We use some of the lower bits of the retire function pointer for
potential flags, which is quite thorny, since the caller needs to
remember to give the function the correct alignment with
__i915_active_call, otherwise we might incorrectly unpack the pointer
and jump to some garbage address later. Instead of all this let's just
pass the flags along as a separate parameter.
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Suggested-by: Daniel Vetter <daniel@ffwll.ch>
References: ca419f407b ("drm/i915: Fix crash in auto_retire")
References: d8e44e4dd2 ("drm/i915/overlay: Fix active retire callback alignment")
References: fd5f262db1 ("drm/i915/selftests: Fix active retire callback alignment")
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Brost <matthew.brost@intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/20210504164136.96456-1-matthew.auld@intel.com
Static analysis reports this problem
amdgpu_pm.c:478:16: warning: The right operand of '<' is a garbage value
for (i = 0; i < data.nums; i++) {
^ ~~~~~~~~~
In some cases data is not set. Initialize to 0 and flag not setting
data as an error with the existing check.
Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Our driver supports overlay planes, and as expected, some userspace
compositor takes advantage of these features. If the userspace is not
enabling the cursor, they can use multiple planes as they please.
Nevertheless, we start to have constraints when userspace tries to
enable hardware cursor with various planes. Basically, we cannot draw
the cursor at the same size and position on two separated pipes since it
uses extra bandwidth and DML only run with one cursor.
For those reasons, when we enable hardware cursor and multiple planes,
our driver should accept variations like the ones described below:
+-------------+ +--------------+
| +---------+ | | |
| |Primary | | | Primary |
| | | | | Overlay |
| +---------+ | | |
|Overlay | | |
+-------------+ +--------------+
In this scenario, we can have the desktop UI in the overlay and some
other framebuffer attached to the primary plane (e.g., video). However,
userspace needs to obey some rules and avoid scenarios like the ones
described below (when enabling hw cursor):
+--------+
|Overlay |
+-------------+ +-----+-------+ +-| |--+
| +--------+ | +--------+ | | +--------+ |
| |Overlay | | |Overlay | | | |
| | | | | | | | |
| +--------+ | +--------+ | | |
| Primary | | Primary | | Primary |
+-------------+ +-------------+ +-------------+
+-------------+ +-------------+
| +--------+ | Primary |
| |Overlay | | |
| | | | |
| +--------+ | +--------+ |
| Primary | | |Overlay | |
+-------------+ +-| |--+
+--------+
If the userspace violates some of the above scenarios, our driver needs
to reject the commit; otherwise, we can have unexpected behavior. Since
we don't have a proper driver validation for the above case, we can see
some problems like a duplicate cursor in applications that use multiple
planes. This commit fixes the cursor issue and others by adding adequate
verification for multiple planes.
Change since V1 (Harry and Sean):
- Remove cursor verification from the equation.
Cc: Louis Li <Ching-shih.Li@amd.com>
Cc: Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com>
Cc: Harry Wentland <Harry.Wentland@amd.com>
Cc: Hersen Wu <hersenxs.wu@amd.com>
Cc: Sean Paul <seanpaul@chromium.org>
Signed-off-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Pull swiotlb updates from Konrad Rzeszutek Wilk:
"Christoph Hellwig has taken a cleaver and trimmed off the not-needed
code and nicely folded duplicate code in the generic framework.
This lays the groundwork for more work to add extra DMA-backend-ish in
the future. Along with that some bug-fixes to make this a nice working
package"
* 'stable/for-linus-5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb:
swiotlb: don't override user specified size in swiotlb_adjust_size
swiotlb: Fix the type of index
swiotlb: Make SWIOTLB_NO_FORCE perform no allocation
ARM: Qualify enabling of swiotlb_init()
swiotlb: remove swiotlb_nr_tbl
swiotlb: dynamically allocate io_tlb_default_mem
swiotlb: move global variables into a new io_tlb_mem structure
xen-swiotlb: remove the unused size argument from xen_swiotlb_fixup
xen-swiotlb: split xen_swiotlb_init
swiotlb: lift the double initialization protection from xen-swiotlb
xen-swiotlb: remove xen_io_tlb_start and xen_io_tlb_nslabs
xen-swiotlb: remove xen_set_nslabs
xen-swiotlb: use io_tlb_end in xen_swiotlb_dma_supported
xen-swiotlb: use is_swiotlb_buffer in is_xen_swiotlb_buffer
swiotlb: split swiotlb_tbl_sync_single
swiotlb: move orig addr and size validation into swiotlb_bounce
swiotlb: remove the alloc_size parameter to swiotlb_tbl_unmap_single
powerpc/svm: stop using io_tlb_start
I don't believe that it ever makes sense to read the EDID when a panel
is not powered and the powering on of the panel is the job of
prepare(). Let's make sure that this happens before we try to read the
EDID. We use the pm_runtime functions directly rather than directly
calling the normal prepare() function because the pm_runtime functions
are definitely refcounted whereas it's less clear if the prepare() one
is.
NOTE: I'm not 100% sure how EDID reading was working for folks in the
past, but I can only assume that it was failing on the initial attempt
and then working only later. This patch, presumably, will fix that. If
some panel out there really can read the EDID without powering up and
it's a big advantage to preserve the old behavior we can add a
per-panel flag. It appears that providing the DDC bus to the panel in
the past was somewhat uncommon in any case.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20210423095743.v5.17.Ibd31b8f7c73255d68c5c9f5b611b4bfaa036f727@changeid
Let's reorganize how we init and turn on the reference clock in the
code to allow us to turn it on early (even before pre_enable()) so
that we can read the EDID early. This is handy for eDP because:
- We always assume that a panel is there.
- Once we report that a panel is there we get asked to read the EDID.
- Pre-enable isn't called until we know what pixel clock we want to
use and we're ready to turn everything on. That's _after_ we get
asked to read the EDID.
NOTE: the above only works out OK if we "refclk" is provided. Though I
don't have access to any hardware that uses ti-sn65dsi86 and _doesn't_
provide a "refclk", I believe that we'll have trouble reading the EDID
at bootup in that case. Specifically I believe that if there's no
"refclk" we need the MIPI source clock to be active before we can
successfully read the EDID. My evidence here is that, in testing, I
couldn't read the EDID until I turned on the DPPLL in the bridge chip
and that the DPPLL needs the input clock to be active.
Since this is hard to support, let's punt trying to handle this case
if there's no "refclk". In that case we'll enable comms in
pre_enable() like we always did.
I don't believe there are any users of the ti-sn65dsi86 bridge chip
that _don't_ use "refclk". The bridge chip is _very_ inflexible in
that mode. The only time I've seen that mode used was for some really
early prototype hardware that was thrown in the e-waste bin years ago
when we realized how inflexible it was.
Even if someone is using the bridge chip without the "refclk" they're
in no worse shape than they were before the (fairly recent) commit
58074b08c0 ("drm/bridge: ti-sn65dsi86: Read EDID blob over DDC").
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20210423095743.v5.13.Ie8cf556114953c6e7634564cc0d3ddbd103cb96c@changeid
When I added support for the hpd-gpio to simple-panel in commit
48834e6084 ("drm/panel-simple: Support hpd-gpios for delaying
prepare()"), I added a special case to handle a circular dependency I
was running into on the ti-sn65dsi86 bridge chip. On my board the
hpd-gpio is actually provided by the bridge chip. That was causing
some circular dependency problems that I had to work around by getting
the hpd-gpio late.
I've now reorganized the ti-sn65dsi86 bridge chip driver to be a
collection of sub-drivers. Now the GPIO part can probe separately and
that breaks the chain. Let's get rid of the old code to clean things
up.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Sean Paul <seanpaul@chromium.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20210423095743.v5.10.I40eeedc23459d1e3fc96fa6cdad775d88c6e706c@changeid
Let's use the newly minted aux bus to break up the driver into sub
drivers. We're not doing a full breakup here: all the code is still in
the same file and remains largely untouched. The big goal here of
using sub-drivers is to allow part of our code to finish probing even
if some other code needs to defer. This can solve some chicken-and-egg
problems. Specifically:
- In commit 48834e6084 ("drm/panel-simple: Support hpd-gpios for
delaying prepare()") we had to add a bit of a hack to simpel-panel
to support HPD showing up late. We can get rid of that hack now
since the GPIO part of our driver can finish probing early.
- We have a desire to expose our DDC bus to simple-panel (and perhaps
to a backlight driver?). That will end up with the same
chicken-and-egg problem. A future patch to move this to a sub-driver
will fix it.
- If/when we support the PWM functionality present in the bridge chip
for a backlight we'll end up with another chicken-and-egg
problem. If we allow the PWM to be a sub-driver too then it solves
this problem.
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patchwork.freedesktop.org/patch/msgid/20210423095743.v5.9.I3e68fa38c4ccbdbdf145cad2b01e83a1e5eac302@changeid
The current calculation based on the required_dma mask can be significantly
off, so that the linear window only overlaps a small part of the DRAM
address space. This can lead to the command buffer being unmappable, which
is obviously bad.
Rework the linear window offset calculation to be based on the command buffer
physical address, making sure that the command buffer is always mappable.
Tested-by: Primoz Fiser <primoz.fiser@norik.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Fix the following coccicheck report:
drivers/gpu/drm/etnaviv/etnaviv_gpu.c:1775:2-9:
line 1775 is redundant because platform_get_irq() already prints an error
Remove dev_err() messages after platform_get_irq() failures.
Signed-off-by: Tian Tao <tiantao6@hisilicon.com>
Signed-off-by: Zihao Tang <tangzihao1@hisilicon.com>
Signed-off-by: Jay Fang <f.fangjian@huawei.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Fix the following coccicheck warnings:
./drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c:622:2-8: WARNING: NULL
check before some freeing functions is not needed.
./drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c:618:2-8: WARNING: NULL
check before some freeing functions is not needed.
./drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c:616:2-8: WARNING: NULL
check before some freeing functions is not needed.
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: Christian Gmeiner <christian.gmeiner@gmail.com>
Signed-off-by: Lucas Stach <l.stach@pengutronix.de>
Registering multiple backlight devices with intel_backlight name will
obviously fail, regardless of whether they're two connectors in the same
drm device or two different drm devices.
It would be preferrable to switch to completely unique names, and sunset
the generic intel_backlight name. However, there are apparently users
out there that hardcode the name, so the change would break backward
compatibility.
As a compromise, register the first device with intel_backlight name. In
the common case, this is the only backlight device anyway. From the
second device on, use card%d-%s-backlight format, for example
card0-eDP-2-backlight, to make the name unique.
This approach does not preclude us from registering the first device
using the same naming scheme in the future.
v2: Keep using intel_backlight name for first backlight device
Cc: Ville Syrjälä <ville.syrjala@linux.intel.com>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/2794
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/7dc3f6974711ce44522189dc9db05d1e6e24e6d8.1619604743.git.jani.nikula@intel.com
