netdev - Re: [EXT] Re: [PATCH net] bnx2x: fix built-in kernel driver load failure

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <d0955a71-4b29-d915-f2ff-c90e8776eb41@molgen.mpg.de>
Date:   Thu, 17 Mar 2022 17:28:09 +0100
From:   Paul Menzel <pmenzel@...gen.mpg.de>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     Manish Chopra <manishc@...vell.com>, netdev@...r.kernel.org,
        Ariel Elior <aelior@...vell.com>, regressions@...ts.linux.dev,
        stable@...r.kernel.org, it+netdev@...gen.mpg.de
Subject: Re: [EXT] Re: [PATCH net] bnx2x: fix built-in kernel driver load
 failure

Dear Jakub,


Am 17.03.22 um 16:33 schrieb Jakub Kicinski:
> On Thu, 17 Mar 2022 14:31:45 +0100 Paul Menzel wrote:
>>>> I think it’s important to document, that the firmware was not present in the
>>>> initrd.
>>>
>>> I believe this problem has nothing to do with initrd module/FW but
>>> rather a module built in the kernel/vmlinuz (CONFIG_BNX2X=y) itself,
>>> A module load from initrd works fine and can access the initrd FW
>>> files present in initrd file system even during the probe. For
>>> example, when I had CONFIG_BNX2X=m, it loads the module fine from
>>> initrd with FW files present in initrd file system. When I had
>>> CONFIG_BNX2X=y, which I believe doesn't install/load module in/from
>>> initrd but in kernel (vmlinuz) itself, that's where it can't access
>>> the firmware file and cause the load failure.
>>
>> I can only say, that adding the firmware to the initrd worked around the
>> problem on our side with `CONFIG_BNX2X=y`.
> 
> I'd like to ship this one to Linus today. It sounds like it's
> okay from functional perspective, can I improve the commit message as
> you were suggesting and leave the comment / print improvements to a
> later patch?

Sure, fine by me.


Kind regards,

Paul