From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-f68.google.com ([74.125.82.68]:35716 "EHLO mail-wm0-f68.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751291AbdB0SmL (ORCPT ); Mon, 27 Feb 2017 13:42:11 -0500 Received: by mail-wm0-f68.google.com with SMTP id u63so14767695wmu.2 for ; Mon, 27 Feb 2017 10:41:50 -0800 (PST) Subject: Re: [PATCH 1/3] blk-mq: make blk_mq_alloc_request_hctx() allocate a scheduler request To: Jens Axboe , Omar Sandoval References: <6137d2c6-816e-83b2-103c-cdf0840f49fd@fb.com> <20170227183021.GG10715@vader> <5d9bc14d-9a90-8915-f538-f61452420677@fb.com> Cc: linux-block@vger.kernel.org, kernel-team@fb.com From: Sagi Grimberg Message-ID: <719e3d15-73e2-0b11-e205-e733b882c2ae@grimberg.me> Date: Mon, 27 Feb 2017 20:41:46 +0200 MIME-Version: 1.0 In-Reply-To: <5d9bc14d-9a90-8915-f538-f61452420677@fb.com> Content-Type: text/plain; charset=windows-1252; format=flowed Sender: linux-block-owner@vger.kernel.org List-Id: linux-block@vger.kernel.org >>>> blk_mq_alloc_request_hctx() allocates a driver request directly, unlike >>>> its blk_mq_alloc_request() counterpart. It also crashes because it >>>> doesn't update the tags->rqs map. >>>> >>>> Fix it by making it allocate a scheduler request. >>> >>> All three look good to me. Would you mind respinning #1, it doesn't >>> apply on top of the reserved tag patch. >> >> Yup, I had those based on Sagi's original patches for some reason. I >> fat-fingered send-email, sent as a reply to the original patch 1 instead >> of this email. > > I got it, applied all 3, thanks Omar! FWIW, you can add my: Tested-by: Sagi Grimberg Although I didn't try to trigger reserved vs. normal tag competition which will never ever happen for nvmf btw, but I assume it can happen in drivers who use reserved for error handling.