This patch (as1243) tries to improve ehci-hcd's scheduling of
interrupt transfers. Instead of trying to cram all transfers with the
same period into the same frame, the new code will spread the
transfers out among lots of different frames. This should reduce the
periodic schedule load in any one frame -- some host controllers have
trouble when there's too much work to do.
A more thorough approach would stagger the uframe values as well. But
this is enough to make a big improvement.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Tested-by: Dwayne Fontenot <dwayne.fontenot@att.net>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Without this patch, the driver won't check that the last fully-occupied
uframe for a new split transaction was vacant beforehand. This can
lead to a situation in which the first 188 bytes of a 192-byte
isochronous transfer are scheduled in the same uframe as an existing
interrupt transfer. The resulting schedule looks like this:
uframe 0: 188-byte isoc-OUT SSPLIT, 8-byte int-IN SSPLIT
uframe 1: 4-byte isoc-OUT SSPLIT
The SSPLITs are intermingled, causing an error in the downstream hub's
TT.
If you are having problems with devices or hub ports resetting, or failed
interrupt transfers, when you start using a USB audio or video (Isochronous)
device, this patch may help.
Signed-off-by: Dan Streetman <ddstreet@ieee.org>
Reported-by: Kung James <kong1191@gmail.com>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Alan Stern <stern@rowland.harvard.edu>
The base versions handle constant folding now.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
ehci-hcd uses usb_get_urb() and usb_put_urb() in an unbalanced way causing
isochronous URB's kref.counts incrementing once per usb_submit_urb() call.
The culprit is *usb being set to NULL when usb_put_urb() is called after URB
is given back.
Due to other fixes there is no need for ehci-hcd to deal with usb_get_urb()
nor usb_put_urb() anymore, so patch removes their usages in ehci-hcd.
Patch also makes ehci_to_hcd(ehci)->self.bandwidth_allocated adjust, if a
stream finishes.
Signed-off-by: Karsten Wiese <fzu@wemgehoertderstaat.de>
Cc: David Brownell <david-b@pacbell.net>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Currently ITDs are immediately recycled whenever their URB completes.
However, EHCI hardware can sometimes remember some ITD state. This
means that when the ITD is reused before end-of-frame it may sometimes
cause the hardware to reference bogus state.
This patch defers reusing such ITDs by moving them into a new ehci member
cached_itd_list. ITDs resting in cached_itd_list are moved back into their
stream's free_list once scan_periodic() detects that the active frame has
elapsed.
This makes the snd_usb_us122l driver (in kernel since .28) work right
when it's hooked up through EHCI.
[ dbrownell@users.sourceforge.net: comment fixups ]
Signed-off-by: Karsten Wiese <fzu@wemgehoertderstaat.de>
Tested-by: Philippe Carriere <philippe-f.carriere@wanadoo.fr>
Tested-by: Federico Briata <federicobriata@gmail.com>
Cc: stable <stable@kernel.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1164) fixes a bug in the EHCI scheduler. The interval
value it uses is already in linear format, not logarithmically coded.
The existing code can sometimes crash the system by trying to divide
by zero.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Cc: David Brownell <david-b@pacbell.net>
Cc: Stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
As noted by Stefan Neis <Stefan.Neis@kobil.com>, we had a recent
regression with EHCI periodic transfers, in some (seemingly not
all that common) cases.
The root cause was that the schedule activation was only loosely
coupled to the addition or removal of transfers, so two different
execution contexts could both think they had to deactivate (or
conversely activate) the schedule. So this fix tightens that
coupling, managing it more like a refcount.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1098) changes the way ehci-hcd schedules its periodic
Iso transfers. That the current scheduling code is wrong is clear on
the face of it: Sometimes it returns -EL2NSYNC (meaning that an URB
couldn't be scheduled because it was submitted too late), but it does
this even when the URB_ISO_ASAP flag is set (meaning the URB should be
scheduled as soon as possible).
The new code properly implements as-soon-as-possible scheduling,
assigning the next unexpired slot as the URB's starting point. It
also is more careful about checking for Iso URB completion: It doesn't
bother to check for activity during frames that are already over,
and it allows for the possibility that some of the URB's packets may
have raced the hardware when they were submitted and so never got used
(the packet status is set to -EXDEV).
This fixes problems several people have experienced with USB video
applications.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
__FUNCTION__ is gcc-specific, use __func__
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Refactor the EHCI "if (handshake()) state = HC_STATE_HALT" idiom,
which appears 4 times, by replacing it with calls to a new function
called handshake_on_error_set_halt(). Saves a few bytes too.
Signed-off-by: Karsten Wiese <fzu@wemgehoertderstaat.de>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This has some bugfixes for the EHCI driver's ISO transfer scanning
logic. It was leaving ITDs and SITDs on the schedule too long, for
a few different reasons, which caused trouble.
(a) Look at all microframes for high speed transfers, not just
the ones we expect to have finished. This way transfers
ending mid-frame will complete without needing another IRQ.
This also minimizes bogus scheduling underruns (e.g. EL2NSYNC).
(b) When we encounter an ISO transfer (either speed, but this
hits mostly at full speed) that's not yet been completed,
immediately stop scanning; we've caught up to the hardware,
no matter what other indications might say.
(c) Always clean up ITDs (for high speed transfers) when the HC
is no longer running.
I'm not sure whether the last one has been observed before, but both
the others have been reported with "real world" audio and video code.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Small updates to the EHCI driver's ISO support:
- Get rid of the Kconfig option for full speed ISO. It may
not be perfect yet, but it hasn't appeared to be dangerous
and pretty much every configuration wants it.
- Instead of two places to disable an empty periodic schedule
after an ISO transfer completes, just have one.
- After the periodic schedule is disabled, we can short-circuit
the schedule scan ... it can't possibly have more work to do.
Assuming a typical config with split iso enabled, the only change
in behavior should be almost unobservable: quicker termination
of periodic scans when the schedule gets emptied.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In the EHCI driver, itd->usecs[8] is used in periodic_usecs(), indexed by
uframe. For an ITD's unused uframes it is 0, else it contains the same
value as itd->stream->usecs. To check if an ITD's uframe is used, we can
instead test itd->hw_transaction[uframe]: if used, it will be nonzero no
matter what endianess is used.
This patch replaces those two uses, eliminates itd->usecs[], and saves
eight bytes from each ITD.
Signed-off-by: Karsten Wiese <fzu@wemgehoertderstaat.de>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This adds a workaround for an issue reported with ISO transfers
on some EHCI controllers, most recently with VIA KT800 and PS3
EHCI silicon.
The issue is that the silicon doesn't necessarily seem to be done
using ISO DMA descriptors (itd, sitd) when it marks them inactive.
(One theory is that the ill-defined mechanism where hardware caches
periodic transfer descriptors isn't invalidating their state...)
With such silicon, quick re-use of those descriptors makes trouble.
Waiting until the next frame seems to be a sufficient workaround.
This patch ensures that the relevant descriptors aren't available
for immediate re-use. It does so by not recycling them until after
issuing the completion callback which would reuse them by enqueueing
an URB and thus (re)allocating ISO DMA descriptors.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Geoff Levand <geoffrey.levand@am.sony.com>
Cc: Masashi Kimoto <Masashi_Kimoto@hq.scei.sony.co.jp>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as974) reorganizes the way ehci-hcd sets urb->status. It
now keeps the information in a local variable until the last moment.
The patch also simplifies the handling of -EREMOTEIO, since the only
use of that code is to set the do_status flag.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as954) implements a suggestion of David Brownell's. Now
the host controller drivers are responsible for linking and unlinking
URBs to/from their endpoint queues. This eliminates the possiblity of
strange situations where usbcore thinks an URB is linked but the HCD
thinks it isn't. It also means HCDs no longer have to check for URBs
being dequeued before they were fully enqueued.
In addition to the core changes, this requires changing every host
controller driver and the root-hub URB handler. For the most part the
required changes are fairly small; drivers have to call
usb_hcd_link_urb_to_ep() in their urb_enqueue method,
usb_hcd_check_unlink_urb() in their urb_dequeue method, and
usb_hcd_unlink_urb_from_ep() before giving URBs back. A few HCDs make
matters more complicated by the way they split up the flow of control.
In addition some method interfaces get changed. The endpoint argument
for urb_enqueue is now redundant so it is removed. The unlink status
is required by usb_hcd_check_unlink_urb(), so it has been added to
urb_dequeue.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: David Brownell <david-b@pacbell.net>
CC: Olav Kongas <ok@artecdesign.ee>
CC: Tony Olech <tony.olech@elandigitalsystems.com>
CC: Yoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This reverts commit 196705c9bb. It was
reported to cause a regression by Daniel Exner, and Arjan van de Ven
points out that we actually already have infrastructure in place for
setting limits on acceptable DMA latency that would be the much more
correct fix for the problem with some Broadcom EHCI controllers.
Fixed up trivial conflicts due to the changes to support big-endian host
controller descriptors in drivers/usb/host/{ehci-sched.c,ehci.h}.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch implements supports for EHCI controllers whose in-memory
data structures are represented in big-endian format. This is needed
(unfortunately) for the AMCC PPC440EPx SoC EHCI controller; the EHCI
spec doesn't specify little-endian format, although that's what most
other implementations use.
The guts of the patch are to introduce the hc32 type and change all
references from le32 to hc32. All access routines are converted from
cpu_to_le32(...) to cpu_to_hc32(ehci, ...) and similar for the other
"direction". (This is the same approach used with OHCI.)
David fixed:
Whitespace fixes; refresh against ehci cpufreq patch; move glue
for that PPC driver to the patch adding it; fix free symbol
capture bugs in modified "constant" macros; and make "hc32" etc
be "le32" unless we really need the BE options, so "sparse" can
do some real good.
Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
EHCI controllers that don't cache enough microframes can get MMF errors
when CPU frequency changes occur between the start and completion of
split interrupt transactions, due to delays in reading main memory
(caused by CPU cache snoop delays).
This patch adds a cpufreq notifier to the EHCI driver that will
inactivate split interrupt transactions during frequency transitions.
It was tested on Intel ICH7 and Serverworks/Broadcom HT1000 EHCI
controllers.
Signed-off-by: Stuart Hayes <stuart_hayes@dell.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch implements supports for EHCI controllers whose MMIO
registers are big endian and enables that functionality for
the Toshiba SCC chip. It does _not_ add support for big endian
in-memory data structures as this is not needed for that chip
and I hope it will never be.
The guts of the patch are to convert readl(...) to
ehci_readl(ehci, ...) and similarly for register writes.
Signed-off-by: Kou Ishizaki <kou.ishizaki@toshiba.co.jp>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Geoff Levand <geoffrey.levand@am.sony.com>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Maintain a per-CPU global "struct pt_regs *" variable which can be used instead
of passing regs around manually through all ~1800 interrupt handlers in the
Linux kernel.
The regs pointer is used in few places, but it potentially costs both stack
space and code to pass it around. On the FRV arch, removing the regs parameter
from all the genirq function results in a 20% speed up of the IRQ exit path
(ie: from leaving timer_interrupt() to leaving do_IRQ()).
Where appropriate, an arch may override the generic storage facility and do
something different with the variable. On FRV, for instance, the address is
maintained in GR28 at all times inside the kernel as part of general exception
handling.
Having looked over the code, it appears that the parameter may be handed down
through up to twenty or so layers of functions. Consider a USB character
device attached to a USB hub, attached to a USB controller that posts its
interrupts through a cascaded auxiliary interrupt controller. A character
device driver may want to pass regs to the sysrq handler through the input
layer which adds another few layers of parameter passing.
I've build this code with allyesconfig for x86_64 and i386. I've runtested the
main part of the code on FRV and i386, though I can't test most of the drivers.
I've also done partial conversion for powerpc and MIPS - these at least compile
with minimal configurations.
This will affect all archs. Mostly the changes should be relatively easy.
Take do_IRQ(), store the regs pointer at the beginning, saving the old one:
struct pt_regs *old_regs = set_irq_regs(regs);
And put the old one back at the end:
set_irq_regs(old_regs);
Don't pass regs through to generic_handle_irq() or __do_IRQ().
In timer_interrupt(), this sort of change will be necessary:
- update_process_times(user_mode(regs));
- profile_tick(CPU_PROFILING, regs);
+ update_process_times(user_mode(get_irq_regs()));
+ profile_tick(CPU_PROFILING);
I'd like to move update_process_times()'s use of get_irq_regs() into itself,
except that i386, alone of the archs, uses something other than user_mode().
Some notes on the interrupt handling in the drivers:
(*) input_dev() is now gone entirely. The regs pointer is no longer stored in
the input_dev struct.
(*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does
something different depending on whether it's been supplied with a regs
pointer or not.
(*) Various IRQ handler function pointers have been moved to type
irq_handler_t.
Signed-Off-By: David Howells <dhowells@redhat.com>
(cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
[ ... when you have an editor set to remind you of whitespace bugs ... ]
Cosmetic EHCI changes: remove end-of-line whitespace, spaces before tabs.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This updates the EHCI driver by adding an improved scheduler for the
transaction translators, found in USB 2.0 hubs and used for low and
full speed devices.
- adds periodic_tt_usecs() and some helper functions, which does
the same thing that "periodic_usecs" does, except on the other
side of the TT, i.e. it calculates the low/fullspeed bandwidth
usage instead of highspeed.
- adds a tt_available() function which is the new implementation
of what tt_no_collision() does ... while tt_no_collision() ensures
that each TT handles only 1 periodic transfer at a time (a very
pessimistic approach) this version instead tracks bandwidth and
allows each TT to handle as many transfers as will fit on each TT's
downstream bus (closer to best-case).
The new scheduler is selected by a config option, marked as EXPERIMENTAL
so it can be tested (and more broadly reviewed) for a while until it
seems safe to remove the original scheduler.
Signed-off-by: Dan Streetman <ddstreet@ieee.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as640) removes several put_device and the corresponding
get_device calls from the USB core and HCDs. Some of the puts were done
in atomic contexts, and none of them are needed since the core now
guarantees that every endpoint will be disabled and every URB completed
before a USB device is released.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Acked-by: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch replaces the split ISO raw_mask calculation code in the
iso_stream_init() function that computed incorrect numbers of high
speed transactions for both input and output transfers.
In the output case, it added a superfluous start-split transaction for
all maxmimum packet sizes that are a multiple of 188.
In the input case, it forgot to add complete-split transactions for all
microframes covered by the full speed transaction, and the additional
complete-split transaction needed for the case when full speed data
starts arriving near the end of a microframe.
These changes don't affect the lack of full speed bandwidth, but at
least it removes the MMF errors that the HC raised with some input
streams.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch adds a reinitializion for the uf variable that got modified
by the preceding start-split bandwidth check.
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch closes the IRQ race and makes various other OHCI & EHCI code
path safer vs. suspend/resume.
I've been able to (finally !) successfully suspend and resume various
Mac models, with or without USB mouse plugged, or plugging while asleep,
or unplugging while asleep etc... all without a crash.
Alan, please verify the UHCI bit I did, I only verified that it builds.
It's very simple so I wouldn't expect any issue there. If you aren't
confident, then just drop the hunks that change uhci-hcd.c
I also made the patch a little bit more "safer" by making sure the store
to the interrupt register that disables interrupts is not posted before
I set the flag and drop the spinlock.
Without this patch, you cannot reliably sleep/wakeup any recent Mac, and
I suspect PCs have some more sneaky issues too (they don't frankly crash
with machine checks because x86 tend to silently swallow PCI errors but
that won't last afaik, at least PCI Express will blow up in those
situations, but the USB code may still misbehave).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This adds the field tt_usecs to ehci_qh and ehci_iso_stream, and sets it
appropriately when setting them up as periodic endpoints. It records
the transation translator's think_time (added in last patch) plus the
downstream (i.e. low or full speed) bustime of the transfer associated
with each interrupt or iso frame, as calculated by usb_calc_bus_time.
Signed-off-by: Dan Streetman <ddstreet@ieee.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch converts kcalloc(1, ...) calls to use the new kzalloc() function.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch has a one line oops fix, plus related cleanups.
- The bugfix uses microframe scheduling data given to the hardware to
test "is this a periodic QH", rather than testing for nonzero period.
(Prevents an oops by providing the correct answer.)
- The cleanup going along with the patch should make it clearer what's
going on whenever those bitfields are accessed.
The bug came about when, around January, two new kinds of EHCI interrupt
scheduling operation were added, involving both the high speed (24 KBytes
per millisec) and low/full speed (1-64 bytes per millisec) microframe
scheduling. A driver for the Edirol UA-1000 Audio Capture Unit ran into
the oops; it used one of the newly supported high speed modes.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Greg,
This patch fixes the kmalloc() flags argument type in USB
subsystem; hopefully all of its occurences. The patch was
made against patch-2.6.12-git2 from Jun 20.
Cleanup of flags for kmalloc() in USB subsystem.
Signed-off-by: Olav Kongas <ok@artecdesign.ee>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The itd_patch() function is responsible for allocating entries in the
buffer page pointer list of the iTD. Particularly, a new page pointer
is needed every time when buffer data crosses a page boundary.
However, there is a bug in the allocation logic: the function does not
allocate a new entry when the current transaction is the first
transaction in the iTD (as indicated by first!=0).
The consequence is that, when the data of the first transaction begins
somewhere at the end of a page so that it actually does cross the page
boundary, no new page pointer is allocated. This means that the data
at the end of the first transaction (beyond the page boundary) will be
accessed by the HC using the second page pointer, which is zero.
Furthermore, the first page pointer will be later overwritten by the
page pointers of the other transactions, which will garble it because
the value is or-ed into the iTD field.
All this particular check (for !first) does is cause incorrect
behaviour, so it should be entirely removed (and with it the variable
first that is not used for anything else).
Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch includes minor "sparse -Wbitwise" updates for the PCI based
HCDs. Almost all of them involve just changing the second parameter of the
suspend() method to a pm_message_t ... the others relate to how the EHCI
code walks in-memory data structures. (There's a minor bug fixed there too
... affecting the big-endian sysfs async schedule dump.)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Index: gregkh-2.6/drivers/usb/core/hcd.h
===================================================================
Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.
Let it rip!
