From mboxrd@z Thu Jan 1 00:00:00 1970 From: jthumshirn@suse.de (Johannes Thumshirn) Date: Tue, 17 Jan 2017 17:27:52 +0100 Subject: [LSF/MM TOPIC][LSF/MM ATTEND] NAPI polling for block drivers In-Reply-To: <53a3fb6c-c75a-519a-f669-2bcab404e01d@grimberg.me> References: <20170111134312.GH6286@linux-x5ow.site> <8b47ca34-d2ff-26dc-721e-2cb1e18f1efc@grimberg.me> <499af528-7810-f82d-1f11-cbf8f3a5b21c@grimberg.me> <53a3fb6c-c75a-519a-f669-2bcab404e01d@grimberg.me> Message-ID: <20170117162752.GE6067@linux-x5ow.site> On Tue, Jan 17, 2017@06:15:43PM +0200, Sagi Grimberg wrote: > Oh, and the current code that was tested can be found at: > > git://git.infradead.org/nvme.git nvme-irqpoll Just for the record, all tests you've run are with the upper irq_poll_budget of 256 [1]? We (Hannes and me) recently stumbed accross this when trying to poll for more than 256 queue entries in the drivers we've been testing. Did your system load reduce with irq polling? In theory it should but I have seen increases with AHCI at least according to fio. IIRC Hannes saw decreases with his SAS HBA tests, as expected. [1] lib/irq_poll.c:13 Byte, Johannes -- Johannes Thumshirn Storage jthumshirn at suse.de +49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 N?rnberg GF: Felix Imend?rffer, Jane Smithard, Graham Norton HRB 21284 (AG N?rnberg) Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850