linux-kernel - Re: [PATCH net-next v1] net: devmem: NULL check netdev_nl_get_dma

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CAHS8izNdAdLv9_ExJFrFxJxpXr2VbHSE9-bTkAhQ4NVco9SSnA@mail.gmail.com>
Date: Tue, 2 Sep 2025 09:14:07 -0700
From: Mina Almasry <almasrymina@...gle.com>
To: Stanislav Fomichev <stfomichev@...il.com>
Cc: Jakub Kicinski <kuba@...nel.org>, netdev@...r.kernel.org, linux-kernel@...r.kernel.org, 
	"David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
	Paolo Abeni <pabeni@...hat.com>, Simon Horman <horms@...nel.org>, 
	Stanislav Fomichev <sdf@...ichev.me>
Subject: Re: [PATCH net-next v1] net: devmem: NULL check netdev_nl_get_dma_dev
 return value

On Tue, Sep 2, 2025 at 8:54 AM Stanislav Fomichev <stfomichev@...il.com> wrote:
>
> On 08/29, Jakub Kicinski wrote:
> > On Fri, 29 Aug 2025 21:59:38 +0000 Mina Almasry wrote:
> > > netdev_nl_get_dma_dev can return NULL. This happens in the unlikely
> > > scenario that netdev->dev.parent is NULL, or all the calls to the
> > > ndo_queue_get_dma_dev return NULL from the driver.
> >
> > I probably have Friday brain but I don't see what you mean..
> > In net-next net_devmem_bind_dmabuf() gets a dma_dev and returns
> > -EOPNOTSUPP PTR if its NULL.
>
> +1, the description and the fix are confusing.
>
> Unless I'm missing something, the intent seems to be to avoid hitting
> a WARN_ON in dma_buf_attach (really dma_buf_dynamic_attach) when the dma_dev
> is NULL. Mina, can we do this in the callers of netdev_queue_get_dma_dev?

Yes looks like I was the one with the Friday brain. I missed the NULL
check in net_devmem_bind_dmabuf. I'll check if this issue is on net
and send a fix there if needed.

-- 
Thanks,
Mina