Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Hipsparselt Integration for Rocm6.3 #1602

Open
wants to merge 29 commits into
base: rocm6.3_internal_testing
Choose a base branch
from
Open

Hipsparselt Integration for Rocm6.3 #1602

wants to merge 29 commits into from
+272 −31

Conversation

petrex
Copy link

@petrex petrex commented Sep 26, 2024

Cherry-pick from pytorch#133934

With hipsparselt && hipblaslt fixes after mainline 14782 : 112 UT expected passing, 7 UT will be skip due to hipsparselt support.

@rocm-mici
Copy link

Jenkins build for 611e3322f8fe4ab6d254c0f90eaa5f7c1e88f3fa commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 7e63cb25749d6e0761c875020b1cd6a8d8ee9a30 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 7e63cb25749d6e0761c875020b1cd6a8d8ee9a30 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 7e63cb25749d6e0761c875020b1cd6a8d8ee9a30 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 7e63cb25749d6e0761c875020b1cd6a8d8ee9a30 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@petrex petrex force-pushed the hipsparselT_rocm6.3_internal_testing branch 2 times, most recently from 47bbe0b to 5603ad5 Compare September 27, 2024 03:13
@rocm-mici
Copy link

Jenkins build for 5603ad59f9fea68d8ee17b467fc0fed2186103a4 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 5603ad59f9fea68d8ee17b467fc0fed2186103a4 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

petrex added 19 commits October 3, 2024 17:12
@petrex
hipSparseLt integration for ROCm 6.2
2f8802b
@petrex
update to support row_major_order and unified hip enum
4e7ae3b 
update support check for hipSparseLt

cusparseLtDenseDescriptorInit update for hipsparselt
@petrex
fix cusparselt/hipsparselt enable
6b0bc02
@petrex
lint
60f961d
@petrex
fix typo function name
85271c8
@petrex
enable cutlass/ck path for rocm
a1b1743
@petrex
add composable kernel submodule
b19eb90
@petrex
add ck layout
0b4fc65
@petrex
Add meta cpu ref / reorder cpu ref with CK wrapper
8e9760a 
add reorder_meta function for CK wrapper

add check for ROCm platform

chore: Remove temporary function used for testing

add TODO and fix typo
@petrex
optimize hipsparselt check
2c7cf70
@petrex
add a check for hipsparselt
c7737c1
@petrex
fix test util
8f111b5
@petrex
drop seperate res buffer (update for new hipsparselt)
dcf1eb8
@petrex
submodule ck / add ck include
46cba31
@petrex
prototype reorder_meta without CK
d697a7f
@petrex
reorder_meta kernel implementation
0a9220f
@petrex
generate sparse tensor with hipsparselt
de729f1
@petrex
tmp drop platform check
4fbbb3f
@petrex
revise hipsparselt support check
dded435
@rocm-mici
Copy link

Jenkins build for 36f5443fb3e69889138d07d87fcc6b33b44eef28 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 36f5443fb3e69889138d07d87fcc6b33b44eef28 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@petrex petrex requested a review from pruthvistony October 10, 2024 20:52
@petrex
Copy link
Author

petrex commented Oct 10, 2024

@pruthvistony CI fails at
0 20:13:49 + wget --no-check-certificate -q -O - https://hub.docker.com/v2/namespaces/rocm/repositories/pytorch/tags?page_size=50 1 script returned exit code 4

any tips?

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for d8e7986f2412e9f7c8042f799a30b2ffaffb1f87 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 0310b9fd7a133a64397d02f0e8172dd9640368c9 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for c4b1db35f0d54e9debf14d912cae375a9d938090 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for c4b1db35f0d54e9debf14d912cae375a9d938090 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for c4b1db35f0d54e9debf14d912cae375a9d938090 commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Detected error during base docker image building:

#29 12.58 The following packages have unmet dependencies:
#29 12.64  rocm-dev : Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 12.64             Depends: rocm-device-libs (= 1.0.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 12.64  rocm-utils : Depends: rocminfo (= 1.0.0.60203-124~22.04)
#29 12.64               Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 12.65 E: Unable to correct problems, you have held broken packages.
#29 ERROR: process "/bin/sh -c bash ./install_rocm.sh" did not complete successfully: exit code: 100
------
 > [24/49] RUN bash ./install_rocm.sh:
12.58 distribution that some required packages have not yet been created
12.58 or been moved out of Incoming.

@rocm-mici
Copy link

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Detected error during base docker image building:

#29 20.24 The following packages have unmet dependencies:
#29 20.31  rocm-dev : Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 20.31             Depends: rocm-device-libs (= 1.0.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 20.31  rocm-utils : Depends: rocminfo (= 1.0.0.60203-124~22.04)
#29 20.31               Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 20.32 E: Unable to correct problems, you have held broken packages.
#29 ERROR: process "/bin/sh -c bash ./install_rocm.sh" did not complete successfully: exit code: 100
------
 > [24/49] RUN bash ./install_rocm.sh:
20.24 distribution that some required packages have not yet been created
20.24 or been moved out of Incoming.

@rocm-mici
Copy link

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Detected error during base docker image building:

#29 15.91 The following packages have unmet dependencies:
#29 15.98  rocm-dev : Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 15.98             Depends: rocm-device-libs (= 1.0.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 15.98  rocm-utils : Depends: rocminfo (= 1.0.0.60203-124~22.04)
#29 15.98               Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 15.99 E: Unable to correct problems, you have held broken packages.
#29 ERROR: process "/bin/sh -c bash ./install_rocm.sh" did not complete successfully: exit code: 100
------
 > [24/49] RUN bash ./install_rocm.sh:
15.91 distribution that some required packages have not yet been created
15.91 or been moved out of Incoming.

@rocm-mici
Copy link

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Detected error during base docker image building:

#29 18.93 The following packages have unmet dependencies:
#29 18.98  rocm-dev : Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 18.98             Depends: rocm-device-libs (= 1.0.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 18.98  rocm-utils : Depends: rocminfo (= 1.0.0.60203-124~22.04)
#29 18.98               Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 18.99 E: Unable to correct problems, you have held broken packages.
#29 ERROR: process "/bin/sh -c bash ./install_rocm.sh" did not complete successfully: exit code: 100
------
 > [24/49] RUN bash ./install_rocm.sh:
18.93 distribution that some required packages have not yet been created
18.93 or been moved out of Incoming.

@okakarpa
Copy link
Collaborator

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Detected error during base docker image building:

#29 12.55 The following packages have unmet dependencies:
#29 12.61  rocm-dev : Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 12.61             Depends: rocm-device-libs (= 1.0.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 12.61  rocm-utils : Depends: rocminfo (= 1.0.0.60203-124~22.04)
#29 12.61               Depends: rocm-cmake (= 0.13.0.60203-124~22.04) but 5.0.0-1 is to be installed
#29 12.62 E: Unable to correct problems, you have held broken packages.
#29 ERROR: process "/bin/sh -c bash ./install_rocm.sh" did not complete successfully: exit code: 100
------
 > [24/49] RUN bash ./install_rocm.sh:
12.55 distribution that some required packages have not yet been created
12.55 or been moved out of Incoming.

@okakarpa
Copy link
Collaborator

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-mici
Copy link

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

@rocm-repo-management-api
Copy link

rocm-repo-management-api bot commented Dec 11, 2024 •
edited
Loading

Jenkins build for 336121897d03bc966e2f256c0a5274e451c96f2c commit finished as FAILURE
Links: Blue Ocean view / Build artifacts

Detected error during base docker image building:

#29 14.03 The following packages have unmet dependencies:
#29 14.10  rocm-dev : Depends: rocm-cmake (= 0.14.0.60300-39~22.04) but 5.0.0-1 is to be installed
#29 14.10             Depends: rocm-device-libs (= 1.0.0.60300-39~22.04) but 5.0.0-1 is to be installed
#29 14.10  rocm-utils : Depends: rocminfo (= 1.0.0.60300-39~22.04) but 5.0.0-1 is to be installed
#29 14.10               Depends: rocm-cmake (= 0.14.0.60300-39~22.04) but 5.0.0-1 is to be installed
#29 14.10 E: Unable to correct problems, you have held broken packages.
#29 ERROR: process "/bin/sh -c bash ./install_rocm.sh" did not complete successfully: exit code: 100
------
 > [24/49] RUN bash ./install_rocm.sh:
14.03 distribution that some required packages have not yet been created
14.03 or been moved out of Incoming.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jeffdaily jeffdaily Awaiting requested review from jeffdaily jeffdaily is a code owner

@jithunnair-amd jithunnair-amd Awaiting requested review from jithunnair-amd jithunnair-amd is a code owner

@pruthvistony pruthvistony Awaiting requested review from pruthvistony

Requested changes must be addressed to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@petrex @rocm-mici @okakarpa @pruthvistony