Stopping Dma after last req transfer.
add an API to return the completed transfer count of a pending, active
or finished DMA request
originally fixed by Gary King <gking@nvidia.com>
It is observed that the dma interrupt has the lower priority then
its client interupt priority. When client's isr calls dma get transfer,
the dma status has not been upated as dma isr have not been served yet.
So before reading the status, explicitly checking the interrupt status and
handling accordingly.
The another issue which is observed is that if dma has transferred the data
of amount = requested -4 and if it moves to invalid requestor before stopping
then status got reset and tarnsfered bytes becomes 0. This seems the apb dma
hw behavior. Following is the suggestion to overcome this issue:
- Disable global enable bit.
- Read status.
- Stop dma.
- Enable global status bit.
Added this workaround and it worked fine.
originally fixed by Laxman Dewangan <ldewangan@nvidia.com>
In continous mode, dma should stop after last transfer completed and
if there is no more req pending.
If there is pending req then it should check whether it has updated
in hw for next transfer or not and if it has not started then stop dma
and start new req immediatley.
originally fixed by Laxman Dewangan <ldewangan@nvidia.com>
Change-Id: I49c97c96eacdf4060de6b21cec0e71d940d33f00
There is no need to poweroff the phy and disable the clocks on shutdown.
This interferes with autopm that may try to disable the clocks after
shutdown.
Change-Id: I3aee19abe6dd11685b3be348e25fc3e195a2a416
Signed-off-by: Benoit Goby <benoit@android.com>
A repeated start should be used for all but the last msg in an xfer. The
NOSTART flag is for skipping the START frame (addr/rw)
Change-Id: If4bc88478521a8c4e1ec761bd42a84d3e398fa26
Signed-off-by: Erik Gilling <konkers@android.com>
There seems to be some interdependency between the two display controllers'
resets and HDMI. If dc0 is put into reset while HDMI is connected to dc1,
HDMI will blank.
Change-Id: I978814628609ebe3d78e7809fccef6a3d19c0f7f
Signed-off-by: Erik Gilling <konkers@android.com>
This makes the default case in fb_find_best_display ("last resort, use
the very first mode") actually use the first mode added.
Change-Id: Ia43ac4a4251522cd30c0da514da23e47b2906e2c
Signed-off-by: Erik Gilling <konkers@android.com>
Print an error message when a DMA channel cannot be allocated.
Change-Id: I93a96851ac12c5ea66b2fb053033aa4260c2178a
Signed-off-by: Mike Corrigan <michael.corrigan@motorola.com>
Handle the case where DMA channels cannot be allocated and PIO must be used.
Fixed a typo for forcing RX to use PIO.
Change-Id: I167184f9ce936d4a08dd9919ae4f8b0d9ad7e0c4
Signed-off-by: Mike Corrigan <michael.corrigan@motorola.com>
Tony Luck reports that the addition of the access_ok() check in commit
0eead9ab41 ("Don't dump task struct in a.out core-dumps") broke the
ia64 compile due to missing the necessary header file includes.
Rather than add yet another include (<asm/unistd.h>) to make everything
happy, just uninline the silly core dump helper functions and move the
bodies to fs/exec.c where they make a lot more sense.
dump_seek() in particular was too big to be an inline function anyway,
and none of them are in any way performance-critical. And we really
don't need to mess up our include file headers more than they already
are.
Reported-and-tested-by: Tony Luck <tony.luck@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
ehea: Fix a checksum issue on the receive path
net: allow FEC driver to use fixed PHY support
tg3: restore rx_dropped accounting
b44: fix carrier detection on bind
net: clear heap allocations for privileged ethtool actions
NET: wimax, fix use after free
ATM: iphase, remove sleep-inside-atomic
ATM: mpc, fix use after free
ATM: solos-pci, remove use after free
net/fec: carrier off initially to avoid root mount failure
r8169: use device model DMA API
r8169: allocate with GFP_KERNEL flag when able to sleep
akiphie points out that a.out core-dumps have that odd task struct
dumping that was never used and was never really a good idea (it goes
back into the mists of history, probably the original core-dumping
code). Just remove it.
Also do the access_ok() check on dump_write(). It probably doesn't
matter (since normal filesystems all seem to do it anyway), but he
points out that it's normally done by the VFS layer, so ...
[ I suspect that we should possibly do "vfs_write()" instead of
calling ->write directly. That also does the whole fsnotify and write
statistics thing, which may or may not be a good idea. ]
And just to be anal, do this all for the x86-64 32-bit a.out emulation
code too, even though it's not enabled (and won't currently even
compile)
Reported-by: akiphie <akiphie@lavabit.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-- Ignore kfifo thresholds on recording and playback and adjust the delays.
-- Take out the code from TEGRA_AUDIO_IN_STOP into a separate function
stop_recording_nosync()
-- Rename stop_recording() to wait_for_recording_to_stop().
-- add ioctl(TEGRA_AUDIO_OUT_FLUSH), which blocks the caller until the output
fifo is drained. While the caller is blocked, pending write() calls will
return immediately with whatever data they had managed to queue up.
-- removed ioctl(TEGRA_AUDIO_OUT_PRELOAD_FIFO)
-- since TEGRA_AUDIO_OUT_FLUSH and TEGRA_AUDIO_IN_STOP act similarly, moved
audio_driver_state::recording_cancelled to audio_stream::stop and changed
the code accordingly. Renamed functions wait_for_recording_to_stop() and
stop_recording_nosync() to wait_till_stopped() and request_stop_nosync()
since they handle both playback and recording.
-- print errors on close() if wakelocks are still held
-- Call request_stop_nosync() on close() of a recording file handle
-- Do not use struct audio_stream::active for playback streams. Instead,
where applicable, use kfifo_len(). As a consequence, playback kfifo
underruns are no longer reported. These were bogus anyway, as we really
need the DMA engine to tell us if there are underruns.
-- Because of above item, had to rework tx_fifo_atn_store(),
rx_fifo_atn_store(), and __attr_fifo_atn_write().
-- Set struct audio_stream::active for a recording stream to true when a
recording starts, and set it to false when recording get stopped. Do not
set/clear it within the body of read(), because just being within read()
does not mean that recording is in progress.
-- In tegra_audio_read(), check for stop == true before calling
start_recording_if_necessary(); this makes sure that if a user calls read()
after calling ioctl(TEGRA_AUDIO_IN_STOP), recording will not resume unless
ioctl(TEGRA_AUDIO_IN_START) gets called, or the file is closed and
re-opened.
-- Fixed TEGRA_AUDIO_IN_START
-- In PIO mode, enabled FIFOs before enabling interrupts as specified in the
TRM.
-- Added missing break in tegra_audio_ioctl().
-- Silenced some debug spew
Signed-off-by: Iliyan Malchev <malchev@google.com>
Adding support for USB-IF High Speed electrical test mode for device mode.
Support added for electrical test modes:
1. TEST_J
2. TEST_K
3. TEST_SE0_NAK
4. TEST_PACKET
5. TEST_FORCE_ENABLE
originally fixed by Venkat Moganty <vmoganty@nvidia.com>
Change-Id: If5a4dcf9eb81dc368f24c660460d35495b6a4253
Signed-off-by: Jay Cheng <jacheng@nvidia.com>
* 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
ring-buffer: Fix typo of time extends per page
perf, MIPS: Support cross compiling of tools/perf for MIPS
perf: Fix incorrect copy_from_user() usage
fsl_build_dtd may be called from atomic context. Don't hardcode the gfp
flag to GFP_KERNEL. Pass down to dma_pool_alloc the flag used by the
function driver to enqueue the request.
Change-Id: Iba1ccf73bbcd648b8c13228157431d2ce52979ef
Signed-off-by: Benoit Goby <benoit@android.com>
* master.kernel.org:/home/rmk/linux-2.6-arm:
ARM: relax ioremap prohibition (309caa9) for -final and -stable
ARM: 6440/1: ep93xx: DMA: fix channel_disable
cpuimx27: fix i2c bus selection
cpuimx27: fix compile when ULPI is selected
ARM: 6435/1: Fix HWCAP_TLS flag for ARM11MPCore/Cortex-A9
ARM: 6436/1: AT91: Fix power-saving in idle-mode on 926T processors
ARM: fix section mismatch warnings in Versatile Express
ARM: 6412/1: kprobes-decode: add support for MOVW instruction
ARM: 6419/1: mmu: Fix MT_MEMORY and MT_MEMORY_NONCACHED pte flags
ARM: 6416/1: errata: faulty hazard checking in the Store Buffer may lead to data corruption
* 'omap-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6:
omap: iommu-load cam register before flushing the entry
* 'drm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/airlied/drm-2.6:
drm/radeon/kms: Silent spurious error message
drm/radeon/kms: fix bad cast/shift in evergreen.c
drm/radeon/kms: make TV/DFP table info less verbose
drm/radeon/kms: leave certain CP int bits enabled
drm/radeon/kms: avoid corner case issue with unmappable vram V2
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86, numa: For each node, register the memory blocks actually used
x86, AMD, MCE thresholding: Fix the MCi_MISCj iteration order
x86, mce, therm_throt.c: Fix missing curly braces in error handling logic
Commit 0793448 "DMAENGINE: generic channel status v2" changed the interface for
how dma channel progress is retrieved. It inadvertently exported an internal
helper function ioat_tx_status() instead of ioat_dma_tx_status(). The latter
polls the hardware to get the latest completion state, while the helper just
evaluates the current state without touching hardware. The effect is that we
end up waiting for completion timeouts or descriptor allocation errors before
the completion state is updated.
iperf (before fix):
[SUM] 0.0-41.3 sec 364 MBytes 73.9 Mbits/sec
iperf (after fix):
[SUM] 0.0- 4.5 sec 499 MBytes 940 Mbits/sec
This is a regression starting with 2.6.35.
Cc: <stable@kernel.org>
Cc: Dave Jiang <dave.jiang@intel.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Linus Walleij <linus.walleij@stericsson.com>
Cc: Maciej Sosnowski <maciej.sosnowski@intel.com>
Reported-by: Richard Scobie <richard@sauce.co.nz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Currently we set all skbs with CHECKSUM_UNNECESSARY, even
those whose protocol we don't know. This patch just
add the CHECKSUM_COMPLETE tag for non TCP/UDP packets.
Reported-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Breno Leitao <leitao@linux.vnet.ibm.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As of commit 43a9aa64a2 "NFSD:
Fill in WCC data for REMOVE, RMDIR, MKNOD, and MKDIR", we sometimes call
fh_unlock on a filehandle that isn't fully initialized.
We should fix up the callers, but as a quick fix it is also sufficient
just to remove this assertion.
Reported-by: Marius Tolzmann <tolzmann@molgen.mpg.de>
Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
At least one board using the FEC driver does not have a conventional
PHY attached to it, it is directly connected to a somewhat simple
ethernet switch (the board is the SnapGear/LITE, and the attached
4-port ethernet switch is a RealTek RTL8305). This switch does not
present the usual register interface of a PHY, it presents nothing.
So a PHY scan will find nothing - it finds ID's of 0 for each PHY
on the attached MII bus.
After the FEC driver was changed to use phylib for supporting PHYs
it no longer works on this particular board/switch setup.
Add code support to use a fixed phy if no PHY is found on the MII bus.
This is based on the way the cpmac.c driver solved this same problem.
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
register the tegradc0, tegradc1, grhost, pwm-backlight and carveout
devices
Change-Id: Ia61c0632470e571cc57279dc3b197ccd1fca80f0
Signed-off-by: Gary King <gking@nvidia.com>
Low mem pages are allocated in larger super pages and their caching
attributes can't be controlled on a per page basis. This patch
forces nvmap to map out of highmem pages which are guaranteed to have
page mappings.
Change-Id: Id3921342ecceb0345d43365d4dd90b82ca8cfd11
Signed-off-by: Rebecca Schultz Zavin <rebecca@android.com>
when allocating uncached pages, the outer cache should be flushed;
the end address should be specified in bytes, not in pages.
Change-Id: I3fe036f4f7e10e009f96567e3afeeef6ea603240
Signed-off-by: Gary King <gking@nvidia.com>
... but produce a big warning about the problem as encouragement
for people to fix their drivers.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
