Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
To: Sergey Shtylyov <s.shtylyov@omp.ru>,
"davem@davemloft.net" <davem@davemloft.net>,
"kuba@kernel.org" <kuba@kernel.org>
Cc: "netdev@vger.kernel.org" <netdev@vger.kernel.org>,
"linux-renesas-soc@vger.kernel.org"
<linux-renesas-soc@vger.kernel.org>
Subject: RE: [PATCH] net: renesas: sh_eth: Fix freeing wrong tx descriptor
Date: Wed, 8 Sep 2021 11:43:24 +0000 [thread overview]
Message-ID: <TY2PR01MB36924BD44CA5B512B96AC2AFD8D49@TY2PR01MB3692.jpnprd01.prod.outlook.com> (raw)
In-Reply-To: <9f3a0195-8218-e73c-eb63-bbd7b6ff9777@omp.ru>
Hi Sergey,
> From: Sergey Shtylyov, Sent: Wednesday, September 8, 2021 6:46 PM
>
> On 08.09.2021 8:45, Yoshihiro Shimoda wrote:
>
> >>> The cur_tx counter must be incremented after TACT bit of
> >>> txdesc->status was set. However, a CPU is possible to reorder
> >>> instructions and/or memory accesses between cur_tx and
> >>> txdesc->status. And then, if TX interrupt happened at such a
> >>> timing, the sh_eth_tx_free() may free the descriptor wrongly.
> >>> So, add wmb() before cur_tx++.
> >>
> >> Not dma_wmb()? :-)
> >
> > On armv8, dma_wmb() is DMB OSHST, and wmb() is DSB ST.
> > IIUC, DMB OSHST is not affected the ordering of instructions.
> > So, we have to use wmb().
>
> I should really read up the ARM manuals on the barrier instructions... :-)
:)
> >>> Otherwise NETDEV WATCHDOG timeout is possible to happen.
> >>>
> >>> Fixes: 86a74ff21a7a ("net: sh_eth: add support for Renesas SuperH Ethernet")
> >>> Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
> >>
> >> Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
> >
> > Thank you for your review!
>
> Out of curiosity: have you really experienced the bug or found it by
> review?
Our team has really experienced the bug when iperf3 runs on both side(server and client),
and this patch could fix the issue.
Best regards,
Yoshihiro Shimoda
> > Best regards,
> > Yoshihiro Shimoda
>
> MBR, Sergey
prev parent reply other threads:[~2021-09-08 11:43 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-07 11:29 Yoshihiro Shimoda
2021-09-07 13:10 ` patchwork-bot+netdevbpf
2021-09-07 19:29 ` Sergey Shtylyov
2021-09-08 5:45 ` Yoshihiro Shimoda
2021-09-08 9:46 ` Sergey Shtylyov
2021-09-08 11:43 ` Yoshihiro Shimoda [this message]
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=TY2PR01MB36924BD44CA5B512B96AC2AFD8D49@TY2PR01MB3692.jpnprd01.prod.outlook.com \
--to=yoshihiro.shimoda.uh@renesas.com \
--cc=davem@davemloft.net \
--cc=kuba@kernel.org \
--cc=linux-renesas-soc@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=s.shtylyov@omp.ru \
--subject='RE: [PATCH] net: renesas: sh_eth: Fix freeing wrong tx descriptor' \
/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
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).