From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sergei Shtylyov Subject: Re: [PATCH v3] libata: support the ata host which implements a queue depth less than 32 Date: Fri, 11 Jul 2014 15:57:01 +0400 Message-ID: <53BFD10D.3080005@cogentembedded.com> References: <1405059026-18673-1-git-send-email-haokexin@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail-la0-f50.google.com ([209.85.215.50]:57715 "EHLO mail-la0-f50.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753037AbaGKL4x (ORCPT ); Fri, 11 Jul 2014 07:56:53 -0400 Received: by mail-la0-f50.google.com with SMTP id pv20so784425lab.37 for ; Fri, 11 Jul 2014 04:56:51 -0700 (PDT) In-Reply-To: <1405059026-18673-1-git-send-email-haokexin@gmail.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Kevin Hao , linux-ide@vger.kernel.org Cc: Tejun Heo , Dan Williams Hello. On 11-07-2014 10:10, Kevin Hao wrote: > The sata on fsl mpc8315e is broken after the commit 8a4aeec8d2d6 > ("libata/ahci: accommodate tag ordered controllers"). The reason is > that the ata controller on this SoC only implement a queue depth of > 16. When issuing the commands in tag order, all the commands in tag > 16 ~ 31 are mapped to tag 0 unconditionally and then causes the sata > malfunction. It makes no senses to use a 32 queue in software while > the hardware has less queue depth. So consider the queue depth > implemented by the hardware when requesting a command tag. > Fixes: 8a4aeec8d2d6 ("libata/ahci: accommodate tag ordered controllers") > Cc: stable@vger.kernel.org > Signed-off-by: Kevin Hao > --- > v3: Use ap->scsi_host->can_queue for the queue depth implemented by hardware. > Patch 2 in v2 is also dropped due to this change. > v2: Remove the changes for the ata tag helper functions > Hi Tejun, > I didn't get explicit objection for the codes at http://marc.info/?l=linux-ide&m=140478830920334&w=2 > So I assume that you are OK wit it. The code in this patch is the same as that, > just add the commit log. > drivers/ata/libata-core.c | 9 ++++++--- > 1 file changed, 6 insertions(+), 3 deletions(-) > diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c > index 8f3043165048..4792fea79acf 100644 > --- a/drivers/ata/libata-core.c > +++ b/drivers/ata/libata-core.c > @@ -4728,14 +4728,17 @@ void swap_buf_le16(u16 *buf, unsigned int buf_words) > static struct ata_queued_cmd *ata_qc_new(struct ata_port *ap) > { > struct ata_queued_cmd *qc = NULL; > - unsigned int i, tag; > + unsigned int i, tag, max_queue; > + > + max_queue = ap->scsi_host->can_queue; > + WARN_ON_ONCE(max_queue > ATA_MAX_QUEUE); > > /* no command while frozen */ > if (unlikely(ap->pflags & ATA_PFLAG_FROZEN)) > return NULL; > > - for (i = 0; i < ATA_MAX_QUEUE; i++) { > - tag = (i + ap->last_tag + 1) % ATA_MAX_QUEUE; > + for (i = 0, tag = ap->last_tag + 1; i < max_queue; i++, tag++) { > + tag = tag < max_queue ? tag : 0; Assigning 'tag' back to 'tag' is quite stupid, don't you think? Why not: if (tag >= max_queue) tag = 0; MBR, Sergei