from smaslov-intel and a team as code owners April 28, 2020 20:02

garimagu requested a review from alexbatashev April 28, 2020 20:02

garimagu assigned AlexeySachkov Apr 28, 2020

smaslov-intel reviewed Apr 28, 2020

View reviewed changes

sycl/plugins/opencl/pi_opencl.cpp Outdated Show resolved Hide resolved

smaslov-intel reviewed Apr 28, 2020

View reviewed changes

sycl/plugins/opencl/pi_opencl.cpp Outdated Show resolved Hide resolved

garimagu force-pushed the private/garimagu/subgroup_extension branch from 102c7eb to 11bcbd9 Compare April 30, 2020 19:55

garimagu requested a review from smaslov-intel May 5, 2020 19:04

garimagu force-pushed the private/garimagu/subgroup_extension branch from eac4935 to ac04c3f Compare May 6, 2020 22:13

alexbatashev suggested changes May 7, 2020

View reviewed changes

garimagu force-pushed the private/garimagu/subgroup_extension branch from ac04c3f to 602c90f Compare May 7, 2020 19:07

smaslov-intel previously approved these changes May 7, 2020

View reviewed changes

garimagu dismissed smaslov-intel’s stale review via b4c6d7b May 11, 2020 17:05

garimagu force-pushed the private/garimagu/subgroup_extension branch from e083664 to b4c6d7b Compare May 11, 2020 17:05

garimagu changed the title ~~[SYCL] Subgroup Extension spec update~~ [SYCL][ABI] Subgroup Extension spec update May 11, 2020

alexbatashev reviewed May 11, 2020

View reviewed changes

smaslov-intel previously approved these changes May 11, 2020

View reviewed changes

alexbatashev previously approved these changes May 11, 2020

View reviewed changes

garimagu mentioned this pull request May 11, 2020

[SYCL][Doc] Update sub-group docs #1565

Merged

garimagu dismissed stale reviews from alexbatashev and smaslov-intel via 21c9f6f May 12, 2020 16:54

garimagu force-pushed the private/garimagu/subgroup_extension branch from cee31e8 to 21c9f6f Compare May 12, 2020 16:54

Edit test pi_opencl_symbol_check.dump

06a3073

Signed-off-by: Garima Gupta <garima.gupta@intel.com>

bader reviewed May 13, 2020

View reviewed changes

smaslov-intel previously approved these changes May 13, 2020

View reviewed changes

Suggested changed to sub_group/info.cpp

2fcb115

Signed-off-by: Garima Gupta <garima.gupta@intel.com>

garimagu dismissed smaslov-intel’s stale review via 2fcb115 May 13, 2020 17:59

alexbatashev approved these changes May 14, 2020

View reviewed changes

bader requested a review from smaslov-intel May 15, 2020 17:00

smaslov-intel approved these changes May 15, 2020

View reviewed changes

bader merged commit 9d4c284 into intel:sycl May 15, 2020

kbenzie added a commit to kbenzie/intel-llvm that referenced this pull request Feb 17, 2025

Merge pull request intel#1600 from winstonzhang-intel/event-bool-refa…

b05ffb5

…ctoring [L0] Refactoring of boolean event parameters

Chenyang-L pushed a commit that referenced this pull request Feb 18, 2025

Merge pull request #1600 from winstonzhang-intel/event-bool-refactoring

9edff97

[L0] Refactoring of boolean event parameters

[SYCL][ABI] Subgroup Extension spec update #1600

[SYCL][ABI] Subgroup Extension spec update #1600

Uh oh!

Conversation

garimagu commented Apr 28, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

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!

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!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

alexbatashev commented Apr 30, 2020

Uh oh!

garimagu commented May 5, 2020

Uh oh!

alexbatashev left a comment

Choose a reason for hiding this comment

Uh oh!

garimagu commented May 7, 2020

Uh oh!

smaslov-intel left a comment

Choose a reason for hiding this comment

Uh oh!

alexbatashev commented May 8, 2020

Uh oh!

garimagu commented May 11, 2020

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!

Choose a reason for hiding this comment

Uh oh!

garimagu commented May 13, 2020

Uh oh!

bader commented May 13, 2020

Uh oh!

bader 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!

garimagu commented May 13, 2020

Uh oh!

garimagu commented May 15, 2020

Uh oh!

Uh oh!

garimagu commented Apr 28, 2020 •

edited

Loading