OMAP: DSS2: OMAPFB: Fix invalid bpp for PAL and NTSC modes
omapfb_mode_to_timings() sets the bpp to 0 when bootarg omapfb.mode is set to
either "pal" or "ntsc". This patch corrects this by setting the bpp to 24, as
would be done if omapdss_default_get_recommended_bpp() would be called.
Signed-off-by: Maurus Cuelenaere <mcuelenaere@gmail.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Afzal Mohammed [Thu, 1 Jul 2010 13:40:01 +0000 (15:40 +0200)]
OMAP: DSS2: OMAPFB: Fix probe error path
Move sysfs entry creation to omapfb_probe() from omapfb_create_framebuffers().
This will make sure that sysfs entry is not left behind in case of unsuccessful
probe due to failure in enabling fb0 of omapfb_create_framebuffers().
Signed-off-by: Afzal Mohammed <lazfamam@gmail.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Vaibhav Hiremath [Fri, 19 Mar 2010 11:21:19 +0000 (11:21 +0000)]
OMAP3EVM: Replace vdvi regulator supply with vdds_dsi
With recent changes happened in OMAP2/3 DSS library for regulator interface, it
is required to define DSI regulator supply, without this DSS (in turn Fbdev)
fails to get regulator.
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Thu, 3 Jun 2010 13:27:46 +0000 (16:27 +0300)]
OMAP: DSS2: adjust YUV overlay width to be even
An overlay in YUV mode has to have an even input width, because data for
each pixel is divided between two adjacent pixels.
The algorithm handling manual update overlay adjusting may adjust the
overlay width to be odd.
This patch adds a check for that situation, and makes the width even.
The width is increased by one if it is possible (the unadjusted input
width is larger than the width), and decreased by one if increasing is
not possible.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Jani Nikula [Tue, 1 Jun 2010 15:08:18 +0000 (18:08 +0300)]
OMAP: DSS2: OMAPFB: Fix sysfs mirror input check
Using bool silently converted input to 0 or 1, making the range check
useless. Use unsigned long instead, and convert to bool later. Found by
Coverity.
Signed-off-by: Jani Nikula <ext-jani.1.nikula@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
OMAP DSS HW cannot send updates with odd widths. Normally the widths are
made even while preparing the update.
This patch adds a BUG_ON() to check if the update width is even. This is
to detect broken updates cleanly, as otherwise the OMAP DSS HW will just
halt, leading to obscure error situations.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Tue, 18 May 2010 10:35:37 +0000 (13:35 +0300)]
OMAP: DSS2: DSI: increase FIFO low threshold
The current FIFO low threshold was too low, and caused the FIFO to run
empty when core domain went to INA state between FIFO fills. This patch
increases the low threshold to keep that from happening.
The threshold values depend quite much on the HW and the use cases, so
this should actually be somehow configurable from board files, perhaps.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Fri, 14 May 2010 14:42:07 +0000 (17:42 +0300)]
OMAP: DSS2: DSI: Remove BTA after set_max_rx_packet_size
SMRPS function always sent BTA after sending the SMRPS packet. This is
not needed, and also caused some (buggy) panels to bug. This patch
removes the BTA.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Wed, 9 Jun 2010 12:31:34 +0000 (15:31 +0300)]
OMAP: DSS2: change manual update scaling setup
Currently the update area on manual update displays is automatically
enlargened to fully cover scaled overlays. This patch makes that
optional, allowing the panel driver to choose if it's used or not.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Wed, 9 Jun 2010 12:31:01 +0000 (15:31 +0300)]
OMAP: DSS2: DSI: use BTA to end the frame transfer
Previously a work was started on FRAMEDONE interrupt, and this work
either sent a BTA synchronously or looped until TE_SIZE was zero, to
wait for the end of the transfer.
This patch changes a BTA to be sent asynchronously from FRAMEDONE
interrupt, and when a BTA interrupt is received, the transfer is
finished. This way we do the whole process asynchronously, and also
inside interrupt context.
This will give us much better latency to handle the end of the frame
than with the previous work based solution.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Fri, 30 Apr 2010 08:24:33 +0000 (11:24 +0300)]
OMAP: DSS2: DSI: change dsi_vc_config_l4/vp()
Change dsi_vc_config_l4/vp() to loop for the VC_BUSY flag to change, and
return an error if it fails.
Busy looping is bad, but there's no interrupt that can be used for all the
cases where VC can be busy. So the caller should first try to make sure
that the VC is not busy, if possible, and then call dsi_vc_config_l4/vp().
Most notable case when the caller cannot be sure if the VC is busy is
after frame has been sent. Usually DSI buffers have been emptied until we
need to reconfig the VC, but in some rare cases the VC can still be busy,
and this patch will handle that case.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Ville Syrjälä [Thu, 22 Apr 2010 20:50:09 +0000 (22:50 +0200)]
OMAP: DSS2: DSI: Disable interface when disabling the display
Once the DSI PLL is separated from the DSI display a full DSI reset will
not be performed every time the display is enabled so the interface and
VCs must be disabled when disabling the display. If the VCs are not
disabled some register accesses will abort.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Wed, 9 Jun 2010 12:28:12 +0000 (15:28 +0300)]
OMAP: DSS2: DSI: Wait for DSI PLL clocks to be active before selecting them
The TRM tells us to wait for the DSI PLL derived clocks to become
active before selecting them for use. I didn't actually have any issues
which this would fix but according to the TRM it seems to be the right
thing to do.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Jani Nikula [Thu, 22 Apr 2010 12:27:36 +0000 (15:27 +0300)]
OMAP: DSS2: Add Nokia DSI command mode panel configuration struct
Introduce a configuration struct for platform/board specific information
of Nokia DSI command mode panels, to be used in addition to struct
omap_dss_device (passed via the 'void *data' member).
Signed-off-by: Jani Nikula <ext-jani.1.nikula@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Thu, 15 Apr 2010 13:41:07 +0000 (16:41 +0300)]
OMAP: DSS2: Taal: add locks to taal_bl_update_status
taal_bl_update_status was missing locks to protect taal_data. This caused
a kernel crash randomly, as the code attempted to set the brightness while
the OMAP's DSI block was actually disabled.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Mon, 12 Apr 2010 07:40:12 +0000 (10:40 +0300)]
OMAP: DSS2: DSI: change DSI timeout functions
Using nanoseconds as arguments to functions that set the DSI timeouts was
perhaps not so good idea. The timeouts are based on different DSI clocks,
so the possible range for the timeouts vary greatly depending on the
clocks. Also, the multipliers used with the timeouts cause big gaps in the
timeout range, meaning that the nanosecond based functions could cause the
timeout to be quite far from the intended value.
This patch changes the functions to take the plain tick values with the
multiplier enable/disable bits, and sets the TA/LP_RX/HS_TX timeouts to
maximum. While the timeouts could be much lower, the fact is that when
TA/LP_RX/HS_TX timeout happens, we are in an error situation and not in a
hurry anyway.
STOP_STATE_COUNTER is a different matter, but it is only used at
initialization time, and won't normally affect the performance.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Mon, 12 Apr 2010 06:57:19 +0000 (09:57 +0300)]
OMAP: DSS2: DSI: use a private workqueue
Using the shared workqueue led to to a deadlock in the case where the
display was unblanked via keyboard.
What happens is something like this:
- User presses a key
context 1:
- drivers/char/keyboard.c calls schedule_console_callback()
- fb_unblank takes the console semaphore
- dsi bus lock is taken, and frame transfer is started (dsi bus lock is
left on)
- Unblank code tries to set the panel backlight, which tries to take dsi
bus lock, but is blocked while the frame transfer is going on
context 2, shared workqueue, console_callback in drivers/char/vt.c:
- Tries to take console semaphore
- Blocks, as console semaphore is being held by context 1
- No other shared workqueue work can be run
context 3, HW irq, caused by FRAMEDONE interrupt:
- Interrupt handler schedules framedone-work in shared workqueue
- Framedone-work is never ran, as the shared workqueue is blocked. This
means that the unblank thread stays blocked, which means that context 2
stays blocked.
While I think the real problem is in keyboard/virtual terminal code, using
a private workqueue in the DSI driver is perhaps safer and more robust
than using the shared one. The DSI works should not be delayed more than a
millisecond or so, and even if the private workqueue gives us no hard
promise of doing so, it's still safer bet than the shared workqueue.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Ville Syrjälä [Wed, 17 Mar 2010 19:42:06 +0000 (21:42 +0200)]
OMAP: DSS2: OMAPFB: Make lockdep happy
When more than one memory region needs to be lockd at the same time use
the memory region id to fix the order in which the locks are taken. Also
one needs to use the _nested() versions of the locking primitives. The
memory region id can serve as the lock class there as well.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Ville Syrjälä [Wed, 17 Mar 2010 18:43:23 +0000 (20:43 +0200)]
OMAP: DSS2: OMAPFB: Add locking for memory regions
Add locking to the memory regions to make sure the memory region size
won't be changed while some other piece of code is performing some
checks or setup based on that information.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Ville Syrjälä [Wed, 17 Mar 2010 18:36:51 +0000 (20:36 +0200)]
OMAP: DSS2: OMAPFB: Add support for switching memory regions
Separate the memory region from the framebuffer device a little bit.
It's now possible to select the memory region used by the framebuffer
device using the new mem_idx parameter of omapfb_plane_info. If the
mem_idx is specified it will be interpreted as an index into the
memory regions array, if it's not specified the framebuffer's index is
used instead. So by default each framebuffer keeps using it's own
memory region which preserves backwards compatibility.
This allows cloning the same memory region to several overlays and yet
each overlay can be controlled independently since they can be
associated with separate framebuffer devices.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Ville Syrjälä [Wed, 17 Mar 2010 18:05:38 +0000 (20:05 +0200)]
OMAP: DSS2: OMAPFB: Check var even if there isn't memory
If video memory hasn't been allocate have check_fb_var() still check
most of the settings, just skip the ones involving the size of the
memory region. Also skip the memory address calculations in
omapfb_setup_overlay() if there's no memory.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Wed, 9 Jun 2010 12:09:30 +0000 (15:09 +0300)]
OMAP: DSS2: Fix update area calculations with multiple scaled overlays
When there are multiple scaled overlays simply checking whether the update
area intersects any of them in order is not enough. If eg. VID1 starts out
completely outside the update area but VID2 causes the update area to
increase in such a way that VID1 now falls partially within the increased
update area VID1 should be rechecked and the update area possibly
increased even further to fully encompass VID1. So simply keep looping
over the overlays until such time that none of the overlays caused the
update area to change.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
When DSS transitions from off mode to on VENC may generate a spurious
SYNC_LOST_DIGIT error. Just ack it when restoring the context. Also
restore IRQENABLE last to avoid triggering interrupts before the
context is fully restored.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Ville Syrjälä [Tue, 2 Mar 2010 19:30:52 +0000 (21:30 +0200)]
OMAP: DSS2: Check if display supports update mode changes
Check whether the display actually has the set_update_mode() function
before calling it. Only the sysfs codepath was broken, the omapfb ioctl
had the necessary protection.
Signed-off-by: Ville Syrjälä <ville.syrjala@nokia.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
Tomi Valkeinen [Fri, 12 Mar 2010 13:11:19 +0000 (15:11 +0200)]
OMAP: DSS2: DSI: Increase HS TX timeout
For some reason screen updates of certain odd widths seem to be triggering
HS TX timeouts on OMAP 3430, even if bigger updates do not. The reason for
this is unknown, but increasing the timeout removes the problem with no
(so far) noticeable problems. I haven't seen this problem on OMAP 3630.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@nokia.com>
* master.kernel.org:/home/rmk/linux-2.6-arm:
cyber2000fb: fix console in truecolor modes
cyber2000fb: fix machine hang on module load
SA1111: Eliminate use after free
ARM: Fix Versatile/Realview/VExpress MMC card detection sense
ARM: 6279/1: highmem: fix SMP preemption bug in kmap_high_l1_vipt
ARM: Add barriers to io{read,write}{8,16,32} accessors as well
ARM: 6273/1: Add barriers to the I/O accessors if ARM_DMA_MEM_BUFFERABLE
ARM: 6272/1: Convert L2x0 to use the IO relaxed operations
ARM: 6271/1: Introduce *_relaxed() I/O accessors
ARM: 6275/1: ux500: don't use writeb() in uncompress.h
ARM: 6270/1: clean files in arch/arm/boot/compressed/
ARM: Fix csum_partial_copy_from_user()
Merge branch 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6
* 'bugfixes' of git://git.linux-nfs.org/projects/trondmy/nfs-2.6:
NFS: Ensure that writepage respects the nonblock flag
NFS: kswapd must not block in nfs_release_page
nfs: include space for the NUL in root path
Debian's ia64 autobuilders have been seeing kernel freeze or reboot
when running the gdb testsuite (Debian bug 588574): dannf bisected to
2.6.32 62eede62dafb4a6633eae7ffbeb34c60dba5e7b1 "mm: ZERO_PAGE without
PTE_SPECIAL"; and reproduced it with gdb's gcore on a simple target.
I'd missed updating the gate_vma handling in __get_user_pages(): that
happens to use vm_normal_page() (nowadays failing on the zero page),
yet reported success even when it failed to get a page - boom when
access_process_vm() tried to copy that to its intermediate buffer.
Fix this, resisting cleanups: in particular, leave it for now reporting
success when not asked to get any pages - very probably safe to change,
but let's not risk it without testing exposure.
Why did ia64 crash with 16kB pages, but succeed with 64kB pages?
Because setup_gate() pads each 64kB of its gate area with zero pages.
Reported-by: Andreas Barth <aba@not.so.argh.org> Bisected-by: dann frazier <dannf@debian.org> Signed-off-by: Hugh Dickins <hughd@google.com> Tested-by: dann frazier <dannf@dannf.org> Cc: stable@kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Howells [Fri, 30 Jul 2010 14:25:19 +0000 (15:25 +0100)]
CIFS: Remove __exit mark from cifs_exit_dns_resolver()
Remove the __exit mark from cifs_exit_dns_resolver() as it's called by the
module init routine in case of error, and so may have been discarded during
linkage.
Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Jeff Layton <jlayton@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ondrej Zary [Thu, 29 Jul 2010 20:40:54 +0000 (22:40 +0200)]
cyber2000fb: fix console in truecolor modes
Return value was not set to 0 in setcolreg() with truecolor modes. This causes
fb_set_cmap() to abort after first color, resulting in blank palette - and
blank console in 24bpp and 32bpp modes.
Signed-off-by: Ondrej Zary <linux@rainbow-software.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Ondrej Zary [Thu, 29 Jul 2010 20:32:20 +0000 (22:32 +0200)]
cyber2000fb: fix machine hang on module load
I was testing two CyberPro 2000 based PCI cards on x86 and the machine always
hanged completely when the cyber2000fb module was loaded. It seems that the
card hangs when some registers are accessed too quickly after writing RAMDAC
control register. With this patch, both card work.
Add delay after RAMDAC control register write to prevent hangs on module load.
Signed-off-by: Ondrej Zary <linux@rainbow-software.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Julia Lawall [Fri, 30 Jul 2010 15:17:28 +0000 (17:17 +0200)]
SA1111: Eliminate use after free
__sa1111_remove always frees its argument, so the subsequent reference to
sachip->saved_state represents a use after free. __sa1111_remove does not
appear to use the saved_state field, so the patch simply frees it first.
A simplified version of the semantic patch that finds this problem is as
follows: (http://coccinelle.lip6.fr/)
// <smpl>
@@
expression E,E2;
@@
__sa1111_remove(E)
...
(
E = E2
|
* E
)
// </smpl>
Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Russell King [Thu, 29 Jul 2010 14:58:59 +0000 (15:58 +0100)]
ARM: Fix Versatile/Realview/VExpress MMC card detection sense
The MMC card detection sense has become really confused with negations
at various levels, leading to some platforms not detecting inserted
cards. Fix this by converting everything to positive logic throughout,
thereby getting rid of these negations.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Gary King [Thu, 29 Jul 2010 16:37:20 +0000 (17:37 +0100)]
ARM: 6279/1: highmem: fix SMP preemption bug in kmap_high_l1_vipt
smp_processor_id() must not be called from a preemptible context (this
is checked by CONFIG_DEBUG_PREEMPT). kmap_high_l1_vipt() was doing so.
This lead to a problem where the wrong per_cpu kmap_high_l1_vipt_depth
could be incremented, causing a BUG_ON(*depth <= 0); in
kunmap_high_l1_vipt().
The solution is to move the call to smp_processor_id() after the call
to preempt_disable().
Originally by: Andrew Howe <ahowe@nvidia.com>
Signed-off-by: Gary King <gking@nvidia.com> Acked-by: Nicolas Pitre <nico.as.pitre@linaro.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
See https://bugzilla.kernel.org/show_bug.cgi?id=16056
If other processes are blocked waiting for kswapd to free up some memory so
that they can make progress, then we cannot allow kswapd to block on those
processes.
Dan Carpenter [Tue, 13 Jul 2010 11:34:59 +0000 (13:34 +0200)]
nfs: include space for the NUL in root path
In root_nfs_name() it does the following:
if (strlen(buf) + strlen(cp) > NFS_MAXPATHLEN) {
printk(KERN_ERR "Root-NFS: Pathname for remote directory too long.\n");
return -1;
}
sprintf(nfs_export_path, buf, cp);
In the original code if (strlen(buf) + strlen(cp) == NFS_MAXPATHLEN)
then the sprintf() would lead to an overflow. Generally the rest of the
code assumes that the path can have NFS_MAXPATHLEN (1024) characters and
a NUL terminator so the fix is to add space to the nfs_export_path[]
buffer.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Merge branch 'fix/hda' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6
* 'fix/hda' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
ALSA: hda - Add a PC-beep workaround for ASUS P5-V
ALSA: hda - Assume PC-beep as default for Realtek
ALSA: hda - Don't register beep input device when no beep is available
ALSA: hda - Fix pin-detection of Nvidia HDMI
David Howells [Thu, 29 Jul 2010 11:45:55 +0000 (12:45 +0100)]
CRED: Fix __task_cred()'s lockdep check and banner comment
Fix __task_cred()'s lockdep check by removing the following validation
condition:
lockdep_tasklist_lock_is_held()
as commit_creds() does not take the tasklist_lock, and nor do most of the
functions that call it, so this check is pointless and it can prevent
detection of the RCU lock not being held if the tasklist_lock is held.
Instead, add the following validation condition:
task->exit_state >= 0
to permit the access if the target task is dead and therefore unable to change
its own credentials.
Fix __task_cred()'s comment to:
(1) discard the bit that says that the caller must prevent the target task
from being deleted. That shouldn't need saying.
(2) Add a comment indicating the result of __task_cred() should not be passed
directly to get_cred(), but rather than get_task_cred() should be used
instead.
Also put a note into the documentation to enforce this point there too.
Signed-off-by: David Howells <dhowells@redhat.com> Acked-by: Jiri Olsa <jolsa@redhat.com> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David Howells [Thu, 29 Jul 2010 11:45:49 +0000 (12:45 +0100)]
CRED: Fix get_task_cred() and task_state() to not resurrect dead credentials
It's possible for get_task_cred() as it currently stands to 'corrupt' a set of
credentials by incrementing their usage count after their replacement by the
task being accessed.
What happens is that get_task_cred() can race with commit_creds():
However, since a tasks credentials are generally not changed very often, we can
reasonably make use of a loop involving reading the creds pointer and using
atomic_inc_not_zero() to attempt to increment it if it hasn't already hit zero.
If successful, we can safely return the credentials in the knowledge that, even
if the task we're accessing has released them, they haven't gone to the RCU
cleanup code.
We then change task_state() in procfs to use get_task_cred() rather than
calling get_cred() on the result of __task_cred(), as that suffers from the
same problem.
Without this change, a BUG_ON in __put_cred() or in put_cred_rcu() can be
tripped when it is noticed that the usage count is not zero as it ought to be,
for example:
ALSA: hda - Add a PC-beep workaround for ASUS P5-V
ASUS P5-V provides a SSID that unexpectedly matches with the value
compilant with Realtek's specification. Thus the driver interprets
it badly, resulting in non-working PC beep.
This patch adds a white-list for such a case; a white-list of known
devices with working PC beep.
Russell King [Thu, 29 Jul 2010 10:38:05 +0000 (11:38 +0100)]
ARM: Add barriers to io{read,write}{8,16,32} accessors as well
The ioread/iowrite accessors also need barriers as they're used in
place of readl/writel et.al. in portable drivers. Create __iormb()
and __iowmb() which are conditionally defined to be barriers dependent
on ARM_DMA_MEM_BUFFERABLE, and always use these macros in the accessors.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
ARM: 6273/1: Add barriers to the I/O accessors if ARM_DMA_MEM_BUFFERABLE
When the coherent DMA buffers are mapped as Normal Non-cacheable
(ARM_DMA_MEM_BUFFERABLE enabled), buffer accesses are no longer ordered
with Device memory accesses causing failures in device drivers that do
not use the mandatory memory barriers before starting a DMA transfer.
LKML discussions led to the conclusion that such barriers have to be
added to the I/O accessors:
This patch introduces a wmb() barrier to the write*() I/O accessors to
handle the situations where Normal Non-cacheable writes are still in the
processor (or L2 cache controller) write buffer before a DMA transfer
command is issued. For the read*() accessors, a rmb() is introduced
after the I/O to avoid speculative loads where the driver polls for a
DMA transfer ready bit.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
ARM: 6272/1: Convert L2x0 to use the IO relaxed operations
This patch is in preparation for a subsequent patch which adds barriers
to the I/O accessors. Since the mandatory barriers may do an L2 cache
sync, this patch avoids a recursive call into l2x0_cache_sync() via the
write*() accessors and wmb() and a call into l2x0_cache_sync() with the
l2x0_lock held.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
This patch introduces readl*_relaxed()/write*_relaxed() as the main I/O
accessors (when __mem_pci is defined). The standard read*()/write*()
macros are now based on the relaxed accessors.
This patch is in preparation for a subsequent patch which adds barriers
to the I/O accessors.
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Rabin Vincent [Thu, 29 Jul 2010 11:13:18 +0000 (12:13 +0100)]
ARM: 6275/1: ux500: don't use writeb() in uncompress.h
Don't use writeb() in uncompress.h, to avoid the following build errors
when the "Add barriers to the I/O accessors" series is applied. Use
__raw_writeb() instead.
arch/arm/boot/compressed/misc.o: In function `putc':
arch/arm/mach-ux500/include/mach/uncompress.h:41:
undefined reference to `outer_cache'
Acked-by: Linus Walleij <linus.walleij@stericsson.com> Signed-off-by: Rabin Vincent <rabin.vincent@stericsson.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
drm/radeon/kms: fix radeon mid power profile reporting
Fix incorrectly reporting 'default' power profile, when it is set to 'mid'.
Signed-off-by: Daniel J Blueman <daniel.blueman@gmail.com> Reviewed-by: Alex Deucher <alexdeucher@gmail.com> Signed-off-by: Dave Airlie <airlied@redhat.com>
* git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi-rc-fixes-2.6:
[SCSI] ibmvscsi: Fix oops when an interrupt is pending during probe
[SCSI] zfcp: Update status read mempool
[SCSI] zfcp: Do not wait for SBALs on stopped queue
[SCSI] zfcp: Fix check whether unchained ct_els is possible
[SCSI] ipr: fix resource path display and formatting
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lrg/voltage-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/lrg/voltage-2.6:
davinci: da850/omap-l138 evm: account for DEFDCDC{2,3} being tied high
regulator: tps6507x: allow driver to use DEFDCDC{2,3}_HIGH register
wm8350-regulator: fix wm8350_register_regulator error handling
ab3100: fix off-by-one value range checking for voltage selector
Andre Osterhues [Tue, 13 Jul 2010 20:59:17 +0000 (15:59 -0500)]
ecryptfs: Bugfix for error related to ecryptfs_hash_buckets
The function ecryptfs_uid_hash wrongly assumes that the
second parameter to hash_long() is the number of hash
buckets instead of the number of hash bits.
This patch fixes that and renames the variable
ecryptfs_hash_buckets to ecryptfs_hash_bits to make it
clearer.
Jason Wessel [Thu, 29 Jul 2010 00:10:30 +0000 (19:10 -0500)]
x86,kgdb: Fix hw breakpoint regression
HW breakpoints events stopped working correctly with kgdb
as a result of commit: 018cbffe6819f6f8db20a0a3acd9bab9bfd667e4
(Merge commit 'v2.6.33' into perf/core).
The regression occurred because the behavior changed for setting
NOTIFY_STOP as the return value to the die notifier if the breakpoint
was known to the HW breakpoint API. Because kgdb is using the HW
breakpoint API to register HW breakpoints slots, it must also now
implement the overflow_handler call back else kgdb does not get to see
the events from the die notifier.
The kgdb_ll_trap function will be changed to be general purpose code
which can allow an easy way to implement the hw_breakpoint API
overflow call back.
Signed-off-by: Jason Wessel <jason.wessel@windriver.com> Acked-by: Dongdong Deng <dongdong.deng@windriver.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
If we don't need a huge amount of memory in ->readdir() then
we can use kmalloc rather than vmalloc to allocate it. This
should cut down on the greater overheads associated with
vmalloc for smaller directories.
We may be able to eliminate vmalloc entirely at some stage,
but this is easy to do right away.
Also using GFP_NOFS to avoid any issues wrt to deleting inodes
while under a glock, and suggestion from Linus to factor out
the alloc/dealloc.
I've given this a test with a variety of different sized
directories and it seems to work ok.
Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Nick Piggin <npiggin@suse.de> Cc: Prarit Bhargava <prarit@redhat.com> Signed-off-by: Steven Whitehouse <swhiteho@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Enable PC-beep as default for hardwares that aren't compliant with the
SSID value Realtek requires. In such a case, better to enable the beep
to avoid a regression.
ALSA: hda - Don't register beep input device when no beep is available
We check now the availability of PC beep and skip the build of beep
mixers, but the driver still registers the input device. This should
be checked as well.
Sekhar Nori [Mon, 12 Jul 2010 12:26:21 +0000 (17:56 +0530)]
davinci: da850/omap-l138 evm: account for DEFDCDC{2,3} being tied high
Per the da850/omap-l138 Beta EVM SOM schematic, the DEFDCDC2 and DEFDCDC3 lines are tied high. This leads to a 3.3V IO and 1.2V CVDD
voltage.
Pass the right platform data to the TPS6507x driver so it can operate
on the DEFDCDC{2,3}_HIGH register to read and change voltage levels.
Signed-off-by: Sekhar Nori <nsekhar@ti.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Liam Girdwood <lrg@slimlogic.co.uk>
projects / linux-beck.git / log