linux-kernel - Re: [PATCH v4 01/15] scripts: checkpatch: Add _

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: <59d9ac4e1f7475f51672222af1fc90d7ec23b5da.camel@perches.com>
Date: Sat, 16 Dec 2023 10:07:24 -0800
From: Joe Perches <joe@...ches.com>
To: Marcelo Schmitt <marcelo.schmitt@...log.com>, apw@...onical.com, 
 dwaipayanray1@...il.com, lukas.bulwahn@...il.com, paul.cercueil@...log.com,
  Michael.Hennerich@...log.com, lars@...afoo.de, jic23@...nel.org, 
 robh+dt@...nel.org, krzysztof.kozlowski+dt@...aro.org, conor+dt@...nel.org,
  dan.carpenter@...aro.org, dlechner@...libre.com,
 marcelo.schmitt1@...il.com,  Andrew Morton <akpm@...ux-foundation.org>
Cc: linux-iio@...r.kernel.org, devicetree@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 01/15] scripts: checkpatch: Add __aligned to the list
 of attribute notes

On Sat, 2023-12-16 at 14:45 -0300, Marcelo Schmitt wrote:
> Checkpatch presumes attributes marked with __aligned(alignment) are part
> of a function declaration and throws a warning stating that those
> compiler attributes should have an identifier name which is not correct.
> Add __aligned compiler attributes to the list of attribute notes
> so they don't cause warnings anymore.
> 
> Signed-off-by: Marcelo Schmitt <marcelo.schmitt@...log.com>
> ---
> Any expression that evaluates to an integer that is a power of 2 can be
> within __aligned parenthesis.
> I can't see how we could use a regex to check code meets such constraint (if possible).

You can't because if a #define is used for the alignment
value it is not necessarily available to a patch fragment
or even a file if the #define is in an #include.

> Some additional exotic uses of __aligned are:
> drivers/net/wireless/quantenna/qtnfmac/bus.h:72:   char bus_priv[] __aligned(sizeof(void *));
> include/linux/netdevice.h:225:} __aligned(4 * sizeof(unsigned long)); 

Right, then there are random uses like:

drivers/firmware/qcom/qcom_qseecom_uefisecapp.c:		size_t __aligned;						\
drivers/firmware/qcom/qcom_qseecom_uefisecapp.c:			*__offset = __aligned;					\

so there's always some false positive/negative issue
with checkpatch.

Anyway:

Acked-by: Joe Perches <joe@...ches.com>

> 
> The regex
> 			__aligned\s*\(.*\)
> seems to match all use cases. 
> 
> We might not catch invalid arguments to __aligned, but it looks like making
> checkpath confidently report those wouldn't be feasible anyway.
> 
> The patch that would trigger the mentioned warning in now
> patch number 13 (iio: adc: Add support for AD7091R-8).
> 
>  scripts/checkpatch.pl | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> index 25fdb7fda112..d56c98146da3 100755
> --- a/scripts/checkpatch.pl
> +++ b/scripts/checkpatch.pl
> @@ -512,6 +512,7 @@ our $Attribute	= qr{
>  			__ro_after_init|
>  			__kprobes|
>  			$InitAttribute|
> +			__aligned\s*\(.*\)|
>  			____cacheline_aligned|
>  			____cacheline_aligned_in_smp|
>  			____cacheline_internodealigned_in_smp|