On Jun 10, 2009, at 8:52 AM, Styner, Douglas W wrote:
Anirban Chakraborty [mailto:anirban.chakraborty@xxxxxxxxxx] write:
You got it right. Basically we are doing:
1. baseline scsi-misc + earlier patch
2. baseline scsi-misc + this patch
And then compare 1 and 2.
Results are as follows:
Total interrupts are down slightly (-0.8%), QLA interrupts are flat
(-0.1%) Cycles for qla2xxx_intr_handler up from .77% to .92%.
So, it looks like using spin_lock_irqsave is doing a better job than
globally disabling the sharing of the interrupt via IRQF_DISABLED.
-Anirban
Linux OLTP Performance summary
Kernel# Speedup(x) Intr/s CtxSw/s us% sy% id
% wa%
2.6.30-rc6_scsi-misc_0001 1.000 29538 43374 75 25
0 0
2.6.30-rc6_scsi-misc_irqf-d 1.000 29286 43125 74 25
0 1
Server configurations:
Intel Xeon Quad-core 2.0GHz 2 cpus/8 cores/8 threads
64GB memory, 3 qle2462 FC HBA, 450 spindles (30 logical units)
======oprofile CPU_CLK_UNHALTED for top 30 functions
Cycles% 2.6.30-rc6_scsi-misc_0001 Cycles% 2.6.30-rc6_scsi-
misc_irqf-d
68.1402 <database> 66.1875 <database>
0.9125 qla24xx_start_scsi 0.9552 qla24xx_start_scsi
0.8729 __schedule 0.9272 __schedule
0.7739 qla24xx_intr_handler 0.9178 qla24xx_intr_handler
0.7161 kmem_cache_alloc 0.7963 kmem_cache_alloc
0.4686 __sigsetjmp 0.4580 __sigsetjmp
0.4043 __blockdev_direct_IO 0.4206 __switch_to
0.3894 __switch_to 0.4169 __blockdev_direct_IO
0.3878 rb_get_reader_page 0.3907 scsi_request_fn
0.3762 task_rq_lock 0.3776 task_rq_lock
0.3729 scsi_request_fn 0.3701 rb_get_reader_page
0.3696 __list_add 0.3664 __list_add
0.3168 <bash> 0.3664 copy_user_generic_string
0.3135 try_to_wake_up 0.3122 lock_timer_base
0.2921 ring_buffer_consume 0.2972 ring_buffer_consume
0.2822 lock_timer_base 0.2879 <bash>
0.2624 memset_c 0.2785 aio_complete
0.2607 aio_complete 0.2524 blk_queue_end_tag
0.2508 copy_user_generic_string 0.2468 try_to_wake_up
0.2294 qla2x00_process_completed_re 0.2430 tcp_sendmsg
0.2294 tcp_sendmsg 0.2393 mod_timer
0.2211 kfree 0.2374
qla2x00_process_completed_re
0.2195 memmove 0.2355 memset_c
0.2195 blk_queue_end_tag 0.2225 e1000_xmit_frame
0.2145 scsi_softirq_done 0.2094 generic_make_request
0.2145 mod_timer 0.2075 scsi_device_unbusy
0.2129 generic_file_aio_read 0.2056 kref_get
0.2096 kref_get 0.2056 mempool_free
0.1997 e1000_xmit_frame 0.2038 qla2xxx_queuecommand
0.1997 scsi_device_unbusy 0.2000 kfree
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
