MIME-Version: 1.0
References: <CANnVG6=a3etRagG+RaSEH-b4_nfzxpEKffQtuMWrttrbgjunZQ@mail.gmail.com>
 <186e23b5668.279b.9b12b7fc0a3841636cfb5e919b41b954@broadcom.com>
 <CANnVG6kjWj02eEFv_OeLiRtjrJ6yn4EsELz_BtrzFHH15GNMLw@mail.gmail.com>
 <186e26dc0a8.279b.9b12b7fc0a3841636cfb5e919b41b954@broadcom.com>
 <CANnVG6n-eqKUQnX_6wncmjG1kyVfhxqs2L82xYQpDmGq89eVAQ@mail.gmail.com>
 <186e4673718.279b.9b12b7fc0a3841636cfb5e919b41b954@broadcom.com>
 <CANnVG6kaGj1SVCqf2y3=Xj_N2G9j+-VhLDN4_WY_ywDswNkO3g@mail.gmail.com>
 <89262772-7358-77e8-8913-c459819d544a@broadcom.com> <a459b9adc5eb7b67a432bd947d47d1df08718762.camel@sipsolutions.net>
 <f430c376-5992-d394-ce55-5994c793c7c7@broadcom.com> <CANnVG6neOX+Bgc3zWA8k=hwrifmXb=xBN95aMUEyJ=orz3+Wkg@mail.gmail.com>
 <186e9109ae8.279b.9b12b7fc0a3841636cfb5e919b41b954@broadcom.com>
 <CANnVG6kvP78+WADH+Z4Yk4k9m-rjwRw0qMprHNsngNaj-jpqJQ@mail.gmail.com> <20fc4c19-6f39-d3bf-b873-ea26a26175cd@broadcom.com>
In-Reply-To: <20fc4c19-6f39-d3bf-b873-ea26a26175cd@broadcom.com>
From:   Michael Stapelberg <michael+lkml@stapelberg.ch>
Date:   Fri, 17 Mar 2023 09:04:52 +0100
Message-ID: <CANnVG6m2re0AmZB8WTbGfNyWA0dy0KEx+nvidGsQ=E2LP0wsnQ@mail.gmail.com>
Subject: Re: wifi breakage due to commit "wifi: brcmfmac: add support for
 vendor-specific firmware api"
To:     Arend van Spriel <arend.vanspriel@broadcom.com>
Cc:     Johannes Berg <johannes@sipsolutions.net>,
        Kalle Valo <kvalo@kernel.org>,
        Franky Lin <franky.lin@broadcom.com>,
        Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com>,
        Hante Meuleman <hante.meuleman@broadcom.com>,
        linux-wireless@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Can you outline under what conditions the race can happen?

I=E2=80=99ll take care of implementing a userspace modprobe helper.

Thanks


On Thu, 16 Mar 2023 at 10:40, Arend van Spriel
<arend.vanspriel@broadcom.com> wrote:
>
> On 3/16/2023 8:54 AM, Michael Stapelberg wrote:
> > Hmm, I=E2=80=99m not sure either.
>
> Actually, the check I referred to only bails out when
> CONFIG_MODPROBE_PATH is unset. In your case it is set but it is not
> existing. I dived into request_module() and it ends up scheduling a work
> to call the modprobe which will fail. However, brcmfmac will wait for
> brcmfmac-wcc to register itself so doing the insmod of brcmfmac-wcc will
> do the trick. However, now I am back at saying this can be racy. It
> would be better to come up with a modprobe.go for your user-space
> environment and put that in kernel config.
>
> Regards,
> Arend
>