Author: jeff Date: Wed Nov 20 01:57:33 2019 New Revision: 354874 URL: https://svnweb.freebsd.org/changeset/base/354874
Log: When we set OFFPAGE to limit fragmentation we should also set VTOSLAB so that we avoid the hashtables. The hashtable is now only required if a zone is created with OFFPAGE specified initially, not internally. This flag signals to UMA that it can't touch the allocated memory and so can't store a slab pointer in the containing page. Reviewed by: markj Differential Revision: https://reviews.freebsd.org/D22453 Modified: head/sys/vm/uma_core.c Modified: head/sys/vm/uma_core.c ============================================================================== --- head/sys/vm/uma_core.c Wed Nov 20 01:14:22 2019 (r354873) +++ head/sys/vm/uma_core.c Wed Nov 20 01:57:33 2019 (r354874) @@ -585,7 +585,7 @@ zone_timeout(uma_zone_t zone) uma_keg_t keg; u_int slabs; - if ((zone->uz_flags & UMA_ZFLAG_CACHE) != 0) + if ((zone->uz_flags & UMA_ZONE_HASH) == 0) goto update_wss; keg = zone->uz_keg; @@ -1568,7 +1568,14 @@ keg_small_init(uma_keg_t keg) "new wasted space = %d\n", keg->uk_name, keg, wastedspace, slabsize / UMA_MAX_WASTE, keg->uk_ipers, slabsize - keg->uk_ipers * keg->uk_rsize); - keg->uk_flags |= UMA_ZONE_OFFPAGE; + /* + * If we had access to memory to embed a slab header we + * also have a page structure to use vtoslab() instead of + * hash to find slabs. If the zone was explicitly created + * OFFPAGE we can't necessarily touch the memory. + */ + if ((keg->uk_flags & UMA_ZONE_OFFPAGE) == 0) + keg->uk_flags |= UMA_ZONE_OFFPAGE | UMA_ZONE_VTOSLAB; } if ((keg->uk_flags & UMA_ZONE_OFFPAGE) && @@ -1608,7 +1615,7 @@ keg_large_init(uma_keg_t keg) * slab header. */ if ((keg->uk_flags & UMA_ZFLAG_INTERNAL) == 0) - keg->uk_flags |= UMA_ZONE_OFFPAGE; + keg->uk_flags |= UMA_ZONE_OFFPAGE | UMA_ZONE_VTOSLAB; else keg->uk_ppera++; } _______________________________________________ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
