From: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org> To: Sagi Grimberg <sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> Cc: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>, Max Gurtovoy <maxg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>, linux-nvme-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, keith.busch-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org, linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, vladimirk-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org Subject: Re: [PATCH v2 3/4] nvme: enable SG gaps support Date: Sun, 23 Apr 2017 10:45:53 +0200 [thread overview] Message-ID: <20170423084553.GA24434@lst.de> (raw) In-Reply-To: <9ff1a2e8-f84f-c772-e294-2339eee8e37d-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> On Sun, Apr 23, 2017 at 10:20:59AM +0300, Sagi Grimberg wrote: > Christoph, > >> All our block I/O must be cache line aligned, so dword aligned SGLs >> should not be an issue. And for PRPs or MRs we'll always use the >> host page size. > > Where is that constraint coming from? Is that new? Back when I wrote > GAP support in rdma I tested a single byte alignment via vectored > direct IO and it seemed to be going through... Each block queue has a ->dma_alignment field, which is set to 511 by default unless changed. Everything not aligned to that will be bounced in __blk_rq_map_user_iov and friends before we send it to the driver. > I also just tested Bart's unaligned test [1] on scsi_debug and it seems > to go through as well (alignment and length are 4)... SCSI reduces the alignment in __scsi_init_queue: /* * set a reasonable default alignment on word boundaries: the * host and device may alter it using * blk_queue_update_dma_alignment() later. */ blk_queue_dma_alignment(q, 0x03); still would fit NVMe dword-alignment SGLs, though :) -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html
WARNING: multiple messages have this Message-ID (diff)
From: hch@lst.de (Christoph Hellwig) Subject: [PATCH v2 3/4] nvme: enable SG gaps support Date: Sun, 23 Apr 2017 10:45:53 +0200 [thread overview] Message-ID: <20170423084553.GA24434@lst.de> (raw) In-Reply-To: <9ff1a2e8-f84f-c772-e294-2339eee8e37d@grimberg.me> On Sun, Apr 23, 2017@10:20:59AM +0300, Sagi Grimberg wrote: > Christoph, > >> All our block I/O must be cache line aligned, so dword aligned SGLs >> should not be an issue. And for PRPs or MRs we'll always use the >> host page size. > > Where is that constraint coming from? Is that new? Back when I wrote > GAP support in rdma I tested a single byte alignment via vectored > direct IO and it seemed to be going through... Each block queue has a ->dma_alignment field, which is set to 511 by default unless changed. Everything not aligned to that will be bounced in __blk_rq_map_user_iov and friends before we send it to the driver. > I also just tested Bart's unaligned test [1] on scsi_debug and it seems > to go through as well (alignment and length are 4)... SCSI reduces the alignment in __scsi_init_queue: /* * set a reasonable default alignment on word boundaries: the * host and device may alter it using * blk_queue_update_dma_alignment() later. */ blk_queue_dma_alignment(q, 0x03); still would fit NVMe dword-alignment SGLs, though :)
next prev parent reply other threads:[~2017-04-23 8:45 UTC|newest] Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-04-13 15:56 [PATCH v2 0/4] arbitrary sg lists support Max Gurtovoy 2017-04-13 15:56 ` Max Gurtovoy [not found] ` <1492098977-5231-1-git-send-email-maxg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> 2017-04-13 15:56 ` [PATCH v2 1/4] IB/core: Add inline function to get sg mr type Max Gurtovoy 2017-04-13 15:56 ` Max Gurtovoy [not found] ` <1492098977-5231-2-git-send-email-maxg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> 2017-04-20 11:24 ` Sagi Grimberg 2017-04-20 11:24 ` Sagi Grimberg 2017-04-13 15:56 ` [PATCH v2 2/4] IB/iser: Use ib_get_sg_mr_type() helper Max Gurtovoy 2017-04-13 15:56 ` Max Gurtovoy 2017-04-13 15:56 ` [PATCH v2 3/4] nvme: enable SG gaps support Max Gurtovoy 2017-04-13 15:56 ` Max Gurtovoy [not found] ` <1492098977-5231-4-git-send-email-maxg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> 2017-04-20 11:27 ` Sagi Grimberg 2017-04-20 11:27 ` Sagi Grimberg [not found] ` <2bd673ee-8455-32b7-1bad-34fa9eebb076-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-04-21 6:44 ` Christoph Hellwig 2017-04-21 6:44 ` Christoph Hellwig [not found] ` <20170421064427.GB6095-jcswGhMUV9g@public.gmane.org> 2017-04-23 7:20 ` Sagi Grimberg 2017-04-23 7:20 ` Sagi Grimberg [not found] ` <9ff1a2e8-f84f-c772-e294-2339eee8e37d-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-04-23 8:45 ` Christoph Hellwig [this message] 2017-04-23 8:45 ` Christoph Hellwig [not found] ` <20170423084553.GA24434-jcswGhMUV9g@public.gmane.org> 2017-04-23 9:45 ` Sagi Grimberg 2017-04-23 9:45 ` Sagi Grimberg [not found] ` <9fe3f5fa-4496-3819-0288-08a073dd1a1a-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-04-23 13:23 ` Max Gurtovoy 2017-04-23 13:23 ` Max Gurtovoy 2017-04-24 7:02 ` Christoph Hellwig 2017-04-24 7:02 ` Christoph Hellwig 2017-04-13 15:56 ` [PATCH v2 4/4] nvme-rdma: add support for arbitrary sg lists mapping Max Gurtovoy 2017-04-13 15:56 ` Max Gurtovoy 2017-04-13 16:31 ` [PATCH v2 0/4] arbitrary sg lists support Bart Van Assche 2017-04-13 16:31 ` Bart Van Assche [not found] ` <1492101079.5176.9.camel-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org> 2017-04-13 16:46 ` Max Gurtovoy 2017-04-13 16:46 ` Max Gurtovoy [not found] ` <9a8bc957-dcc0-2e54-d631-9f05f96e0f18-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> 2017-04-13 16:59 ` Laurence Oberman 2017-04-13 16:59 ` Laurence Oberman 2017-04-20 11:30 ` Sagi Grimberg 2017-04-20 11:30 ` Sagi Grimberg [not found] ` <c52b5ce7-5346-419d-26f2-0d4d45d20471-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-04-23 10:34 ` Leon Romanovsky 2017-04-23 10:34 ` Leon Romanovsky [not found] ` <20170423103439.GC14088-U/DQcQFIOTAAJjI8aNfphQ@public.gmane.org> 2017-04-23 11:49 ` Sagi Grimberg 2017-04-23 11:49 ` Sagi Grimberg [not found] ` <d697d360-99ca-c3ed-08e2-b89cc50a15b0-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-04-23 12:12 ` Leon Romanovsky 2017-04-23 12:12 ` Leon Romanovsky [not found] ` <20170423121254.GD14088-U/DQcQFIOTAAJjI8aNfphQ@public.gmane.org> 2017-04-23 12:28 ` Sagi Grimberg 2017-04-23 12:28 ` Sagi Grimberg [not found] ` <a3264968-2d9e-bda5-1f23-74da248cf4d5-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-04-23 13:35 ` Max Gurtovoy 2017-04-23 13:35 ` Max Gurtovoy [not found] ` <9a9668fa-87a7-e730-27e2-8a2b3e7fbc36-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> 2017-04-23 14:35 ` Sagi Grimberg 2017-04-23 14:35 ` Sagi Grimberg [not found] ` <8ed506ba-ae7b-af27-2d2a-9a434c22353c-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org> 2017-05-01 18:50 ` Doug Ledford 2017-05-01 18:50 ` Doug Ledford [not found] ` <1493664645.3041.184.camel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> 2017-05-02 16:14 ` Max Gurtovoy 2017-05-02 16:14 ` Max Gurtovoy [not found] ` <4a1c94b5-fe13-8b34-762f-e7465182ed27-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org> 2017-05-02 16:39 ` Laurence Oberman 2017-05-02 16:39 ` Laurence Oberman
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20170423084553.GA24434@lst.de \ --to=hch-jcswghmuv9g@public.gmane.org \ --cc=keith.busch-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \ --cc=leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \ --cc=linux-nvme-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org \ --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ --cc=maxg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \ --cc=sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org \ --cc=vladimirk-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.