From ab28f2aaa30f2511a7035d1c3d55c45555df574d Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Fri, 7 Sep 2012 12:20:05 +0000 Subject: [PATCH] srpt/README: Recommend to disable CONFIG_SCHED_DEBUG and CONFIG_SCHEDSTATS git-svn-id: http://svn.code.sf.net/p/scst/svn/trunk@4507 d57e44dd-8a1f-0410-8b47-8ef2f437770f --- srpt/README | 29 ++++++++++++++++------------- 1 file changed, 16 insertions(+), 13 deletions(-) diff --git a/srpt/README b/srpt/README index e14081c1b..e56df4cd6 100644 --- a/srpt/README +++ b/srpt/README @@ -321,20 +321,23 @@ For more information, see also: Performance Notes - Target Side ------------------------------- -When using high-latency storage devices (hard disks), the default value -choosen by SCST for DEVICE.threads_num should be fine. When using low-latency -storage devices though (SSDs), DEVICE.threads_num should be set to 1 or 2 in -/etc/scst.conf in order to reach optimal performance for small block sizes -(e.g. 4 KB). +* When using high-latency storage devices (hard disks), the default value + choosen by SCST for DEVICE.threads_num should be fine. When using + low-latency storage devices though (SSDs), DEVICE.threads_num should be set + to 1 or 2 in /etc/scst.conf in order to reach optimal performance for small + block sizes (e.g. 4 KB). -When multiple InfiniBand HCA's are present in a target system the Linux -kernel by default will assign the associated interrupt handlers to CPU 0. -Even irqbalance will often assign the interrupt handlers of multiple HCA's -to the same CPU. That is unfortunate because it leads to unfair handling of -SRP sessions. The solution is to assign InfiniBand HCA interrupts manually -to different CPU's. That's possible by writing looking up the InfiniBand -interrupt numbers in /proc/interrupts and by writing proper bitmasks into -/proc/irq//smp_affinity. +* When multiple InfiniBand HCA's are present in a target system the Linux + kernel by default will assign the associated interrupt handlers to CPU 0. + Even irqbalance will often assign the interrupt handlers of multiple HCA's + to the same CPU. That is unfortunate because it leads to unfair handling of + SRP sessions. The solution is to assign InfiniBand HCA interrupts manually + to different CPU's. That's possible by writing looking up the InfiniBand + interrupt numbers in /proc/interrupts and by writing proper bitmasks into + /proc/irq//smp_affinity. + +* Disabling CONFIG_SCHED_DEBUG and CONFIG_SCHEDSTATS in the kernel config + helps. Performance Notes - Initiator Side