From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sagi Grimberg Subject: Re: selective block polling and preadv2/pwritev2 revisited Date: Thu, 7 Jan 2016 16:41:13 +0200 Message-ID: <568E7909.5030203@dev.mellanox.co.il> References: <1450966464-6847-1-git-send-email-hch@lst.de> <568A889E.4020204@dev.mellanox.co.il> <20160104163949.GA17409@lst.de> <568D4859.4010206@dev.mellanox.co.il> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: linux-api-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: "Martin K. Petersen" Cc: Christoph Hellwig , viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org, axboe-b10kYP2dOMg@public.gmane.org, milosz-B5zB6C1i6pkAvxtiuMwx3w@public.gmane.org, linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-block-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-api@vger.kernel.org Hi Martin, > Sagi> This means that the user would need to pass iovec entries of 8 > Sagi> bytes correct? Seems like a waste for large IOs (sorry for > Sagi> diverging from the subject) > > The assumption was that there would be a 1:1 mapping between the number > of data buffers and the metadata ditto. But nothing says that a data > iovec entry is limited in size to a single sector. Yea... I meant 1:1, I got confused on the 8 bytes comment... > The other option to have a single iovec at the end representing the > metadata for all data buffers. I think there are valid use cases for > either approach and we may end up having to support both via a flag. Either approach presents limitation, but I guess user-space can deal with it... From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-f49.google.com ([74.125.82.49]:34191 "EHLO mail-wm0-f49.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752255AbcAGOli (ORCPT ); Thu, 7 Jan 2016 09:41:38 -0500 Received: by mail-wm0-f49.google.com with SMTP id u188so101046430wmu.1 for ; Thu, 07 Jan 2016 06:41:18 -0800 (PST) Subject: Re: selective block polling and preadv2/pwritev2 revisited To: "Martin K. Petersen" References: <1450966464-6847-1-git-send-email-hch@lst.de> <568A889E.4020204@dev.mellanox.co.il> <20160104163949.GA17409@lst.de> <568D4859.4010206@dev.mellanox.co.il> Cc: Christoph Hellwig , viro@zeniv.linux.org.uk, axboe@fb.com, milosz@adfin.com, linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, linux-api@vger.kernel.org From: Sagi Grimberg Message-ID: <568E7909.5030203@dev.mellanox.co.il> Date: Thu, 7 Jan 2016 16:41:13 +0200 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-fsdevel-owner@vger.kernel.org List-ID: Hi Martin, > Sagi> This means that the user would need to pass iovec entries of 8 > Sagi> bytes correct? Seems like a waste for large IOs (sorry for > Sagi> diverging from the subject) > > The assumption was that there would be a 1:1 mapping between the number > of data buffers and the metadata ditto. But nothing says that a data > iovec entry is limited in size to a single sector. Yea... I meant 1:1, I got confused on the 8 bytes comment... > The other option to have a single iovec at the end representing the > metadata for all data buffers. I think there are valid use cases for > either approach and we may end up having to support both via a flag. Either approach presents limitation, but I guess user-space can deal with it...