On Mon, Jun 04 2018 at 12:06pm -0400,
Randy Dunlap <rdunlap@xxxxxxxxxxxxx> wrote:
> On 06/04/2018 04:59 AM, Stephen Rothwell wrote:
> > Hi all,
> >
> > Changes since 20180601:
> >
>
> on x86_64:
>
> In file included from ../drivers/md/dm-writecache.c:8:0:
> ../drivers/md/dm-writecache.c: In function ‘writecache_status’:
> ../include/linux/compiler.h:61:17: warning: format ‘%ld’ expects argument of type ‘long int’, but argument 4 has type ‘int’ [-Wformat=]
> static struct ftrace_branch_data \
> ^
> ../include/linux/device-mapper.h:550:46: note: in definition of macro ‘DMEMIT’
> 0 : scnprintf(result + sz, maxlen - sz, x))
> ^
> ../include/linux/compiler.h:48:24: note: in expansion of macro ‘__branch_check__’
> # define unlikely(x) (__branch_check__(x, 0, __builtin_constant_p(x)))
> ^
> ../drivers/md/dm-writecache.c:414:35: note: in expansion of macro ‘unlikely’
> #define writecache_has_error(wc) (unlikely(READ_ONCE((wc)->error)))
> ^
> ../include/linux/compiler.h:56:23: note: in expansion of macro ‘__trace_if’
> #define if(cond, ...) __trace_if( (cond , ## __VA_ARGS__) )
> ^
> ../include/linux/compiler.h:251:2: note: in expansion of macro ‘if’
> if (check) \
> ^
> ../include/linux/compiler.h:258:22: note: in expansion of macro ‘__READ_ONCE’
> #define READ_ONCE(x) __READ_ONCE(x, 1)
> ^
> ../drivers/md/dm-writecache.c:414:44: note: in expansion of macro ‘READ_ONCE’
> #define writecache_has_error(wc) (unlikely(READ_ONCE((wc)->error)))
> ^
> ../drivers/md/dm-writecache.c:2196:32: note: in expansion of macro ‘writecache_has_error’
> DMEMIT("%ld %llu %llu %llu", writecache_has_error(wc),
> ^
>
> s/%ld/%d/
AFAICT, this is the correct fix:
https://patchwork.kernel.org/patch/10438865/
Not seen any response to it (Steven and Ingo were emailed, Arnd and LKML
were cc'd).
Mike
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
