commit 85fbd722ad upstream.
Freezable kthreads and workqueues are fundamentally problematic in
that they effectively introduce a big kernel lock widely used in the
kernel and have already been the culprit of several deadlock
scenarios. This is the latest occurrence.
During resume, libata rescans all the ports and revalidates all
pre-existing devices. If it determines that a device has gone
missing, the device is removed from the system which involves
invalidating block device and flushing bdi while holding driver core
layer locks. Unfortunately, this can race with the rest of device
resume. Because freezable kthreads and workqueues are thawed after
device resume is complete and block device removal depends on
freezable workqueues and kthreads (e.g. bdi_wq, jbd2) to make
progress, this can lead to deadlock - block device removal can't
proceed because kthreads are frozen and kthreads can't be thawed
because device resume is blocked behind block device removal.
839a8e8660 ("writeback: replace custom worker pool implementation
with unbound workqueue") made this particular deadlock scenario more
visible but the underlying problem has always been there - the
original forker task and jbd2 are freezable too. In fact, this is
highly likely just one of many possible deadlock scenarios given that
freezer behaves as a big kernel lock and we don't have any debug
mechanism around it.
I believe the right thing to do is getting rid of freezable kthreads
and workqueues. This is something fundamentally broken. For now,
implement a funny workaround in libata - just avoid doing block device
hot[un]plug while the system is frozen. Kernel engineering at its
finest. :(
v2: Add EXPORT_SYMBOL_GPL(pm_freezing) for cases where libata is built
as a module.
v3: Comment updated and polling interval changed to 10ms as suggested
by Rafael.
v4: Add #ifdef CONFIG_FREEZER around the hack as pm_freezing is not
defined when FREEZER is not configured thus breaking build.
Reported by kbuild test robot.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Tomaž Šolc <tomaz.solc@tablix.org>
Reviewed-by: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=62801
Link: http://lkml.kernel.org/r/20131213174932.GA27070@htj.dyndns.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Len Brown <len.brown@intel.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 966fbe193f upstream.
Some device require DMADIR to be enabled, but are not detected as such
by atapi_id_dmadir. One such example is "Asus Serillel 2"
SATA-host-to-PATA-device bridge: the bridge itself requires DMADIR,
even if the bridged device does not.
As atapi_dmadir module parameter can cause problems with some devices
(as per Tejun Heo's memory), enabling it globally may not be possible
depending on the hardware.
This patch adds atapi_dmadir in the form of a "force" horkage value,
allowing global, per-bus and per-device control.
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 87809942d3 upstream.
We've received multiple reports in Fedora via (BZ 907193)
that the Seagate Momentus SpinPoint M8 errors out when enabling AA:
[ 2.555905] ata2.00: failed to enable AA (error_mask=0x1)
[ 2.568482] ata2.00: failed to enable AA (error_mask=0x1)
Add the ATA_HORKAGE_BROKEN_FPDMA_AA for this specific harddisk.
Reported-by: Nicholas <arealityfarbetween@googlemail.com>
Signed-off-by: Michele Baldessari <michele@acksyn.org>
Tested-by: Nicholas <arealityfarbetween@googlemail.com>
Acked-by: Alan Cox <gnomes@lxorguk.ukuu.org.uk>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 54b2b50c20 upstream.
Some host adapters do not pass commands through to the target disk
directly. Instead they provide an emulated target which may or may not
accurately report its capabilities. In some cases the physical device
characteristics are reported even when the host adapter is processing
commands on the device's behalf. This can lead to adapter firmware hangs
or excessive I/O errors.
This patch disables WRITE SAME for devices connected to host adapters
that provide an emulated target. Driver writers can disable WRITE SAME
by setting the no_write_same flag in the host adapter template.
[jejb: fix up rejections due to eh_deadline patch]
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 89dafa20f3 upstream.
Tested with Marvell 88se9125, attached with one port mulitplier(5 ports)
and one disk, we will get following boot log messages if using current
code:
ata8: SATA link up 6.0 Gbps (SStatus 133 SControl 330)
ata8.15: Port Multiplier 1.2, 0x1b4b:0x9715 r160, 5 ports, feat 0x1/0x1f
ahci 0000:03:00.0: FBS is enabled
ata8.00: hard resetting link
ata8.00: SATA link down (SStatus 0 SControl 330)
ata8.01: hard resetting link
ata8.01: SATA link down (SStatus 0 SControl 330)
ata8.02: hard resetting link
ata8.02: SATA link down (SStatus 0 SControl 330)
ata8.03: hard resetting link
ata8.03: SATA link up 6.0 Gbps (SStatus 133 SControl 133)
ata8.04: hard resetting link
ata8.04: failed to resume link (SControl 133)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.04: failed to read SCR 1 (Emask=0x40)
ata8.04: failed to read SCR 0 (Emask=0x40)
ata8.03: native sectors (2) is smaller than sectors (976773168)
ata8.03: ATA-8: ST3500413AS, JC4B, max UDMA/133
ata8.03: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32)
ata8.03: configured for UDMA/133
ata8.04: failed to IDENTIFY (I/O error, err_mask=0x100)
ata8.15: hard resetting link
ata8.15: SATA link up 6.0 Gbps (SStatus 133 SControl 330)
ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133'
ata8.15: PMP revalidation failed (errno=-19)
ata8.15: hard resetting link
ata8.15: SATA link up 6.0 Gbps (SStatus 133 SControl 330)
ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133'
ata8.15: PMP revalidation failed (errno=-19)
ata8.15: limiting SATA link speed to 3.0 Gbps
ata8.15: hard resetting link
ata8.15: SATA link up 3.0 Gbps (SStatus 123 SControl 320)
ata8.15: Port Multiplier vendor mismatch '0x1b4b' != '0x133'
ata8.15: PMP revalidation failed (errno=-19)
ata8.15: failed to recover PMP after 5 tries, giving up
ata8.15: Port Multiplier detaching
ata8.03: disabled
ata8.00: disabled
ata8: EH complete
The reason is that current detection code doesn't follow AHCI spec:
First,the port multiplier detection process look like this:
ahci_hardreset(link, class, deadline)
if (class == ATA_DEV_PMP) {
sata_pmp_attach(dev) /* will enable FBS */
sata_pmp_init_links(ap, nr_ports);
ata_for_each_link(link, ap, EDGE) {
sata_std_hardreset(link, class, deadline);
if (link_is_online) /* do soft reset */
ahci_softreset(link, class, deadline);
}
}
But, according to chapter 9.3.9 in AHCI spec: Prior to issuing software
reset, software shall clear PxCMD.ST to '0' and then clear PxFBS.EN to
'0'.
The patch test ok with kernel 3.11.1.
tj: Patch white space contaminated, applied manually with trivial
updates.
Signed-off-by: Xiangliang Yu <yuxiangl@marvell.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f13e220161 upstream.
libata EH decrements scmd->retries when the command failed for reasons
unrelated to the command itself so that, for example, commands aborted
due to suspend / resume cycle don't get penalized; however,
decrementing scmd->retries isn't enough for ATA passthrough commands.
Without this fix, ATA passthrough commands are not resend to the
drive, and no error is signalled to the caller because:
- allowed retry count is 1
- ata_eh_qc_complete fill the sense data, so result is valid
- sense data is filled with untouched ATA registers.
Signed-off-by: Gwendal Grignou <gwendal@google.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8ffff94d20 upstream.
Fixing support for the Silicon Image 3826 port multiplier, by applying
to it the same quirks applied to the Silicon Image 3726. Specifically
fixes the repeated timeout/reset process which previously afflicted
the 3726, as described from line 290. Slightly based on notes from:
https://bugzilla.redhat.com/show_bug.cgi?id=890237
Signed-off-by: Terry Suereth <terry.suereth@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit bb96961928 upstream.
sata_inic162x never reached a state where it's reliable enough for
production use and data corruption is a relatively common occurrence.
Make the driver generate warning about the issues and mark the Kconfig
option as experimental.
If the situation doesn't improve, we'd be better off making it depend
on CONFIG_BROKEN. Let's wait for several cycles and see if the kernel
message draws any attention.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Martin Braure de Calignon <braurede@free.fr>
Reported-by: Ben Hutchings <ben@decadent.org.uk>
Reported-by: risc4all@yahoo.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit eac27f04a7 upstream.
There is a patch b55f84e2d5 "ata_piix: Fix DVD
not dectected at some Haswell platforms" to fix an issue of DVD not
recognized on Haswell Desktop platform with Lynx Point.
Recently, it is also found the same issue at some platformas with Wellsburg PCH.
So deliver a similar patch to fix it by disables 32bit PIO in IDE mode.
Signed-off-by: Youquan Song <youquan.song@intel.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit ddfef5de3d upstream.
Increase the retry count for the hard reset function to 100 but
shorten the time out period to 500 ms. See the comment for
ahci_highbank_hardreset for the reasons why those vaulues were
chosen.
Signed-off-by: Mark Langsdorf <mark.langsdorf@calxeda.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7a87718d92 upstream.
For some reason, a lot of port-multipliers have issues with softreset.
SIMG [34]7x series port-multipliers have been quite erratic in this
regard. I recall that it was better with some firmware revisions and
the current list of quirks worked fine for a while. I think it got
worse with later firmwares or maybe my test coverage wasn't good
enough. Anyways, HPA is reporting that his 3726 setup suffers SRST
failures and then the PMP gets confused and fails to probe the last
port.
The hope was that we try to stick to the standard as much as possible
and soonish the PMPs and their firmwares will improve in quality, so
the quirk list was kept to minimum. Well, it seems like that's never
gonna happen.
Let's set NO_SRST for all [34]7x PMPs so that whatever remaining
userbase of the device suffer the least. Maybe we should do the same
for 57xx's but unfortunately I don't have any device left to test and
I'm not even sure 57xx's have ever been made widely available, so
let's leave those alone for now.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit d0887c43f5 upstream.
There are some SATA controllers which have both devices 0 and 1 but this module
just zeroes out taskfile and sets then ATA_TFLAG_DEVICE (not sure that's needed)
which could lead to a wrong device being selected just before issuing command.
Thus we should call ata_tf_init() which sets up the device register value
properly, like all other users of ata_exec_internal() do...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 30dcf76acc "libata: migrate ACPI code over to new bindings"
mistakenly dropped the code to register hotplug notificaion handler
for ATA port/devices, causing regression for people using ATA bay,
as kernel bug #59871 shows.
Fix this by adding back the hotplug notification handler registration
code. Since this code has to be run once and notification needs to
be installed on every ATA port/devices handle no matter if there is
actual device attached, we can't do this in binding time for ATA
device ACPI handle, as the binding only occurs when a SCSI device is
created, i.e. there is device attached. So introduce the
ata_acpi_hotplug_init() function to loop scan all ATA ACPI handles
and if it is available, install the notificaion handler for it during
ATA init time.
With the ATA ACPI handle binding to SCSI device tree, it is possible
now that when the SCSI hotplug work removes the SCSI device, the ACPI
unbind function will find that the corresponding ACPI device has
already been deleted by dock driver, causing a scaring message like:
[ 128.263966] scsi 4:0:0:0: Oops, 'acpi_handle' corrupt
Fix this by waiting for SCSI hotplug task finish in our notificaion
handler, so that the removal of ACPI device done in ACPI unbind
function triggered by the removal of SCSI device is run earlier when
ACPI device is still available.
[rjw: Rebased]
References: https://bugzilla.kernel.org/show_bug.cgi?id=59871
Reported-bisected-and-tested-by: Dirk Griesbach <spamthis@freenet.de>
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: 3.6+ <stable@vger.kernel.org>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Pull libata changes from Tejun Heo:
"Nothing too interesting. PCI ID additions, some sata_rcar fixes and a
fringe bug fix for DMADIR handling which shouldn't affect any device
remotely modern."
* 'for-3.10-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata:
sata_rcar: fix interrupt handling
ahci: add an observed PCI ID for Marvell 88se9172 SATA controller
sata_rcar: clear STOP bit in bmdma_start() method
libata: make ata_exec_internal_sg honor DMADIR
ata_piix: add PCI IDs for Intel BayTail
libata: update "Maintained by:" tags
The driver's interrupt handling code is too picky in deciding whether it should
handle an interrupt or not which causes completely unneeded spurious interrupts.
Thus make sata_rcar_{ata|serr}_interrupt() *void*; add ATA status register read
to sata_rcar_ata_interrupt() to clear an unexpected ATA interrupt -- it doesn't
get cleared by writing to the SATAINTSTAT register in the interrupt mode we use.
Also, in sata_rcar_ata_interrupt() we should check SATAINTSTAT register only for
enabled interrupts and we should clear only those interrupts that we have read
as active first time around, because else we have a race and risk clearing an
interrupt that can occur between read and write of the SATAINTSTAT register
and never registering it...
Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: stable@vger.kernel.org
