which is rather confusing and can cause gcc warnings:
s5k5baf.c: In function 's5k5baf_load_setfile.isra':
s5k5baf.c:390:13: warning: array subscript 65535 is outside the bounds of an interior zero-length array 'struct <anonymous>[0]' [-Wzero-length-bounds]
390 | if (f->seq[i].offset + d <= end)
| ~~~~~~^~~
It turns out that 'data[]' is used in only one place and it can
easily be replaced by &fw->seq[0].id and 'seq[0]' can be replaced by
'seq[]'.
This is both more readable and solved that warnings.
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Cc: Kyungmin Park <kyungmin.park@samsung.com> Cc: Sylwester Nawrocki <snawrocki@kernel.org> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
media: dt-bindings: media: venus: Add an optional power domain for perf voting
Venus needs to vote for the performance state of a power domain (cx)
to be able to support DVFS. This 'cx' power domain is controlled by
rpmh and is a common power domain (scalable) not specific to
venus alone. This is optional in the sense that, leaving this power
domain out does not really impact the functionality but just makes
the platform a little less power efficient.
Lad Prabhakar [Fri, 2 Oct 2020 10:26:52 +0000 (12:26 +0200)]
media: rcar-vin: rcar-dma: Fix setting VNIS_REG for RAW8 formats
pixelformat in vin priv structure holds V4L2_PIX_FMT_* and not
MEDIA_BUS_FMT_* so make sure we check against V4L2_PIX_FMT_* formats
while setting the VNIS_REG.
Fixes: 8c3e0f67df6c9 ("media: rcar-vin: Extend RAW8 support to all RGB layouts") Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> Reviewed-by: Biju Das <biju.das.jz@bp.renesas.com> Reviewed-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
* tag 'v5.9-rc7': (683 commits)
Linux 5.9-rc7
mm/thp: Split huge pmds/puds if they're pinned when fork()
mm: Do early cow for pinned pages during fork() for ptes
mm/fork: Pass new vma pointer into copy_page_range()
mm: Introduce mm_struct.has_pinned
mm: validate pmd after splitting
mm: don't rely on system state to detect hot-plug operations
mm: replace memmap_context by meminit_context
arch/x86/lib/usercopy_64.c: fix __copy_user_flushcache() cache writeback
lib/memregion.c: include memregion.h
lib/string.c: implement stpcpy
mm/migrate: correct thp migration stats
mm/gup: fix gup_fast with dynamic page table folding
mm: memcontrol: fix missing suffix of workingset_restore
mm, THP, swap: fix allocating cluster for swapfile by mistake
mm: slab: fix potential double free in ___cache_free
Documentation/llvm: Fix clang target examples
io_uring: ensure async buffered read-retry is setup properly
KVM: SVM: Add a dedicated INVD intercept routine
io_uring: don't unconditionally set plug->nowait = true
...
Hans Verkuil [Mon, 28 Sep 2020 13:16:37 +0000 (15:16 +0200)]
media: zoran: fix sparse warnings
The output is not fully supported yet, so some ops are
commented out. Also comment out the corresponding callbacks to prevent
these sparse warnings:
drivers/staging/media/zoran/zoran_driver.c:656:12: warning: 'zoran_s_output' defined but not used [-Wunused-function]
656 | static int zoran_s_output(struct file *file, void *__fh, unsigned int output)
| ^~~~~~~~~~~~~~
drivers/staging/media/zoran/zoran_driver.c:649:12: warning: 'zoran_g_output' defined but not used [-Wunused-function]
649 | static int zoran_g_output(struct file *file, void *__fh, unsigned int *output)
| ^~~~~~~~~~~~~~
drivers/staging/media/zoran/zoran_driver.c:635:12: warning: 'zoran_enum_output' defined but not used [-Wunused-function]
635 | static int zoran_enum_output(struct file *file, void *__fh,
| ^~~~~~~~~~~~~~~~~
drivers/staging/media/zoran/zoran_driver.c:302:12: warning: 'zoran_enum_fmt_vid_overlay' defined but not used [-Wunused-function]
302 | static int zoran_enum_fmt_vid_overlay(struct file *file, void *__fh,
| ^~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/staging/media/zoran/zoran_driver.c:294:12: warning: 'zoran_enum_fmt_vid_out' defined but not used [-Wunused-function]
294 | static int zoran_enum_fmt_vid_out(struct file *file, void *__fh,
| ^~~~~~~~~~~~~~~~~~~~~~
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
This reverts commit 8dce4b265a5357731058f69645840dabc718c687.
The revert is slighly modified:
- Documentation/media/v4l-drivers/index.rst is ,ot restored since removed from tree
- drivers/staging/media/Makefile is not restored since the zoran driver
is not compilable yet.
Marek Szyprowski [Mon, 28 Sep 2020 08:48:44 +0000 (10:48 +0200)]
media: platform: s5p-mfc: Fix adding a standard frame skip mode control
Add proper control type to the recently introduced
V4L2_CID_MPEG_VIDEO_FRAME_SKIP_MODE control. This will forward it to
v4l2_ctrl_new_std_menu() not v4l2_ctrl_new_std(), what causes the
failure. This fixes the following warning during driver initialization:
s5p_mfc_enc_ctrls_setup:2671: Adding control (18) failed
s5p_mfc_open:811: Failed to setup mfc controls
Fixes: ef56b3ee4f54 ("media: s5p-mfc: Use standard frame skip mode control") Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
When compared with a stream generated via ffmpeg, it can be
noticed that the PES doesn't contain any PCR info. That could
cause problems with userspace decoding. So, rewrite the
logic that fills the adaptation info, in order to allow it
to add PCR frames without breaking frame alignment.
In order to better test userspace applications, add an encoding
for the strings. The original channel string is also too big.
As we're just playing the first compasses of the 5th Beethoven
Symphony, change the service name to just "Beethoven". Also,
add a provider's name. For such purpose, let's point it to
the Linux media community website.
With such changes, the SDT table now looks more like one got
from a real digital TV channel:
SDT
| table_id 0x42
| section_length 44
| one 2
| zero 1
| syntax 1
| transport_stream_id 1860
| current_next 1
| version 0
| one2 3
| section_number 0
| last_section_number 0
| network_id 65281
| reserved 255
|\
|- service 0x0880
| EIT schedule 0
| EIT present following 0
| free CA mode 0
| running status 4
| descriptor length 27
| 0x48: service_descriptor
| service type 1
| provider 'LinuxTV.org'
| name 'Beethoven'
|_ 1 services
There's no reason to use static vars to store PSI version
numbers.
Also, currently, version numbers are starting with 0x01,
because there's a code being called that increases it to
1 for all table initializer code, as the code may support
dynamic changes at the PS tables on some future.
So, let's just initialize them to 0x1f, in order for the
versions to be reported as starting from 0.
media: vidtv: fix initialization of the network_id field at SDT
This field should point to the network ID, and has different
ranges for cable, terrestrial or satellite. It also has
an special range for temporary private usage.
For now, let's use the temporary private one. Once the
Network Information Table (NIT) gets added, this should be
reviewed.
media: vidtv: remove more ENDIAN_BITFIELD nonsense
Changeset 870e350d4e39 ("media: vidtv: get rid of ENDIAN_BITFIELD nonsense")
was incomplete. There are still some wrong endannes logic at
the driver. Get rid of them.
media: hist-v4l2.rst: remove :c:`type` from structs and enums
Using struct|enum :c:`type` here sounds a bad idea, as it may be
documenting data types that were already removed. Also, it makes
the file harder to read.
As automarkup.py should be able to handle those cases, let's
just get rid of the markup.
While not required by automarkup, in order to make easier for
grep, both type and names are now at the same line.
media: docs: ipu3.rst: rely at automarkup extension
There are several :c:type: definitions there, in order to
do cross-references with the driver's documentation.
Those are broken when docs are built with Sphinx 3.x, as
it would require :c:struct: instead.
For Sphinx < 3.x, the automarkup.py extension is able to do the
replacement already, and a future improvement on it should make
it also work with Sphinx 3.x.
So, get rid of the usage of the :c:type: macro there.
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
media: glossary.rst: use the right case for glossary entries
Sphinx 3.x is pedantic with glossary entries:
Documentation/userspace-api/media/glossary.rst:17: WARNING: term device driver not found in case sensitive match.made a reference to Device Driver instead.
Documentation/userspace-api/media/glossary.rst:59: WARNING: term media hardware not found in case sensitive match.made a reference to Media Hardware instead.
Documentation/userspace-api/media/glossary.rst:59: WARNING: term IP block not found in case sensitive match.made a reference to IP Block instead.
Documentation/userspace-api/media/glossary.rst:64: WARNING: term hardware component not found in case sensitive match.made a reference to Hardware Component instead.
Documentation/userspace-api/media/glossary.rst:64: WARNING: term ip block not found in case sensitive match.made a reference to IP Block instead.
Documentation/userspace-api/media/glossary.rst:70: WARNING: term peripheral not found in case sensitive match.made a reference to Peripheral instead.
Documentation/userspace-api/media/glossary.rst:130: WARNING: term V4L2 hardware not found in case sensitive match.made a reference to V4L2 Hardware instead.
Documentation/userspace-api/media/glossary.rst:151: WARNING: term hardware peripheral not found in case sensitive match.made a reference to Hardware Peripheral instead.
Documentation/userspace-api/media/glossary.rst:183: WARNING: term device node not found in case sensitive match.made a reference to Device Node instead.
Documentation/userspace-api/media/glossary.rst:191: WARNING: term bridge driver not found in case sensitive match.made a reference to Bridge Driver instead.
While it works with case-insensitive entires, it complains.
Let's fix it, in order to cleanup the warnings. Also, I won't
doubt that a later change on Sphinx will end breaking support
for it.
Merge tag 'x86-urgent-2020-09-27' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Thomas Gleixner:
"Two fixes for the x86 interrupt code:
- Unbreak the magic 'search the timer interrupt' logic in IO/APIC
code which got wreckaged when the core interrupt code made the
state tracking logic stricter.
That caused the interrupt line to stay masked after switching from
IO/APIC to PIC delivery mode, which obviously prevents interrupts
from being delivered.
- Make run_on_irqstack_code() typesafe. The function argument is a
void pointer which is then cast to 'void (*fun)(void *).
This breaks Control Flow Integrity checking in clang. Use proper
helper functions for the three variants reuqired"
* tag 'x86-urgent-2020-09-27' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/ioapic: Unbreak check_timer()
x86/irq: Make run_on_irqstack_cond() typesafe
Merge tag 'timers-urgent-2020-09-27' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer updates from Thomas Gleixner:
"A set of clocksource/clockevents updates:
- Reset the TI/DM timer before enabling it instead of doing it the
other way round.
- Initialize the reload value for the GX6605s timer correctly so the
hardware counter starts at 0 again after overrun.
- Make error return value negative in the h8300 timer init function"
* tag 'timers-urgent-2020-09-27' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
clocksource/drivers/timer-gx6605s: Fixup counter reload
clocksource/drivers/timer-ti-dm: Do reset before enable
clocksource/drivers/h8300_timer8: Fix wrong return value in h8300_8timer_init()
Peter Xu [Fri, 25 Sep 2020 22:26:00 +0000 (18:26 -0400)]
mm/thp: Split huge pmds/puds if they're pinned when fork()
Pinned pages shouldn't be write-protected when fork() happens, because
follow up copy-on-write on these pages could cause the pinned pages to
be replaced by random newly allocated pages.
For huge PMDs, we split the huge pmd if pinning is detected. So that
future handling will be done by the PTE level (with our latest changes,
each of the small pages will be copied). We can achieve this by let
copy_huge_pmd() return -EAGAIN for pinned pages, so that we'll
fallthrough in copy_pmd_range() and finally land the next
copy_pte_range() call.
Huge PUDs will be even more special - so far it does not support
anonymous pages. But it can actually be done the same as the huge PMDs
even if the split huge PUDs means to erase the PUD entries. It'll
guarantee the follow up fault ins will remap the same pages in either
parent/child later.
This might not be the most efficient way, but it should be easy and
clean enough. It should be fine, since we're tackling with a very rare
case just to make sure userspaces that pinned some thps will still work
even without MADV_DONTFORK and after they fork()ed.
Signed-off-by: Peter Xu <peterx@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Peter Xu [Fri, 25 Sep 2020 22:25:59 +0000 (18:25 -0400)]
mm: Do early cow for pinned pages during fork() for ptes
This allows copy_pte_range() to do early cow if the pages were pinned on
the source mm.
Currently we don't have an accurate way to know whether a page is pinned
or not. The only thing we have is page_maybe_dma_pinned(). However
that's good enough for now. Especially, with the newly added
mm->has_pinned flag to make sure we won't affect processes that never
pinned any pages.
It would be easier if we can do GFP_KERNEL allocation within
copy_one_pte(). Unluckily, we can't because we're with the page table
locks held for both the parent and child processes. So the page
allocation needs to be done outside copy_one_pte().
Some trick is there in copy_present_pte(), majorly the wrprotect trick
to block concurrent fast-gup. Comments in the function should explain
better in place.
Oleg Nesterov reported a (probably harmless) bug during review that we
didn't reset entry.val properly in copy_pte_range() so that potentially
there's chance to call add_swap_count_continuation() multiple times on
the same swp entry. However that should be harmless since even if it
happens, the same function (add_swap_count_continuation()) will return
directly noticing that there're enough space for the swp counter. So
instead of a standalone stable patch, it is touched up in this patch
directly.
Peter Xu [Fri, 25 Sep 2020 22:25:57 +0000 (18:25 -0400)]
mm: Introduce mm_struct.has_pinned
(Commit message majorly collected from Jason Gunthorpe)
Reduce the chance of false positive from page_maybe_dma_pinned() by
keeping track if the mm_struct has ever been used with pin_user_pages().
This allows cases that might drive up the page ref_count to avoid any
penalty from handling dma_pinned pages.
Future work is planned, to provide a more sophisticated solution, likely
to turn it into a real counter. For now, make it atomic_t but use it as
a boolean for simplicity.
Suggested-by: Jason Gunthorpe <jgg@ziepe.ca> Signed-off-by: Peter Xu <peterx@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
media: staging: rkisp1: cap: protect access to buf with the spin lock
The function 'rkisp1_stream_start' calls 'rkisp1_set_next_buf'
which access the buffers, so the call should be protected by
taking the cap->buf.lock.
After this patch, all locks are reviewed and commented so remove
the TODO item "review and comment every lock"
media: staging: rkisp1: use the right variants of spin_lock
When locking, use either 'spin_lock' or 'spin_lock_irq'
according to the context. There is nowhere need to
use 'spin_lock_irqsave'.
Outside of irq context, always use 'spin_lock_irq'
to be on the safe side.
media: staging: rkisp1: params: no need to lock default config
In the function 'rkisp1_params_config_parameter' the lock
is taken while applying the default config. But the lock
only needs to protect the buffers list and the 'is_streaming'
field, so move the locking to lock only what is needed.
Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com> Acked-by: Helen Koike <helen.koike@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
media: staging: rkisp1: stats: protect write to 'is_streaming' in start_streaming cb
The field stats->is_streaming is written in 'start_streaming' callback
without the stats->lock protection.
The isr might run together with the callback so 'spin_lock_irq'
should be used.
Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com> Acked-by: Helen Koike <helen.koike@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
media: staging: rkisp1: isp: add a warning and debugfs var for irq delay
The signal RKISP1_CIF_ISP_FRAME is set when the ISP completes
outputting the frame to the next block in the pipeline.
In order to keep buffer synchronization we assume that the
RKISP1_CIF_ISP_V_START signal never arrives together with the
RKISP1_CIF_ISP_FRAME signal.
In case those signals arrive together then the code is not able to
tell if the RKISP1_CIF_ISP_FRAME signal relates to the frame of
the current v-start or the previous. This patch adds a WARN_ONCE
and a debugfs var to catch it.
media: staging: rkisp1: remove atomic operations for frame sequence
The isp.frame_sequence is now read only from the irq handlers
that are all fired from the same interrupt, so there is not need
for atomic operation.
In addition, the frame seq incrementation is moved from
rkisp1_isp_queue_event_sof to rkisp1_isp_isr to make the code
clearer and the incorrect inline comment is removed.
Signed-off-by: Dafna Hirschfeld <dafna.hirschfeld@collabora.com> Acked-by: Helen Koike <helen.koike@collabora.com> Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
projects / thirdparty / kernel / linux.git / log
