From: Bernd S. <ber...@it...> - 2013-06-03 16:02:55
|
On 06/03/2013 05:46 PM, Miklos Szeredi wrote: > On Mon, Jun 3, 2013 at 5:29 PM, Bernd Schubert > <ber...@it...> wrote: >> On 06/03/2013 03:35 PM, Miklos Szeredi wrote: > >>> >>> https://git.kernel.org/cgit/linux/kernel/git/mszeredi/fuse.git/commit/?h=for-linus&id=28420dad233520811c0e0860e7fb4975ed863fc4 >>> >>> I'll send it off to Linus tomorrow and then it'll automatically go >>> into a 3.9 stable kernel. >> >> >> What is the 2nd part in inode.c process_init_reply() for? >> There is a dependcy of FUSE_READDIRPLUS_AUTO to FUSE_DO_READDIRPLUS? > > Yes. > >> And if >> that part is right, shouldn't it be a new commit with a commit description? >> At least so far I don't see how it relates to the NULL pointer. > > The dependency needs to be made explicit otherwise checking > fc->readdirplus_auto alone is ambiguous (it doesn't tell whether we > are actually using readdirplus or not). Ah, the meaning is even documented in fuse_i.h I first only checked the defines and didn't find anything there and then forgot to lookup the variables... Thanks, Bernd |