netdev - Re: stable-rc: 6.1: mlx5: params.c:994:53: error: 'MLX5_IPSEC_CAP_CRYPTO' undeclared (first use in this function)

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: <2024013024-overripe-serve-4e45@gregkh>
Date: Tue, 30 Jan 2024 10:30:40 -0800
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: Leon Romanovsky <leonro@...dia.com>
Cc: Florian Fainelli <f.fainelli@...il.com>,
	Naresh Kamboju <naresh.kamboju@...aro.org>,
	linux-stable <stable@...r.kernel.org>,
	Netdev <netdev@...r.kernel.org>, linux-rdma@...r.kernel.org,
	lkft-triage@...ts.linaro.org, Sasha Levin <sashal@...nel.org>,
	Saeed Mahameed <saeedm@...dia.com>,
	"David S. Miller" <davem@...emloft.net>,
	Dan Carpenter <dan.carpenter@...aro.org>,
	Arnd Bergmann <arnd@...db.de>
Subject: Re: stable-rc: 6.1: mlx5: params.c:994:53: error:
 'MLX5_IPSEC_CAP_CRYPTO' undeclared (first use in this function)

On Tue, Jan 30, 2024 at 09:14:34AM +0200, Leon Romanovsky wrote:
> On Mon, Jan 29, 2024 at 08:25:42PM -0800, Florian Fainelli wrote:
> > 
> > 
> > On 1/29/2024 6:52 PM, Naresh Kamboju wrote:
> > > On Mon, 29 Jan 2024 at 21:58, Greg Kroah-Hartman
> > > <gregkh@...uxfoundation.org> wrote:
> > > > 
> > > > On Mon, Jan 29, 2024 at 09:17:31PM +0530, Naresh Kamboju wrote:
> > > > > Following build errors noticed on stable-rc linux-6.1.y for arm64.
> > > > > 
> > > > > arm64:
> > > > > --------
> > > > >    * build/gcc-13-lkftconfig
> > > > >    * build/gcc-13-lkftconfig-kunit
> > > > >    * build/clang-nightly-lkftconfig
> > > > >    * build/clang-17-lkftconfig-no-kselftest-frag
> > > > >    * build/gcc-13-lkftconfig-devicetree
> > > > >    * build/clang-lkftconfig
> > > > >    * build/gcc-13-lkftconfig-perf
> > > > > 
> > > > > Reported-by: Linux Kernel Functional Testing <lkft@...aro.org>
> > > > > 
> > > > > Build errors:
> > > > > ------
> > > > > drivers/net/ethernet/mellanox/mlx5/core/en/params.c: In function
> > > > > 'mlx5e_build_sq_param':
> > > > > drivers/net/ethernet/mellanox/mlx5/core/en/params.c:994:53: error:
> > > > > 'MLX5_IPSEC_CAP_CRYPTO' undeclared (first use in this function)
> > > > >    994 |                     (mlx5_ipsec_device_caps(mdev) &
> > > > > MLX5_IPSEC_CAP_CRYPTO);
> > > > >        |
> > > > > ^~~~~~~~~~~~~~~~~~~~~
> > > > > 
> > > > > Suspecting commit:
> > > > >    net/mlx5e: Allow software parsing when IPsec crypto is enabled
> > > > >    [ Upstream commit 20f5468a7988dedd94a57ba8acd65ebda6a59723 ]
> > > > 
> > > > Something looks very odd here, as the proper .h file is being included,
> > > > AND this isn't a build failure on x86, so why is this only arm64 having
> > > > problems?  What's causing this not to show up?
> > > 
> > > As per the Daniel report on stable-rc review on 6.1, these build failures also
> > > reported on System/390.
> > 
> > The build failure is legitimate here since
> > drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h guards all of the
> > definitions and enumerations under a CONFIG_MLX5_EN_IPSEC which is not
> > enabled in the build configuration that failed.
> > 
> > This is implicitly fixed upstream with
> > 8c582ddfbb473c1d799c40b5140aed81278e2837 ("net/mlx5e: Handle hardware IPsec
> > limits events") which relocates the #ifdef CONFIG_MLX5_EN_IPSEC below and
> > allows the MLX5_IPSEC_CAP_CRYPTO enum value, amongst others to be visible to
> > code that is not guarded with CONFIG_MLX5_EN_IPSEC. This specific commit
> > does not apply cleanly to the stable-6.1 branch, so maybe the best we can
> > come up with is this targeted change that does the same thing against 6.1:
> 
> Thanks for taking look into it. This fix looks as a best solution for me.

Thanks, will queue this up now and push out a new -rc