Received-SPF: pass (google.com: domain of linux-kernel+bounces-101466-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Message-ID: <ccb4d341-fe9e-4920-8fd5-7467aadd2ea6@mydatex.cz>
Date: Wed, 13 Mar 2024 13:21:11 +0100
Precedence: bulk
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: Regressions can bus with MCP2515 on sama5d3 connected over SPI
 stop working on kernel > 5.4.271
Content-Language: cs-CZ, en-US
To: Marc Kleine-Budde <mkl@pengutronix.de>
Cc: "Linux regression tracking (Thorsten Leemhuis)"
 <regressions@leemhuis.info>,
 Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>,
 Linux regressions mailing list <regressions@lists.linux.dev>,
 linux-can@vger.kernel.org, LKML <linux-kernel@vger.kernel.org>
References: <2ede8f49-41f8-4d9a-83e4-ea5d4d1677fe@mydatex.cz>
 <a2e64be0-e07d-4c55-aba7-87c7e4c876e0@leemhuis.info>
 <734cf096-3769-4610-b72f-394c31a8d942@mydatex.cz>
 <91d8a6b6-6186-4aa0-8462-56b4751854e9@leemhuis.info>
 <6918321b-038d-40b9-8149-d535bf9d3d52@mydatex.cz>
 <5f69d6ee-a07a-42a9-a238-7dbe1f82cc3f@mydatex.cz>
 <3e46d70b-196d-45c7-bfdf-869c78cdc81f@leemhuis.info>
 <20240313-polio-jinx-bc5fd5df7c06-mkl@pengutronix.de>
 <b6c9b015-86bc-475b-a190-e35fa76c1ceb@mydatex.cz>
 <20240313-overload-eloquence-3e504ee0d00b-mkl@pengutronix.de>
From: Daniel Smolik <smolik@mydatex.cz>
In-Reply-To: <20240313-overload-eloquence-3e504ee0d00b-mkl@pengutronix.de>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Hm,
that's what I was worried about.  Main problem is that I can't easy 
compile kernel around version 5.5 because that always failed with this:

   HOSTLD  scripts/dtc/dtc
/usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x20): multiple 
definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first 
defined here
collect2: error: ld returned 1 exit status
make[1]: *** [scripts/Makefile.host:116: scripts/dtc/dtc] Error 1
make: *** [Makefile:1260: scripts_dtc] Error 2
make: *** Waiting for unfinished jobs....

I solve this problem  copying  whole scripts dir from 5.5.14 where this 
issue was solved.
After this  git reset --hard and after this   git bisect good | bad.

Prepare with this:

git clone -o mainline --no-checkout \
   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git  linux/
cd  linux/
git remote add -t master stable \
   https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
git checkout --detach v5.4


git bisect start
git bisect good v5.4
git bisect bad v5.5

On each iteration I do git reset --hard and copy sctipt direcotry.

Regards
                 Dan


Dne 13. 03. 24 v 12:43 Marc Kleine-Budde napsal(a):
> On 13.03.2024 12:22:25, Daniel Smolik wrote:
>> this is my first bisection please be lenient :-) Threse is a result:
>>
>> marvin@marvin:/usr/src/linux$ git bisect good
>> Bisecting: 0 revisions left to test after this (roughly 0 steps)
>> [52f96cd135b160d44db4cb62a5b614b3bca20fbc] net: stmmac: xgmac: Remove
>> uneeded computation for RFA/RFD
> I think you still miss a bisection step.
>
> The commit 52f96cd135b1 ("net: stmmac: xgmac: Remove uneeded computation
> for RFA/RFD") is unrelated to CAN problem you described. It's in the
> stmmac Ethernet driver.
>
> Marc
>