netdev - Re: [PATCH] brcmfmac: sdio: Fix OOB interrupt initialization on brcm43362

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: <141f055a-cd1d-66cb-7052-007cda629d3a@gmail.com>
Date:   Thu, 26 Dec 2019 17:37:58 +0300
From:   Dmitry Osipenko <digetx@...il.com>
To:     Arend Van Spriel <arend.vanspriel@...adcom.com>,
        Jean-Philippe Brucker <jean-philippe@...aro.org>,
        linux-wireless@...r.kernel.org,
        brcm80211-dev-list.pdl@...adcom.com,
        brcm80211-dev-list@...ress.com, netdev@...r.kernel.org,
        hdegoede@...hat.com
Cc:     franky.lin@...adcom.com, hante.meuleman@...adcom.com,
        chi-hsien.lin@...ress.com, wright.feng@...ress.com,
        kvalo@...eaurora.org, davem@...emloft.net
Subject: Re: [PATCH] brcmfmac: sdio: Fix OOB interrupt initialization on
 brcm43362

26.12.2019 12:47, Arend Van Spriel пишет:
> On December 26, 2019 10:23:41 AM Jean-Philippe Brucker
> <jean-philippe@...aro.org> wrote:
> 
>> Commit 262f2b53f679 ("brcmfmac: call brcmf_attach() just before calling
>> brcmf_bus_started()") changed the initialization order of the brcmfmac
>> SDIO driver. Unfortunately since brcmf_sdiod_intr_register() is now
>> called before the sdiodev->bus_if initialization, it reads the wrong
>> chip ID and fails to initialize the GPIO on brcm43362. Thus the chip
>> cannot send interrupts and fails to probe:
>>
>> [   12.517023] brcmfmac: brcmf_sdio_bus_rxctl: resumed on timeout
>> [   12.531214] ieee80211 phy0: brcmf_bus_started: failed: -110
>> [   12.536976] ieee80211 phy0: brcmf_attach: dongle is not responding:
>> err=-110
>> [   12.566467] brcmfmac: brcmf_sdio_firmware_callback: brcmf_attach
>> failed
>>
>> Initialize the bus interface earlier to ensure that
>> brcmf_sdiod_intr_register() properly sets up the OOB interrupt.
>>
>> BugLink: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=908438
>> Fixes: 262f2b53f679 ("brcmfmac: call brcmf_attach() just before
>> calling brcmf_bus_started()")
> 
> Reviewed-by: Arend van Spriel <arend.vanspriel@...adcom.com>
> 
>> Signed-off-by: Jean-Philippe Brucker <jean-philippe@...aro.org>
>> ---
>> A workaround [1] disabling the OOB interrupt is being discussed. It
>> works for me, but this patch fixes the wifi problem on my cubietruck.
> 
> I missed that one. Too bad it was not sent to linux-wireless as well.
> Good find here. I did see another patch dealing with the OOB interrupt
> on Nvidia Tegra. Now I wonder if this is the same issue.
> 
> Regards,
> Arend
> 
>> [1]
>> https://lore.kernel.org/linux-arm-kernel/20180930150927.12076-1-hdegoede@redhat.com/
>>
>> ---
>> .../net/wireless/broadcom/brcm80211/brcmfmac/sdio.c  | 12 ++++++------
>> 1 file changed, 6 insertions(+), 6 deletions(-)

I haven't seen any driver probe failures due to OOB on NVIDIA Tegra,
only suspend-resume was problematic due to the unbalanced OOB
interrupt-wake enabling.

But maybe checking whether OOB interrupt-wake works by invoking
enable_irq_wake() during brcmf_sdiod_intr_register() causes trouble for
the cubietruck board.

@Jean-Philippe, could you please try this change (on top of recent
linux-next):

diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
index b684a5b6d904..80d7106b10a9 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/bcmsdh.c
@@ -115,13 +115,6 @@ int brcmf_sdiod_intr_register(struct brcmf_sdio_dev
*sdiodev)
                }
                sdiodev->oob_irq_requested = true;

-               ret = enable_irq_wake(pdata->oob_irq_nr);
-               if (ret != 0) {
-                       brcmf_err("enable_irq_wake failed %d\n", ret);
-                       return ret;
-               }
-               disable_irq_wake(pdata->oob_irq_nr);
-
                sdio_claim_host(sdiodev->func1);

                if (sdiodev->bus_if->chip == BRCM_CC_43362_CHIP_ID) {