> -----Original Message----- > From: Shivasharan Srikanteshwara > [mailto:shivasharan.srikanteshwara@xxxxxxxxxxxx] > Sent: Wednesday, July 12, 2017 1:51 PM > To: Kashyap Desai; 'Christoph Hellwig' > Cc: 'linux-scsi@xxxxxxxxxxxxxxx'; 'martin.petersen@xxxxxxxxxx'; > 'thenzl@xxxxxxxxxx'; 'jejb@xxxxxxxxxxxxxxxxxx'; Sumit Saxena; > 'hare@xxxxxxxx' > Subject: RE: [PATCH v2 11/15] megaraid_sas: Set device queue_depth same as > HBA can_queue value in scsi-mq mode > > > -----Original Message----- > > From: Kashyap Desai [mailto:kashyap.desai@xxxxxxxxxxxx] > > Sent: Tuesday, July 11, 2017 9:18 PM > > To: Christoph Hellwig; Shivasharan Srikanteshwara > > Cc: linux-scsi@xxxxxxxxxxxxxxx; martin.petersen@xxxxxxxxxx; > > thenzl@xxxxxxxxxx; jejb@xxxxxxxxxxxxxxxxxx; Sumit Saxena; hare@xxxxxxxx > > Subject: RE: [PATCH v2 11/15] megaraid_sas: Set device queue_depth same > > as > > HBA can_queue value in scsi-mq mode > > > > > -----Original Message----- > > > From: Christoph Hellwig [mailto:hch@xxxxxx] > > > Sent: Tuesday, July 11, 2017 7:28 PM > > > To: Shivasharan S > > > Cc: linux-scsi@xxxxxxxxxxxxxxx; martin.petersen@xxxxxxxxxx; > > > thenzl@xxxxxxxxxx; jejb@xxxxxxxxxxxxxxxxxx; > > > kashyap.desai@xxxxxxxxxxxx; sumit.saxena@xxxxxxxxxxxx; > > hare@xxxxxxxx; > > > hch@xxxxxx > > > Subject: Re: [PATCH v2 11/15] megaraid_sas: Set device queue_depth > > > same > > as > > > HBA can_queue value in scsi-mq mode > > > > > > On Wed, Jul 05, 2017 at 05:00:25AM -0700, Shivasharan S wrote: > > > > Currently driver sets default queue_depth for VDs at 256 and JBODs > > > > based on interface type, ie., for SAS JBOD QD will be 64, for SATA > > JBOD QD > > > will be 32. > > > > During performance runs with scsi-mq enabled, we are seeing better > > > > results by setting QD same as HBA queue_depth. > > > > > > Please no scsi-mq specifics. just do this unconditionally. > > > > Chris - Intent for mq specific check is mainly because of sequential > > work load > > for HDD is having penalty due to mq scheduler issue. > > We did this exercise prior to mq-deadline support. > > > > Making generic change for non-mq and mq was good, but we may see some > > user may not like to see regression. > > E.a In case of, QD = 32 for SATA PD file system creation may be faster > compare > > to large QD. There may be a soft merger at block layer due to queue > > depth > > throttling. Eventually, FS creation goes fast due to IO merges, but same > > will > not > > be true if we change queue depth logic (means, increase device queue > > depth > to > > HBA QD.) > > > > We have choice to completely remove this patch and ask users to do sysfs > > settings in case of scsi-mq performance issue for HDD sequential work > > load. > > Having this patch, we want to provide better QD settings as default from > driver. > > > > > > Thanks, Kashyap > > Hi Christoph, > As Kashyap mentioned, the performance issues seen were specific to scsi-mq > enabled case when running sequential workloads with HDDs. > Making this generic might result in regressions in some scenarios for > non-mq. > That was the idea behind making the change specific to scsi-mq only. > > Let us know if you are ok with having this as is or we could remove > this patch completely and have users manually tune queue depth settings if > they > are seeing performance issues with scsi-mq enabled. > > Thanks, > Shivasharan Hi Christoph, Can you please let us know your thoughts on this patch? Are we good to keep the changes as is? Thanks, Shivasharan
