netdev - Re: [PATCH net-next v2 06/10] drivers/net/ethernet: handle one warning explicitly

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

Message-ID: <e15b85af416c7257aaa601901b18c7c9bc9586e0.camel@kernel.org>
Date:   Mon, 14 Sep 2020 20:25:09 -0700
From:   Saeed Mahameed <saeed@...nel.org>
To:     Jesse Brandeburg <jesse.brandeburg@...el.com>,
        netdev@...r.kernel.org
Cc:     intel-wired-lan@...ts.osuosl.org
Subject: Re: [PATCH net-next v2 06/10] drivers/net/ethernet: handle one
 warning explicitly

On Mon, 2020-09-14 at 18:44 -0700, Jesse Brandeburg wrote:
> While fixing the W=1 builds, this warning came up because the
> developers used a very tricky way to get structures initialized
> to a non-zero value, but this causes GCC to warn about an
> override. In this case the override was intentional, so just
> disable the warning for this code with a macro that results
> in disabling the warning for compiles on GCC versions after 8.
> 
> NOTE: the __diag_ignore macro currently only accepts a second
> argument of 8 (version 80000)
> 
> Signed-off-by: Jesse Brandeburg <jesse.brandeburg@...el.com>
> ---
>  drivers/net/ethernet/renesas/sh_eth.c | 10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/drivers/net/ethernet/renesas/sh_eth.c
> b/drivers/net/ethernet/renesas/sh_eth.c
> index 586642c33d2b..c63304632935 100644
> --- a/drivers/net/ethernet/renesas/sh_eth.c
> +++ b/drivers/net/ethernet/renesas/sh_eth.c
> @@ -45,6 +45,15 @@
>  #define SH_ETH_OFFSET_DEFAULTS			\
>  	[0 ... SH_ETH_MAX_REGISTER_OFFSET - 1] = SH_ETH_OFFSET_INVALID
>  
> +/* use some intentionally tricky logic here to initialize the whole
> struct to
> + * 0xffff, but then override certain fields, requiring us to
> indicate that we
> + * "know" that there are overrides in this structure, and we'll need
> to disable
> + * that warning from W=1 builds. GCC has supported this option since
> 4.2.X, but
> + * the macros available to do this only define GCC 8.
> + */
> +__diag_push();
> +__diag_ignore(GCC, 8, "-Woverride-init",
> +	      "logic to initialize all and then override some is OK");
>  static const u16 sh_eth_offset_gigabit[SH_ETH_MAX_REGISTER_OFFSET] =
> {
>  	SH_ETH_OFFSET_DEFAULTS,
>  
> @@ -332,6 +341,7 @@ static const u16
> sh_eth_offset_fast_sh3_sh2[SH_ETH_MAX_REGISTER_OFFSET] = {
>  
>  	[TSU_ADRH0]	= 0x0100,
>  };
> +__diag_pop();
>  

I don't have any strong feeling against disabling compiler warnings,
but maybe the right thing to do here is to initialize the gaps to the
invalid value instead of pre-initializing the whole thing first and
then setting up the valid values on the 2nd pass.

I don't think there are too many gaps to fill, it is doable, so maybe
add this as a comment to this driver maintainer so they could pickup
the work from here.