Bug 2258306

Summary:	Fix for issue related to ROCm HIP
Product:	[Fedora] Fedora	Reporter:	Luya Tshimbalanga <luya_tfz>
Component:	llvm	Assignee:	Tom Stellard <tstellar>
Status:	CLOSED EOL	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	medium	Docs Contact:
Priority:	unspecified
Version:	39	CC:	dmalcolm, garrett.mitchener, jakub, jchecahi, jistone, kkleine, npopov, scottt.tw, sergesanspaille, siddharth.kde, tbaeder, tstellar, tuliom
Target Milestone:	---	Keywords:	Desktop
Target Release:	---
Hardware:	x86_64
OS:	Linux
URL:	https://bugzilla.redhat.com/show_bug.cgi?id=2216594#c21
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2024-11-27 22:39:02 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	2216594
Bug Blocks:

Description Luya Tshimbalanga 2024-01-14 07:21:42 UTC

From the link, upstream ROCm has a patch addressing the issue related to HIP, build from LLVM 17.x, causing a crash to Blender while sometime freezing the entire system. 

Reproducible: Always

Steps to Reproduce:
1.Use HIP to render with Blender Cycles
2.
3.
Actual Results:  
Without the patch from ROCm, rendering with enabled HIP brought down an application like Blender.

Expected Results:  
With the patch from ROCm, rendering with enabledHIP is stable.

Comment 1 Tulio Magno Quites Machado Filho 2024-01-15 20:00:10 UTC

According to bug 2216594 comment 21:

> So in theory it should be as simple as doing a diff of llvm/lib/Target/AMDGPU between rocm/llvm-project (amd-mainline-open) and llvm/llvm-project (release/17.x). As I said prior, upstream thinks that a bugfix patch is missing in the upstream llvm tree, which is what Fedora uses.

I tried this, but the difference is too large to be integrated downstream:

$ git diff --shortstat origin/release/17.x..ROCm/amd-mainline-open -- llvm/lib/Target/AMDGPU
 77 files changed, 4047 insertions(+), 1940 deletions(-)

According to bug 2216594 comment 16:

> it seems it's fixed in LLVM 18

LLVM 18 received 520 commits to llvm/lib/Target/AMDGPU so far.

Has this issue already been reported to https://github.com/llvm/llvm-project/ ?

@luya_tfz , is there a way to automate how this issue is reproduced in order to bisect which commit fixed this?

Comment 2 Luya Tshimbalanga 2024-01-18 08:13:34 UTC

(In reply to Tulio Magno Quites Machado Filho from comment #1)

> > it seems it's fixed in LLVM 18
> 
> LLVM 18 received 520 commits to llvm/lib/Target/AMDGPU so far.
> 
> Has this issue already been reported to
> https://github.com/llvm/llvm-project/ ?

As far as I know, @alexjnewt, the maintainer of ROCCL stack can provide the details.
> 
> @luya_tfz , is there a way to automate how this issue is
> reproduced in order to bisect which commit fixed this?

Alas not yet. Easy way to do so is running Blender with HIP enabled on an AMD hardware and attempt to render.

Comment 3 Aoife Moloney 2024-11-13 10:27:50 UTC

This message is a reminder that Fedora Linux 39 is nearing its end of life.
Fedora will stop maintaining and issuing updates for Fedora Linux 39 on 2024-11-26.
It is Fedora's policy to close all bug reports from releases that are no longer
maintained. At that time this bug will be closed as EOL if it remains open with a
'version' of '39'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, change the 'version' 
to a later Fedora Linux version. Note that the version field may be hidden.
Click the "Show advanced fields" button if you do not see it.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora Linux 39 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora Linux, you are encouraged to change the 'version' to a later version
prior to this bug being closed.

Comment 4 Aoife Moloney 2024-11-27 22:39:02 UTC

Fedora Linux 39 entered end-of-life (EOL) status on 2024-11-26.

Fedora Linux 39 is no longer maintained, which means that it
will not receive any further security or bug fix updates. As a result we
are closing this bug.

If you can reproduce this bug against a currently maintained version of Fedora Linux
please feel free to reopen this bug against that version. Note that the version
field may be hidden. Click the "Show advanced fields" button if you do not see
the version field.

If you are unable to reopen this bug, please file a new report against an
active release.

Thank you for reporting this bug and we are sorry it could not be fixed.