On Wed, Sep 06, 2017 at 11:34:26PM +0200, Arnd Bergmann wrote: > After changing INIT_UDATA_BUF_OR_NULL() to an inline function, > this does the same change to INIT_UDATA for consistency. > I'm keeping it separate as this part is much larger and > we wouldn't want to backport this to stable kernels if we > ever want to address the gcc warnings by backporting the > first patch. > > Again, using an inline function gives us better type > safety here among other issues with macros. I'm using > u64_to_user_ptr() to convert the user pointer to simplify > the logic rather than adding lots of new type casts. > > Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx> > --- > drivers/infiniband/core/uverbs.h | 22 +++--- > drivers/infiniband/core/uverbs_cmd.c | 125 +++++++++++++++-------------------- > 2 files changed, 67 insertions(+), 80 deletions(-) > Thanks, for doing it and for spotting light on u64_to_user_ptr().
Attachment:
signature.asc
Description: PGP signature
