Martin Svec 8f67835f1e target/rd: improve sg_table lookup scalability
Sequential scan of rd_dev->sg_table_array in rd_get_sg_table is
a serious I/O performance bottleneck for large rd LUNs. Fix this
by computing the sg_table index directly from page offset because
all sg_tables (except the last one) have the same number of pages.

Tested with 90 GiB rd_mcp LUN, where the patch improved maximal
random R/W IOPS by more than 100-150%, depending on actual
hardware and SAN setup.

Signed-off-by: Martin Svec<martin.svec@zoner.cz>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
2013-02-13 11:27:31 -08:00
..
2013-02-08 14:22:05 -05:00
2013-01-23 20:30:52 -08:00
2013-01-30 11:38:25 +01:00
2013-01-14 09:08:38 -08:00
2012-12-11 14:08:47 -08:00
2012-12-18 15:19:06 +10:30
2013-01-22 16:36:23 -08:00
2012-12-12 11:45:16 -08:00
2013-01-08 18:53:56 -08:00
2012-12-19 08:19:07 -08:00
2012-12-11 13:13:55 -08:00
2013-01-16 15:57:54 +01:00
2013-01-18 14:05:25 -08:00
2012-12-12 12:05:15 -08:00
2013-01-15 10:45:26 -07:00
2013-02-08 11:55:27 +11:00