From: zhichen@codeaurora.org
To: Adrian Chadd <adrian@freebsd.org>
Cc: Peter Oh <peter.oh@eero.com>, ath10k <ath10k@lists.infradead.org>,
linux-wireless@vger.kernel.org
Subject: Re: [PATCH RFC] Revert "ath10k: fix DMA related firmware crashes on multiple devices"
Date: Wed, 30 Oct 2019 14:28:41 +0800 [thread overview]
Message-ID: <fb2b3b1fc5b3fd00a23e97413777a21e@codeaurora.org> (raw)
In-Reply-To: <CAJ-VmonXbUQ=Gv9fBbpN+ez25c3Pz+xxLoL67etMdC0Q+bwyXg@mail.gmail.com>
On 2019-10-23 02:24, Adrian Chadd wrote:
> On Tue, 22 Oct 2019 at 10:17, Peter Oh <peter.oh@eero.com> wrote:
>>
>>
>> On 10/22/19 1:57 AM, Zhi Chen wrote:
>> > This reverts commit 76d164f582150fd0259ec0fcbc485470bcd8033e.
>> > PCIe hung issue was observed on multiple platforms. The issue was reproduced
>> > when DUT was configured as AP and associated with 50+ STAs.
>> >
>> > With PCIe protocol analyzer, we can see DMA Read crossing 4KB boundary when
>> > issue happened. It broke PCIe spec and caused PCIe stuck.
>>
>> How can you say value 0 (I believe it's 64 bytes) DMA burst size
>> causes
>> the symptom and 1 fixes it?
>
>
> +1 to this question.
>
> Also, shouldn't the DMA engine be doing what the firmware says? Is the
> firmware/copy engine actually somehow bursting / prefetching across a
> 4K page boundary?
>
> Surely this is something that can be fixed in software/firmware by
> correctly configuring up buffer size/offsets?
>
>
>
> -Adrian
DMA engine is working as expected as the configuration. It's copy engine
which actually
splits the RD/WR requests and accesses host memory.
And yes it's platform related configuration. We have never hit this
issue on x86 platform.
Zhi
WARNING: multiple messages have this Message-ID (diff)
From: zhichen@codeaurora.org
To: Adrian Chadd <adrian@freebsd.org>
Cc: linux-wireless@vger.kernel.org, Peter Oh <peter.oh@eero.com>,
ath10k <ath10k@lists.infradead.org>
Subject: Re: [PATCH RFC] Revert "ath10k: fix DMA related firmware crashes on multiple devices"
Date: Wed, 30 Oct 2019 14:28:41 +0800 [thread overview]
Message-ID: <fb2b3b1fc5b3fd00a23e97413777a21e@codeaurora.org> (raw)
In-Reply-To: <CAJ-VmonXbUQ=Gv9fBbpN+ez25c3Pz+xxLoL67etMdC0Q+bwyXg@mail.gmail.com>
On 2019-10-23 02:24, Adrian Chadd wrote:
> On Tue, 22 Oct 2019 at 10:17, Peter Oh <peter.oh@eero.com> wrote:
>>
>>
>> On 10/22/19 1:57 AM, Zhi Chen wrote:
>> > This reverts commit 76d164f582150fd0259ec0fcbc485470bcd8033e.
>> > PCIe hung issue was observed on multiple platforms. The issue was reproduced
>> > when DUT was configured as AP and associated with 50+ STAs.
>> >
>> > With PCIe protocol analyzer, we can see DMA Read crossing 4KB boundary when
>> > issue happened. It broke PCIe spec and caused PCIe stuck.
>>
>> How can you say value 0 (I believe it's 64 bytes) DMA burst size
>> causes
>> the symptom and 1 fixes it?
>
>
> +1 to this question.
>
> Also, shouldn't the DMA engine be doing what the firmware says? Is the
> firmware/copy engine actually somehow bursting / prefetching across a
> 4K page boundary?
>
> Surely this is something that can be fixed in software/firmware by
> correctly configuring up buffer size/offsets?
>
>
>
> -Adrian
DMA engine is working as expected as the configuration. It's copy engine
which actually
splits the RD/WR requests and accesses host memory.
And yes it's platform related configuration. We have never hit this
issue on x86 platform.
Zhi
_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k
next prev parent reply other threads:[~2019-10-30 6:28 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-22 8:57 [PATCH RFC] Revert "ath10k: fix DMA related firmware crashes on multiple devices" Zhi Chen
2019-10-22 8:57 ` Zhi Chen
2019-10-22 10:07 ` Tom Psyborg
2019-10-22 10:07 ` Tom Psyborg
2019-10-30 2:44 ` zhichen
2019-10-30 2:44 ` zhichen
2019-11-06 10:01 ` Tom Psyborg
2019-11-06 10:01 ` Tom Psyborg
2019-11-07 5:35 ` zhichen
2019-11-07 5:35 ` zhichen
2019-10-22 17:16 ` Peter Oh
2019-10-22 17:16 ` Peter Oh
2019-10-22 18:24 ` Adrian Chadd
2019-10-22 18:24 ` Adrian Chadd
2019-10-30 6:04 ` Peter Oh
2019-10-30 6:04 ` Peter Oh
2019-10-30 6:28 ` zhichen [this message]
2019-10-30 6:28 ` zhichen
2019-10-30 6:16 ` zhichen
2019-10-30 6:16 ` zhichen
2019-10-30 23:01 ` Peter Oh
2019-10-30 23:01 ` Peter Oh
2019-11-03 15:36 ` Tom Psyborg
2019-11-03 15:36 ` Tom Psyborg
2019-11-04 18:41 ` Peter Oh
2019-11-04 18:41 ` Peter Oh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=fb2b3b1fc5b3fd00a23e97413777a21e@codeaurora.org \
--to=zhichen@codeaurora.org \
--cc=adrian@freebsd.org \
--cc=ath10k@lists.infradead.org \
--cc=linux-wireless@vger.kernel.org \
--cc=peter.oh@eero.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.