Display corruption with Kernel 5.10 and 5.12

AmigaOne X5000 platform specific issues related to Linux only.
Post Reply
Roland
Posts: 824
Joined: Tue May 02, 2017 7:23 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by Roland »

xeno74 wrote: Sat Oct 02, 2021 12:55 pm Download: uImage-rb10
rb10 works without issues! "git bisect good"
- Roland -
User avatar
xeno74
Posts: 9320
Joined: Fri Mar 23, 2012 7:58 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by xeno74 »

Roland wrote: Sat Oct 02, 2021 6:37 pm rb10 works without issues! "git bisect good"
Great! Next step.

"git bisect good"

Output:

Code: Select all

[b18c1f0aa4775bf1ecfbe5942e4e7c6bffdeff6a] bpf: selftest: Adapt sock_fields test to use skel and global variables
Download: uImage-rb11
http://www.amigalinux.org
http://www.supertuxkart-amiga.de

Running Linux on AmigaONEs can require some tinkering.
Roland
Posts: 824
Joined: Tue May 02, 2017 7:23 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by Roland »

xeno74 wrote: Sat Oct 02, 2021 7:29 pm Download: uImage-rb11
The issue present in rb11. "git bisect bad"
- Roland -
User avatar
xeno74
Posts: 9320
Joined: Fri Mar 23, 2012 7:58 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by xeno74 »

Roland wrote: Sat Oct 02, 2021 8:32 pm The issue present in rb11. "git bisect bad"
OK, next ...

"git bisect bad"

Output:

Code: Select all

[27e5203bd9c5cc6d54dcac48c3027f3f04522b8b] bpf: Change bpf_sk_assign to accept ARG_PTR_TO_BTF_ID_SOCK_COMMON
Download: uImage-rb12
http://www.amigalinux.org
http://www.supertuxkart-amiga.de

Running Linux on AmigaONEs can require some tinkering.
Roland
Posts: 824
Joined: Tue May 02, 2017 7:23 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by Roland »

xeno74 wrote: Sun Oct 03, 2021 6:44 am Download: uImage-rb12
The problem is present in sb 12. "git bisect bad"
- Roland -
User avatar
xeno74
Posts: 9320
Joined: Fri Mar 23, 2012 7:58 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by xeno74 »

Roland wrote: Sun Oct 03, 2021 12:47 pm The problem is present in sb 12. "git bisect bad"

OK, "git bisect bad". Output:

Code: Select all

[c0df236e1394970f3503a8fb103de95d000014ca] bpf: Change bpf_tcp_*_syncookie to accept ARG_PTR_TO_BTF_ID_SOCK_COMMON
Download: uImage-rb13
http://www.amigalinux.org
http://www.supertuxkart-amiga.de

Running Linux on AmigaONEs can require some tinkering.
Roland
Posts: 824
Joined: Tue May 02, 2017 7:23 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by Roland »

xeno74 wrote: Sun Oct 03, 2021 4:13 pm Download: uImage-rb13
The issue is visible in rb13, too. "git bisect bad"
- Roland -
User avatar
xeno74
Posts: 9320
Joined: Fri Mar 23, 2012 7:58 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by xeno74 »

Roland wrote: Sun Oct 03, 2021 5:12 pm The issue is visible in rb13, too. "git bisect bad"
OK, "git bisect bad". Output:

Code: Select all

[592a3498648af000e93dff2d36229ab11cd8c7f6] bpf: Change bpf_sk_storage_*() to accept ARG_PTR_TO_BTF_ID_SOCK_COMMON
Download: uImage-rb14
http://www.amigalinux.org
http://www.supertuxkart-amiga.de

Running Linux on AmigaONEs can require some tinkering.
Roland
Posts: 824
Joined: Tue May 02, 2017 7:23 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by Roland »

xeno74 wrote: Sun Oct 03, 2021 7:32 pm Download: uImage-rb14
The issue is present in rb14. "git bisect bad"
- Roland -
User avatar
xeno74
Posts: 9320
Joined: Fri Mar 23, 2012 7:58 am

Re: Display corruption with Kernel 5.10 and 5.12

Post by xeno74 »

Roland wrote: Mon Oct 04, 2021 12:07 pm The issue is present in rb14. "git bisect bad"
Thanks for testing! OK, "git bisect bad". Output:

Code: Select all

592a3498648af000e93dff2d36229ab11cd8c7f6 is the first bad commit
commit 592a3498648af000e93dff2d36229ab11cd8c7f6
Author: Martin KaFai Lau <kafai@fb.com>
Date:   Thu Sep 24 17:04:02 2020 -0700

    bpf: Change bpf_sk_storage_*() to accept ARG_PTR_TO_BTF_ID_SOCK_COMMON
    
    This patch changes the bpf_sk_storage_*() to take
    ARG_PTR_TO_BTF_ID_SOCK_COMMON such that they will work with the pointer
    returned by the bpf_skc_to_*() helpers also.
    
    A micro benchmark has been done on a "cgroup_skb/egress" bpf program
    which does a bpf_sk_storage_get().  It was driven by netperf doing
    a 4096 connected UDP_STREAM test with 64bytes packet.
    The stats from "kernel.bpf_stats_enabled" shows no meaningful difference.
    
    The sk_storage_get_btf_proto, sk_storage_delete_btf_proto,
    btf_sk_storage_get_proto, and btf_sk_storage_delete_proto are
    no longer needed, so they are removed.
    
    Signed-off-by: Martin KaFai Lau <kafai@fb.com>
    Signed-off-by: Alexei Starovoitov <ast@kernel.org>
    Acked-by: Lorenz Bauer <lmb@cloudflare.com>
    Link: https://lore.kernel.org/bpf/20200925000402.3856307-1-kafai@fb.com

:040000 040000 19ce1b29a6b4dc97f460c6978ae79db8d05ae254 f2eb6cc62df6bd1a44183589f444c506e5f45437 M	include
:040000 040000 02401e0c3d0515bc7ffc3aec2d01885477d9df16 b601a0afbb254a0c431f55d602ef4f29bd9f2fc2 M	kernel
:040000 040000 137c0c5f284595759964b23f96cf4d72f74eb0d9 ba6f2f4a729d5de69793cffa9c05ed23beffdb48 M	net
:040000 040000 f83e0116c20d5993d9b632d322ad85bfcbacb123 5ca5db317985395a77aa8f0bd6c51202673b86f0 M	tools
Result 592a3498648af000e93dff2d36229ab11cd8c7f6 is the first bad commit.

I was able to revert the first bad commit with the following commands.

Code: Select all

git checkout 592a3498648af000e93dff2d36229ab11cd8c7f6

Code: Select all

git revert 592a3498648af000e93dff2d36229ab11cd8c7f6
Please test the rb15.

Download: uImage-rb15

Thanks
http://www.amigalinux.org
http://www.supertuxkart-amiga.de

Running Linux on AmigaONEs can require some tinkering.
Post Reply