Reading, including through mmap from sparse files always returns zeroes, so there is no need to memset the shm queue. Zeroing it causes backing pages to be allocated to it in tmpfs, and thus increases the memory usage for a not fully used queue. Signed-off-by: Christoph Hellwig <hch at lst.de> --- sheep/cluster/local.c | 1 - 1 file changed, 1 deletion(-) Index: sheepdog/sheep/cluster/local.c =================================================================== --- sheepdog.orig/sheep/cluster/local.c 2012-05-20 16:18:51.094558848 +0200 +++ sheepdog/sheep/cluster/local.c 2012-05-20 19:39:17.768023684 +0200 @@ -200,7 +200,6 @@ static void shm_queue_init(void) else { /* initialize shared memory */ event_pos = 0; - memset(shm_queue, 0, sizeof(*shm_queue)); ret = ftruncate(shmfd, 0); assert(ret == 0); ret = ftruncate(shmfd, sizeof(*shm_queue)); |