lobo/isync
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

remove legacy (bsd-style) locking

Browse source 
flock() may be implemented via fcntl(), which may cause the process to
deadlock itself when trying to apply both types of locks. this is the
case even on linux when the file lives on NFS.

it's unlikely that anything except mbsync would try to access the
.uidvalidity files anyway, so there is no point in trying to be
compatible with anything else ...

REFMAIL: uddy4g589ym.fsf@eismej-u14.spgear.lab.emc.com
This commit is contained in:
Oswald Buddenhagen 2015-09-27 11:47:45 +02:00
parent 682a05a676
commit cda596d530
1 changed files with 0 additions and 16 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

16
src/drv_maildir.c
View file

@ -33,15 +33,10 @@
#include <stdio.h>
#include <unistd.h>
#include <sys/stat.h>
#include <sys/file.h>
#include <errno.h>
#include <time.h>
#include <utime.h>
#ifdef __linux__
# define LEGACY_FLOCK 1
#endif
#if !defined(_POSIX_SYNCHRONIZED_IO) || _POSIX_SYNCHRONIZED_IO <= 0
# define fdatasync fsync
#endif
@ -550,13 +545,6 @@ maildir_uidval_lock( maildir_store_t *ctx )
/* The unlock timer is active, so we are obviously already locked. */
return DRV_OK;
}
#ifdef LEGACY_FLOCK
/* This is legacy only */
if (flock( ctx->uvfd, LOCK_EX ) < 0) {
error( "Maildir error: cannot flock UIDVALIDITY.\n" );
return DRV_BOX_BAD;
}
#endif
/* This (theoretically) works over NFS. Let's hope nobody else did
the same in the opposite order, as we'd deadlock then. */
#if SEEK_SET != 0
@ -628,10 +616,6 @@ maildir_uidval_unlock( maildir_store_t *ctx )
#endif /* USE_DB */
lck.l_type = F_UNLCK;
fcntl( ctx->uvfd, F_SETLK, &lck );
#ifdef LEGACY_FLOCK
/* This is legacy only */
flock( ctx->uvfd, LOCK_UN );
#endif
}
static void