Hi Linus,
Here's a PR with an MMC fix intended for v5.13-rc8. Details about the
highlights are as usual found in the signed tag.
Please pull this in!
Kind regards
Ulf Hansson
The following changes since commit 009c9aa5be652675a06d5211e1640e02bbb1c33d:
Linux 5.13-rc6 (2021-06-13 14:43:10 -0700)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc.git tags/mmc-v5.13-rc6
for you to fetch changes up to 103a5348c22c3fca8b96c735a9e353b8a0801842:
mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk (2021-06-14 14:02:33 +0200)
----------------------------------------------------------------
MMC host:
- meson-gx: Use memcpy_to/fromio for dram-access-quirk
----------------------------------------------------------------
Neil Armstrong (1):
mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk
drivers/mmc/host/meson-gx-mmc.c | 50 ++++++++++++++++++++++++++++++++++++-----
1 file changed, 45 insertions(+), 5 deletions(-)
On Thu, Jun 24, 2021 at 5:32 AM Ulf Hansson <[email protected]> wrote:
>
> Neil Armstrong (1):
> mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk
Ugh. How horribly ugly.
Why is that 'host->dram_access_quirk' test _inside_ the loop, rather
than be something like
if (host->dram_access_quirk)
return sg_copy_to_buffer(data->sg, data->sg_len,
host->bounce_buf, xfer_bytes);
at the top of the function, with meson_mmc_copy_buffer() then only
handling the mmio case?
No, I don't know this code, I'm just looking at the patch and going
"that looks really ugly".
Anyway, I've pulled it, but I thought I'd voice my reaction to it..
Linus
The pull request you sent on Thu, 24 Jun 2021 14:32:51 +0200:
> git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc.git tags/mmc-v5.13-rc6
has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/4a09d388f2ab382f217a764e6a152b3f614246f6
Thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html
+ Neil, Robin
On Thu, 24 Jun 2021 at 20:00, Linus Torvalds
<[email protected]> wrote:
>
> On Thu, Jun 24, 2021 at 5:32 AM Ulf Hansson <[email protected]> wrote:
> >
> > Neil Armstrong (1):
> > mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk
>
> Ugh. How horribly ugly.
>
> Why is that 'host->dram_access_quirk' test _inside_ the loop, rather
> than be something like
>
> if (host->dram_access_quirk)
> return sg_copy_to_buffer(data->sg, data->sg_len,
> host->bounce_buf, xfer_bytes);
>
> at the top of the function, with meson_mmc_copy_buffer() then only
> handling the mmio case?
>
> No, I don't know this code, I'm just looking at the patch and going
> "that looks really ugly".
>
> Anyway, I've pulled it, but I thought I'd voice my reaction to it..
Thanks, good point!
Robin and Neil did discuss some cleanup that could be made on top [1],
but it looks like that was targeting a different part.
In any case, Neil, would you mind sending a cleanup according to the
comment from Linus?
Kind regards
Uffe
[1]
https://www.spinics.net/lists/arm-kernel/msg901045.html
Hi,
On 28/06/2021 09:49, Ulf Hansson wrote:
> + Neil, Robin
>
> On Thu, 24 Jun 2021 at 20:00, Linus Torvalds
> <[email protected]> wrote:
>>
>> On Thu, Jun 24, 2021 at 5:32 AM Ulf Hansson <[email protected]> wrote:
>>>
>>> Neil Armstrong (1):
>>> mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk
>>
>> Ugh. How horribly ugly.
>>
>> Why is that 'host->dram_access_quirk' test _inside_ the loop, rather
>> than be something like
>>
>> if (host->dram_access_quirk)
>> return sg_copy_to_buffer(data->sg, data->sg_len,
>> host->bounce_buf, xfer_bytes);
>>
>> at the top of the function, with meson_mmc_copy_buffer() then only
>> handling the mmio case?
>>
>> No, I don't know this code, I'm just looking at the patch and going
>> "that looks really ugly".
>>
>> Anyway, I've pulled it, but I thought I'd voice my reaction to it..
>
> Thanks, good point!
>
> Robin and Neil did discuss some cleanup that could be made on top [1],
> but it looks like that was targeting a different part.
Yes this was a quick hack (and obviously not the optimal way to) to address the fatal system crash.
>
> In any case, Neil, would you mind sending a cleanup according to the
> comment from Linus?
Sure, as discussed with Robin, the plan is to introduce sg_copy_to_buffer_to/fromio
and remove this crappy hack, but this requires multiple subsystems involved.
>
> Kind regards
> Uffe
Thanks,
Neil
>
> [1]
> https://www.spinics.net/lists/arm-kernel/msg901045.html
>