[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <34e3e90e.1f22.19be37eea29.Coremail.slark_xiao@163.com>
Date: Thu, 22 Jan 2026 10:18:24 +0800 (CST)
From: "Slark Xiao" <slark_xiao@....com>
To: "Gustavo A. R. Silva" <gustavo@...eddedor.com>,
"Loic Poulain" <loic.poulain@...aro.org>
Cc: ryazanov.s.a@...il.com, johannes@...solutions.net, andrew+netdev@...n.ch,
davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org,
pabeni@...hat.com, gustavoars@...nel.org, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-hardening@...r.kernel.org
Subject: Re:Re: [net] Revert "net: wwan: mhi_wwan_mbim: Avoid
-Wflex-array-member-not-at-end warning"
At 2026-01-22 10:14:41, "Gustavo A. R. Silva" <gustavo@...eddedor.com> wrote:
>
>
>On 1/22/26 10:30, Slark Xiao wrote:
>>
>>
>> At 2026-01-20 17:28:42, "Gustavo A. R. Silva" <gustavo@...eddedor.com> wrote:
>>>
>>>
>>> On 1/20/26 18:16, Slark Xiao wrote:
>>>>
>>>>
>>>> At 2026-01-20 15:51:57, "Gustavo A. R. Silva" <gustavo@...eddedor.com> wrote:
>>>>> Hi Slark,
>>>>>
>>>>> On 1/20/26 16:20, Slark Xiao wrote:
>>>>>> This reverts commit eeecf5d3a3a484cedfa3f2f87e6d51a7390ed960.
>>>>>>
>>>>>> This change lead to MHI WWAN device can't connect to internet.
>>>>>> I found a netwrok issue with kernel 6.19-rc4, but network works
>>>>>> well with kernel 6.18-rc1. After checking, this commit is the
>>>>>> root cause.
>>>>>
>>>>> Thanks for the report.
>>>>>
>>>>> Could you please apply the following patch on top of this revert,
>>>>> and let us know if the problem still manifests? Thank you!
>>>>>
>>>>> diff --git a/drivers/net/wwan/mhi_wwan_mbim.c b/drivers/net/wwan/mhi_wwan_mbim.c
>>>>> index 1d7e3ad900c1..a271a72fed63 100644
>>>>> --- a/drivers/net/wwan/mhi_wwan_mbim.c
>>>>> +++ b/drivers/net/wwan/mhi_wwan_mbim.c
>>>>> @@ -78,9 +78,12 @@ struct mhi_mbim_context {
>>>>>
>>>>> struct mbim_tx_hdr {
>>>>> struct usb_cdc_ncm_nth16 nth16;
>>>>> - struct usb_cdc_ncm_ndp16 ndp16;
>>>>> - struct usb_cdc_ncm_dpe16 dpe16[2];
>>>>> + __TRAILING_OVERLAP(struct usb_cdc_ncm_ndp16, ndp16, dpe16, __packed,
>>>>> + struct usb_cdc_ncm_dpe16 dpe16[2];
>>>>> + );
>>>>> } __packed;
>>>>> +static_assert(offsetof(struct mbim_tx_hdr, ndp16.dpe16) ==
>>>>> + offsetof(struct mbim_tx_hdr, dpe16));
>>>>>
>>>>> static struct mhi_mbim_link *mhi_mbim_get_link_rcu(struct mhi_mbim_context *mbim,
>>>>> unsigned int session)
>>>> This patch won't introduce previous problem.
>>>>
>>>
>>> Thanks for confirming this.
>>>
>>> I'll turn it into a proper patch.
>>>
>>> -Gustavo
>> Hi,
>> Are you planing to add this revert commit to your coming patch?
>
>No. I think the revert should be applied by the maintainers, first.
>
>Thanks
>-Gustavo
>
Hi Loic,
Could you help check this commit status?
Thanks
>> Actually it's urgent to revert this change since the release deadline of v6.19 is coming soon
>> and all my MHI WWAN device have been impacted.
>> I don't know why this commit has been tagged as "Not Applicable" and who did this.
>>
>> I hope we can fix it regression before the official v6.19 release.
>>
>> Thanks
Powered by blists - more mailing lists