[PATCH] nilfs2: Use __field_struct() for a bitwise field

[PATCH] nilfs2: Use __field_struct() for a bitwise field

[Bart Van Assche]

As one can see in include/trace/stages/stage4_event_fields.h, the
implementation of __field() uses the is_signed_type() macro. As one can see
in commit dcf8e5633e2e ("tracing: Define the is_signed_type() macro once"),
there has been an attempt to not make is_signed_type() trigger sparse
warnings for bitwise types. Despite that change, sparse complains when
passing a bitwise type to is_signed_type(). It is not clear to me why.

Follow the example of <trace/events/initcall.h> and suppress the following
sparse warnings by changing __field() into __field_struct():

 fs/nilfs2/segment.c: note: in included file (through
   include/trace/trace_events.h, include/trace/define_trace.h,
   include/trace/events/nilfs2.h):
 ./include/trace/events/nilfs2.h:191:1: warning: cast to restricted
   blk_opf_t
 ./include/trace/events/nilfs2.h:191:1: warning: restricted blk_opf_t
   degrades to integer
 ./include/trace/events/nilfs2.h:191:1: warning: restricted blk_opf_t
   degrades to integer

Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Reported-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
Closes: https://lore.kernel.org/all/20240430080019.4242-2-konishi.ryusuke@gmail.com/
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
---
 include/trace/events/nilfs2.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/include/trace/events/nilfs2.h b/include/trace/events/nilfs2.h
index 8efc6236f57c..8880c11733dd 100644
--- a/include/trace/events/nilfs2.h
+++ b/include/trace/events/nilfs2.h
@@ -200,7 +200,11 @@ TRACE_EVENT(nilfs2_mdt_submit_block,
 		    __field(struct inode *, inode)
 		    __field(unsigned long, ino)
 		    __field(unsigned long, blkoff)
-		    __field(enum req_op, mode)
+		    /*
+		     * Use field_struct() to avoid is_signed_type() on the
+		     * bitwise type enum req_op.
+		     */
+		    __field_struct(enum req_op, mode)
 	    ),
 
 	    TP_fast_assign(

Re: [PATCH] nilfs2: Use __field_struct() for a bitwise field

[Ryusuke Konishi]

On Tue, May 7, 2024 at 8:24 AM Bart Van Assche wrote:
>
> As one can see in include/trace/stages/stage4_event_fields.h, the
> implementation of __field() uses the is_signed_type() macro. As one can see
> in commit dcf8e5633e2e ("tracing: Define the is_signed_type() macro once"),
> there has been an attempt to not make is_signed_type() trigger sparse
> warnings for bitwise types. Despite that change, sparse complains when
> passing a bitwise type to is_signed_type(). It is not clear to me why.
>
> Follow the example of <trace/events/initcall.h> and suppress the following
> sparse warnings by changing __field() into __field_struct():
>
>  fs/nilfs2/segment.c: note: in included file (through
>    include/trace/trace_events.h, include/trace/define_trace.h,
>    include/trace/events/nilfs2.h):
>  ./include/trace/events/nilfs2.h:191:1: warning: cast to restricted
>    blk_opf_t
>  ./include/trace/events/nilfs2.h:191:1: warning: restricted blk_opf_t
>    degrades to integer
>  ./include/trace/events/nilfs2.h:191:1: warning: restricted blk_opf_t
>    degrades to integer
>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
> Cc: Ryusuke Konishi <konishi.ryusuke@gmail.com>
> Reported-by: Ryusuke Konishi <konishi.ryusuke@gmail.com>
> Closes: https://lore.kernel.org/all/20240430080019.4242-2-konishi.ryusuke@gmail.com/
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
>  include/trace/events/nilfs2.h | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/include/trace/events/nilfs2.h b/include/trace/events/nilfs2.h
> index 8efc6236f57c..8880c11733dd 100644
> --- a/include/trace/events/nilfs2.h
> +++ b/include/trace/events/nilfs2.h
> @@ -200,7 +200,11 @@ TRACE_EVENT(nilfs2_mdt_submit_block,
>                     __field(struct inode *, inode)
>                     __field(unsigned long, ino)
>                     __field(unsigned long, blkoff)
> -                   __field(enum req_op, mode)
> +                   /*
> +                    * Use field_struct() to avoid is_signed_type() on the
> +                    * bitwise type enum req_op.
> +                    */
> +                   __field_struct(enum req_op, mode)
>             ),
>
>             TP_fast_assign(

Bart, thank you very much for all your help.

This is a bit technical and may be debatable.  But this can actually
eliminate the sparse warnings.
So I'm thinking of having Andrew pick this up instead of my patch
currently pending in the mm tree.

If anyone has any objections, please let us know.


Thanks,
Ryusuke Konishi

Re: [PATCH] nilfs2: Use __field_struct() for a bitwise field

[Bart Van Assche]

On 5/6/24 21:10, Ryusuke Konishi wrote:
> So I'm thinking of having Andrew pick this up instead of my patch
> currently pending in the mm tree.
(+Andrew)

The patch Ryusuke is referring to is available here:
https://lore.kernel.org/linux-nilfs/20240506232437.21264-1-bvanassche@acm.org/

Thanks,

Bart.	

Re: [PATCH] nilfs2: Use __field_struct() for a bitwise field

[Andrew Morton]

On Tue, 7 May 2024 05:55:27 -0700 Bart Van Assche <bvanassche@acm.org> wrote:

> On 5/6/24 21:10, Ryusuke Konishi wrote:
> > So I'm thinking of having Andrew pick this up instead of my patch
> > currently pending in the mm tree.

Please identify "my patch"?

> (+Andrew)
> 
> The patch Ryusuke is referring to is available here:
> https://lore.kernel.org/linux-nilfs/20240506232437.21264-1-bvanassche@acm.org/
> 

I'm not subscribed to linux-nilfs.  It's good to cc linux-kernel on
everything for this reason.  Resend, please?

Re: [PATCH] nilfs2: Use __field_struct() for a bitwise field

[Bart Van Assche]

On 5/7/24 9:01 AM, Andrew Morton wrote:
> On Tue, 7 May 2024 05:55:27 -0700 Bart Van Assche <bvanassche@acm.org> wrote:
> 
>> On 5/6/24 21:10, Ryusuke Konishi wrote:
>>> So I'm thinking of having Andrew pick this up instead of my patch
>>> currently pending in the mm tree.
> 
> Please identify "my patch"?

I think it's this patch from one week ago: "[PATCH -mm 1/2] nilfs2: use 
integer type instead of enum req_op for event tracing header" 
(https://lore.kernel.org/linux-nilfs/CAKFNMokwnkku4tHcaAJPmFaBONZqoTMOLtKh_A7kcjzoxj3QZw@mail.gmail.com/T/#mfbbd7edb4a3ecc2d819869e58fb5a460dcb5b4bd)

Thanks,

Bart.