Sasha, please review the patchset. -- Best regards,
Konstantin Khorenko, Virtuozzo Linux Kernel Team On 27.01.2023 12:34, Nikolay Borisov wrote:
The per-cpu allocator guarantees the returned memory is going to be zeroed out so it's redundant to do our own zeroing when initialising cbt. Simply remove this code. Signed-off-by: Nikolay Borisov <nikolay.bori...@virtuozzo.com> --- block/blk-cbt.c | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) diff --git a/block/blk-cbt.c b/block/blk-cbt.c index c7cf7b0ae264..e2a9be83ac0d 100644 --- a/block/blk-cbt.c +++ b/block/blk-cbt.c @@ -265,12 +265,8 @@ inline void blk_cbt_bio_queue(struct request_queue *q, struct bio *bio) static struct cbt_info* do_cbt_alloc(struct request_queue *q, __u8 *uuid, loff_t size, loff_t blocksize) { - struct cbt_info *cbt; - struct cbt_extent *ex; - int i; - + struct cbt_info *cbt = kzalloc(sizeof(*cbt), GFP_KERNEL);- cbt = kzalloc(sizeof(*cbt), GFP_KERNEL);if (!cbt) return ERR_PTR(-ENOMEM);@@ -282,11 +278,6 @@ static struct cbt_info* do_cbt_alloc(struct request_queue *q, __u8 *uuid,if (!cbt->cache) goto err_cbt;- for_each_possible_cpu(i) {- ex = per_cpu_ptr(cbt->cache, i); - memset(ex, 0, sizeof (*ex)); - } - cbt->map = vmalloc(NR_PAGES(cbt->block_max) * sizeof(void*)); if (!cbt->map) goto err_pcpu;
_______________________________________________ Devel mailing list Devel@openvz.org https://lists.openvz.org/mailman/listinfo/devel
