From: zhoumin <zhoumin@loongson.cn>
To: "bibo, mao" <maobibo@loongson.cn>,
qiming.yang@intel.com, wenjun1.wu@intel.com
Cc: dev@dpdk.org
Subject: Re: [PATCH 1/1] net/ixgbe: add a proper memory barrier for LoongArch
Date: Mon, 24 Apr 2023 20:30:26 +0800 [thread overview]
Message-ID: <07cd7dde-89ca-663c-08b4-9a83b836f275@loongson.cn> (raw)
In-Reply-To: <c4d4e0fd-3810-f35e-b8af-0d087dcf0456@loongson.cn>
Hi Bibo,
On Sat, Apr 22, 2023 at 8:29AM, bibo, mao wrote:
>
>
> 在 2023/4/21 9:12, zhoumin 写道:
>> On Fri, Apr 7, 2023 at 4:50PM, Min Zhou wrote:
>>> Segmentation fault has been observed while running the
>>> ixgbe_recv_pkts_lro() function to receive packets on the Loongson
>>> 3C5000 processor which has 64 cores and 4 NUMA nodes.
>>>
>>> Reason is the read ordering of the status and the rest of the
>>> descriptor fields in this function may not be correct on the
>>> LoongArch processor. We should add rte_rmb() to ensure the read
>>> ordering be correct.
>>>
>>> We also did the same thing in the ixgbe_recv_pkts() function.
>>>
>>> Signed-off-by: Min Zhou <zhoumin@loongson.cn>
>>> ---
>>> drivers/net/ixgbe/ixgbe_rxtx.c | 6 ++++++
>>> 1 file changed, 6 insertions(+)
>>>
>>> diff --git a/drivers/net/ixgbe/ixgbe_rxtx.c
>>> b/drivers/net/ixgbe/ixgbe_rxtx.c
>>> index c9d6ca9efe..16391a42f9 100644
>>> --- a/drivers/net/ixgbe/ixgbe_rxtx.c
>>> +++ b/drivers/net/ixgbe/ixgbe_rxtx.c
>>> @@ -1823,6 +1823,9 @@ ixgbe_recv_pkts(void *rx_queue, struct
>>> rte_mbuf **rx_pkts,
>>> staterr = rxdp->wb.upper.status_error;
>>> if (!(staterr & rte_cpu_to_le_32(IXGBE_RXDADV_STAT_DD)))
>>> break;
>>> +#if defined(RTE_ARCH_LOONGARCH)
>>> + rte_rmb();
>>> +#endif
>>> rxd = *rxdp;
> Hi Min,
>
> Could you add more detailed analysis aboout the issu? Althrough rxdp
> is declared as volatile, which is only in order for compiler. However
> some architectures like LoongArch are weak-ordered. For this piece of
> code:
>
> 1: staterr = rxdp->wb.upper.status_error;
> Sentence 1 can be execute after sentence 1, dd indicated that packet
> is ready with new value.
>
> 2: rxd = *rxdp;
> Sentence 2 can be execute first and get old value.
>
> .......Balabala
>
>
> Regards
> Bibo, Mao
>
Thanks for your kind comments.
I have sent the v2 patch and give the more detailed analysis for the
segmentation fault.
Regards
Min
>>> /*
>>> @@ -2122,6 +2125,9 @@ ixgbe_recv_pkts_lro(void *rx_queue, struct
>>> rte_mbuf **rx_pkts, uint16_t nb_pkts,
>>> if (!(staterr & IXGBE_RXDADV_STAT_DD))
>>> break;
>>> +#if defined(RTE_ARCH_LOONGARCH)
>>> + rte_rmb();
>>> +#endif
>>> rxd = *rxdp;
>>> PMD_RX_LOG(DEBUG, "port_id=%u queue_id=%u rx_id=%u "
>>
>> Kindly ping.
>>
>> Any comments or suggestions will be appreciated.
>>
>>
>> Min
>>
prev parent reply other threads:[~2023-04-24 12:30 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-07 8:50 Min Zhou
2023-04-21 1:12 ` zhoumin
2023-04-22 0:29 ` bibo, mao
2023-04-24 12:30 ` zhoumin [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=07cd7dde-89ca-663c-08b4-9a83b836f275@loongson.cn \
--to=zhoumin@loongson.cn \
--cc=dev@dpdk.org \
--cc=maobibo@loongson.cn \
--cc=qiming.yang@intel.com \
--cc=wenjun1.wu@intel.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 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).