Received-SPF: pass (google.com: domain of linux-bluetooth+bounces-3823-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Message-ID: <e723ec51-a1de-4d4d-870f-34c9427fffb9@leemhuis.info>
Date: Mon, 22 Apr 2024 07:41:22 +0200
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: Bluetooth kernel BUG with Intel AX211 (regression in 6.1.83)
To: =?UTF-8?Q?Jeremy_Lain=C3=A9?= <jeremy.laine@m4x.org>,
 Paul Menzel <pmenzel@molgen.mpg.de>
Cc: linux-bluetooth@vger.kernel.org, regressions@lists.linux.dev
References: <CADRbXaDqx6S+7tzdDPPEpRu9eDLrHQkqoWTTGfKJSRxY=hT5MQ@mail.gmail.com>
 <8eeb980a-f04a-4e94-8065-25566cfef4dd@molgen.mpg.de>
 <CADRbXaD5Fsy0dxjrVG2JAn1PA36hpoTe6=23=zbhx8NWNf_9mg@mail.gmail.com>
From: "Linux regression tracking (Thorsten Leemhuis)"
 <regressions@leemhuis.info>
Content-Language: en-US, de-DE
Reply-To: Linux regressions mailing list <regressions@lists.linux.dev>
In-Reply-To: <CADRbXaD5Fsy0dxjrVG2JAn1PA36hpoTe6=23=zbhx8NWNf_9mg@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

On 22.04.24 01:17, Jeremy Lainé wrote:
> 
> On Sun, Apr 21, 2024 at 11:01 PM Paul Menzel <pmenzel@molgen.mpg.de> wrote:
>>
>> Would you be so kind to go the extra mail, and bisect the commit between
>> 6.1.82 and 6.1.83 [1]?
>>
> 
> Thanks for the link to the instructions, here's the bisect log:

Thx! Did you also test if mainline (e.g. 6.9-rc5) is affected? Without
this we won't know if this is something the stable team or the regular
bluetooth developers have to handle.

Ciao, Thorsten

> git bisect start
> # status: waiting for both good and bad commits
> # good: [d7543167affd372819a94879b8b1e8b9b12547d9] Linux 6.1.82
> git bisect good d7543167affd372819a94879b8b1e8b9b12547d9
> # status: waiting for bad commit, 1 good commit known
> # bad: [e5cd595e23c1a075359a337c0e5c3a4f2dc28dd1] Linux 6.1.83
> git bisect bad e5cd595e23c1a075359a337c0e5c3a4f2dc28dd1
> # bad: [440e278cb53b8dd6627c32e84950350083c39d35] net: kcm: fix
> incorrect parameter validation in the kcm_getsockopt) function
> git bisect bad 440e278cb53b8dd6627c32e84950350083c39d35
> # good: [a4116bd6ee5e1c1b65a61ed9221657615a2f45bf] arm64: dts:
> imx8mm-kontron: Disable pull resistors for SD card signals on BL OSM-S
> board
> git bisect good a4116bd6ee5e1c1b65a61ed9221657615a2f45bf
> # good: [e16c33dd9967b7f20987bf653acc4f605836127b] net: mctp: copy skb
> ext data when fragmenting
> git bisect good e16c33dd9967b7f20987bf653acc4f605836127b
> # bad: [6083089ab00631617f9eac678df3ab050a9d837a] Bluetooth: hci_conn:
> Consolidate code for aborting connections
> git bisect bad 6083089ab00631617f9eac678df3ab050a9d837a
> # good: [934212a623cbab851848b6de377eb476718c3e4c] SUNRPC: fix some
> memleaks in gssx_dec_option_array
> git bisect good 934212a623cbab851848b6de377eb476718c3e4c
> # good: [8499af0616cf76e6cbe811107e3f5b33bd472041] igb: Fix missing
> time sync events
> git bisect good 8499af0616cf76e6cbe811107e3f5b33bd472041
> # good: [653a17a99d752ffde175d4bc96154f2a3642f400] Bluetooth: Remove
> superfluous call to hci_conn_check_pending()
> git bisect good 653a17a99d752ffde175d4bc96154f2a3642f400
> # good: [1023de27cd1d0d692e70fe6d6d5cee9fff9b9c84] Bluetooth: Cancel
> sync command before suspend and power off
> git bisect good 1023de27cd1d0d692e70fe6d6d5cee9fff9b9c84
> # good: [ac7a47aaa7944efc94e4fc23cc438b7bd9cc222c] Bluetooth:
> hci_sync: Only allow hci_cmd_sync_queue if running
> git bisect good ac7a47aaa7944efc94e4fc23cc438b7bd9cc222c
> # first bad commit: [6083089ab00631617f9eac678df3ab050a9d837a]
> Bluetooth: hci_conn: Consolidate code for aborting connections
> 
> 
>> You can pipe the output through `scripts/decodecode` and it should show
>> more information.
> 
> This was the output of running the dmesg snippet through `scripts/decodecode`:
> 
> All code
> ========
>    0:   fe                      (bad)
>    1:   ff 0f                   decl   (%rdi)
>    3:   0b 48 89                or     -0x77(%rax),%ecx
>    6:   f2 48 89 fe             repnz mov %rdi,%rsi
>    a:   48 c7 c7 48 18 7a 9f    mov    $0xffffffff9f7a1848,%rdi
>   11:   e8 14 a1 fe ff          call   0xfffffffffffea12a
>   16:   0f 0b                   ud2
>   18:   48 89 fe                mov    %rdi,%rsi
>   1b:   48 89 ca                mov    %rcx,%rdx
>   1e:   48 c7 c7 10 18 7a 9f    mov    $0xffffffff9f7a1810,%rdi
>   25:   e8 00 a1 fe ff          call   0xfffffffffffea12a
>   2a:*  0f 0b                   ud2             <-- trapping instruction
>   2c:   48 89 fe                mov    %rdi,%rsi
>   2f:   48 c7 c7 d8 17 7a 9f    mov    $0xffffffff9f7a17d8,%rdi
>   36:   e8 ef a0 fe ff          call   0xfffffffffffea12a
>   3b:   0f 0b                   ud2
>   3d:   48 89 fe                mov    %rdi,%rsi
> 
> Code starting with the faulting instruction
> ===========================================
>    0:   0f 0b                   ud2
>    2:   48 89 fe                mov    %rdi,%rsi
>    5:   48 c7 c7 d8 17 7a 9f    mov    $0xffffffff9f7a17d8,%rdi
>    c:   e8 ef a0 fe ff          call   0xfffffffffffea100
>   11:   0f 0b                   ud2
>   13:   48 89 fe                mov    %rdi,%rsi
> 
> Best regards,
> Jeremy
> 
>