-
Notifications
You must be signed in to change notification settings - Fork 769
[SYCL][Matrix] Add spec document for the matrix extension interface and its first implementation for AMX #3551
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
That seems quite interesting. |
@keryell They are defined cl::sycl::ext::intel::matrix;
They are free functions to match the group library algorithms. But I will add your comment to the open questions section. |
…d its first implementation for AMX Signed-off-by: Dounia <dounia.khaldi@intel.com>
On the long term, using either member functions or ADL could avoid this requirement that adds to the namespace clutter... |
…espace sycl::ext::intel::experimental::matrix, remove the C++ specific API from the document, better formatting Signed-off-by: Dounia <dounia.khaldi@intel.com>
ping @gmlueck |
…escription to a different section towards the end Signed-off-by: Dounia <dounia.khaldi@intel.com>
…espace, some formatting and rewording Signed-off-by: Dounia <dounia.khaldi@intel.com>
- Add the implementation status to doc/extensions/README.md - Add "API description vs what is actually implemented" question like dynamic_ extent and Group to the open questions - Add more clarification about packed_a and packed_b layout, and difference between layouts on matrix and in load/store functions Signed-off-by: Dounia <dounia.khaldi@intel.com>
…tation capabilities in each of the API sections, and some improvement to the layouts explanation text
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is looking a lot better! Now that I understand the intent, I have some more comments.
- Remove all the comments from the asciidoc file - Add more explicitly in two other places that "a kernel using this extension must be decorated with the [[sycl::reqd_sub_group_size(1)]] attribute" - Add 8 bit example for VNNI transform - Incorporate rewording suggestions from Greg.
* upstream/sycl: (649 commits) [SYCL][Driver][NFC] Update integration footer test for 32-bit host (intel#4039) [SYCL][L0] Initialize descriptor .stype and .pNext (intel#4032) [SYCL] Add sycl::kernel::get_kernel_bundle method (intel#3855) [SYCL] Add support for device UUID as a SYCL extension. (intel#3696) [SYCL][Matrix] Add spec document for the matrix extension interface and its first implementation for AMX (intel#3551) Fix debug build mangler test after PR#3992 (8f38045). (intel#4033) [Driver][SYCL] Restrict user -include file in final integration footer step (intel#4036) [SYCL] [Tests] Do not copy device binary image mocks (intel#4023) [SYCL][Doc] Update docs to reflect new compiler features (intel#4030) [SYCL][CUDA] cl_khr_fp16 extension connected to cuda PI. (intel#4029) [SYCL][NFC] Refactor RT unit tests (intel#4021) [SYCL] Switch to using integration footer by default (intel#3777) [SYCL][CUDA] Add the Use Default Stream property (intel#4004) Uplift GPU RT version for Linux to 21.24.20098 (intel#4003) [SYCL][CUDA] atomic_ref.fetch_add used for fp64 reduction if device.has(atomic64) (intel#3950) [Driver][SYCL] Differentiate host dependency link from regular host link (intel#4002) [SYCL][ESIMD] Support device half type in intrinsics. (intel#4024) [SYCL] Allow fpga_reg only for PODs and Trivially-copyable structs (intel#3643) [SYCL][FPGA] Restore legacy debug info version for the hardware (intel#3991) [SYCL][PI][L0] Force reset of memcpy command-list. (intel#4001) ...
Signed-off-by: Dounia dounia.khaldi@intel.com