Re: Missing UNLOCK in gid_cache_lookup() ?

[Yaniv Kaul <ykaul@xxxxxxxxxx>]

Replying to myself:

Apparently, the callers are expected to manually perform the UNLOCK later.

I've added a comment in the code for future reference.

Y.

On Sat, Dec 1, 2018 at 8:53 PM Yaniv Kaul <ykaul@xxxxxxxxxx> wrote:

I'm looking at the code and I'm not sure if there's a missing UNLOCK there (location marked in bold red) for cache->gc_lock :

    LOCK(&cache->gc_lock);

    bucket = id % cache->gc_nbuckets;

    agl = BUCKET_START(cache->gc_cache, bucket);

    for (i = 0; i < AUX_GID_CACHE_ASSOC; i++, agl++) {

        if (!agl->gl_list || agl->gl_id != id)

            continue;

        /*

          @uid and @gid reflect the latest UID/GID of the

           process performing the syscall (taken from frame->root).

           If the UID and GID has changed for the PID since the

           time we cached it, we should treat the cache as having

           stale values and query them freshly.

        */

        if (agl->gl_uid != uid || agl->gl_gid != gid)

            break;

        /*

         * We don't put new entries in the cache when expiration=0, but

         * there might be entries still in there if expiration was

         * changed very recently.  Writing the check this way ensures

         * that they're not used.

         */

        if (now < agl->gl_deadline) {

            return agl;

        }

        /*

         * We're not going to find any more UID matches, and reaping

         * is handled further down to maintain LRU order.

         */

        break;

    }

    UNLOCK(&cache->gc_lock);

    return NULL;

}

TIA,

Y.

_______________________________________________
Gluster-devel mailing list
Gluster-devel@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-devel