[media] Fix build when drivers are builtin and frontend modules
There are a large number of reports that the media build is
not compiling when some drivers are compiled as builtin, while
the needed frontends are compiled as module.
On the last one of such reports:
From: kbuild test robot <fengguang.wu@intel.com>
Subject: saa7134-dvb.c:undefined reference to `zl10039_attach'
The .config file has:
CONFIG_VIDEO_SAA7134=y
CONFIG_VIDEO_SAA7134_DVB=y
# CONFIG_MEDIA_ATTACH is not set
CONFIG_DVB_ZL10039=m
And it produces all those errors:
drivers/built-in.o: In function `set_type':
tuner-core.c:(.text+0x2f263e): undefined reference to `tea5767_attach'
tuner-core.c:(.text+0x2f273e): undefined reference to `tda9887_attach'
drivers/built-in.o: In function `tuner_probe':
tuner-core.c:(.text+0x2f2d20): undefined reference to `tea5767_autodetection'
drivers/built-in.o: In function `av7110_attach':
av7110.c:(.text+0x330bda): undefined reference to `ves1x93_attach'
av7110.c:(.text+0x330bf7): undefined reference to `stv0299_attach'
av7110.c:(.text+0x330c63): undefined reference to `tda8083_attach'
av7110.c:(.text+0x330d09): undefined reference to `ves1x93_attach'
av7110.c:(.text+0x330d33): undefined reference to `tda8083_attach'
av7110.c:(.text+0x330d5d): undefined reference to `stv0297_attach'
av7110.c:(.text+0x330dbe): undefined reference to `stv0299_attach'
drivers/built-in.o: In function `tuner_attach_dtt7520x':
ngene-cards.c:(.text+0x3381cb): undefined reference to `dvb_pll_attach'
drivers/built-in.o: In function `demod_attach_lg330x':
ngene-cards.c:(.text+0x33828a): undefined reference to `lgdt330x_attach'
drivers/built-in.o: In function `demod_attach_stv0900':
ngene-cards.c:(.text+0x3383d5): undefined reference to `stv090x_attach'
drivers/built-in.o: In function `cineS2_probe':
ngene-cards.c:(.text+0x338b7f): undefined reference to `drxk_attach'
drivers/built-in.o: In function `configure_tda827x_fe':
saa7134-dvb.c:(.text+0x346ae7): undefined reference to `tda10046_attach'
drivers/built-in.o: In function `dvb_init':
saa7134-dvb.c:(.text+0x347283): undefined reference to `mt352_attach'
saa7134-dvb.c:(.text+0x3472cd): undefined reference to `mt352_attach'
saa7134-dvb.c:(.text+0x34731c): undefined reference to `tda10046_attach'
saa7134-dvb.c:(.text+0x34733c): undefined reference to `tda10046_attach'
saa7134-dvb.c:(.text+0x34735c): undefined reference to `tda10046_attach'
saa7134-dvb.c:(.text+0x347378): undefined reference to `tda10046_attach'
saa7134-dvb.c:(.text+0x3473db): undefined reference to `tda10046_attach'
drivers/built-in.o:saa7134-dvb.c:(.text+0x347502): more undefined references to `tda10046_attach' follow
drivers/built-in.o: In function `dvb_init':
saa7134-dvb.c:(.text+0x347812): undefined reference to `mt352_attach'
saa7134-dvb.c:(.text+0x347951): undefined reference to `mt312_attach'
saa7134-dvb.c:(.text+0x3479a9): undefined reference to `mt312_attach'
>> saa7134-dvb.c:(.text+0x3479c1): undefined reference to `zl10039_attach'
This is happening because a builtin module can't use directly a symbol
found on a module. By enabling CONFIG_MEDIA_ATTACH, the configuration
becomes valid, as dvb_attach() macro loads the module if needed, making
the symbol available to the builtin module.
While this bug started to appear after the patches that use IS_DEFINED
macro (like changeset 7b34be71db533f3e0cf93d53cf62d036cdb5418a), this
bug is a way ancient than that.
The thing is that, before the IS_DEFINED() patches, the logic used to be:
The above code, with the .config file used, was evoluting to FALSE
(instead of TRUE as it should be, as CONFIG_DVB_ZL10039 is 'm'),
and were adding the static inline code at saa7134-dvb, instead
of the external call. So, while it weren't producing any compilation
error, the code weren't working either.
So, as the overhead for using CONFIG_MEDIA_ATTACH is minimal, just
enable it, if MODULES is defined.
[media] s5p makefiles: don't override other selections on obj-[ym]
The $obj-m/$obj-y vars should be adding new modules to build, not
overriding it. So, it should never use
$obj-y := foo.o
instead, it should use:
$obj-y += foo.o
Failing to do that is very bad, as it will suppress needed modules.
The ISP clock register content is not preserved over the ISP power domain
off/on cycle. Instead of setting the clock frequencies once at probe time
the clock rates set up is moved to the runtime_resume handler, which is
invoked after the related power domain is already enabled, ensuring the
clocks are properly configured when the device is actively used.
This fixes the FIMC-IS malfunctions and STREAM ON timeout errors accuring
on some boards:
[ 59.860000] fimc_is_general_irq_handler:583 ISR_NDONE: 5: 0x800003e8, IS_ERROR_UNKNOWN
[ 59.860000] fimc_is_general_irq_handler:586 IS_ERROR_TIME_OUT
[media] rtl28xxu: fix buffer overflow when probing Rafael Micro r820t tuner
As suggested by Antti, this patch replaces:
https://patchwork.kernel.org/patch/2649861/
The buffer overflow is fixed by reading only the r820t ID register.
Signed-off-by: Gianluca Gennari <gennarone@gmail.com> Acked-by: Antti Palosaari <crope@iki.fi> Reviewed-by: Antti Palosaari <crope@iki.fi> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Wenbing Wang [Wed, 5 Jun 2013 09:37:14 +0000 (06:37 -0300)]
[media] soc_camera: error dev remove and v4l2 call
in soc_camera_close(), if ici->ops->remove() removes device firstly,
and then call __soc_camera_power_off(), it has logic error. Since
if remove device, it should disable subdev clk. but in __soc_camera_
power_off(), it will callback v4l2 s_power function which will
read/write subdev registers to control power by i2c. and then
i2c read/write will fail because of clk disable.
So suggest to re-sequence two functions call.
Change-Id: Iee7a6d4fc7c7c1addb5d342621eb8dcd00fa2745
The 'bytesperline' value only indicates the stride of the Y plane
if the color format is planar, such as NV12. When calculating
the total plane size, the size of CbCr plane must also be considered.
[media] sh_veu: keep power supply until the m2m context is released
In the sh_veu driver, only the interrupt handler 'sh_veu_bh'
can invoke the v4l2_m2m_job_finish() function.
So the hardware must be alive for handling interrupts
until returning from v4l2_m2m_ctx_release().
[media] sh_veu: invoke v4l2_m2m_job_finish() even if a job has been aborted
v4l2_m2m_job_finish() should be invoked even if the current
ongoing job has been aborted since v4l2_m2m_ctx_release() which
has issued the job abort may wait until the finish function is invoked.
Hans Verkuil [Mon, 3 Jun 2013 07:35:22 +0000 (04:35 -0300)]
[media] v4l2-ioctl: don't print the clips list
The clips pointer is a userspace pointer, not a kernelspace pointer,
so you can't dereference the clips pointer.
Also add a few missing commas and newlines.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 29 May 2013 13:44:22 +0000 (10:44 -0300)]
[media] cx88: fix NULL pointer dereference
This fixes a NULL pointer deference when loading the cx88_dvb module for a
Hauppauge HVR4000.
The bugzilla bug report is here:
https://bugzilla.kernel.org/show_bug.cgi?id=56271
The cause is that the wm8775 is optional, so even though the board info says
there is one, it doesn't have to be there. Checking whether the module was
actually loaded is much safer.
Note that this driver is quite buggy when it comes to unloading and reloading
modules. Unloading cx8800 and reloading it again will still cause a crash,
most likely because either the i2c bus isn't unloaded at the right time and/or
the v4l2_device_unregister isn't called at the right time.
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Reported-by: Sebastian Frei <sebastian@familie-frei.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
The "sysreg" clock is required by multiple subsystems and none of the
other drivers handles this clock explicitly. It is currently assumed
that this clock is always on, left in its default state after system
reset.
Remove handling of this clock from the FIMC-IS driver to avoid breaking
other subsystems.
[media] exynos4-is: Ensure fimc-is clocks are not enabled until properly configured
Use clk_prepare_enable/clk_unprepare_disable instead of preparing the
clocks during the driver initalization and then using just clk_disable/
clk_enable. The clock framework doesn't guarantee a clock will not get
enabled during e.g. clk_set_parent if clk_prepare has been called on it.
So we ensure clk_prepare() is called only when it is safe to enable
the clocks, i.e. the parent clocks and the clocks' frequencies are set.
It must be ensured the FIMC-IS clocks have proper frequencies before they
are enabled, otherwise the whole system will hang.
Andrzej Hajda [Tue, 28 May 2013 06:26:15 +0000 (03:26 -0300)]
[media] s5p-mfc: v4l2 controls setup routine moved to initialization code
Callback .start_streaming is called once for every queue,
so v4l2_ctrl_handler_setup was called twice during stream start.
Moving v4l2_ctrl_handler_setup to context initialization
reduces numbers of calls and seems to be more consistent with API.
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com> Signed-off-by: Kamil Debski <k.debski@samsung.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Arun Kumar K [Mon, 27 May 2013 06:39:29 +0000 (03:39 -0300)]
[media] s5p-mfc: Remove special clock usage in driver
MFC uses two clocks - MFC gate clock and special clock
which is named as "sclk_mfc" in exynos4 and "aclk_333" in
exynos5 SoC. The driver was doing just a clk_prepare on
this special clock without a clk_enable call. As this
sclk is the parent of gate clock, it gets prepared and
enabled along with the gate clock. So there is no need
for the driver to use this sclk. This patch removes the
sclk usage from driver.
[media] s5p-mfc: Remove unused s5p_mfc_get_decoded_status_v6() function
This patch fixes following compilation warning:
CC [M] drivers/media/platform/s5p-mfc/s5p_mfc_opr_v6.o
drivers/media/platform/s5p-mfc/s5p_mfc_opr_v6.c:1733:12: warning: ‘s5p_mfc_get_decoded_status_v6’ defined but not used
It assigns existing but not used s5p_mfc_get_dec_status_v6() function to the
get_dec_status callback. It seems the get_dec_status callback is not used
anyway, as there is no corresponding s5p_mfc_hw_call().
Seung-Woo Kim [Tue, 21 May 2013 02:47:30 +0000 (23:47 -0300)]
[media] media: v4l2-mem2mem: return for polling if a buffer is available
The v4l2_m2m_poll() does not need to wait if there is already a buffer in
done_list of source and destination queues, but current v4l2_m2m_poll() always
waits. So done_list of each queue is checked before calling poll_wait().
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com> Acked-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Kamil Debski <k.debski@samsung.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Seung-Woo Kim [Tue, 21 May 2013 02:47:29 +0000 (23:47 -0300)]
[media] media: vb2: return for polling if a buffer is available
The vb2_poll() does not need to wait next vb_buffer_done() if there is already
a buffer in done_list of queue, but current vb2_poll() always waits.
So done_list is checked before calling poll_wait().
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com> Acked-by: Marek Szyprowski <m.szyprowski@samsung.com> Signed-off-by: Kamil Debski <k.debski@samsung.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
MFC v6 needs minimum number of output buffers to be queued
for encoder depending on the stream type and profile.
The patch modifies the driver so that encoding cannot be
started with lesser number of OUTPUT buffers than required.
This also fixes the crash happeninig during multi instance
encoder-decoder simultaneous run due to memory allocation
happening from interrupt context.
Lad, Prabhakar [Mon, 13 May 2013 09:39:07 +0000 (06:39 -0300)]
[media] drivers/staging: davinci: vpfe: fix dependency for building the driver
from commit 3778d05036cc7ddd983ae2451da579af00acdac2
[media: davinci: kconfig: fix incorrect selects]
VIDEO_VPFE_CAPTURE was removed but there was a negative
dependancy for building the DM365 VPFE MC based capture driver
(VIDEO_DM365_VPFE), This patch fixes this dependency by replacing
the VIDEO_VPFE_CAPTURE with VIDEO_DM365_ISIF, so as when older DM365
ISIF v4l driver is selected the newer media controller driver for
DM365 isnt visible.
Reported-by: Paul Bolle <pebolle@tiscali.nl> Signed-off-by: Lad, Prabhakar <prabhakar.csengg@gmail.com> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Tue, 7 May 2013 11:06:11 +0000 (08:06 -0300)]
[media] DocBook: media: update codec section, drop obsolete 'suspended' state
The Codec section in the V4L2 specification was marked as 'suspended', even
though codec support has been around for quite some time. Update this
section, explaining a bit about memory-to-memory devices and pointing to
the MPEG controls section.
Arnd Bergmann [Thu, 2 May 2013 14:16:26 +0000 (11:16 -0300)]
[media] radio-si476x: depend on SND_SOC
It is not possible to select SND_SOC_SI476X if we have not also
enabled SND_SOC.
warning: (RADIO_SI476X) selects SND_SOC_SI476X which has unmet
direct dependencies (SOUND && !M68K && !UML && SND && SND_SOC)
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[hans.verkuil@cisco.com: fixed wrong driver name in subject] Acked-by: Mauro Carvalho Chehab <mchehab@redhat.com> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Lad, Prabhakar [Fri, 10 May 2013 03:48:38 +0000 (00:48 -0300)]
[media] davinci: vpfe: fix error path in probe
The error path on failure was calling mutex_unlock(), but there was
no actuall call before for mutex_lock(). This patch fixes this issue
by pointing it to proper go label.
Reported-by: Jose Pablo Carballo <jose.carballo@ridgerun.com> Signed-off-by: Lad, Prabhakar <prabhakar.csengg@gmail.com> Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Lad, Prabhakar [Tue, 7 May 2013 04:07:25 +0000 (01:07 -0300)]
[media] media: davinci: vpbe: fix layer availability for NV12 format
For NV12 format, even if display data is single image,
both VIDWIN0 and VIDWIN1 parameters must be used. The start
address of Y data plane and C data plane is configured in
VIDEOWIN0ADH/L and VIDEOWIN1ADH/L respectively.
cuurently only one layer was requested, which is suffice
for yuv422, but for yuv420(NV12) two layers are required and
fix the same by requesting for other layer if pix fmt is NV12
during set_fmt.
Hans de Goede [Mon, 29 Apr 2013 11:54:14 +0000 (08:54 -0300)]
[media] gspca-sonixb: Adjust hstart on sn9c103 + pas202
For some unknown reason we need to increase hstart by 1 on when using the
PAS202 on the sn9c103 (versus on the sn9c102), otherwise we get the wrong
colors, due to shifting of the bayer pattern.
Reported-by: Patrizio Bassi <patrizio.bassi@gmail.com> Tested-by: Patrizio Bassi <patrizio.bassi@gmail.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans de Goede [Mon, 1 Apr 2013 21:04:21 +0000 (18:04 -0300)]
[media] pwc: Fix comment wrt lock ordering
With all the changes to handle the locking in the v4l2-core rather then at
the driver level, the order in which the 2 pwc locks need to be taken has
changed, update the comment in the header file to correctly reflect this.
Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
* patchwork: (831 commits)
[media] cx88: make core less verbose
[media] em28xx: fix oops at em28xx_dvb_bus_ctrl()
[media] s5c73m3: fix indentation of the help section in Kconfig
[media] cx25821-alsa: get rid of a __must_check warning
[media] cx25821-video: declare cx25821_vidioc_s_std as static
[media] cx25821-video: remove maxw from cx25821_vidioc_try_fmt_vid_cap
[media] r820t: Remove a warning for an unused value
[media] dib0090: Fix a warning at dib0090_set_EFUSE
[media] dib8000: fix a warning
[media] dib8000: Fix sub-channel range
[media] dib8000: store dtv_property_cache in a temp var
[media] dib8000: warning fix: declare internal functions as static
[media] r820t: quiet gcc warning on n_ring
[media] r820t: memory leak in release()
[media] r820t: precendence bug in r820t_xtal_check()
[media] videodev2.h: Remove the unused old V4L1 buffer types
[media] anysee: Grammar s/report the/report to/
[media] anysee: Initialize ret = 0 in anysee_frontend_attach()
[media] media: videobuf2: fix the length check for mmap
[media] em28xx: save isoc endpoint number for DVB only if endpoint has alt settings with xMaxPacketSize != 0
...
* tag 'v3.9': (1099 commits)
Linux 3.9
vm: add no-mmu vm_iomap_memory() stub
efivars: only check for duplicates on the registered list
TTY: fix atime/mtime regression
aio: fix possible invalid memory access when DEBUG is enabled
parisc: use spin_lock_irqsave/spin_unlock_irqrestore for PTE updates
parisc: disable -mlong-calls compiler option for kernel modules
parisc: uaccess: fix compiler warnings caused by __put_user casting
parisc: Change kunmap macro to static inline function
parisc: Provide __ucmpdi2 to resolve undefined references in 32 bit builds.
sparc64: Fix missing put_cpu_var() in tlb_batch_add_one() when not batching.
Revert "gpio: pxa: set initcall level to module init"
efi: Check EFI revision in setup_efi_vars
x86, efi: Fix a build warning
Revert "MIPS: page.h: Provide more readable definition for PAGE_MASK."
kernel/hz.bc: ignore.
Linux 3.9-rc8
events: Protect access via task_subsys_state_check()
net: fix incorrect credentials passing
x86, microcode: Verify the family before dispatching microcode patching
...
Along the time, several debug messages were added at cx88-cards.
While those are still useful to track some troubles with
tuners, they're too verbose:
[ 5768.281801] cx88[0]: Calling XC2028/3028 callback
[ 5768.287388] cx88[0]: Calling XC2028/3028 callback
[ 5768.292575] cx88[0]: Calling XC2028/3028 callback
[ 5768.299408] cx88[0]: Calling XC2028/3028 callback
[ 5768.306244] cx88[0]: Calling XC2028/3028 callback
...
and, most of the time, useless.
So, disable them, except if core_debug modprobe parameter
is used.
Merge tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
Pull ARM SoC fix from Olof Johansson:
"A late-arriving fix for musb on OMAP4, resolving an issue where the
musb IP won't be clocked and thus not functional. Small in scope,
most of the lines changed is a longish comment."
* tag 'fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
ARM: OMAP4: hwmod data: make 'ocp2scp_usb_phy_phy_48m" as the main clock
I think we could just move the full vm_iomap_memory() function into
util.h or similar, but I didn't get any reply from anybody actually
using nommu even to this trivial patch, so I'm not going to touch it any
more than required.
Here's the fairly minimal stub to make the nommu case at least
potentially work. It doesn't seem like anybody cares, though.
Olof Johansson [Sat, 27 Apr 2013 00:35:13 +0000 (17:35 -0700)]
Merge tag 'omap-for-v3.9-rc6/fixes-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap into fixes
From Tony Lindgren:
One MUSB regression fix that I forgot to send earlier. Without
this MUSB no longer works on omap4 based devices.
* tag 'omap-for-v3.9-rc6/fixes-signed' of git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap:
ARM: OMAP4: hwmod data: make 'ocp2scp_usb_phy_phy_48m" as the main clock
Merge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media
Pull media fixes from Mauro Carvalho Chehab:
"Two driver fixes.
One avoids reading any file at a system with a cx25821 board
(fortunately, this is not a common device). The other one prevents
reading after a buffer with ISDB-T devices based on mb86a20s."
* 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
[media] cx25821: do not expose broken video output streams
[media] mb86a20s: Fix estimate_rate setting
Merge branch 'fixes-3.9-late' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux
Pull late parisc fixes from Helge Deller:
"I know it's *very* late in the 3.9 release cycle, but since there
aren't that many people testing the parisc linux kernel, a few (for
our port) critical issues just showed up a few days back for the first
time.
What's in it?
- add missing __ucmpdi2 symbol, which is required for btrfs on 32bit
kernel.
- change kunmap() macro to static inline function. This fixes a
debian/gcc-4.4 build error.
- add locking when doing PTE updates. This fixes random userspace
crashes.
- disable (optional) -mlong-calls compiler option for modules, else
modules can't be loaded at runtime.
- a smart patch by Will Deacon which fixes 64bit put_user() warnings
on 32bit kernel."
* 'fixes-3.9-late' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: use spin_lock_irqsave/spin_unlock_irqrestore for PTE updates
parisc: disable -mlong-calls compiler option for kernel modules
parisc: uaccess: fix compiler warnings caused by __put_user casting
parisc: Change kunmap macro to static inline function
parisc: Provide __ucmpdi2 to resolve undefined references in 32 bit builds.
Matt Fleming [Fri, 26 Apr 2013 09:10:55 +0000 (10:10 +0100)]
efivars: only check for duplicates on the registered list
variable_is_present() accesses '__efivars' directly, but when called via
gsmi_init() Michel reports observing the following crash,
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: variable_is_present+0x55/0x170
Call Trace:
register_efivars+0x106/0x370
gsmi_init+0x2ad/0x3da
do_one_initcall+0x3f/0x170
The reason for the crash is that '__efivars' hasn't been initialised nor
has it been registered with register_efivars() by the time the google
EFI SMI driver runs. The gsmi code uses its own struct efivars, and
therefore, a different variable list. Fix the above crash by passing
the registered struct efivars to variable_is_present(), so that we
traverse the correct list.
Reported-by: Michel Lespinasse <walken@google.com> Tested-by: Michel Lespinasse <walken@google.com> Cc: Mike Waychison <mikew@google.com> Cc: Matthew Garrett <matthew.garrett@nebula.com> Cc: Seiji Aguchi <seiji.aguchi@hds.com> Signed-off-by: Matt Fleming <matt.fleming@intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In commit b0de59b5733d ("TTY: do not update atime/mtime on read/write")
we removed timestamps from tty inodes to fix a security issue and waited
if something breaks. Well, 'w', the utility to find out logged users
and their inactivity time broke. It shows that users are inactive since
the time they logged in.
To revert to the old behaviour while still preventing attackers to
guess the password length, we update the timestamps in one-minute
intervals by this patch.
H. Peter Anvin [Thu, 25 Apr 2013 21:00:22 +0000 (14:00 -0700)]
Merge tag 'efi-urgent' into x86/urgent
* The EFI variable anti-bricking algorithm merged in -rc8 broke booting
on some Apple machines because they implement EFI spec 1.10, which
doesn't provide a QueryVariableInfo() runtime function and the logic
used to check for the existence of that function was insufficient.
Fix from Josh Boyer.
* The anti-bricking algorithm also introduced a compiler warning on
32-bit. Fix from Borislav Petkov.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
parisc: use spin_lock_irqsave/spin_unlock_irqrestore for PTE updates
User applications running on SMP kernels have long suffered from instability
and random segmentation faults. This patch improves the situation although
there is more work to be done.
One of the problems is the various routines in pgtable.h that update page table
entries use different locking mechanisms, or no lock at all (set_pte_at). This
change modifies the routines to all use the same lock pa_dbit_lock. This lock
is used for dirty bit updates in the interruption code. The patch also purges
the TLB entries associated with the PTE to ensure that inconsistent values are
not used after the page table entry is updated. The UP and SMP code are now
identical.
The change also includes a minor update to the purge_tlb_entries function in
cache.c to improve its efficiency.
Signed-off-by: John David Anglin <dave.anglin@bell.net> Cc: Helge Deller <deller@gmx.de> Signed-off-by: Helge Deller <deller@gmx.de>
parisc: disable -mlong-calls compiler option for kernel modules
CONFIG_MLONGCALLS was introduced in commit ec758f98328da3eb933a25dc7a2eed01ef44d849 to overcome linker issues when linking
huge linux kernels, e.g. with many modules linked in.
But in the kernel module loader there is no support yet for the new relocation
types, which is why modules built with -mlong-calls can't be loaded.
Furthermore, for modules long calls are not really necessary, since we already
use stub sections which resolve long distance calls.
So, let's just disable this compiler option when compiling kernel modules.
Will Deacon [Mon, 22 Apr 2013 12:53:43 +0000 (12:53 +0000)]
parisc: uaccess: fix compiler warnings caused by __put_user casting
When targetting 32-bit processors, __put_user emits a pair of stw
instructions for the 8-byte case. If the type of __val is a pointer, the
marshalling code casts it to the wider integer type of u64, resulting
in the following compiler warnings:
kernel/signal.c: In function 'copy_siginfo_to_user':
kernel/signal.c:2752:11: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
kernel/signal.c:2752:11: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]
[...]
This patch fixes the warnings by removing the marshalling code and using
the correct output modifiers in the __put_{user,kernel}_asm64 macros
so that GCC will allocate the right registers without the need to
extract the two words explicitly.
parisc: Change kunmap macro to static inline function
Change kunmap macro to static inline function to fix build error
compiling drivers/base/dma-buf.c.
Without the change, the following error can occur:
CC drivers/base/dma-buf.o
drivers/base/dma-buf.c: In function 'dma_buf_kunmap':
drivers/base/dma-buf.c:427:46:
error: macro "kunmap" passed 3 arguments, but takes just 1
I believe parisc is the only arch to implement kunmap using a macro.
Signed-off-by: John David Anglin <dave.anglin@bell.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: Helge Deller <deller@gmx.de> Signed-off-by: Helge Deller <deller@gmx.de>
parisc: Provide __ucmpdi2 to resolve undefined references in 32 bit builds.
The Debian experimental linux source package (3.8.5-1) build fails
with the following errors:
...
MODPOST 2016 modules
ERROR: "__ucmpdi2" [fs/btrfs/btrfs.ko] undefined!
ERROR: "__ucmpdi2" [drivers/md/dm-verity.ko] undefined!
The attached patch resolves this problem. It is based on the s390
implementation of ucmpdi2.c.
Signed-off-by: John David Anglin <dave.anglin@bell.net> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Signed-off-by: Helge Deller <deller@gmx.de>
[media] s5c73m3: fix indentation of the help section in Kconfig
The 'help' section of the Kconfig entry for this driver is missing
an extra alignment. That seems to violate what's stated at:
Documentation/kbuild/kconfig-language.txt
Even if it works, the better is to add 2 extra spaces there, as
this is the common practice and helps human reading of the file.
[media] cx25821-alsa: get rid of a __must_check warning
The hole reason for __must_check is to not ignore an error.
However, a "ret" value is used at cx25821 just to avoid the
Kernel compilation to compain about it.
That, however, produces another warning (with W=1):
drivers/media/pci/cx25821/cx25821-alsa.c: In function 'cx25821_audio_fini':
drivers/media/pci/cx25821/cx25821-alsa.c:727:6: warning: variable 'ret' set but not used [-Wunused-but-set-variable]
With the current implementation of driver_for_each_device() and
cx25821_alsa_exit_callback(), there's actually just one
very unlikely condition where it will currently produce
an error: if driver_find() returns NULL.
Ok, there's not much that can be done, as it is on a driver's
function that returns void, but it can at least print some message
if the error happens.
[media] cx25821-video: declare cx25821_vidioc_s_std as static
Fixes the following warning:
drivers/media/pci/cx25821/cx25821-video.c: At top level:
drivers/media/pci/cx25821/cx25821-video.c:766:5: warning: no previous prototype for 'cx25821_vidioc_s_std' [-Wmissing-prototypes]
[media] cx25821-video: remove maxw from cx25821_vidioc_try_fmt_vid_cap
After cx25821-video cleanup, this var is not used anymore:
drivers/media/pci/cx25821/cx25821-video.c: In function 'cx25821_vidioc_try_fmt_vid_cap':
drivers/media/pci/cx25821/cx25821-video.c:591:15: warning: variable 'maxw' set but not used [-Wunused-but-set-variable]
as the code now checks the max width as the default case for the
range check.
[media] r820t: Remove a warning for an unused value
Currently, the driver complains about the pre_detect var:
drivers/media/tuners/r820t.c: In function 'r820t_sysfreq_sel':
drivers/media/tuners/r820t.c:722:31: warning: variable 'pre_dect' set but not used [-Wunused-but-set-variable]
While rtl8232 code comments it, perhaps some other driver may use.
So, the better is to keep the code there, allowing to enable it
via r820t config data.
[media] dib0090: Fix a warning at dib0090_set_EFUSE
The check if the values for c, h and n are within the range is
always true, as, if one of this values is out of range, the
previous "if" clauses will default to a value within the
range.
That fixes the following warning:
drivers/media/dvb-frontends/dib0090.c: In function 'dib0090_set_EFUSE':
drivers/media/dvb-frontends/dib0090.c:1545:5: warning: comparison is always true due to limited range of data type [-Wtype-limits]
drivers/media/dvb-frontends/dib8000.c: In function 'dib8000_wait_lock':
drivers/media/dvb-frontends/dib8000.c:3972:1: warning: 'value' may be used uninitialized in this function [-Wmaybe-uninitialized]
drivers/media/dvb-frontends/dib8000.c:2419:6: note: 'value' was declared here
isdbt_sb_subchannel is unsigned with 8 bits. So, it will
never be -1. Instead, any value bigger than 13 is invalid.
As is, the current code generates the following warnings:
drivers/media/dvb-frontends/dib8000.c: In function 'dib8000_set_isdbt_common_channel':
drivers/media/dvb-frontends/dib8000.c:2358:3: warning: comparison is always true due to limited range of data type [-Wtype-limits]
drivers/media/dvb-frontends/dib8000.c: In function 'dib8000_tune':
drivers/media/dvb-frontends/dib8000.c:3107:8: warning: comparison is always false due to limited range of data type [-Wtype-limits]
drivers/media/dvb-frontends/dib8000.c:3153:9: warning: comparison is always false due to limited range of data type [-Wtype-limits]
drivers/media/dvb-frontends/dib8000.c:3160:5: warning: comparison is always false
It should also be noticed that ARIB STD-B31, item
"3.15.6.8 Number of segments" at TMCC table defines the
value 15 for unused segment, and 14 as reserved.
So, better to change the check to consider any value
bigger than 13 to mean that sub-channels should be
disabled, fixing the warning and doing the right thing
even if an invalid value is filled by userspace.
[media] dib8000: store dtv_property_cache in a temp var
dtv_property_cache is used on several places on very long lines.
On all places it is used, a long list of struct reference is done.
Instead of doing it, at the routines where it is used more than once,
replace it by one temporary var. That may help the compiler to
use a better code. It also makes easier to review the code, as the
lines becomes closer to 80 columns, making them a way clearer
to read.
[media] dib8000: warning fix: declare internal functions as static
drivers/media/dvb-frontends/dib8000.c:2412:5: warning: no previous prototype for 'dib8000_wait_lock' [-Wmissing-prototypes]
drivers/media/dvb-frontends/dib8000.c:2688:5: warning: no previous prototype for 'dib8000_get_symbol_duration' [-Wmissing-prototypes]
drivers/media/tuners/r820t.c: In function 'r820t_imr':
drivers/media/tuners/r820t.c:1871:8: warning: 'n_ring' may be used uninitialized in this function [-Wmaybe-uninitialized]
Mauro: This is a FALSE POSITIVE: the loop will always return a value
for n_ring, as the last test will fill it with 15, if the loop fails.
* topic/r820t: (31 commits)
[media] r820t: Don't divide the IF by two
[media] r820t: disable auto gain/VGA setting
[media] rtl2832: Fix IF calculus
[media] r820t: put it into automatic gain mode
[media] r820t: Fix hp_cor filter mask
[media] r820t: fix PLL calculus
[media] r820t: Don't put it in standby if not initialized yet
[media] r820t: avoid rewrite all regs when not needed
[media] r820t: Allow disabling IMR callibration
[media] r820t: add a commented code for GPIO
[media] r820t: add IMR calibrate code
[media] r820t: proper initialize the PLL register
[media] r820t: use usleep_range()
[media] r820t: fix prefix of the r820t_read() function
[media] r820t: split the function that read cached regs
[media] r820t: better report signal strength
[media] r820t: add support for diplexer
[media] r820t: Show the read data in the bit-reversed order
[media] r820t: use the second table for 7MHz
[media] r820t: Invert bits for read ops
...
[media] anysee: Initialize ret = 0 in anysee_frontend_attach()
drivers/media/usb/dvb-usb-v2/anysee.c: In function ‘anysee_frontend_attach’:
drivers/media/usb/dvb-usb-v2/anysee.c:641: warning: ‘ret’ may be used uninitialized in this function
And gcc is right (see the ANYSEE_HW_507T case), so initialize ret to zero
to fix this.
[media] media: videobuf2: fix the length check for mmap
Memory maps typically require that the buffer size to be page
aligned. Currently, two memops drivers do such alignment
internally, but videobuf-vmalloc doesn't.
Also, the buffer overflow check doesn't take it into account.
So, instead of doing it at each memops driver, enforce it at
VB2 core.
Frank Schaefer [Mon, 8 Apr 2013 16:06:59 +0000 (13:06 -0300)]
[media] em28xx: save isoc endpoint number for DVB only if endpoint has alt settings with xMaxPacketSize != 0
X-Patchwork-Delegate: mchehab@redhat.com
In addition to commit 72cc9ba3 "em28xx: ignore isoc DVB USB endpoints with
wMaxPacketSize = 0 bytes for all alt settings" we should not save the endpoint
number of the isoc DVB endpoint before it has been validated.
While the current code works fine, dev->dvb_ep_isoc != 0 could be interpreted
as indicator that the device provides DVB support.
Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Hans Verkuil [Wed, 17 Apr 2013 06:04:10 +0000 (03:04 -0300)]
[media] mem2mem_testdev: set timestamp_type and add debug param
While testing v4l2-ctl I noticed that this m2m driver didn't set timestamp_type
and that it spammed the kernel log with debug messages. Set timestamp_type
correctly and add debug module option to enable debug messages.
Since the introduction of the timestamp_type field, it is necessary that
the driver chooses which type it will use. This patch adds support for
the timestamp_type.
[media] m2m-deinterlace: Add copy time stamp handling
Since the introduction of the timestamp_type field, it is necessary that
the driver chooses which type it will use. This patch adds support for
the timestamp_type.
Since the introduction of the timestamp_type field, it is necessary that
the driver chooses which type it will use. This patch adds support for
the timestamp_type.
Since the introduction of the timestamp_type field, it is necessary that
the driver chooses which type it will use. This patch adds support for
the timestamp_type.
Since the introduction of the timestamp_type field, it is necessary that
the driver chooses which type it will use. This patch adds support for
the timestamp_type.
Since the introduction of the timestamp_type field, it is necessary that
the driver chooses which type it will use. This patch adds support for
the timestamp_type.
[media] exynos4-is: Copy timestamps from M2M OUTPUT to CAPTURE buffer queue
Add copying of buffer timestamps and set the timestamp_type to
V4L2_BUF_FLAG_TIMESTAMP_COPY to avoid warnings about UNDEFINED
timestamp type like:
WARNING: at drivers/media/v4l2-core/videobuf2-core.c:2042 vb2_queue_init+0xe0/0x18c()
Modules linked in:
[<c0016ef0>] (unwind_backtrace+0x0/0x13c) from [<c0029b3c>] (warn_slowpath_common+0x54/0x64)
[<c0029b3c>] (warn_slowpath_common+0x54/0x64) from [<c0029b68>] (warn_slowpath_null+0x1c/0x24)
[<c0029b68>] (warn_slowpath_null+0x1c/0x24) from [<c03b7018>] (vb2_queue_init+0xe0/0x18c)
[<c03b7018>] (vb2_queue_init+0xe0/0x18c) from [<c03b4e08>] (v4l2_m2m_ctx_init+0xa0/0xc4)
[<c03b4e08>] (v4l2_m2m_ctx_init+0xa0/0xc4) from [<c03ca6c4>] (fimc_m2m_open+0x130/0x1f8)
[<c03ca6c4>] (fimc_m2m_open+0x130/0x1f8) from [<c03a5dd4>] (v4l2_open+0xac/0xe8)
[<c03a5dd4>] (v4l2_open+0xac/0xe8) from [<c0113920>] (chrdev_open+0x9c/0x158)
[<c0113920>] (chrdev_open+0x9c/0x158) from [<c010e488>] (do_dentry_open+0x1f8/0x280)
[<c010e488>] (do_dentry_open+0x1f8/0x280) from [<c010e600>] (finish_open+0x34/0x50)
[<c010e600>] (finish_open+0x34/0x50) from [<c011cc58>] (do_last+0x5bc/0xc00)
[<c011cc58>] (do_last+0x5bc/0xc00) from [<c011d34c>] (path_openat+0xb0/0x484)
[<c011d34c>] (path_openat+0xb0/0x484) from [<c011d824>] (do_filp_open+0x30/0x84)
[<c011d824>] (do_filp_open+0x30/0x84) from [<c010e0f8>] (do_sys_open+0xe8/0x170)
[<c010e0f8>] (do_sys_open+0xe8/0x170) from [<c000f040>] (ret_fast_syscall+0x0/0x30)
[media] exynos4-is: Fix TRY format propagation at MIPI-CSIS subdev
Ensure TRY format is propagated from the sink to source pad.
The format at both pads is always same so the TRY format buffer
for pad 0 is used to hold format for both pads.
While at it remove redundant fmt->pad checking.
[media] exynos4-is: Fix driver name reported in vidioc_querycap
Originally struct v4l2_capability driver and card name was filled
with name of the platform device. After switching to the device tree
the device names have changed and now are 4 different driver names
reported, depending on the video device opened. So instead of e.g.
"exynos4-fimc" there is now one of: 11800000.fimc, 11810000.fimc, 11820000.fimc, 11830000.fimc.
Fix this by using dev->driver_name, rather than platform device name.
A common vidioc_querycap function is created for both M2M and capture
video node.
This fixes any breakage at user space should any application/library
rely on the driver's name.
projects / mirror_ubuntu-disco-kernel.git / log