Re: [PATCH 2/3] y2038: linux: Provide __utimes64 implementation

[Lukasz Majewski <lukma at denx dot de>]

Hi Adhemerval,

> On 20/02/2020 12:23, Lukasz Majewski wrote:
> > Hi Adhemerval,
> >   
> >> On 07/02/2020 10:00, Lukasz Majewski wrote:  
> >>> This patch provides new __utimes64 explicit 64 bit function for
> >>> setting file's 64 bit attributes for access and modification time.
> >>>
> >>> Internally, the __utimensat64_helper function is used. This patch
> >>> is necessary for having architectures with __WORDSIZE == 32 Y2038
> >>> safe.
> >>>
> >>> Moreover, a 32 bit version - __utimes has been refactored to
> >>> internally use __utimes64.
> >>>
> >>> The __utimes is now supposed to be used on systems still
> >>> supporting 32 bit time (__TIMESIZE != 64) - hence the necessary
> >>> conversion of struct timeval to 64 bit struct __timeval64.
> >>>
> >>> Build tests:
> >>> ./src/scripts/build-many-glibcs.py glibcs
> >>>
> >>> Run-time tests:
> >>> - Run specific tests on ARM/x86 32bit systems (qemu):
> >>>   https://github.com/lmajewski/meta-y2038 and run tests:
> >>>   https://github.com/lmajewski/y2038-tests/commits/master
> >>>
> >>> Above tests were performed with Y2038 redirection applied as well
> >>> as without to test proper usage of both __utimes64 and __utimes.
> >>>   
> >>
> >> LGTM with some smalls changes below.
> >>
> >> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
> >>  
> >>> ---
> >>>  include/time.h                   |  3 +++
> >>>  sysdeps/unix/sysv/linux/utimes.c | 37
> >>> ++++++++++++++++++++++---------- 2 files changed, 29
> >>> insertions(+), 11 deletions(-)
> >>>
> >>> diff --git a/include/time.h b/include/time.h
> >>> index e38f5e32e6..b04747889a 100644
> >>> --- a/include/time.h
> >>> +++ b/include/time.h
> >>> @@ -211,8 +211,11 @@ libc_hidden_proto (__clock_getres64);
> >>>  #endif
> >>>  
> >>>  #if __TIMESIZE == 64
> >>> +# define __utimes64 __utimes
> >>>  # define __utimensat64 __utimensat
> >>>  #else
> >>> +extern int __utimes64 (const char *file, const struct __timeval64
> >>> tvp[2]); +libc_hidden_proto (__utimes64)
> >>>  extern int __utimensat64 (int fd, const char *file,
> >>>                            const struct __timespec64 tsp[2], int
> >>> flags); libc_hidden_proto (__utimensat64);    
> >>
> >> Ok.
> >>  
> >>> diff --git a/sysdeps/unix/sysv/linux/utimes.c
> >>> b/sysdeps/unix/sysv/linux/utimes.c index 121d883469..09c4e56f18
> >>> 100644 --- a/sysdeps/unix/sysv/linux/utimes.c
> >>> +++ b/sysdeps/unix/sysv/linux/utimes.c
> >>> @@ -16,22 +16,37 @@
> >>>     License along with the GNU C Library; if not, see
> >>>     <https://www.gnu.org/licenses/>.  */
> >>>  
> >>> -#include <errno.h>
> >>> -#include <stddef.h>
> >>> -#include <utime.h>
> >>> -#include <sys/time.h>
> >>> -#include <sysdep.h>
> >>> +#include <time.h>
> >>>  
> >>> +int
> >>> +__utimes64 (const char *file, const struct __timeval64 tvp[2])
> >>> +{
> >>> +  struct __timespec64 ts64[2];
> >>> +
> >>> +  if (tvp)    
> >>
> >> No implicit checks.  
> > 
> > The documentation [1] and [2] explicitly says that times (here tvp)
> > can be NULL:
> > "If times is NULL, then the access and modification times of the
> > file are set to the current time. "
> > 
> > Hence, it is perfectly valid to pass the NULL to the
> > __utimensat64_helper(). Without this check we will segfault earlier
> > (before we reach proper syscall) and introduce regression in glibc.
> >  
> 
> I meant to use:
> 
>   if (tvp != NULL)

Ach ... :-) Thanks for the clarification.

I thought that this check is not needed at all. My misunderstanding :-)


Best regards,

Lukasz Majewski

--

DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma@denx.de

Attachment: pgp_3Ta0ZyrM6.pgp
Description: OpenPGP digital signature