From: Mike Christie <michaelc@cs.wisc.edu>
To: Vasu Dev <vasu.dev@linux.intel.com>
Cc: Robert Love <robert.w.love@intel.com>,
James.Bottomley@suse.de, linux-scsi@vger.kernel.org,
devel@open-fcoe.org
Subject: Re: [PATCH 03/28] libfc: IO errors on link down due to cable unplug
Date: Wed, 28 Jul 2010 02:32:06 -0500 [thread overview]
Message-ID: <4C4FDCF6.2070105@cs.wisc.edu> (raw)
In-Reply-To: <1280266320.31061.2590.camel@vi2.jf.intel.com>
[-- Attachment #1: Type: text/plain, Size: 995 bytes --]
On 07/27/2010 04:32 PM, Vasu Dev wrote:
>> This isn't one of those races where you are blocking the rport, but the
>> IO keeps coming around so the retries are used really quickly right (the
>> driver looks like it has the fc class and internal state checks to
>> prevent this but I wanted to make sure).
>
I think I am hitting the above problem.
>
> This tiny time windows can be further reduced by use of wmb() on rport
> state change to block and lport getting out of ready since their states
> are checked w/o lock in queuecommand path.
I think you need something like this for the lport queue ready check.
It looks like the initial failure from __fc_linkdown->fc_fcp_cleanup
burned a timeout. Then because it missed the lport ready check, it will
use an extra retry just sitting there timing out.
For the rport state check, I think you are supposed to call the fc
chkready function under the host lock. The port state and flags are set
under it. I did the attached patch for that.
[-- Attachment #2: libfc-hold-host-lock-when-checking-rport.patch --]
[-- Type: text/x-patch, Size: 910 bytes --]
libfc: call fc_remote_port_chkready under the host lock.
The rport port state and flags are set under the host lock,
so this patch calls fc_remote_port_chkready with the host lock
held like is also done in the other fc drivers.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
diff --git a/drivers/scsi/libfc/fc_fcp.c b/drivers/scsi/libfc/fc_fcp.c
index eac4d09..a116846 100644
--- a/drivers/scsi/libfc/fc_fcp.c
+++ b/drivers/scsi/libfc/fc_fcp.c
@@ -1765,14 +1768,14 @@ int fc_queuecommand(struct scsi_cmnd *sc_cmd, void (*done)(struct scsi_cmnd *))
struct fcoe_dev_stats *stats;
lport = shost_priv(sc_cmd->device->host);
- spin_unlock_irq(lport->host->host_lock);
rval = fc_remote_port_chkready(rport);
if (rval) {
sc_cmd->result = rval;
done(sc_cmd);
- goto out;
+ return 0;
}
+ spin_unlock_irq(lport->host->host_lock);
if (!*(struct fc_remote_port **)rport->dd_data) {
/*
next prev parent reply other threads:[~2010-07-28 7:32 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-20 22:19 [PATCH 00/28] libfc, libfcoe and fcoe updates for scsi-misc Robert Love
2010-07-20 22:19 ` [PATCH 01/28] libfc: fix slowpath error from WARN_ON in fc_fcp_send_data Robert Love
2010-07-20 22:19 ` [PATCH 02/28] fcoe: make it possible to verify fcoe with sparse Robert Love
2010-07-20 22:19 ` [PATCH 03/28] libfc: IO errors on link down due to cable unplug Robert Love
2010-07-21 3:29 ` Mike Christie
2010-07-27 21:32 ` Vasu Dev
2010-07-27 22:08 ` Mike Christie
2010-07-28 7:32 ` Mike Christie [this message]
2010-07-30 20:34 ` Vasu Dev
2010-07-20 22:19 ` [PATCH 04/28] fcoe: cleans up fcoe_disable and fcoe_enable Robert Love
2010-07-20 22:19 ` [PATCH 05/28] fcoe: adds src and dest mac address checking for fcoe frames Robert Love
2010-07-20 22:19 ` [PATCH 06/28] libfc: convert rport lookup to be RCU safe Robert Love
2010-07-20 22:19 ` [PATCH 07/28] libfc: provide space for LLD after remote port structure Robert Love
2010-07-20 22:19 ` [PATCH 08/28] libfcoe: convert FIP to lock with mutex instead of spin lock Robert Love
2010-07-20 22:19 ` [PATCH 09/28] libfc: add discovery-private pointer for LLD Robert Love
2010-07-20 22:19 ` [PATCH 10/28] libfcoe: fcoe: fnic: change fcoe_ctlr_init interface to specify mode Robert Love
2010-07-20 22:20 ` [PATCH 11/28] libfc: Add local port point-to-multipoint flag Robert Love
2010-07-20 22:20 ` [PATCH 12/28] libfc: add FLOGI state to rport for VN2VN Robert Love
2010-07-20 22:20 ` [PATCH 13/28] libfc: track FIP exchanges Robert Love
2010-07-20 22:20 ` [PATCH 14/28] libfcoe: add protocol description of FIP VN2VN mode Robert Love
2010-07-20 22:20 ` [PATCH 15/28] libfcoe: add state change debugging Robert Love
2010-07-20 22:20 ` [PATCH 16/28] libfcoe: fcoe: fnic: add FIP VN2VN point-to-multipoint support Robert Love
2010-07-20 22:20 ` [PATCH 17/28] libfcoe: Fix FIP ELS encapsulation details for FLOGI responses Robert Love
2010-07-20 22:20 ` [PATCH 18/28] fcoe libfcoe: use correct FC-MAP for VN2VN mode Robert Love
2010-07-20 22:20 ` [PATCH 19/28] fcoe: config via separate create_vn2vn module parameter Robert Love
2010-07-20 22:20 ` [PATCH 20/28] libfc: eliminate rport LOGO state Robert Love
2010-07-20 22:20 ` [PATCH 21/28] libfc: add fc_frame_sid() and fc_frame_did() functions Robert Love
2010-07-20 22:21 ` [PATCH 22/28] libfc: add fc_fill_reply_hdr() and fc_fill_hdr() Robert Love
2010-07-20 22:21 ` [PATCH 23/28] libfc: add interface to allocate a sequence for incoming requests Robert Love
2010-07-20 22:21 ` [PATCH 24/28] libfc: don't require a local exchange " Robert Love
2010-07-20 22:21 ` [PATCH 25/28] fcoe: fix offload feature flag change from netdev Robert Love
2010-07-20 22:21 ` [PATCH 26/28] Revert "[SCSI] fcoe: Fix using VLAN ID in creating lport's WWWN/WWPN" Robert Love
2010-07-20 22:21 ` [PATCH 27/28] libfc: Add retry logic to lport state machine when receiving LS_RJT Robert Love
2010-07-20 22:21 ` [PATCH 28/28] fcoe: remove check for zero fabric name Robert Love
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=4C4FDCF6.2070105@cs.wisc.edu \
--to=michaelc@cs.wisc.edu \
--cc=James.Bottomley@suse.de \
--cc=devel@open-fcoe.org \
--cc=linux-scsi@vger.kernel.org \
--cc=robert.w.love@intel.com \
--cc=vasu.dev@linux.intel.com \
/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: link
Be 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.