linux-kernel - Re: [PATCH] tracing: Remove pointer (asterisk) from cpumask

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: <20221213090411.676021b8fff2c9bc95fba53f@kernel.org>
Date:   Tue, 13 Dec 2022 09:04:11 +0900
From:   Masami Hiramatsu (Google) <mhiramat@...nel.org>
To:     Steven Rostedt <rostedt@...dmis.org>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Linux Trace Kernel <linux-trace-kernel@...r.kernel.org>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Douglas Raillard <douglas.raillard@....com>,
        Valentin Schneider <vschneid@...hat.com>
Subject: Re: [PATCH] tracing: Remove pointer (asterisk) from cpumask_t field

On Mon, 12 Dec 2022 13:03:52 -0500
Steven Rostedt <rostedt@...dmis.org> wrote:

> From: "Steven Rostedt (Google)" <rostedt@...dmis.org>
> 
> To differentiate between long arrays and cpumasks, the __cpumask() field
> was created. Part of the TRACE_EVENT() macros test if the type is signed
> or not by using the is_signed_type() macro. The __cpumask() field used the
> __dynamic_array() helper but because cpumask_t is a structure, it could
> not be used in the is_signed_type() macro as that would fail to build, so
> instead it passed in the pointer to cpumask_t.
> 
> Unfortunately, that creates in the format file:
> 
>   field:__data_loc cpumask_t *[] mask;    offset:36;      size:4; signed:0;
> 
> Which looks like an array of pointers to cpumask_t and not a cpumask_t
> type, which is misleading to user space parsers.
> 
> Instead, create another helper called __dynamic_array_sign() that can be
> used directly by the __cpumaks() field macro where it passes in an
> unsigned value to avoid using the is_signed_type() and have the
> __dynamic_array() macro call it by passing it the is_signed_type() on the
> type.

What about renaming it as __dynamic_array_struct(), because the root issue
is that we can not pass the data structure to is_signed_tyep() macro?

Thank you,

> 
> This now produces:
> 
>   field:__data_loc cpumask_t[] mask;    offset:36;      size:4; signed:0;
> 
> Which is the correct type of the field.
> 
> Link: https://lore.kernel.org/lkml/6dda5e1d-9416-b55e-88f3-31d148bc925f@arm.com/
> 
> Fixes: 8230f27b1ccc ("tracing: Add __cpumask to denote a trace event field that is a cpumask_t")
> Reported-by: Douglas Raillard <douglas.raillard@....com>
> Signed-off-by: Steven Rostedt (Google) <rostedt@...dmis.org>
> ---
>  include/trace/stages/stage4_event_fields.h | 12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
> 
> diff --git a/include/trace/stages/stage4_event_fields.h b/include/trace/stages/stage4_event_fields.h
> index f2990d22313c..9f0049c309f0 100644
> --- a/include/trace/stages/stage4_event_fields.h
> +++ b/include/trace/stages/stage4_event_fields.h
> @@ -28,11 +28,15 @@
>  	.size = sizeof(_type[_len]), .align = ALIGN_STRUCTFIELD(_type),	\
>  	.is_signed = is_signed_type(_type), .filter_type = FILTER_OTHER },
>  
> -#undef __dynamic_array
> -#define __dynamic_array(_type, _item, _len) {				\
> +#undef __dynamic_array_sign
> +#define __dynamic_array_sign(_type, _item, _len, _sign) {		\
>  	.type = "__data_loc " #_type "[]", .name = #_item,		\
>  	.size = 4, .align = 4,						\
> -	.is_signed = is_signed_type(_type), .filter_type = FILTER_OTHER },
> +	.is_signed = _sign, .filter_type = FILTER_OTHER },
> +
> +#undef __dynamic_array
> +#define __dynamic_array(_type, _item, _len)				\
> +	__dynamic_array_sign(_type, item, len, is_signed_type(_type))
>  
>  #undef __string
>  #define __string(item, src) __dynamic_array(char, item, -1)
> @@ -47,7 +51,7 @@
>  #define __bitmask(item, nr_bits) __dynamic_array(unsigned long, item, -1)
>  
>  #undef __cpumask
> -#define __cpumask(item) __dynamic_array(cpumask_t *, item, -1)
> +#define __cpumask(item) __dynamic_array_sign(cpumask_t, item, -1, 0)
>  
>  #undef __sockaddr
>  #define __sockaddr(field, len) __dynamic_array(u8, field, len)
> -- 
> 2.35.1
> 


-- 
Masami Hiramatsu (Google) <mhiramat@...nel.org>