GitLab

Merge branch 'conv_dlops/quantization' of github.com:ROCmSoftwarePlatform/composable_kernel into conv_dlops/quantization

Co-authored-by: Rosty Geyyer <rosty.geyyer@amd.com>

* Change gridwise gemm mD blockwise gemm to naive

* RRR Gemm fix

* Fix RCR gemm bug

* Isolate wmma instructions

* Update amd_inline_asm.hpp

* Update amd_wmma.hpp

* Update amd_wmma.hpp

* fix syntax and update Jenkinsfile

---------
Co-authored-by: zjing14 <zhangjing14@gmail.com>
Co-authored-by: Illia Silin <98187287+illsilin@users.noreply.github.com>
Co-authored-by: illsilin <Illia.Silin@amd.com>

Co-authored-by: zjing14 <zhangjing14@gmail.com>

* enable building on Nav31

* fix syntax

* replace GPU_TARGETS with offload-arch

* add gfx1102 rachitecture

* fix typo

* update changelog

* Grouped gemm + Gelu instances.

* Device Instance Factory for GroupedGemm+Gelu

* Client example

* Rangify fill helper functions.

* Fix name clash.

* Profiler for grouped_gemm+gelu

* No need to use full namespace name.

* Add check for MRaw divisible by vector load.

* Ugly fix for big errors.

* Add grouped_gemm+gelu to profiler CMakelists.

* Store in argument additional info.

* Information about Mraw, Nraw, Kraw values.

* Use FastGelu instead of Gelu.

* Change client ex to use FastGelu

* Remove relaxed error precision.

* Remove duplicate output elementwise-op

---------
Co-authored-by: Adam Osewski <aosewski@amd.com>
Co-authored-by: zjing14 <zhangjing14@gmail.com>

Co-authored-by: Rosty Geyyer <rosty.geyyer@amd.com>

* Modify Doxygen config to pick up include directories recursively

* Add DeviceMem struct to API Reference guide

* Add classes that are used in Flash Attention kernel

* Add a reference and config for generating bibliography
Co-authored-by: Philip Maybank <Philip.Maybank@amd.com>

* add new parallel stage on navi node

* dont run performance tests on navi, get rid of 9110 compiler

* only run navi build when not doing QA

* fix syntax

* use navi21 label

* dont stash profiler on navi nodes, scp deb package to ginger

* disable tests on navi nodes

* test posting a binary to ginger

* add sshpass and use it to copy deb package

* fix the scp example

* fix syntax

* debug the scp issues

* add jenkins user to docker

* dont try whoami

* change jenkins uid and add user with uid=1002

* try scp from the last stage on micimaster

* rename and stash the package, scp from micimaster

* suppress the reserved-identifier warnings

* keep BUILD_DEV=On and use -Werror by default

* fix a bug blocking wmma_gemm_multipleD

* Utilize matrix padder in device_wmma_op

* cosmetic change for gemmpadding format

* clang format

* Change gridwise gemm from FIFO to KMN loop fashion

* clean up

* fast gelu using builtin function

* clean

* clean

* clean

* clean:

* clean

* fix compilation

* clean

* clean

---------
Co-authored-by: zjing14 <zhangjing14@gmail.com>

* Add DeviceOp and examples

* Format DeviceOp template arguments

* Remove bf16 example

* Format

* Format

* Update MakeABCGridDescriptor_A_K0_M_K1_B_K0_N_K1_C_M_N

* Refactor argument preparation

* Update conv_bwd_weight_dl to grouped_conv_bwd_weight_dl

* Rename device op file

* Update include directive in the example file

* Update descriptor preparation for grouped op

* Update the argument

* Update batch handling

* Add gridwise gemm supporting batched input

* Update blockwise indexing, working version

* Update copyright year

* Update check if argument is supported

* Refactor and make consistent with xdl examples

* Update check if argument is supported

* Add changelog entry

* Added comments on Dl op split_k>1 support

---------
Co-authored-by: Rosty Geyyer <rosty.geyyer@amd.com>
Co-authored-by: zjing14 <zhangjing14@gmail.com>

* add conv1d fwd client example

* change 07_grouped_conv2d_fwd to 07_grouped_convnd_fwd

* add conv1d bwd weight

---------
Co-authored-by: zjing14 <zhangjing14@gmail.com>

* fix a bug while building for gfx1030 and add gfx1030 to targets

* fix syntax

* build and archive deb packages

* fix syntax

* run QA to test building packages

* apply cron to develop branch again

* New docs directory with minimal config

* Based on docs directory of rocBLAS

* Config for running Doxygen then Sphinx to generate HTML

* Add minimal content - intro to doc

* Add some boilerplate sections to doc

* content still needs to be done,
* e.g., need to generate API documentation using Doxygen
* need to write contributor guide

* Start Softmax section of Support Primitives doc

* Written as a test bed for typesetting math content

* Need to decide how much detail to go into

* add doc directories to git ignore file.

* Minor edits - new line at EOF, change year in copyright notices

* Port Markdown files to ReStructuredText

* Copy Markdown files from pre-existing doc directory to docs directory

* Convert to reStructured Text (rst) - section headings, links, tables
  have a different syntax in rst

* New rst files added to index - can generate HTML with same style as
  HTML generated from rst files in previous commits

* Intention is to make all the content in doc redundant and use rst
  throughout rather than mix of md and rst

* Extend Softmax section of Primitives Guide

* rename l to z

* add material on applying softmax row-wise to matrix

* define macro for diag operator (represents diagonal matrix)

---------
Co-authored-by: zjing14 <zhangjing14@gmail.com>