Fupan Li [Tue, 4 Aug 2015 01:51:21 +0000 (09:51 +0800)]
efi: fix 32bit kernel boot failed problem using efi
Commit 35d5134b7d5a
("x86/efi: Correct EFI boot stub use of code32_start")
imported a bug, which will cause 32bit kernel boot failed
using efi method. It should use the label's address instead
of the value stored in the label to caculate the address of
code32_start.
Signed-off-by: Fupan Li <fupan.li@windriver.com> Reviewed-by: Matt Fleming <matt.fleming@intel.com>
This patch fixes a regression introduced with the following commit
in v4.0-rc1 code, where an explicit iser-target logout would result
in ->tx_thread_active being incorrectly cleared by the logout post
handler, and subsequent TX kthread leak:
iscsi-target: Convert iscsi_thread_set usage to kthread.h
To address this bug, change iscsit_logout_post_handler_closesession()
and iscsit_logout_post_handler_samecid() to only cmpxchg() on
->tx_thread_active for traditional iscsi/tcp connections.
This is required because iscsi/tcp connections are invoking logout
post handler logic directly from TX kthread context, while iser
connections are invoking logout post handler logic from a seperate
workqueue context.
This patch fixes a use-after-free bug in iscsit_release_sessions_for_tpg()
where se_portal_group->session_lock was incorrectly released/re-acquired
while walking the active se_portal_group->tpg_sess_list.
The can result in a NULL pointer dereference when iscsit_close_session()
shutdown happens in the normal path asynchronously to this code, causing
a bogus dereference of an already freed list entry to occur.
To address this bug, walk the session list checking for the same state
as before, but move entries to a local list to avoid dropping the lock
while walking the active list.
As before, signal using iscsi_session->session_restatement=1 for those
list entries to be released locally by iscsit_free_session() code.
Fixes: 3e0249f9c05c ("RDS/IB: add refcount tracking to struct rds_ib_device")
There lacks a dropping on rds_ib_device.refcount in case rds_ib_alloc_fmr
failed(mr pool running out). this lead to the refcount overflow.
A complain in line 117(see following) is seen. From vmcore:
s_ib_rdma_mr_pool_depleted is 2147485544 and rds_ibdev->refcount is -2147475448.
That is the evidence the mr pool is used up. so rds_ib_alloc_fmr is very likely
to return ERR_PTR(-EAGAIN).
Port link change with port in resume state should not be
reported to usbcore, as this is an internal state to be
handled by xhci driver. Reporting PLC to usbcore may
cause usbcore clearing PLC first and port change event irq
won't be generated.
Signed-off-by: Zhuang Jin Can <jin.can.zhuang@intel.com> Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the link is just waken, it's in Resume state, and driver sets PLS to
U0. This refers to Phase 1. Phase 2 refers to when the link has completed
the transition from Resume state to U0.
With the fix of xhci: report U3 when link is in resume state, it also
exposes an issue that usb3 roothub and controller can suspend right
after phase 1, and this causes a hard hang in controller.
To fix the issue, we need to prevent usb3 bus suspend if any port is
resuming in phase 1.
[merge separate USB2 and USB3 port resume checking to one -Mathias] Signed-off-by: Zhuang Jin Can <jin.can.zhuang@intel.com> Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
xhci_hub_report_usb3_link_state() returns pls as U0 when the link
is in resume state, and this causes usb core to think the link is in
U0 while actually it's in resume state. When usb core transfers
control request on the link, it fails with TRB error as the link
is not ready for transfer.
To fix the issue, report U3 when the link is in resume state, thus
usb core knows the link it's not ready for transfer.
Signed-off-by: Zhuang Jin Can <jin.can.zhuang@intel.com> Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When resetting a device the number of active TTs may need to be
corrected by xhci_update_tt_active_eps, but the number of old active
endpoints supplied to it was always zero, so the number of TTs and the
bandwidth reserved for them was not updated, and could rise
unnecessarily.
This affected systems using Intel's Patherpoint chipset, which rely on
software bandwidth checking. For example, a Lenovo X230 would lose the
ability to use ports on the docking station after enough suspend/resume
cycles because the bandwidth calculated would rise with every cycle when
a suitable device is attached.
The correct number of active endpoints is calculated in the same way as
in xhci_reserve_bandwidth.
This device automatically switches itself to another mode (0x1405)
unless the specific access pattern of Windows is followed in its
initial mode. That makes a dirty unmount of the internal storage
devices inevitable if they are mounted. So the card reader of
such a device should be ignored, lest an unclean removal become
inevitable.
This replaces an earlier patch that ignored all LUNs of this device.
That patch was overly broad.
Signed-off-by: Oliver Neukum <oneukum@suse.com> Reviewed-by: Lars Melin <larsm17@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit adds the necessary quirk to make the Marvell 4140 SATA PMP
work properly. This PMP doesn't like SRST on port number 4 (the host
port) so this commit marks this port as not supporting SRST.
When a blkcg configuration is targeted to a partition rather than a
whole device, blkg_conf_prep fails with -EINVAL; unfortunately, it
forgets to put the gendisk ref in that case. Fix it.
This patch fixes a problem in the usbtouchscreen driver for DMC TSC-30
touch screen. Due to a missing delay between the RESET and SET_RATE
commands, the touch screen may become unresponsive during system startup or
driver loading.
According to the DMC documentation, a delay is needed after the RESET
command to allow the chip to complete its internal initialization. As this
delay is not guaranteed, we had a system where the touch screen
occasionally did not send any touch data. There was no other indication of
the problem.
The patch fixes the problem by adding a 150ms delay between the RESET and
SET_RATE commands.
Suggested-by: Jakob Mustafa <jakob.mustafa@bytecmed.com> Signed-off-by: Bernhard Bender <bernhard.bender@bytecmed.com> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When we get a read error from the last working device, we don't
try to repair it, and don't fail the device. We simple report a
read error to the caller.
However the current test for 'is this the last working device' is
wrong.
When there is only one fully working device, it assumes that a
non-faulty device is that device. However a spare which is rebuilding
would be non-faulty but so not the only working device.
So change the test from "!Faulty" to "In_sync". If ->degraded says
there is only one fully working device and this device is in_sync,
this must be the one.
This bug has existed since we allowed read_balance to read from
a recovering spare in v3.0
Reported-and-tested-by: Alexander Lyakas <alex.bolshoy@gmail.com> Fixes: 76073054c95b ("md/raid1: clean up read_balance.") Signed-off-by: NeilBrown <neilb@suse.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
pdev->dev.platform_data is not initialized if match is true in function
sdhci_pxav3_probe. Just local variable pdata is assigned the return value
from function pxav3_get_mmc_pdata().
static int sdhci_pxav3_probe(struct platform_device *pdev) {
struct sdhci_pxa_platdata *pdata = pdev->dev.platform_data;
...
if (match) {
ret = mmc_of_parse(host->mmc);
if (ret)
goto err_of_parse;
sdhci_get_of_property(pdev);
pdata = pxav3_get_mmc_pdata(dev);
}
...
}
Support for 8BIT bus with was added some time ago to sdhci-esdhc but
then missed to remove the 8BIT from the reserved bit mask which made
8BIT non functional.
Fixes: 66b50a00992d ("mmc: esdhc: Add support for 8-bit bus width and..") Signed-off-by: Joakim Tjernlund <joakim.tjernlund@transmode.se> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we don't do this, and we then fail to recreate the debugfs
directory during a mode change, then we will fail later trying
to add stations to this now bogus directory:
Analysis reveals the cause of the crash to be due to STp->device
being NULL. The pointer was NULLed via scsi_tape_put(STp) when it
calls scsi_tape_release(). In st_open() we jump to err_out after
scsi_block_when_processing_errors() completes and returns the
device as offline (sdev_state was SDEV_DEL):
1180 /* Open the device. Needs to take the BKL only because of incrementing the SCSI host
1181 module count. */
1182 static int st_open(struct inode *inode, struct file *filp)
1183 {
1184 int i, retval = (-EIO);
1185 int resumed = 0;
1186 struct scsi_tape *STp;
1187 struct st_partstat *STps;
1188 int dev = TAPE_NR(inode);
1189 char *name;
...
1217 if (scsi_autopm_get_device(STp->device) < 0) {
1218 retval = -EIO;
1219 goto err_out;
1220 }
1221 resumed = 1;
1222 if (!scsi_block_when_processing_errors(STp->device)) {
1223 retval = (-ENXIO);
1224 goto err_out;
1225 }
...
1264 err_out:
1265 normalize_buffer(STp->buffer);
1266 spin_lock(&st_use_lock);
1267 STp->in_use = 0;
1268 spin_unlock(&st_use_lock);
1269 scsi_tape_put(STp); <-- STp->device = 0 after this
1270 if (resumed)
1271 scsi_autopm_put_device(STp->device);
1272 return retval;
The ref count for the struct scsi_tape had already been reduced
to 1 when the .remove method of the st module had been called.
The kref_put() in scsi_tape_put() caused scsi_tape_release()
to be called:
MacBook Pro 5,2 with ALC889 codec had already a fixup entry, but this
seems not working correctly, a fix for pin NID 0x15 is needed in
addition. It's equivalent with the fixup for MacBook Air 1,1, so use
this instead.
The Steinberg MI2 and MI4 interfaces are compatible with the USB class
audio spec, but the MIDI part of the devices is reported as a vendor
specific interface.
This patch adds entries to quirks-table.h to recognize the MIDI
endpoints. Audio functionality was already working and is unaffected by
this change.
Signed-off-by: Dominic Sacré <dominic.sacre@gmx.de> Signed-off-by: Albert Huitsing <albert@huitsing.nl> Acked-by: Clemens Ladisch <clemens@ladisch.de> Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The resend mechanism happily calls the interrupt handler of interrupts
which are marked IRQ_NESTED_THREAD from softirq context. This can
result in crashes because the interrupt handler is not the proper way
to invoke the device handlers. They must be invoked via
handle_nested_irq.
Prevent the resend even if the interrupt has no valid parent irq
set. Its better to have a lost interrupt than a crashing machine.
Currently instruction_pointer() returns pt_regs->ret and so return value
is of type "long", which implicitly stands for "signed long".
While that's perfectly fine when dealing with 32-bit values if return
value of instruction_pointer() gets assigned to 64-bit variable sign
extension may happen.
And at least in one real use-case it happens already.
In perf_prepare_sample() return value of perf_instruction_pointer()
(which is an alias to instruction_pointer() in case of ARC) is assigned
to (struct perf_sample_data)->ip (which type is "u64").
And what we see if instuction pointer points to user-space application
that in case of ARC lays below 0x8000_0000 "ip" gets set properly with
leading 32 zeros. But if instruction pointer points to kernel address
space that starts from 0x8000_0000 then "ip" is set with 32 leadig
"f"-s. I.e. id instruction_pointer() returns 0x8100_0000, "ip" will be
assigned with 0xffff_ffff__8100_0000. Which is obviously wrong.
In particular that issuse broke output of perf, because perf was unable
to associate addresses like 0xffff_ffff__8100_0000 with anything from
/proc/kallsyms.
That's what we used to see:
----------->8----------
6.27% ls [unknown] [k] 0xffffffff8046c5cc
2.96% ls libuClibc-0.9.34-git.so [.] memcpy
2.25% ls libuClibc-0.9.34-git.so [.] memset
1.66% ls [unknown] [k] 0xffffffff80666536
1.54% ls libuClibc-0.9.34-git.so [.] 0x000224d6
1.18% ls libuClibc-0.9.34-git.so [.] 0x00022472
----------->8----------
With that change perf output looks much better now:
----------->8----------
8.21% ls [kernel.kallsyms] [k] memset
3.52% ls libuClibc-0.9.34-git.so [.] memcpy
2.11% ls libuClibc-0.9.34-git.so [.] malloc
1.88% ls libuClibc-0.9.34-git.so [.] memset
1.64% ls [kernel.kallsyms] [k] _raw_spin_unlock_irqrestore
1.41% ls [kernel.kallsyms] [k] __d_lookup_rcu
----------->8----------
If the kernel is compiled with gcc 5.1 and the XZ compression option
the decompress_kernel function calls _sclp_print_early in 64-bit mode
while the content of the upper register half of %r6 is non-zero.
This causes a specification exception on the servc instruction in
_sclp_servc.
The _sclp_print_early function saves and restores the upper registers
halves but it fails to clear them for the 31-bit code of the mini sclp
driver.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Normally opening a file, unlinking it and then closing will have
the inode freed upon close() (provided that it's not otherwise busy and
has no remaining links, of course). However, there's one case where that
does *not* happen. Namely, if you open it by fhandle with cold dcache,
then unlink() and close().
In normal case you get d_delete() in unlink(2) notice that dentry
is busy and unhash it; on the final dput() it will be forcibly evicted from
dcache, triggering iput() and inode removal. In this case, though, we end
up with *two* dentries - disconnected (created by open-by-fhandle) and
regular one (used by unlink()). The latter will have its reference to inode
dropped just fine, but the former will not - it's considered hashed (it
is on the ->s_anon list), so it will stay around until the memory pressure
will finally do it in. As the result, we have the final iput() delayed
indefinitely. It's trivial to reproduce -
main()
{
int fd;
union {
struct file_handle f;
char buf[MAX_HANDLE_SZ];
} x;
int m;
x.f.handle_bytes = sizeof(x);
chdir("/root");
mkdir("foo", 0700);
fd = open("foo/bar", O_CREAT | O_RDWR, 0600);
close(fd);
name_to_handle_at(AT_FDCWD, "foo/bar", &x.f, &m, 0);
flush_dcache();
fd = open_by_handle_at(AT_FDCWD, &x.f, O_RDWR);
unlink("foo/bar");
write(fd, buf, sizeof(buf));
system("df ."); /* 20Mb eaten */
close(fd);
system("df ."); /* should've freed those 20Mb */
flush_dcache();
system("df ."); /* should be the same as #2 */
}
will spit out something like
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/root 322023 303843 1131 100% /
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/root 322023 303843 1131 100% /
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/root 322023 283282 21692 93% /
- inode gets freed only when dentry is finally evicted (here we trigger
than by remount; normally it would've happened in response to memory
pressure hell knows when).
Acked-by: J. Bruce Fields <bfields@fieldses.org> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reading page fault handler code I've noticed that under right
circumstances kernel would map anonymous pages into file mappings: if
the VMA doesn't have vm_ops->fault() and the VMA wasn't fully populated
on ->mmap(), kernel would handle page fault to not populated pte with
do_anonymous_page().
Let's change page fault handler to use do_anonymous_page() only on
anonymous VMA (->vm_ops == NULL) and make sure that the VMA is not
shared.
For file mappings without vm_ops->fault() or shred VMA without vm_ops,
page fault on pte_none() entry would lead to SIGBUS.
The firmware class uevent function accessed the "fw_priv->buf" buffer
without the proper locking and testing for NULL. This is an old bug
(looks like it goes back to 2012 and commit 1244691c73b2: "firmware
loader: introduce firmware_buf"), but for some reason it's triggering
only now in 4.2-rc1.
Shuah Khan is trying to bisect what it is that causes this to trigger
more easily, but in the meantime let's just fix the bug since others are
hitting it too (at least Ingo reports having seen it as well).
Reported-and-tested-by: Shuah Khan <shuahkh@osg.samsung.com> Acked-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current pmd_huge() and pud_huge() functions simply check if the table
bit is not set and reports the entries as huge in that case. This is
counter-intuitive as a clear pmd/pud cannot also be a huge pmd/pud, and
it is inconsistent with at least arm and x86.
To prevent others from making the same mistake as me in looking at code
that calls these functions and to fix an issue with KVM on arm64 that
causes memory corruption due to incorrect page reference counting
resulting from this mistake, let's change the behavior.
Signed-off-by: Christoffer Dall <christoffer.dall@linaro.org> Reviewed-by: Steve Capper <steve.capper@linaro.org> Acked-by: Marc Zyngier <marc.zyngier@arm.com> Fixes: 084bd29810a5 ("ARM64: mm: HugeTLB support.") Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In needs_ilk_vtd_wa(), we pass in the GPU device but compared it against
the ids for the mobile GPU and the mobile host bridge. That latter is
impossible and so likely was just a typo for the desktop GPU device id
(which is also buggy).
drm/i915: Disable WC PTE updates to w/a buggy IOMMU on ILK
Reported-by: Ting-Wei Lan <lantw44@gmail.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91127
References: https://bugzilla.freedesktop.org/show_bug.cgi?id=60391 Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
rbd_obj_request_create() is called on the main I/O path, so we need to
use GFP_NOIO to make sure allocation doesn't blow back on us. Not all
callers need this, but I'm still hardcoding the flag inside rather than
making it a parameter because a) this is going to stable, and b) those
callers shouldn't really use rbd_obj_request_create() and will be fixed
in the future.
If we'd already sent a request and decide to abort it, we *must*
issue TFLUSH properly and not just blindly reuse the tag, or
we'll get seriously screwed when response eventually arrives
and we confuse it for response to later request that had reused
the same tag.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Problem: When an operation like WRITE receives a BAD_STATEID, even though
recovery code clears the RECLAIM_NOGRACE recovery flag before recovering
the open state, because of clearing delegation state for the associated
inode, nfs_inode_find_state_and_recover() gets called and it makes the
same state with RECLAIM_NOGRACE flag again. As a results, when we restart
looking over the open states, we end up in the infinite loop instead of
breaking out in the next test of state flags.
Solution: unset the RECLAIM_NOGRACE set because of
calling of nfs_inode_find_state_and_recover() after returning from calling
recover_open() function.
When encoding the NFSACL SETACL operation, reserve just the estimated
size of the ACL rather than a fixed maximum. This eliminates needless
zero padding on the wire that the server ignores.
Fixes: ee5dc7732bd5 ('NFS: Fix "kernel BUG at fs/nfs/nfs3xdr.c:1338!"') Signed-off-by: Chuck Lever <chuck.lever@oracle.com> Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The omap watchdog has the annoying behaviour that writes to most
registers don't have any effect when the watchdog is already running.
Quoting the AM335x reference manual:
To modify the timer counter value (the WDT_WCRR register),
prescaler ratio (the WDT_WCLR[4:2] PTV bit field), delay
configuration value (the WDT_WDLY[31:0] DLY_VALUE bit field), or
the load value (the WDT_WLDR[31:0] TIMER_LOAD bit field), the
watchdog timer must be disabled by using the start/stop sequence
(the WDT_WSPR register).
Currently the timer is stopped in the .probe callback but still there
are possibilities that yield to a situation where omap_wdt_start is
entered with the timer running (e.g. when /dev/watchdog is closed
without stopping and then reopened). In such a case programming the
timeout silently fails!
To circumvent this stop the timer before reprogramming.
Assuming one of the first things the watchdog user does is setting the
timeout explicitly nothing too bad should happen because this explicit
setting works fine.
This fixes a data corruption bug when using discard on top of MD linear,
raid0 and raid10 personalities.
Commit 20d0189b1012 "block: Introduce new bio_split()" permits sharing
the bio_vec between the two resulting bios. That is fine for read/write
requests where the bio_vec is immutable. For discards, however, we need
to be able to attach a payload and update the bio_vec so the page can
get mapped to a scatterlist entry. Therefore the bio_vec can not be
shared when splitting discards and we must do a full clone.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Reported-by: Seunguk Shin <seunguk.shin@samsung.com> Tested-by: Seunguk Shin <seunguk.shin@samsung.com> Cc: Seunguk Shin <seunguk.shin@samsung.com> Cc: Jens Axboe <axboe@fb.com> Cc: Kent Overstreet <kent.overstreet@gmail.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The usbfs API has a peculiar hole: Users are not allowed to reap their
URBs after the device has been disconnected. There doesn't seem to be
any good reason for this; it is an ad-hoc inconsistency.
The patch allows users to issue the USBDEVFS_REAPURB and
USBDEVFS_REAPURBNDELAY ioctls (together with their 32-bit counterparts
on 64-bit systems) even after the device is gone. If no URBs are
pending for a disconnected device then the ioctls will return -ENODEV
rather than -EAGAIN, because obviously no new URBs will ever be able
to complete.
The patch also adds a new capability flag for
USBDEVFS_GET_CAPABILITIES to indicate that the reap-after-disconnect
feature is supported.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Tested-by: Chris Dickens <christopher.a.dickens@gmail.com> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit fix kernel crash when probing for rfkill devices in dell-laptop
driver failed. Function free_page() was incorrectly used on struct page *
instead of virtual address of SMI buffer.
This commit also simplify allocating page for SMI buffer by using
__get_free_page() function instead of sequential call of functions
alloc_page() and page_address().
Signed-off-by: Pali Rohár <pali.rohar@gmail.com> Acked-by: Michal Hocko <mhocko@suse.cz> Signed-off-by: Darren Hart <dvhart@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes a several year old regression that I found while trying
to get the Yoga 3 11 to work. The ideapad_rfk_set function is meant
to send a command to the embedded controller through ACPI, but
as of c1f73658ed, it sends the index of the rfkill device instead
of the command, and ignores the opcode field.
This changes it back to the original behavior, which indeed
flips the rfkill state as seen in the debugfs interface.
Whilst testing cpu hotplug events on kernel configured with
DEBUG_PREEMPT and DEBUG_ATOMIC_SLEEP we get following BUG message,
caused by calling request_irq() and free_irq() in the context of
hotplug notification (which is in this case atomic context).
[ 40.785859] CPU1: Software reset
[ 40.786660] BUG: sleeping function called from invalid context at mm/slub.c:1241
[ 40.786668] in_atomic(): 1, irqs_disabled(): 128, pid: 0, name: swapper/1
[ 40.786678] Preemption disabled at:[< (null)>] (null)
[ 40.786681]
[ 40.786692] CPU: 1 PID: 0 Comm: swapper/1 Not tainted 3.19.0-rc4-00024-g7dca860 #36
[ 40.786698] Hardware name: SAMSUNG EXYNOS (Flattened Device Tree)
[ 40.786728] [<c0014a00>] (unwind_backtrace) from [<c0011980>] (show_stack+0x10/0x14)
[ 40.786747] [<c0011980>] (show_stack) from [<c0449ba0>] (dump_stack+0x70/0xbc)
[ 40.786767] [<c0449ba0>] (dump_stack) from [<c00c6124>] (kmem_cache_alloc+0xd8/0x170)
[ 40.786785] [<c00c6124>] (kmem_cache_alloc) from [<c005d6f8>] (request_threaded_irq+0x64/0x128)
[ 40.786804] [<c005d6f8>] (request_threaded_irq) from [<c0350b8c>] (exynos4_local_timer_setup+0xc0/0x13c)
[ 40.786820] [<c0350b8c>] (exynos4_local_timer_setup) from [<c0350ca8>] (exynos4_mct_cpu_notify+0x30/0xa8)
[ 40.786838] [<c0350ca8>] (exynos4_mct_cpu_notify) from [<c003b330>] (notifier_call_chain+0x44/0x84)
[ 40.786857] [<c003b330>] (notifier_call_chain) from [<c0022fd4>] (__cpu_notify+0x28/0x44)
[ 40.786873] [<c0022fd4>] (__cpu_notify) from [<c0013714>] (secondary_start_kernel+0xec/0x150)
[ 40.786886] [<c0013714>] (secondary_start_kernel) from [<40008764>] (0x40008764)
Interrupts cannot be requested/freed in the CPU_STARTING/CPU_DYING
notifications which run on the hotplugged cpu with interrupts and
preemption disabled.
To avoid the issue, request the interrupts for all possible cpus in
the boot code. The interrupts are marked NO_AUTOENABLE to avoid a racy
request_irq/disable_irq() sequence. The flag prevents the
request_irq() code from enabling the interrupt immediately.
The interrupt is then enabled in the CPU_STARTING notifier of the
hotplugged cpu and again disabled with disable_irq_nosync() in the
CPU_DYING notifier.
[ tglx: Massaged changelog to match the patch ]
Fixes: 7114cd749a12 ("clocksource: exynos_mct: use (request/free)_irq calls for local timer registration") Reported-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Reviewed-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Krzysztof Kozlowski <k.kozlowski@samsung.com> Tested-by: Marcin Jabrzyk <m.jabrzyk@samsung.com> Signed-off-by: Damian Eppel <d.eppel@samsung.com> Cc: m.szyprowski@samsung.com Cc: kyungmin.park@samsung.com Cc: daniel.lezcano@linaro.org Cc: kgene@kernel.org Cc: linux-arm-kernel@lists.infradead.org Link: http://lkml.kernel.org/r/1435324984-7328-1-git-send-email-d.eppel@samsung.com Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There was a possible race between
ieee80211_reconfig() and
ieee80211_delayed_tailroom_dec(). This could
result in inability to transmit data if driver
crashed during roaming or rekeying and subsequent
skbs with insufficient tailroom appeared.
This race was probably never seen in the wild
because a device driver would have to crash AND
recover within 0.5s which is very unlikely.
I was able to prove this race exists after
changing the delay to 10s locally and crashing
ath10k via debugfs immediately after GTK
rekeying. In case of ath10k the counter went below
0. This was harmless but other drivers which
actually require tailroom (e.g. for WEP ICV or
MMIC) could end up with the counter at 0 instead
of >0 and introduce insufficient skb tailroom
failures because mac80211 would not resize skbs
appropriately anymore.
Fixes: 8d1f7ecd2af5 ("mac80211: defer tailroom counter manipulation when roaming") Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The final version of commit 637241a900cb ("kmsg: honor dmesg_restrict
sysctl on /dev/kmsg") lost few hooks, as result security_syslog() are
processed incorrectly:
- open of /dev/kmsg checks syslog access permissions by using
check_syslog_permissions() where security_syslog() is not called if
dmesg_restrict is set.
- syslog syscall and /proc/kmsg calls do_syslog() where security_syslog
can be executed twice (inside check_syslog_permissions() and then
directly in do_syslog())
With this patch security_syslog() is called once only in all
syslog-related operations regardless of dmesg_restrict value.
Fixes: 637241a900cb ("kmsg: honor dmesg_restrict sysctl on /dev/kmsg") Signed-off-by: Vasily Averin <vvs@virtuozzo.com> Cc: Kees Cook <keescook@chromium.org> Cc: Josh Boyer <jwboyer@redhat.com> Cc: Eric Paris <eparis@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
bitmap_parselist("", &mask, nmaskbits) will erroneously set bit zero in
the mask. The same bug is visible in cpumask_parselist() since it is
layered on top of the bitmask code, e.g. if you boot with "isolcpus=",
you will actually end up with cpu zero isolated.
The bug was introduced in commit 4b060420a596 ("bitmap, irq: add
smp_affinity_list interface to /proc/irq") when bitmap_parselist() was
generalized to support userspace as well as kernelspace.
Fixes: 4b060420a596 ("bitmap, irq: add smp_affinity_list interface to /proc/irq") Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com> Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Many harddisks (mostly WD ones) have firmware problems and take too
long, more than 10 seconds, to resume from suspend. And this often
exceeds the default DPM watchdog timeout (12 seconds), resulting in a
kernel panic out of sudden.
Since most distros just take the default as is, we should give a bit
more safer value. This patch increases the default value from 12
seconds to one minute, which has been confirmed to be long enough for
such problematic disks.
Link: https://bugzilla.kernel.org/show_bug.cgi?id=91921 Fixes: 70fea60d888d (PM / Sleep: Detect device suspend/resume lockup and log event) Signed-off-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The current handler of MMC_BLK_CMD_ERR in mmc_blk_issue_rw_rq function
may cause new coming request permanent missing when the ongoing
request (previoulsy started) complete end.
The problem scenario is as follows:
(1) Request A is ongoing;
(2) Request B arrived, and finally mmc_blk_issue_rw_rq() is called;
(3) Request A encounters the MMC_BLK_CMD_ERR error;
(4) In the error handling of MMC_BLK_CMD_ERR, suppose mmc_blk_cmd_err()
end request A completed and return zero. Continue the error handling,
suppose mmc_blk_reset() reset device success;
(5) Continue the execution, while loop completed because variable ret
is zero now;
(6) Finally, mmc_blk_issue_rw_rq() return without processing request B.
The process related to the missing request may wait that IO request
complete forever, possibly crashing the application or hanging the system.
Fix this issue by starting new request when reset success.
When receiving a new iser connect request we serialize
the pending requests by adding the newly created iser connection
to the np accept list and let the login thread process the connect
request one by one (np_accept_wait).
In case we received a disconnect request before the iser_conn
has begun processing (still linked in np_accept_list) we should
detach it from the list and clean it up and not have the login
thread process a stale connection. We do it only when the connection
state is not already terminating (initiator driven disconnect) as
this might lead us to access np_accept_mutex after the np was released
in live shutdown scenarios.
Before we reach to connection established we may get an
error event. In this case the core won't teardown this
connection (never established it), so we take care of freeing
it ourselves.
This patch converts iscsi-target code to use modern kthread.h API
callers for creating RX/TX threads for each new iscsi_conn descriptor,
and releasing associated RX/TX threads during connection shutdown.
This is done using iscsit_start_kthreads() -> kthread_run() to start
new kthreads from within iscsi_post_login_handler(), and invoking
kthread_stop() from existing iscsit_close_connection() code.
Also, convert iscsit_logout_post_handler_closesession() code to use
cmpxchg when determing when iscsit_cause_connection_reinstatement()
needs to sleep waiting for completion.
This patch adds a new FACS initialization flag for acpi_tb_initialize().
acpi_enable_subsystem() might be invoked several times in OS bootup process,
and we don't want FACS initialization to be invoked twice. Lv Zheng.
Link: https://github.com/acpica/acpica/commit/90f5332a Signed-off-by: Lv Zheng <lv.zheng@intel.com> Signed-off-by: Bob Moore <robert.moore@intel.com> Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
struct crush_bucket_tree::num_nodes is u8, so ceph_decode_8_safe()
should be used. -Wconversion catches this, but I guess it went
unnoticed in all the noise it spews. The actual problem (at least for
common crushmaps) isn't the u32 -> u8 truncation though - it's the
advancement by 4 bytes instead of 1 in the crushmap buffer.
We were allocating memory with memdup_user() but we were never releasing
that memory. This affected pretty much every call to the ioctl, whether
it deduplicated extents or not.
This issue was reported on IRC by Julian Taylor and on the mailing list
by Marcel Ritter, credit goes to them for finding the issue.
Reported-by: Julian Taylor <jtaylor.debian@googlemail.com> Reported-by: Marcel Ritter <ritter.marcel@gmail.com> Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: Mark Fasheh <mfasheh@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The free space entries are allocated using kmem_cache_zalloc(),
through __btrfs_add_free_space(), therefore we should use
kmem_cache_free() and not kfree() to avoid any confusion and
any potential problem. Looking at the kfree() definition at
mm/slab.c it has the following comment:
/*
* (...)
*
* Don't free memory not originally allocated by kmalloc()
* or you will run into trouble.
*/
So better be safe and use kmem_cache_free().
Signed-off-by: Filipe Manana <fdmanana@suse.com> Reviewed-by: David Sterba <dsterba@suse.cz> Signed-off-by: Chris Mason <clm@fb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/md/md.c: In function "update_array_info":
drivers/md/md.c:6394:26: warning: logical not is only applied
to the left hand side of comparison [-Wlogical-not-parentheses]
!mddev->persistent != info->not_persistent||
Fix it as Neil Brown said:
mddev->persistent != !info->not_persistent ||
The mcp3021 scaling code is dividing the VDD (full-scale) value in
millivolts by the A2D resolution to obtain the scaling factor. When VDD
is 3300mV (the standard value) and the resolution is 12-bit (4096
divisions), the result is a scale factor of 3300/4096, which is always
one. Effectively, the raw A2D reading is always being returned because
no scaling is applied.
This patch fixes the issue and simplifies the register-to-volts
calculation, removing the unneeded "output_scale" struct member.
Signed-off-by: Nick Stevens <Nick.Stevens@digi.com>
[Guenter Roeck: Dropped unnecessary value check] Signed-off-by: Guenter Roeck <linux@roeck-us.net> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes a bug in the XOR driver where the cleanup function can be
called and free descriptors that never been processed by the engine (which
result in data errors).
The cleanup function will free descriptors based on the ownership bit in
the descriptors.
Fengguang Wu's tests triggered a bug in the branch tracer's start up
test when CONFIG_DEBUG_PREEMPT set. This was because that config
adds some debug logic in the per cpu field, which calls back into
the branch tracer.
The branch tracer has its own recursive checks, but uses a per cpu
variable to implement it. If retrieving the per cpu variable calls
back into the branch tracer, you can see how things will break.
Instead of using a per cpu variable, use the trace_recursion field
of the current task struct. Simply set a bit when entering the
branch tracing and clear it when leaving. If the bit is set on
entry, just don't do the tracing.
There's also the case with lockdep, as the local_irq_save() called
before the recursion can also trigger code that can call back into
the function. Changing that to a raw_local_irq_save() will protect
that as well.
This prevents the recursion and the inevitable crash that follows.
The filter_parse() function can call infix_get_op() which calls
infix_advance() that updates the infix filter pointers for the cnt
and tail without checking if the filter is already at the end, which
will put the cnt to zero and the tail beyond the end. The loop then calls
infix_next() that has
The cnt will now be below zero, and the tail that is returned is
already passed the end of the filter string. So far the allocation
of the filter string usually has some buffer that is zeroed out, but
if the filter string is of the exact size of the allocated buffer
there's no guarantee that the charater after the nul terminating
character will be zero.
Luckily, only root can write to the filter.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When testing the fix for the trace filter, I could not come up with
a scenario where the operand count goes below zero, so I added a
WARN_ON_ONCE(cnt < 0) to the logic. But there is legitimate case
that it can happen (although the filter would be wrong).
That is, a single operation without any operands will hit the path
where the WARN_ON_ONCE() can trigger. Although this is harmless,
and the filter is reported as a error. But instead of spitting out
a warning to the kernel dmesg, just fail nicely and report it via
the proper channels.
To prevent offline stripping of existing file xattrs and relabeling of
them at runtime, EVM allows only newly created files to be labeled. As
pseudo filesystems are not persistent, stripping of xattrs is not a
concern.
Some LSMs defer file labeling on pseudo filesystems. This patch
permits the labeling of existing files on pseudo files systems.
__key_link_end is not freeing the associated array edit structure
and this leads to a 512 byte memory leak each time an identical
existing key is added with add_key().
The reason the add_key() system call returns okay is that
key_create_or_update() calls __key_link_begin() before checking to see
whether it can update a key directly rather than adding/replacing - which
it turns out it can. Thus __key_link() is not called through
__key_instantiate_and_link() and __key_link_end() must cancel the edit.
CVE-2015-1333
Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: David Howells <dhowells@redhat.com> Signed-off-by: James Morris <james.l.morris@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
tpm_ibmvtpm_probe() calls ibmvtpm_reset_crq(ibmvtpm) without having yet
set the virtual device in the ibmvtpm structure. So in ibmvtpm_reset_crq,
the phype call contains empty unit addresses, ibmvtpm->vdev->unit_address.
Signed-off-by: Hon Ching(Vicky) Lo <honclo@linux.vnet.ibm.com> Signed-off-by: Joy Latten <jmlatten@linux.vnet.ibm.com> Reviewed-by: Ashley Lai <ashley@ahsleylai.com> Fixes: 132f76294744 ("drivers/char/tpm: Add new device driver to support IBM vTPM") Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If we create a CRC filesystem, mount it, and create a symlink with
a path long enough that it can't live in the inode, we get a very
strange result upon remount:
# ls -l mnt
total 4
lrwxrwxrwx. 1 root root 929 Jun 15 16:58 link -> XSLM
XSLM is the V5 symlink block header magic (which happens to be
followed by a NUL, so the string looks terminated).
xfs_readlink_bmap() advanced cur_chunk by the size of the header
for CRC filesystems, but never actually used that pointer; it
kept reading from bp->b_addr, which is the start of the block,
rather than the start of the symlink data after the header.
Looks like this problem goes back to v3.10.
Fixing this gets us reading the proper link target, again.
Signed-off-by: Eric Sandeen <sandeen@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This can be the case when the GPU is powered off, e.g. via vgaswitcheroo
or runpm. When the GPU is powered up again, radeon_gart_table_vram_pin
flushes the TLB after setting rdev->gart.ptr to non-NULL.
Fixes panic on powering off R7xx GPUs.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=61529 Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Michel Dänzer <michel.daenzer@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In order for hibernation to reliably work we need to properly turn
off the SDMA block, sadly after numerous attemps i haven't not found
proper sequence for clean and full shutdown. So simply reset both
SDMA block, this makes hibernation works reliably on sea island GPU
family (CI)
Hibernation and suspend to ram were tested (several times) on :
Bonaire
Hawaii
Mullins
Kaveri
Kabini
Signed-off-by: Jérôme Glisse <jglisse@redhat.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In order for hibernation to reliably work we need to cleanup more
thoroughly the compute ring. Hibernation is different from suspend
resume as when we resume from hibernation the hardware is first
fully initialize by regular kernel then freeze callback happens
(which correspond to a suspend inside the radeon kernel driver)
and turn off each of the block. It turns out we were not cleanly
shutting down the compute ring. This patch fix that.
Hibernation and suspend to ram were tested (several times) on :
Bonaire
Hawaii
Mullins
Kaveri
Kabini
Changed since v1:
- Factor the ring stop logic into a function taking ring as arg.
Signed-off-by: Jérôme Glisse <jglisse@redhat.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the function fails reference counter to the object is not decremented
causing leaks.
This is hard to spot as it happens only on very low memory situations.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com> Reviewed-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If objects are moved back from system memory to VRAM (and spice id
created again) memory is already initialized so we need to set flag
to not clear memory.
If you don't do it after a while using desktop many images turns to
black or transparents.
Signed-off-by: Frediano Ziglio <fziglio@redhat.com> Reviewed-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Allocate memory using GFP_NOIO when deleting a btree. dm_btree_del()
can be called via an ioctl and we don't want to recurse into the FS or
block layer.
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
redistribute3() shares entries out across 3 nodes. Some entries were
being moved the wrong way, breaking the ordering. This manifested as a
BUG() in dm-btree-remove.c:shift() when entries were removed from the
btree.
For additional context see:
https://www.redhat.com/archives/dm-devel/2015-May/msg00113.html
Signed-off-by: Dennis Yang <shinrairis@gmail.com> Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The metadata space map has a simplified 'bootstrap' mode that is
operational when extending the space maps. Whilst in this mode it's
possible for some refcount decrement operations to become queued (eg, as
a result of shadowing one of the bitmap indexes). These decrements were
not being applied when switching out of bootstrap mode.
The effect of this bug was the leaking of a 4k metadata block. This is
detected by the latest version of thin_check as a non fatal error.
Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
virt_dev->num_cached_rings counts on freed ring and is not updated
correctly. In xhci_free_or_cache_endpoint_ring() function, the free ring
is added into cache and then num_rings_cache is incremented as below:
virt_dev->ring_cache[rings_cached] =
virt_dev->eps[ep_index].ring;
virt_dev->num_rings_cached++;
here, free ring pointer is added to a current index and then
index is incremented.
So current index always points to empty location in the ring cache.
For getting available free ring, current index should be decremented
first and then corresponding ring buffer value should be taken from ring
cache.
But In function xhci_endpoint_init(), the num_rings_cached index is
accessed before decrement.
virt_dev->eps[ep_index].new_ring =
virt_dev->ring_cache[virt_dev->num_rings_cached];
virt_dev->ring_cache[virt_dev->num_rings_cached] = NULL;
virt_dev->num_rings_cached--;
This is bug in manipulating the index of ring cache.
And it should be as below:
virt_dev->num_rings_cached--;
virt_dev->eps[ep_index].new_ring =
virt_dev->ring_cache[virt_dev->num_rings_cached];
virt_dev->ring_cache[virt_dev->num_rings_cached] = NULL;
Signed-off-by: Aman Deep <aman.deep@samsung.com> Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Currently, we're calling musb_start() twice for DRD ports
in some situations. This has been observed to cause enumeration
issues after suspend/resume cycles with AM335x.
In order to fix the problem, we just have to fix the check
on musb_has_gadget() so that it only returns true if
current mode is Host and ignore the fact that we have or
not a gadget driver loaded.
Fixes: ae44df2e21b5 (usb: musb: call musb_start() only once in OTG mode) Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Tested-by: Sekhar Nori <nsekhar@ti.com> Signed-off-by: Felipe Balbi <balbi@ti.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Static checkers complain that the current condition is never true. It
seems pretty likely that it's a typo and "URB" was intended instead of
"USB".
Fixes: 3d97ff63f899 ('usbdevfs: Use scatter-gather lists for large bulk transfers') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Acked-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes an issue introduced in commit b23c843992b6 (usb: dwc3:
gadget: fix DEPSTARTCFG for non-EP0 EPs) that made sure we would
only use DEPSTARTCFG once per SetConfig.
The trick is that we should use one DEPSTARTCFG per SetConfig *OR*
SetInterface. SetInterface was completely missed from the original
patch.
This problem became aparent after commit 76e838c9f776 (usb: dwc3:
gadget: return error if command sent to DEPCMD register fails)
added checking of the return status of device endpoint commands.
'Set Endpoint Transfer Resource' command was caught failing
occasionally. This is because the Transfer Resource
Index was not getting reset during a SET_INTERFACE request.
Finally, to fix the issue, was we have to do is make sure that
our start_config_issued flag gets reset whenever we receive a
SetInterface request.
To verify the problem (and its fix), all we have to do is run
test 9 from testusb with 'testusb -t 9 -s 2048 -a -c 5000'.
I have a ST4000DM000 disk. If Linux is booted while the disk is spun down,
the command that sets transfer mode causes the disk to spin up. The
spin-up takes longer than the default 5s timeout, so the command fails and
timeout is reported.
Fix this by increasing the timeout to 15s, which is enough for the disk to
spin up.