linux-stable.git/drivers/usb/host, branch linux-2.6.37.y Linux kernel stable tree ehci-hcd: Bug fix: don't set a QH's Halt bit 2011-03-27T19:00:26+00:00 Alan Stern stern@rowland.harvard.edu 2011-03-16T14:57:15+00:00 2f10bd1ca416eec79654d65ca9888cc1db3ca743 commit b5a3b3d985493c173925907adfebf3edab236fe7 upstream. This patch (as1453) fixes a long-standing bug in the ehci-hcd driver. There is no need to set the Halt bit in the overlay region for an unlinked or blocked QH. Contrary to what the comment says, setting the Halt bit does not cause the QH to be patched later; that decision (made in qh_refresh()) depends only on whether the QH is currently pointing to a valid qTD. Likewise, setting the Halt bit does not prevent completions from activating the QH while it is "stopped"; they are prevented by the fact that qh_completions() temporarily changes qh->qh_state to QH_STATE_COMPLETING. On the other hand, there are circumstances in which the QH will be reactivated _without_ being patched; this happens after an URB beyond the head of the queue is unlinked. Setting the Halt bit will then cause the hardware to see the QH with both the Active and Halt bits set, an invalid combination that will prevent the queue from advancing and may even crash some controllers. Apparently the only reason this hasn't been reported before is that unlinking URBs from the middle of a running queue is quite uncommon. However Test 17, recently added to the usbtest driver, does exactly this, and it confirms the presence of the bug. In short, there is no reason to set the Halt bit for an unlinked or blocked QH, and there is a very good reason not to set it. Therefore the code that sets it is removed. Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Tested-by: Andiry Xu <andiry.xu@amd.com> CC: David Brownell <david-b@pacbell.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit b5a3b3d985493c173925907adfebf3edab236fe7 upstream.

This patch (as1453) fixes a long-standing bug in the ehci-hcd driver.

There is no need to set the Halt bit in the overlay region for an
unlinked or blocked QH.  Contrary to what the comment says, setting
the Halt bit does not cause the QH to be patched later; that decision
(made in qh_refresh()) depends only on whether the QH is currently
pointing to a valid qTD.  Likewise, setting the Halt bit does not
prevent completions from activating the QH while it is "stopped"; they
are prevented by the fact that qh_completions() temporarily changes
qh->qh_state to QH_STATE_COMPLETING.

On the other hand, there are circumstances in which the QH will be
reactivated _without_ being patched; this happens after an URB beyond
the head of the queue is unlinked.  Setting the Halt bit will then
cause the hardware to see the QH with both the Active and Halt bits
set, an invalid combination that will prevent the queue from
advancing and may even crash some controllers.

Apparently the only reason this hasn't been reported before is that
unlinking URBs from the middle of a running queue is quite uncommon.
However Test 17, recently added to the usbtest driver, does exactly
this, and it confirms the presence of the bug.

In short, there is no reason to set the Halt bit for an unlinked or
blocked QH, and there is a very good reason not to set it.  Therefore
the code that sets it is removed.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested-by: Andiry Xu <andiry.xu@amd.com>
CC: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Fix cycle bit calculation during stall handling. 2011-03-23T19:50:20+00:00 Sarah Sharp sarah.a.sharp@linux.intel.com 2011-02-24T02:12:29+00:00 729bc845266c888b2b9ad6ceaea28312ec0ad39e commit 01a1fdb9a7afa5e3c14c9316d6f380732750b4e4 upstream. When an endpoint stalls, we need to update the xHCI host's internal dequeue pointer to move it past the stalled transfer. This includes updating the cycle bit (TRB ownership bit) if we have moved the dequeue pointer past a link TRB with the toggle cycle bit set. When we're trying to find the new dequeue segment, find_trb_seg() is supposed to keep track of whether we've passed any link TRBs with the toggle cycle bit set. However, this while loop's body while (cur_seg->trbs > trb || &cur_seg->trbs[TRBS_PER_SEGMENT - 1] < trb) { Will never get executed if the ring only contains one segment. find_trb_seg() will return immediately, without updating the new cycle bit. Since find_trb_seg() has no idea where in the segment the TD that stalled was, make the caller, xhci_find_new_dequeue_state(), check for this special case and update the cycle bit accordingly. This patch should be queued to kernels all the way back to 2.6.31. Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Tested-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit 01a1fdb9a7afa5e3c14c9316d6f380732750b4e4 upstream.

When an endpoint stalls, we need to update the xHCI host's internal
dequeue pointer to move it past the stalled transfer.  This includes
updating the cycle bit (TRB ownership bit) if we have moved the dequeue
pointer past a link TRB with the toggle cycle bit set.

When we're trying to find the new dequeue segment, find_trb_seg() is
supposed to keep track of whether we've passed any link TRBs with the
toggle cycle bit set.  However, this while loop's body

	while (cur_seg->trbs > trb ||
			&cur_seg->trbs[TRBS_PER_SEGMENT - 1] < trb) {

Will never get executed if the ring only contains one segment.
find_trb_seg() will return immediately, without updating the new cycle
bit.  Since find_trb_seg() has no idea where in the segment the TD that
stalled was, make the caller, xhci_find_new_dequeue_state(), check for
this special case and update the cycle bit accordingly.

This patch should be queued to kernels all the way back to 2.6.31.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Tested-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Update internal dequeue pointers after stalls. 2011-03-23T19:50:19+00:00 Sarah Sharp sarah.a.sharp@linux.intel.com 2011-02-23T23:46:42+00:00 5f1e1e95d04fc2159484ab3d8295d9167de846b1 commit bf161e85fb153c0dd5a95faca73fd6a9d237c389 upstream. When an endpoint stalls, the xHCI driver must move the endpoint ring's dequeue pointer past the stalled transfer. To do that, the driver issues a Set TR Dequeue Pointer command, which will complete some time later. Takashi was having issues with USB 1.1 audio devices that stalled, and his analysis of the code was that the old code would not update the xHCI driver's ring dequeue pointer after the command completes. However, the dequeue pointer is set in xhci_find_new_dequeue_state(), just before the set command is issued to the hardware. Setting the dequeue pointer before the Set TR Dequeue Pointer command completes is a dangerous thing to do, since the xHCI hardware can fail the command. Instead, store the new dequeue pointer in the xhci_virt_ep structure, and update the ring's dequeue pointer when the Set TR dequeue pointer command completes. While we're at it, make sure we can't queue another Set TR Dequeue Command while the first one is still being processed. This just won't work with the internal xHCI state code. I'm still not sure if this is the right thing to do, since we might have a case where a driver queues multiple URBs to a control ring, one of the URBs Stalls, and then the driver tries to cancel the second URB. There may be a race condition there where the xHCI driver might try to issue multiple Set TR Dequeue Pointer commands, but I would have to think very hard about how the Stop Endpoint and cancellation code works. Keep the fix simple until when/if we run into that case. This patch should be queued to kernels all the way back to 2.6.31. Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Tested-by: Takashi Iwai <tiwai@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit bf161e85fb153c0dd5a95faca73fd6a9d237c389 upstream.

When an endpoint stalls, the xHCI driver must move the endpoint ring's
dequeue pointer past the stalled transfer.  To do that, the driver issues
a Set TR Dequeue Pointer command, which will complete some time later.

Takashi was having issues with USB 1.1 audio devices that stalled, and his
analysis of the code was that the old code would not update the xHCI
driver's ring dequeue pointer after the command completes.  However, the
dequeue pointer is set in xhci_find_new_dequeue_state(), just before the
set command is issued to the hardware.

Setting the dequeue pointer before the Set TR Dequeue Pointer command
completes is a dangerous thing to do, since the xHCI hardware can fail the
command.  Instead, store the new dequeue pointer in the xhci_virt_ep
structure, and update the ring's dequeue pointer when the Set TR dequeue
pointer command completes.

While we're at it, make sure we can't queue another Set TR Dequeue Command
while the first one is still being processed.  This just won't work with
the internal xHCI state code.  I'm still not sure if this is the right
thing to do, since we might have a case where a driver queues multiple
URBs to a control ring, one of the URBs Stalls, and then the driver tries
to cancel the second URB.  There may be a race condition there where the
xHCI driver might try to issue multiple Set TR Dequeue Pointer commands,
but I would have to think very hard about how the Stop Endpoint and
cancellation code works.  Keep the fix simple until when/if we run into
that case.

This patch should be queued to kernels all the way back to 2.6.31.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Tested-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
ehci: Check individual port status registers on resume 2011-03-23T19:50:18+00:00 Matthew Garrett mjg@redhat.com 2011-01-11T17:26:48+00:00 f2d3ab3688b54f1e1cb7998c599d806e0de0b800 commit 294d95f2cbc2aef5346258f216cd9df570e271a5 upstream. If a device plug/unplug is detected on an ATI SB700 USB controller in D3, it appears to set the port status register but not the controller status register. As a result we'll fail to detect the plug event. Check the port status register on resume as well in order to catch this case. Signed-off-by: Matthew Garrett <mjg@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit 294d95f2cbc2aef5346258f216cd9df570e271a5 upstream.

If a device plug/unplug is detected on an ATI SB700 USB controller in D3,
it appears to set the port status register but not the controller status
register. As a result we'll fail to detect the plug event. Check the port
status register on resume as well in order to catch this case.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
USB: isp1760: Implement solution for erratum 2 2011-03-23T19:50:17+00:00 Sebastian Andrzej Siewior bigeasy@linutronix.de 2011-02-08T20:07:40+00:00 dd4dd522c7fc2c1d55713c27ca30820f22e0a225 commit b14e840d04dba211fbdc930247e379085623eacd upstream. The document says: |2.1 Problem description | When at least two USB devices are simultaneously running, it is observed that | sometimes the INT corresponding to one of the USB devices stops occurring. This may | be observed sometimes with USB-to-serial or USB-to-network devices. | The problem is not noticed when only USB mass storage devices are running. |2.2 Implication | This issue is because of the clearing of the respective Done Map bit on reading the ATL | PTD Done Map register when an INT is generated by another PTD completion, but is not | found set on that read access. In this situation, the respective Done Map bit will remain | reset and no further INT will be asserted so the data transfer corresponding to that USB | device will stop. |2.3 Workaround | An SOF INT can be used instead of an ATL INT with polling on Done bits. A time-out can | be implemented and if a certain Done bit is never set, verification of the PTD completion | can be done by reading PTD contents (valid bit). | This is a proven workaround implemented in software. Russell King run into this with an USB-to-serial converter. This patch implements his suggestion to enable the high frequent SOF interrupt only at the time we have ATL packages queued. It goes even one step further and enables the SOF interrupt only if we have more than one ATL packet queued at the same time. Tested-by: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit b14e840d04dba211fbdc930247e379085623eacd upstream.

The document says:
|2.1 Problem description
|    When at least two USB devices are simultaneously running, it is observed that
|    sometimes the INT corresponding to one of the USB devices stops occurring. This may
|    be observed sometimes with USB-to-serial or USB-to-network devices.
|    The problem is not noticed when only USB mass storage devices are running.
|2.2 Implication
|    This issue is because of the clearing of the respective Done Map bit on reading the ATL
|    PTD Done Map register when an INT is generated by another PTD completion, but is not
|    found set on that read access. In this situation, the respective Done Map bit will remain
|    reset and no further INT will be asserted so the data transfer corresponding to that USB
|    device will stop.
|2.3 Workaround
|    An SOF INT can be used instead of an ATL INT with polling on Done bits. A time-out can
|    be implemented and if a certain Done bit is never set, verification of the PTD completion
|    can be done by reading PTD contents (valid bit).
|    This is a proven workaround implemented in software.

Russell King run into this with an USB-to-serial converter. This patch
implements his suggestion to enable the high frequent SOF interrupt only
at the time we have ATL packages queued. It goes even one step further
and enables the SOF interrupt only if we have more than one ATL packet
queued at the same time.

Tested-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Fix an error in count_sg_trbs_needed() 2011-03-07T23:05:13+00:00 Paul Zimmerman Paul.Zimmerman@synopsys.com 2011-02-12T22:07:57+00:00 929364a96776f45edc9b1d71680bf87852dfc39b commit bcd2fde05341cef0052e49566ec88b406a521cf3 upstream. The expression while (running_total < sg_dma_len(sg)) does not take into account that the remaining data length can be less than sg_dma_len(sg). In that case, running_total can end up being greater than the total data length, so an extra TRB is counted. Changing the expression to while (running_total < sg_dma_len(sg) && running_total < temp) fixes that. This patch should be queued for stable kernels back to 2.6.31. Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit bcd2fde05341cef0052e49566ec88b406a521cf3 upstream.

The expression

	while (running_total < sg_dma_len(sg))

does not take into account that the remaining data length can be less
than sg_dma_len(sg). In that case, running_total can end up being
greater than the total data length, so an extra TRB is counted.
Changing the expression to

	while (running_total < sg_dma_len(sg) && running_total < temp)

fixes that.

This patch should be queued for stable kernels back to 2.6.31.

Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Fix errors in the running total calculations in the TRB math 2011-03-07T23:05:13+00:00 Paul Zimmerman Paul.Zimmerman@synopsys.com 2011-02-12T22:07:20+00:00 286681ef90a00ad6fdab8d47ac794d1f59004305 commit 5807795bd4dececdf553719cc02869e633395787 upstream. Calculations like running_total = TRB_MAX_BUFF_SIZE - (sg_dma_address(sg) & (TRB_MAX_BUFF_SIZE - 1)); if (running_total != 0) num_trbs++; are incorrect, because running_total can never be zero, so the if() expression will never be true. I think the intention was that running_total be in the range of 0 to TRB_MAX_BUFF_SIZE-1, not 1 to TRB_MAX_BUFF_SIZE. So adding a running_total &= TRB_MAX_BUFF_SIZE - 1; fixes the problem. This patch should be queued for stable kernels back to 2.6.31. Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit 5807795bd4dececdf553719cc02869e633395787 upstream.

Calculations like

	running_total = TRB_MAX_BUFF_SIZE -
		(sg_dma_address(sg) & (TRB_MAX_BUFF_SIZE - 1));
	if (running_total != 0)
		num_trbs++;

are incorrect, because running_total can never be zero, so the if()
expression will never be true. I think the intention was that
running_total be in the range of 0 to TRB_MAX_BUFF_SIZE-1, not 1
to TRB_MAX_BUFF_SIZE. So adding a

	running_total &= TRB_MAX_BUFF_SIZE - 1;

fixes the problem.

This patch should be queued for stable kernels back to 2.6.31.

Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Clarify some expressions in the TRB math 2011-03-07T23:05:13+00:00 Paul Zimmerman Paul.Zimmerman@synopsys.com 2011-02-12T22:06:44+00:00 2094b73f358ba57e57e11a775965d66f36a3647d commit a2490187011cc2263117626615a581927d19f1d3 upstream. This makes it easier to spot some problems, which will be fixed by the next patch in the series. Also change dev_dbg to dev_err in check_trb_math(), so any math errors will be visible even when running with debug disabled. Note: This patch changes the expressions containing "((1 << TRB_MAX_BUFF_SHIFT) - 1)" to use the equivalent "(TRB_MAX_BUFF_SIZE - 1)". No change in behavior is intended for those expressions. This patch should be queued for stable kernels back to 2.6.31. Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit a2490187011cc2263117626615a581927d19f1d3 upstream.

This makes it easier to spot some problems, which will be fixed by the
next patch in the series. Also change dev_dbg to dev_err in
check_trb_math(), so any math errors will be visible even when running
with debug disabled.

Note: This patch changes the expressions containing
"((1 << TRB_MAX_BUFF_SHIFT) - 1)" to use the equivalent
"(TRB_MAX_BUFF_SIZE - 1)". No change in behavior is intended for
those expressions.

This patch should be queued for stable kernels back to 2.6.31.

Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Avoid BUG() in interrupt context 2011-03-07T23:05:13+00:00 Paul Zimmerman Paul.Zimmerman@synopsys.com 2011-02-12T22:06:06+00:00 07570c3e32779255a8585ce05290ad056ef8b05b commit 68e41c5d032668e2905404afbef75bc58be179d6 upstream. Change the BUGs in xhci_find_new_dequeue_state() to WARN_ONs, to avoid bringing down the box if one of them is hit This patch should be queued for stable kernels back to 2.6.31. Signed-off-by: Paul Zimmerman <paulz@synopsys.com> Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit 68e41c5d032668e2905404afbef75bc58be179d6 upstream.

Change the BUGs in xhci_find_new_dequeue_state() to WARN_ONs, to avoid
bringing down the box if one of them is hit

This patch should be queued for stable kernels back to 2.6.31.

Signed-off-by: Paul Zimmerman <paulz@synopsys.com>
Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
xhci: Use GFP_NOIO during device reset. 2011-02-17T23:15:22+00:00 Sarah Sharp sarah.a.sharp@linux.intel.com 2010-12-28T21:08:42+00:00 61e1444816df473bb5648541ab6e5d8897972053 commit a6d940dd759bf240d28624198660ed34582a327b upstream. When xhci_discover_or_reset_device() is called after a host controller power loss, the virtual device may need to be reallocated. Make sure xhci_alloc_dev() uses GFP_NOIO. This avoid causing a deadlock by allowing the kernel to flush pending I/O while reallocating memory for a virtual device for a USB mass storage device that's holding the backing store for dirty memory buffers. This patch should be queued for the 2.6.37 stable tree. Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> 
commit a6d940dd759bf240d28624198660ed34582a327b upstream.

When xhci_discover_or_reset_device() is called after a host controller
power loss, the virtual device may need to be reallocated.  Make sure
xhci_alloc_dev() uses GFP_NOIO.  This avoid causing a deadlock by allowing
the kernel to flush pending I/O while reallocating memory for a virtual
device for a USB mass storage device that's holding the backing store for
dirty memory buffers.

This patch should be queued for the 2.6.37 stable tree.

Signed-off-by: Sarah Sharp <sarah.a.sharp@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>