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  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:   Sun, 1 Aug 2021 19:23:37 -0700
From:   Axel Rasmussen <axel.rasmussen1@...il.com>
To:     "ojab //" <ojab@...b.ru>
Cc:     Kalle Valo <kvalo@...eaurora.org>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>, ath10k@...ts.infradead.org,
        Linux Wireless <linux-wireless@...r.kernel.org>,
        netdev@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH V2] ath10k: don't fail if IRAM write fails

On Thu, Jul 22, 2021 at 12:42 PM ojab // <ojab@...b.ru> wrote:
>
> See also: https://lists.infradead.org/pipermail/ath10k/2021-May/012626.html
>
> On Thu, 22 Jul 2021 at 22:36, ojab <ojab@...b.ru> wrote:
> >
> > After reboot with kernel & firmware updates I found `failed to copy
> > target iram contents:` in dmesg and missing wlan interfaces for both
> > of my QCA9984 compex cards. Rolling back kernel/firmware didn't fixed
> > it, so while I have no idea what's actually happening, I don't see why
> > we should fail in this case, looks like some optional firmware ability
> > that could be skipped.
> >
> > Also with additional logging there is
> > ```
> > [    6.839858] ath10k_pci 0000:04:00.0: No hardware memory
> > [    6.841205] ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
> > [    6.873578] ath10k_pci 0000:07:00.0: No hardware memory
> > [    6.875052] ath10k_pci 0000:07:00.0: failed to copy target iram contents: -12
> > ```
> > so exact branch could be seen.
> >
> > Signed-off-by: Slava Kardakov <ojab@...b.ru>
> > ---
> >  Of course I forgot to sing off, since I don't use it by default because I
> >  hate my real name and kernel requires it

Thanks for working on this! And sorry for the slow response. I've been
unexpectedly very busy lately, but I plan to test out this patch next
week.

> >
> >  drivers/net/wireless/ath/ath10k/core.c | 9 ++++++---
> >  1 file changed, 6 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/net/wireless/ath/ath10k/core.c b/drivers/net/wireless/ath/ath10k/core.c
> > index 2f9be182fbfb..d9fd5294e142 100644
> > --- a/drivers/net/wireless/ath/ath10k/core.c
> > +++ b/drivers/net/wireless/ath/ath10k/core.c
> > @@ -2691,8 +2691,10 @@ static int ath10k_core_copy_target_iram(struct ath10k *ar)
> >         u32 len, remaining_len;
> >
> >         hw_mem = ath10k_coredump_get_mem_layout(ar);
> > -       if (!hw_mem)
> > +       if (!hw_mem) {
> > +               ath10k_warn(ar, "No hardware memory");
> >                 return -ENOMEM;
> > +       }
> >
> >         for (i = 0; i < hw_mem->region_table.size; i++) {
> >                 tmp = &hw_mem->region_table.regions[i];
> > @@ -2702,8 +2704,10 @@ static int ath10k_core_copy_target_iram(struct ath10k *ar)
> >                 }
> >         }
> >
> > -       if (!mem_region)
> > +       if (!mem_region) {
> > +               ath10k_warn(ar, "No memory region");
> >                 return -ENOMEM;
> > +       }
> >
> >         for (i = 0; i < ar->wmi.num_mem_chunks; i++) {
> >                 if (ar->wmi.mem_chunks[i].req_id ==
> > @@ -2917,7 +2921,6 @@ int ath10k_core_start(struct ath10k *ar, enum ath10k_firmware_mode mode,
> >                 if (status) {
> >                         ath10k_warn(ar, "failed to copy target iram contents: %d",
> >                                     status);
> > -                       goto err_hif_stop;
> >                 }
> >         }
> >
> > --
> > 2.32.0

Powered by blists - more mailing lists