mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
	akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
	Tom Peng <tom_peng@usish.com>, Jack Wang <jack_wang@usish.com>,
	Lindar Liu <lindar_liu@usish.com>,
	James Bottomley <James.Bottomley@HansenPartnership.com>
Subject: [patch 04/28] SCSI: libsas: reuse the original port when hotplugging phys in wide ports
Date: Thu, 13 Aug 2009 12:40:25 -0700	[thread overview]
Message-ID: <20090813194336.011177297@mini.kroah.org> (raw)
In-Reply-To: <20090813194554.GA13947@kroah.com>

[-- Attachment #1: libsas-reuse-the-original-port-when-hotplugging-phys-in-wide-ports.patch --]
[-- Type: text/plain, Size: 2335 bytes --]

2.6.27-stable review patch.  If anyone has any objections, please let us know.

------------------

From: Tom Peng <tom_peng@usish.com>

commit 5381837f125cc62ad703fbcdfcd7566fc81fd404 upstream.

There's a hotplug problem in the way libsas allocates ports: it loops over the
available ports first trying to add to an existing for a wide port and
otherwise allocating the next free port.  This scheme only works if the port
array is packed from zero, which fails if a port gets hot unplugged and the
array becomes sparse.  In that case, a new port is formed even if there's a
wide port it should be part of.  Fix this by creating two loops over all the
ports:  the first to see if the phy should be part of a wide port and the
second to form a new port in an empty port slot.

Signed-off-by: Tom Peng <tom_peng@usish.com>
Signed-off-by: Jack Wang <jack_wang@usish.com>
Signed-off-by: Lindar Liu <lindar_liu@usish.com>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>

---
 drivers/scsi/libsas/sas_port.c |   19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

--- a/drivers/scsi/libsas/sas_port.c
+++ b/drivers/scsi/libsas/sas_port.c
@@ -56,7 +56,7 @@ static void sas_form_port(struct asd_sas
 		}
 	}
 
-	/* find a port */
+	/* see if the phy should be part of a wide port */
 	spin_lock_irqsave(&sas_ha->phy_port_lock, flags);
 	for (i = 0; i < sas_ha->num_phys; i++) {
 		port = sas_ha->sas_port[i];
@@ -69,12 +69,23 @@ static void sas_form_port(struct asd_sas
 			SAS_DPRINTK("phy%d matched wide port%d\n", phy->id,
 				    port->id);
 			break;
-		} else if (*(u64 *) port->sas_addr == 0 && port->num_phys==0) {
-			memcpy(port->sas_addr, phy->sas_addr, SAS_ADDR_SIZE);
-			break;
 		}
 		spin_unlock(&port->phy_list_lock);
 	}
+	/* The phy does not match any existing port, create a new one */
+	if (i == sas_ha->num_phys) {
+		for (i = 0; i < sas_ha->num_phys; i++) {
+			port = sas_ha->sas_port[i];
+			spin_lock(&port->phy_list_lock);
+			if (*(u64 *)port->sas_addr == 0
+				&& port->num_phys == 0) {
+				memcpy(port->sas_addr, phy->sas_addr,
+					SAS_ADDR_SIZE);
+				break;
+			}
+			spin_unlock(&port->phy_list_lock);
+		}
+	}
 
 	if (i >= sas_ha->num_phys) {
 		printk(KERN_NOTICE "%s: couldn't find a free port, bug?\n",



  parent reply	other threads:[~2009-08-13 19:47 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20090813194021.446758568@mini.kroah.org>
2009-08-13 19:45 ` [patch 00/28] 2.6.27.30-stable review Greg KH
2009-08-13 19:40   ` [patch 01/28] hugetlbfs: fix i_blocks accounting Greg KH
2009-08-13 19:40   ` [patch 02/28] hwmon: (smsc47m1) Differentiate between LPC47M233 and LPC47M292 Greg KH
2009-08-13 19:40   ` [patch 03/28] i2c/tsl2550: Fix lux value in dark environment Greg KH
2009-08-13 19:40   ` Greg KH [this message]
2009-08-13 19:40   ` [patch 05/28] Make SCSI SG v4 driver enabled by default and remove EXPERIMENTAL dependency, since udev depends on BSG Greg KH
2009-08-13 19:40   ` [patch 06/28] page-allocator: preserve PFN ordering when __GFP_COLD is set Greg KH
2009-08-13 19:40   ` [patch 07/28] sysfs: fix hardlink count on device_move Greg KH
2009-08-13 19:40   ` [patch 08/28] thinkpad-acpi: disable broken bay and dock subdrivers Greg KH
2009-08-13 19:40   ` [patch 09/28] USB: storage: raise timeout in usb_stor_Bulk_max_lun Greg KH
2009-08-13 19:40   ` [patch 10/28] x86: fix assembly constraints in native_save_fl() Greg KH
2009-08-13 19:40   ` [patch 11/28] parisc: ensure broadcast tlb purge runs single threaded Greg KH
2009-08-13 19:40   ` [patch 12/28] ieee1394: sbp2: add support for disks >2 TB (and 16 bytes long CDBs) Greg KH
2009-08-13 19:40   ` [patch 13/28] firewire: " Greg KH
2009-08-13 19:40   ` [patch 14/28] x86: enable GART-IOMMU only after setting up protection methods Greg KH
2009-08-13 19:40   ` [patch 15/28] asix: new device ids Greg KH
2009-08-13 19:40   ` [patch 16/28] compat_ioctl: hook up compat handler for FIEMAP ioctl Greg KH
2009-08-13 19:40   ` [patch 17/28] execve: must clear current->clear_child_tid Greg KH
2009-08-13 19:40   ` [patch 18/28] flat: fix uninitialized ptr with shared libs Greg KH
2009-08-13 19:40   ` [patch 19/28] USB: devio: Properly do access_ok() checks Greg KH
2009-08-13 19:40   ` [patch 20/28] USB: ftdi_sio: add vendor and product id for Bayer glucose meter serial converter cable Greg KH
2009-08-13 19:40   ` [patch 21/28] USB: ftdi_sio: add product_id for Marvell OpenRD Base, Client Greg KH
2009-08-13 19:40   ` [patch 22/28] USB: storage: include Prolific Technology USB drive in unusual_devs list Greg KH
2009-08-13 19:40   ` [patch 23/28] USB: usbfs: fix -ENOENT error code to be -ENODEV Greg KH
2009-08-13 19:40   ` [patch 24/28] mm_for_maps: simplify, use ptrace_may_access() Greg KH
2009-08-13 19:40   ` [patch 25/28] mm_for_maps: shift down_read(mmap_sem) to the caller Greg KH
2009-08-13 19:40   ` [patch 26/28] Make sock_sendpage() use kernel_sendpage() Greg KH
2009-08-13 19:40   ` [patch 27/28] ALSA: hda - Add missing vmaster initialization for ALC269 Greg KH
2009-08-13 19:59     ` Linus Torvalds
2009-08-13 20:12       ` Greg KH
2009-08-14  6:56         ` Takashi Iwai
2009-08-14 17:14           ` Greg KH
2009-08-13 19:40   ` [patch 28/28] NFS: Fix an O_DIRECT Oops Greg KH

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=20090813194336.011177297@mini.kroah.org \
    --to=gregkh@suse.de \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=jack_wang@usish.com \
    --cc=lindar_liu@usish.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable-review@kernel.org \
    --cc=stable@kernel.org \
    --cc=tom_peng@usish.com \
    --cc=torvalds@linux-foundation.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: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®