As suggested by Nick Piggin it seems we can drop the -ffunction-sections
compile flag, now that the kernel uses thin archives. Testing with 32-
and 64-bit kernel showed no difference in kernel size.
Suggested-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Helge Deller <deller@gmx.de>
This was introduced more than a decade ago when sg chaining was
added, but we never really caught anything with it. The scatterlist
entry size can be critical, since drivers allocate it, so remove
the magic member. Recently it's been triggering allocation stalls
and failures in NVMe.
Tested-by: Jordan Glover <Golden_Miller83@protonmail.ch>
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
- Fix the initialization time error handling in the recently added
Kryo cpufreq driver (Dan Carpenter).
- Fix up the recently added coverage of performance states in the
generic power domains (genpd) framework (Viresh Kumar).
- Add missing documentation of the new hwp_dynamic_boost sysfs knob
in the intel_pstate driver (Rafael Wysocki).
- Fix incorrect sysfs path in the intel_pstate driver documentation
(Rafael Wysocki).
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=842O
-----END PGP SIGNATURE-----
Merge tag 'pm-4.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael Wysocki:
"These fix up recently added features (the Kryo cpufreq driver and
performance states coverage in the generic power domains framework),
add missing documentation for a recently added sysfs knob in the
intel_pstate driver and fix an error in its documentation.
Specifics:
- Fix the initialization time error handling in the recently added
Kryo cpufreq driver (Dan Carpenter).
- Fix up the recently added coverage of performance states in the
generic power domains (genpd) framework (Viresh Kumar).
- Add missing documentation of the new hwp_dynamic_boost sysfs knob
in the intel_pstate driver (Rafael Wysocki).
- Fix incorrect sysfs path in the intel_pstate driver documentation
(Rafael Wysocki)"
* tag 'pm-4.18-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
Documentation: intel_pstate: Describe hwp_dynamic_boost sysfs knob
Documentation: admin-guide: intel_pstate: Fix sysfs path
PM / Domains: Rename opp_node to np
PM / Domains: Fix return value of of_genpd_opp_to_performance_state()
cpufreq: qcom-kryo: Fix error handling in probe()
-----BEGIN PGP SIGNATURE-----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=Em+E
-----END PGP SIGNATURE-----
Merge tag 'drm-fixes-2018-06-29' of git://anongit.freedesktop.org/drm/drm
Pull drm fixes from Dave Airlie:
"Nothing too major this round:
- small set of mali-dp fixes
- single meson fix
- a bunch of amdgpu fixes (one makes non-4k page sizes not be a bad
experience)"
* tag 'drm-fixes-2018-06-29' of git://anongit.freedesktop.org/drm/drm:
drm/amd/display: release spinlock before committing updates to stream
drm/amdgpu:Support new VCN FW version naming convention
drm/amdgpu: fix UBSAN: Undefined behaviour for amdgpu_fence.c
drm/meson: Fix an un-handled error path in 'meson_drv_bind_master()'
drm/amdgpu: GPU vs CPU page size fixes in amdgpu_vm_bo_split_mapping
drm/amdgpu: Count disabled CRTCs in commit tail earlier
drm/mali-dp: Rectify the width and height passed to rotmem_required()
drm/arm/malidp: Preserve LAYER_FORMAT contents when setting format
drm: mali-dp: Enable Global SE interrupts mask for DP500
drm/arm/malidp: Ensure that the crtcs are shutdown before removing any encoder/connector
bio_clone_bioset(). Also fixes splitting bio that has integrity
payload.
- Three fixes related to properly validating DAX capabilities of a
stacked DM device that will advertise DAX support.
- Update DM writecache target to use 2-factor allocator arguments. Kees
says this is the last related change for 4.18.
- Fix DM zoned target to use GFP_NOIO to avoid triggering reclaim during
IO submission (caught by lockdep).
- Fix DM thinp to gracefully recover from running out of data space
while a previous async discard completes (whereby freeing space).
- Fix DM thinp's metadata transaction commit to avoid needless work.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJbNURXAAoJEMUj8QotnQNaPVYIAISzFYsSLEEYiGF5CDn2A07/
vd4TBIk7sAVQ/zouNGSNUHepoPJZR8E75SLO4lqZ214fyeYl9/48Za3dlyXeHldR
ziCCYFyHKvKgfZpBpbO43HbZbWvcVplIAt+GhhXepHy9jJC7+XQzMOqGwcvK2pW4
qCOqLTLaTGdzx5UaoLCG2Yg3E1TrYH0kZei/WLRBGW12WAsCzqNqUsZ5TgeilxCt
r/5ajPsXkcNjV9BrUNwY43J+WX8eKIuFKRSbnVdjHBEoVCeCwnP1R8WJnMm5/BTh
KZpyYTlFsBZ+Gvzojt2XI80ttviXtSUoJJNZgze9JT2xR4jMfn4yFQr/odxJucE=
=ZGTy
-----END PGP SIGNATURE-----
Merge tag 'for-4.18/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper fixes from Mike Snitzer:
- Fix dm core to use more efficient bio_split() instead of
bio_clone_bioset(). Also fixes splitting bio that has integrity
payload.
- Three fixes related to properly validating DAX capabilities of a
stacked DM device that will advertise DAX support.
- Update DM writecache target to use 2-factor allocator arguments. Kees
says this is the last related change for 4.18.
- Fix DM zoned target to use GFP_NOIO to avoid triggering reclaim
during IO submission (caught by lockdep).
- Fix DM thinp to gracefully recover from running out of data space
while a previous async discard completes (whereby freeing space).
- Fix DM thinp's metadata transaction commit to avoid needless work.
* tag 'for-4.18/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm: prevent DAX mounts if not supported
dax: check for QUEUE_FLAG_DAX in bdev_dax_supported()
pmem: only set QUEUE_FLAG_DAX for fsdax mode
dm thin: handle running out of data space vs concurrent discard
dm raid: don't use 'const' in function return
dm zoned: avoid triggering reclaim from inside dmz_map()
dm writecache: use 2-factor allocator arguments
dm thin metadata: remove needless work from __commit_transaction
dm: use bio_split() when splitting out the already processed bio
Pull single NVMe fix from Christoph.
* 'nvme-4.18' of git://git.infradead.org/nvme:
nvme-rdma: fix possible double free of controller async event buffer
Fix the test that verifies whether bio_op(bio) represents a discard
or write zeroes operation. Compile-tested only.
Cc: Philipp Reisner <philipp.reisner@linbit.com>
Cc: Lars Ellenberg <lars.ellenberg@linbit.com>
Fixes: 7435e9018f ("drbd: zero-out partial unaligned discards on local backend")
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Some devices have different queue limits depending on the type of IO. A
classic case is SATA NCQ, where some commands can queue, but others
cannot. If we have NCQ commands inflight and encounter a non-queueable
command, the driver returns busy. Currently we attempt to dispatch more
from the scheduler, if we were able to queue some commands. But for the
case where we ended up stopping due to BUSY, we should not attempt to
retrieve more from the scheduler. If we do, we can get into a situation
where we attempt to queue a non-queueable command, get BUSY, then
successfully retrieve more commands from that scheduler and queue those.
This can repeat forever, starving the non-queuable command indefinitely.
Fix this by NOT attempting to pull more commands from the scheduler, if
we get a BUSY return. This should also be more optimal in terms of
letting requests stay in the scheduler for as long as possible, if we
get a BUSY due to the regular out-of-tags condition.
Reviewed-by: Omar Sandoval <osandov@fb.com>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
io_pgetevents() will not change the signal mask. Mark it const to make
it clear and to reduce the need for casts in user code.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Avi Kivity <avi@scylladb.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
[hch: reapply the patch that got incorrectly reverted]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(contains build fixup for the deferred console takeover feature)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=WVWE
-----END PGP SIGNATURE-----
Merge tag 'ib-fbdev-drm-v4.19-deferred-console-takeover-fixup' of https://github.com/bzolnier/linux into drm-misc-next
Immutable branch between fbdev and drm for the v4.19 merge window
(contains build fixup for the deferred console takeover feature)
Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.com>
# gpg: Signature made Fri 29 Jun 2018 06:47:23 AM -03
# gpg: using RSA key 7E33B63FA047C20B
# gpg: Can't check signature: public key not found
Link: https://patchwork.freedesktop.org/patch/msgid/3340294.YySDL1Tsl7@amdc3058
Export dummycon_[un]register_output_notifier, the fbcon code needs this
and may be build as a module.
Fixes: 83d83bebf4 ("console/fbcon: Add support for deferred console takeover")
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
The limit_id_fallback array uses enum drm_ipp_size_id to index its
content. The content itself is of type enum drm_exynos_ipp_limit_type.
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Stefan Agner <stefan@agner.ch>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
The only bits that should be preserved in decon_win_set_fmt() is
WINCONx_ENWIN_F. All other bits depends on the selected pixel formats and
are set by the mentioned function.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Set per-plane global alpha to maximum value to get proper blending of
XRGB and ARGB planes. This fixes the strange order of overlapping planes.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
DMA hardware should respect buffer pitch, so use the width calculated from
the buffer pitch instead of the virtual one.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Fix following issues related to planar YUV pixel format configuration:
- NV16/61 modes were incorrectly programmed as NV12/21,
- YVU420 was programmed as YUV420 on source,
- YVU420 and YUV422 were programmed as YUV420 on output.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Horizontal (DRM_MODE_REFLECT_Y) and vertical (DMR_MODE_REFLECT_Y) flip
were swapped in GScaler driver. Fix this by swapping code for interpreting
them.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Investigation revealed that GScaler hardware requires the real buffer width
(pitch) to be aligned to 16 pixels.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
DMA hardware should respect buffer pitch, so use the width calculated from
the buffer pitch instead of the virtual one.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Fix Cb/CR components order in two-planar YUV420, YUV422 and YUV444 modes.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Ensure that Scaler hardware is properly reset and interrupts are cleared
before processing next image.
Signed-off-by: Andrzej Pietrasiewicz <andrzej.p@samsung.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Horizontal (DRM_MODE_REFLECT_Y) and vertical (DMR_MODE_REFLECT_Y) flip
were swapped in Rotator driver. Fix this by swapping code for interpreting
them.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
Prepare a common function for size and scale checks and call it for
source and destination buffers. Then also move there the state-less checks
from exynos_drm_ipp_task_setup_buffer, so the format information is already
available in limits processing. Finally perform the IPP_LIMIT_BUFFER check
on the real width of the buffer (the width calculated from the provided
buffer pitch).
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
The only time we should start FBC is when we have waited a vblank
after the atomic update. We've already forced a vblank wait by doing
wait_for_flip_done before intel_post_plane_update(), so we don't need
to wait a second time before enabling.
Removing the worker simplifies the code and removes possible race
conditions, like happening in 103167.
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103167
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180625163758.10871-2-maarten.lankhorst@linux.intel.com
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Merge fixups for the recent extenstion of the generic power domains
(genpd) framework covering performance states.
* pm-domains:
PM / Domains: Rename opp_node to np
PM / Domains: Fix return value of of_genpd_opp_to_performance_state()
It seems that during the conversion from gpio* to gpiod*, the initial
state of SCL was wrongly switched to LOW. Fix it to be HIGH again.
Fixes: 7bb75029ef ("i2c: gpio: Enforce open drain through gpiolib")
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Cc: stable@kernel.org
If DMA safe memory was allocated, but the subsequent I2C transfer
fails the memory is leaked. Plug this leak.
Fixes: 8a77821e74 ("i2c: smbus: use DMA safe buffers for emulated SMBus transactions")
Signed-off-by: Peter Rosin <peda@axentia.se>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Cc: stable@kernel.org
So, if somebody wants to re-implement this in the future, we pinpoint to
a problem case.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
This reverts commit 3e5f06bed7. As per
bugzilla #200045, this caused a regression. I don't really see a way to
fix it without having the hardware. So, revert the patch and I will fix
the issue I was seeing originally in the i2c-gpio driver itself. I
couldn't find new users of this algorithm since, so there should be no
one depending on the new behaviour.
Reported-by: Sergey Larin <cerg2010cerg2010@mail.ru>
Fixes: 3e5f06bed7 ("i2c: algo-bit: init the bus to a known state")
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Tested-by: Sergey Larin <cerg2010cerg2010@mail.ru>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Cc: stable@kernel.org
If a user is accessing a file in selinuxfs with a pointer to a userspace
buffer that is backed by e.g. a userfaultfd, the userspace access can
stall indefinitely, which can block fsi->mutex if it is held.
For sel_read_policy(), remove the locking, since this method doesn't seem
to access anything that requires locking.
For sel_read_bool(), move the user access below the locked region.
For sel_write_bool() and sel_commit_bools_write(), move the user access
up above the locked region.
Cc: stable@vger.kernel.org
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Signed-off-by: Jann Horn <jannh@google.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
[PM: removed an unused variable in sel_read_policy()]
Signed-off-by: Paul Moore <paul@paul-moore.com>
Back in commit 27af5eea54 ("drm/i915: Move execlists irq handler to a
bottom half"), we came to the conclusion that running our CSB processing
and ELSP submission from inside the irq handler was a bad idea. A really
bad idea as we could impose nearly 1s latency on other users of the
system, on average! Deferring our work to a tasklet allowed us to do the
processing with irqs enabled, reducing the impact to an average of about
50us.
We have since eradicated the use of forcewaked mmio from inside the CSB
processing and ELSP submission, bringing the impact down to around 5us
(on Kabylake); an order of magnitude better than our measurements 2
years ago on Broadwell and only about 2x worse on average than the
gem_syslatency on an unladen system.
In this iteration of the tasklet-vs-direct submission debate, we seek a
compromise where by we submit new requests immediately to the HW but
defer processing the CS interrupt onto a tasklet. We gain the advantage
of low-latency and ksoftirqd avoidance when waking up the HW, while
avoiding the system-wide starvation of our CS irq-storms.
Comparing the impact on the maximum latency observed (that is the time
stolen from an RT process) over a 120s interval, repeated several times
(using gem_syslatency, similar to RT's cyclictest) while the system is
fully laden with i915 nops, we see that direct submission an actually
improve the worse case.
Maximum latency in microseconds of a third party RT thread
(gem_syslatency -t 120 -f 2)
x Always using tasklets (a couple of >1000us outliers removed)
+ Only using tasklets from CS irq, direct submission of requests
+------------------------------------------------------------------------+
| + |
| + |
| + |
| + + |
| + + + |
| + + + + x x x |
| +++ + + + x x x x x x |
| +++ + ++ + + *x x x x x x |
| +++ + ++ + * *x x * x x x |
| + +++ + ++ * * +*xxx * x x xx |
| * +++ + ++++* *x+**xx+ * x x xxxx x |
| **x++++*++**+*x*x****x+ * +x xx xxxx x x |
|x* ******+***************++*+***xxxxxx* xx*x xxx + x+|
| |__________MA___________| |
| |______M__A________| |
+------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 118 91 186 124 125.28814 16.279137
+ 120 92 187 109 112.00833 13.458617
Difference at 95.0% confidence
-13.2798 +/- 3.79219
-10.5994% +/- 3.02677%
(Student's t, pooled s = 14.9237)
However the mean latency is adversely affected:
Mean latency in microseconds of a third party RT thread
(gem_syslatency -t 120 -f 1)
x Always using tasklets
+ Only using tasklets from CS irq, direct submission of requests
+------------------------------------------------------------------------+
| xxxxxx + ++ |
| xxxxxx + ++ |
| xxxxxx + +++ ++ |
| xxxxxxx +++++ ++ |
| xxxxxxx +++++ ++ |
| xxxxxxx +++++ +++ |
| xxxxxxx + ++++++++++ |
| xxxxxxxx ++ ++++++++++ |
| xxxxxxxx ++ ++++++++++ |
| xxxxxxxxxx +++++++++++++++ |
| xxxxxxxxxxx x +++++++++++++++ |
|x xxxxxxxxxxxxx x + + ++++++++++++++++++ +|
| |__A__| |
| |____A___| |
+------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 120 3.506 3.727 3.631 3.6321417 0.02773109
+ 120 3.834 4.149 4.039 4.0375167 0.041221676
Difference at 95.0% confidence
0.405375 +/- 0.00888913
11.1608% +/- 0.244735%
(Student's t, pooled s = 0.03513)
However, since the mean latency corresponds to the amount of irqsoff
processing we have to do for a CS interrupt, we only need to speed that
up to benefit not just system latency but our own throughput.
v2: Remember to defer submissions when under reset.
v4: Only use direct submission for new requests
v5: Be aware that with mixing direct tasklet evaluation and deferred
tasklets, we may end up idling before running the deferred tasklet.
v6: Remove the redudant likely() from tasklet_is_enabled(), restrict the
annotation to reset_in_progress().
v7: Take the full timeline.lock when enabling perf_pmu stats as the
tasklet is no longer a valid guard. A consequence is that the stats are
now only valid for engines also using the timeline.lock to process
state.
Testcase: igt/gem_exec_latency/*rthog*
References: 27af5eea54 ("drm/i915: Move execlists irq handler to a bottom half")
Suggested-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-9-chris@chris-wilson.co.uk
Now that we use the CSB stored in the CPU friendly HWSP, we do not need
to track interrupts for when the mmio CSB registers are valid and can
just check where we read up to last from the cached HWSP. This means we
can forgo the atomic bit tracking from interrupt, and in the next patch
it means we can check the CSB at any time.
v2: Change the splitting inside reset_prepare, we only want to lose
testing the interrupt in this patch, the next patch requires the change
in locking
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-8-chris@chris-wilson.co.uk
As we now never read back our current head position from the CSB
pointers register, and the HW itself doesn't use it to prevent
overwriting unread CSB entries, we do not need to keep updating the
register. As it turns out this register is not listed as being shadowed,
and so requires forcewake -- but we haven't been taking forcewake around
it so the writes has probably been regularly dropped. Fortuitously, we
only read the value after a reset where it did not matter, and zero was
the right answer (well, close enough).
Mika pointed out that this was how we used to do it (accidentally!)
before he fixed it in commit cc53699b25 ("drm/i915: Use masked write
for Context Status Buffer Pointer").
References: cc53699b25 ("drm/i915: Use masked write for Context Status Buffer Pointer")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-7-chris@chris-wilson.co.uk
On HW reset, the HW clears the write pointer (to 0). But since it also
writes its first CSB entry to slot 0, we need to reset the write pointer
back to the element before (so the first entry we read is 0).
This is required for the next patch, where we trust the CSB completely!
v2: Use _MASKED_FIELD
v3: Store the reset value, so that we differentiate between mmio/hwsp
transparently and without pretense.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-6-chris@chris-wilson.co.uk
Following the removal of the last workarounds, the only CSB mmio access
is for the old vGPU interface. The mmio registers presented by vGPU do
not require forcewake and can be treated as ordinary volatile memory,
i.e. they behave just like the HWSP access just at a different location.
We can reduce the CSB access to a set of read/write/buffer pointers and
treat the various paths identically and not worry about forcewake.
(Forcewake is nightmare for worstcase latency, and we want to process
this all with irqsoff -- no latency allowed!)
v2: Comments, comments, comments. Well, 2 bonus comments.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-5-chris@chris-wilson.co.uk
In the next patch, we will process the CSB events directly from the
submission path, rather than only after a CS interrupt. Hence, we will
no longer have the need for a loop until the has-interrupt bit is clear,
and in the meantime can remove that small optimisation.
v2: Tvrtko pointed out it was safer to unconditionally kick the tasklet
after each irq, when assuming that the tasklet is called for each irq.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-4-chris@chris-wilson.co.uk
In the following patch, we will process the CSB events under the
timeline.lock and not serialised by the tasklet. This also means that we
will need to protect access to common variables such as
execlists->csb_head with the timeline.lock during reset.
v2: Move sync_irq to avoid deadlocks between taking timeline.lock from
our interrupt handler.
v3: Kill off the synchronize_hardirq as it raises more questions than
answered; now we use the timeline.lock entirely for CSB serialisation
between the irq and elsewhere, we don't need to be so heavy handed with
flushing
v4: Treat request cancellation (wedging after failed reset) similarly
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-3-chris@chris-wilson.co.uk
In the next patch, we will begin processing the CSB from inside the
submission path (underneath an irqsoff section, and even from inside
interrupt handlers). This means that updating the execlists->port[] will
no longer be serialised by the tasklet but needs to be locked by the
engine->timeline.lock instead. Pull dequeue and submit under the same
lock for protection. (An alternate future plan is to keep the in/out
arrays separate for concurrent processing and reduced lock coverage.)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-2-chris@chris-wilson.co.uk
We do not need to do a posting read of our uncached mmio write to
re-enable the master interrupt lines after handling an interrupt, so
don't. This saves us a slow UC read before we can process the interrupt,
most noticeable in execlists where any stalls imposes extra latency on
GPU command execution.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ville Syrjala <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628201211.13837-1-chris@chris-wilson.co.uk
We're fetching GuC/HuC firmwares directly from uc level during
init_early stage but this breaks guc/huc struct isolation and
also strict SW-only initialization rule for init_early. Move fw
fetching to init phase and do it separately per guc/huc struct.
v2: don't forget to move wopcm_init - Michele
v3: fetch in init_misc phase - Michal
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com> #2
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628141522.62788-2-michal.wajdeczko@intel.com
We will add more init steps to misc phase and there is no need
to expose them separately for use in uc_init_misc function.
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Michel Thierry <michel.thierry@intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628141522.62788-1-michal.wajdeczko@intel.com
- A single fix in meson for an unhandled error path in meson_drv_bind_master().
-----BEGIN PGP SIGNATURE-----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=zPj3
-----END PGP SIGNATURE-----
Merge tag 'drm-misc-fixes-2018-06-28' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes
drm-misc-fixes for v4.18-rc3:
- A single fix in meson for an unhandled error path in meson_drv_bind_master().
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/fa740f31-5a8d-ed45-5e8a-aecd3f6f11b7@linux.intel.com
A few fixes for 4.18:
- fix a read past the end of an array due to vega20 changes
- fix driver on systems with non-4K pages
- fix locking with pageflipping in DC that could lead to a sleep while atomic
- fix VCN firmware version reporting for upcoming firmware
Signed-off-by: Dave Airlie <airlied@redhat.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180628032641.2765-1-alexander.deucher@amd.com
Currently device_supports_dax() just checks to see if the QUEUE_FLAG_DAX
flag is set on the device's request queue to decide whether or not the
device supports filesystem DAX. Really we should be using
bdev_dax_supported() like filesystems do at mount time. This performs
other tests like checking to make sure the dax_direct_access() path works.
We also explicitly clear QUEUE_FLAG_DAX on the DM device's request queue if
any of the underlying devices do not support DAX. This makes the handling
of QUEUE_FLAG_DAX consistent with the setting/clearing of most other flags
in dm_table_set_restrictions().
Now that bdev_dax_supported() explicitly checks for QUEUE_FLAG_DAX, this
will ensure that filesystems built upon DM devices will only be able to
mount with DAX if all underlying devices also support DAX.
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Fixes: commit 545ed20e6d ("dm: add infrastructure for DAX support")
Cc: stable@vger.kernel.org
Acked-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Toshi Kani <toshi.kani@hpe.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
