Re: [PATCH 06/21] powerpc: Avoid comparison of unsigned long >= 0 in __access_ok

[Christophe LEROY]

Le 26/02/2018 Ã 07:34, Christophe LEROY a ÃcritÂ:
> Le 25/02/2018 Ã 18:22, Mathieu Malaterre a ÃcritÂ:
> > Rewrite check size - 1 <= Y as size < Y since `size` is unsigned value.
> > Fix warning (treated as error in W=1):
> >
> > ÂÂ CCÂÂÂÂÂ arch/powerpc/kernel/signal_32.o
> > In file included from ./include/linux/uaccess.h:14:0,
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ from ./include/asm-generic/termios-base.h:8,
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ from ./arch/powerpc/include/asm/termios.h:20,
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ from ./include/uapi/linux/termios.h:6,
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ from ./include/linux/tty.h:7,
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ from arch/powerpc/kernel/signal_32.c:36:
> > ./include/asm-generic/termios-base.h: In function 
> > âuser_termio_to_kernel_termiosâ:
> > ./arch/powerpc/include/asm/uaccess.h:52:35: error: comparison of 
> > unsigned expression >= 0 is always true [-Werror=type-limits]
> > ÂÂÂ (((size) == 0) || (((size) - 1) <= ((segment).seg - (addr)))))
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ ^
> > ./arch/powerpc/include/asm/uaccess.h:58:3: note: in expansion of macro 
> > â__access_okâ
> > ÂÂÂ __access_ok((__force unsigned long)(addr), (size), get_fs()))
> > ÂÂÂ ^~~~~~~~~~~
> > ./arch/powerpc/include/asm/uaccess.h:262:6: note: in expansion of 
> > macro âaccess_okâ
> > ÂÂ if (access_ok(VERIFY_READ, __gu_addr, (size)))ÂÂ \
> > ÂÂÂÂÂÂ ^~~~~~~~~
> > ./arch/powerpc/include/asm/uaccess.h:80:2: note: in expansion of macro 
> > â__get_user_checkâ
> > ÂÂ __get_user_check((x), (ptr), sizeof(*(ptr)))
> > ÂÂ ^~~~~~~~~~~~~~~~
> > ./include/asm-generic/termios-base.h:36:6: note: in expansion of macro 
> > âget_userâ
> > ÂÂ if (get_user(termios->c_line, &termio->c_line) < 0)
> > ÂÂÂÂÂÂ ^~~~~~~~
> > [...]
> > cc1: all warnings being treated as errors
> >
> > Signed-off-by: Mathieu Malaterre <malat@xxxxxxxxxx>
> > ---
> > Â arch/powerpc/include/asm/uaccess.h | 2 +-
> > Â 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/arch/powerpc/include/asm/uaccess.h 
> > b/arch/powerpc/include/asm/uaccess.h
> > index 51bfeb8777f0..fadc406bd39d 100644
> > --- a/arch/powerpc/include/asm/uaccess.h
> > +++ b/arch/powerpc/include/asm/uaccess.h
> > @@ -49,7 +49,7 @@
> > Â #define __access_ok(addr, size, segment)ÂÂÂ \
> > ÂÂÂÂÂ (((addr) <= (segment).seg) &&ÂÂÂÂÂÂÂ \
> > -ÂÂÂÂ (((size) == 0) || (((size) - 1) <= ((segment).seg - (addr)))))
> > +ÂÂÂÂ (((size) == 0) || ((size) < ((segment).seg - (addr)))))
>
> IIUC, ((2 - 1) <= 1) is the same as (2 < 1) ?????


Note that I already try to submit a fix for this warning 3 years ago 
(https://patchwork.ozlabs.org/patch/418075/) and it was rejected with 
the following comment:

Again, I don't think Linux enables this warning.  What did you do to
produce this?  In any case, it's a bad warning that doesn't take macros
into account, and the answer is not to make the code less clear by hiding
the fact that zero is a special case.

Christophe


> Christophe
>
> > Â #endif