Maíra Canal [Sun, 12 May 2024 22:23:24 +0000 (19:23 -0300)]
drm/v3d: Add Performance Counters descriptions for V3D 4.2 and 7.1
Add name, category and description for each one of the 93 performance
counters available on V3D.
Note that V3D 4.2 has 87 performance counters, while V3D 7.1 has 93.
Therefore, there are two performance counters arrays. The index of the
performance counter for each V3D version is represented by its position
on the array.
Dmitry Baryshkov [Sat, 11 May 2024 23:00:23 +0000 (02:00 +0300)]
drm/panel: novatek-nt36672e: use wrapped MIPI DCS functions
Remove conditional code and always use mipi_dsi_dcs_*multi() wrappers to
simplify driver's init/exit code. This also includes passing context to
the init_sequence() function instead of passing the DSI device.
Dmitry Baryshkov [Sat, 11 May 2024 23:00:19 +0000 (02:00 +0300)]
drm/mipi-dsi: wrap more functions for streamline handling
Follow the pattern of mipi_dsi_dcs_*_multi() and wrap several existing
MIPI DSI functions to use the context for processing. This simplifies
and streamlines driver code to use simpler code pattern.
Note, msleep function is also wrapped in this way as it is frequently
called inbetween other mipi_dsi_dcs_*() functions.
Thomas Zimmermann [Mon, 13 May 2024 12:51:14 +0000 (14:51 +0200)]
drm/mgag200: Acquire I/O-register lock in DDC code
The modeset lock protects the DDC code from concurrent modeset
operations, which use the same registers. Move that code from the
connector helpers into the DDC helpers .pre_xfer() and .post_xfer().
Both, .pre_xfer() and .post_xfer(), enclose the transfer of data blocks
over the I2C channel in the internal I2C function bit_xfer(). Both
calls are executed unconditionally if present. Invoking DDC transfers
from any where within the driver now takes the lock.
Align the names of the algo-bit helpers with mgag200's convention of
using an mgag200 prefix plus the struct's name plus the callback's name
for such function symbols.
Thomas Zimmermann [Mon, 13 May 2024 12:51:11 +0000 (14:51 +0200)]
drm/mgag200: Replace struct mga_i2c_chan with struct mgag200_ddc
Rename struct mga_i2c_chan to struct mgag200_ddc, define it in the
source file mgag200_i2c.c, and reorder its fields. Rename all related
variables from i2c to ddc. Also rename the i2c adapter accordingly.
Using the term 'ddc' documents the purpose of the code clearly. The
old term 'i2c' could refer to any functionality on an i2c bus. No
functional changes.
Thomas Zimmermann [Mon, 13 May 2024 12:51:10 +0000 (14:51 +0200)]
drm/mgag200: Inline mgag200_i2c_init()
The function mgag200_i2c_init() is an internal helper that sets
up the i2c data structure. Inline its code into the only caller.
Rearrange the individual steps to separate among i2c algorithm,
adapter and fields in struct mga_i2c_chan.
Thomas Zimmermann [Mon, 13 May 2024 12:51:09 +0000 (14:51 +0200)]
drm/mgag200: Allocate instance of struct mga_i2c_chan dynamically
Allocate instances of struct mga_i2c_chan in mgag200_ddc_create()
and return a pointer to the contained i2c adapter. The callers of
the function are now independent from struct mga_i2c_chan.
Thomas Zimmermann [Mon, 13 May 2024 12:51:07 +0000 (14:51 +0200)]
drm/mgag200: Bind I2C lifetime to DRM device
Managed cleanup with devm_add_action_or_reset() will release the I2C
adapter when the underlying Linux device goes away. But the connector
still refers to it, so this cleanup leaves behind a stale pointer
in struct drm_connector.ddc.
Bind the lifetime of the I2C adapter to the connector's lifetime by
using DRM's managed release. When the DRM device goes away (after
the Linux device) DRM will first clean up the connector and then
clean up the I2C adapter.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Reviewed-by: Jocelyn Falempe <jfalempe@redhat.com> Fixes: b279df242972 ("drm/mgag200: Switch I2C code to managed cleanup") Cc: Thomas Zimmermann <tzimmermann@suse.de> Cc: Jocelyn Falempe <jfalempe@redhat.com> Cc: Dave Airlie <airlied@redhat.com> Cc: dri-devel@lists.freedesktop.org Cc: <stable@vger.kernel.org> # v6.0+ Link: https://patchwork.freedesktop.org/patch/msgid/20240513125620.6337-3-tzimmermann@suse.de
Thomas Zimmermann [Mon, 13 May 2024 12:51:06 +0000 (14:51 +0200)]
drm/mgag200: Set DDC timeout in milliseconds
Compute the i2c timeout in jiffies from a value in milliseconds. The
original values of 2 jiffies equals 2 milliseconds if HZ has been
configured to a value of 1000. This corresponds to 2.2 milliseconds
used by most other DRM drivers. Update mgag200 accordingly.
Cong Yang [Thu, 16 May 2024 07:20:39 +0000 (15:20 +0800)]
drm/panel: himax-hx83102: Support for IVO t109nw41 MIPI-DSI panel
The IVO t109nw41 is a 11.0" WUXGA TFT LCD panel, use hx83102 controller
which fits in nicely with the existing panel-himax-hx83102 driver. Hence,
we add a new compatible with panel specific config.
Cong Yang [Thu, 16 May 2024 07:20:37 +0000 (15:20 +0800)]
drm/panel: himax-hx83102: Support for BOE nv110wum-l60 MIPI-DSI panel
The BOE nv110wum-l60 is a 11.0" WUXGA TFT LCD panel, use hx83102 controller
which fits in nicely with the existing panel-himax-hx83102 driver. Hence,
we add a new compatible with panel specific config.
Cong Yang [Thu, 16 May 2024 07:20:35 +0000 (15:20 +0800)]
drm/panel: himax-hx83102: Break out as separate driver
The Starry HX83102 based mipi panel should never have been part of the boe
tv101wum-n16 driver. Discussion with Doug and Linus in V1 [1], we need a
separate driver to enable the hx83102 controller.
In hx83102 driver, add DSI commands as macros. So it can add some panels
with same control model in the future.
In the old boe-tv101wum-nl6 driver inital cmds was invoked at the end of
prepare() function , and call 0x11 and 0x29 at end of inital. For
himax-hx83102 driver, we move 0x11 and 0x29 cmds invoked at prepare()
function.
Note:0x11 is mipi_dsi_dcs_exit_sleep_mode
0x29 is mipi_dsi_dcs_set_display_on
In V1, discussed with Doug and Linus [1], we need break out as separate
driver for the himax83102-j02 controller. Beacuse "starry,himax83102-j02"
and in this series "BOE nv110wum-l60" "IVO t109nw41" panels use same
controller, they have some common CMDS. So add new documentation for
this panels.
For himax83102-j02 controller, no need 3v3 supply, so remove it.
Wachowski, Karol [Wed, 15 May 2024 11:30:05 +0000 (13:30 +0200)]
accel/ivpu: Split IP and buttress code
The NPU device consists of two parts: NPU buttress and NPU IP.
Buttress is a platform specific part that integrates the NPU IP with
the CPU.
NPU IP is the platform agnostic part that does the inference.
This separation enables support for multiple platforms using
a single NPU IP, so for example NPU IP 37XX could be integrated into
MTL and LNL platforms.
Add support for Microtips Technology USA MF-103HIEB0GA0 10.25"[0],
1920x720, 8-bit TFT LCD with LVDS interface. Its a Dual-LVDS Panel and
does not support touch.
Add support for Microtips Technology USA 13-101HIECAF0-C 10.1",
1920x1200, 8-bit TFT LCD with LVDS interface, LED backlight and touch
support (ILITEK 2511).
Aradhya Bhatia [Wed, 15 May 2024 09:51:31 +0000 (15:21 +0530)]
drm/panel: simple: Add Lincoln Tech Sol LCD185-101CT panel
Add support for Lincoln Technology Solutions LCD185-101CT, 10.1",
1920x1200, 8-bit TFT LCD with LVDS interface, LED backlight and PCAP
touch support (Goodix GT928).
Douglas Anderson [Tue, 14 May 2024 17:20:59 +0000 (10:20 -0700)]
drm/panel: innolux-p079zca: Don't use a table for initting panels
Consensus on the mailing lists is that panels shouldn't use a table of
init commands but should instead use init functions. We'll use the
same concepts as the recently introduced
mipi_dsi_generic_write_seq_multi() to make this clean/easy and also
not bloat the driver too much. Measuring before/after this change:
Note that, unlike some other drivers, we actually make this panel
driver _bigger_ by using the new functions. This is because the
innolux-p079zca panel driver didn't have as complex of a table and
thus the old table was more efficient than the code. The bloat is
still not giant (only 412 bytes).
Also note that we can't direclty use
mipi_dsi_generic_write_seq_multi() here because we need to deal with
the crazy "nop" that this driver sends after all commands. This means
that we have to write code that is "inspired" by the new macros.
Since we're touching all the tables, let's also convert hex numbers to
lower case as per kernel conventions.
Douglas Anderson [Tue, 14 May 2024 17:20:58 +0000 (10:20 -0700)]
drm/panel: ili9882t: Don't use a table for initting panels
Consensus on the mailing lists is that panels shouldn't use a table of
init commands but should instead use init functions. With the recently
introduced mipi_dsi_dcs_write_seq_multi() this is not only clean/easy
but also saves space. Measuring before/after this change:
Douglas Anderson [Tue, 14 May 2024 17:20:57 +0000 (10:20 -0700)]
drm/panel: boe-tv101wum-nl6: Don't use a table for initting panels
Consensus on the mailing lists is that panels shouldn't use a table of
init commands but should instead use init functions. With the recently
introduced mipi_dsi_dcs_write_seq_multi() this is not only clean/easy
but also saves space. Measuring before/after this change:
Douglas Anderson [Tue, 14 May 2024 17:20:56 +0000 (10:20 -0700)]
drm/panel: novatek-nt36672e: Switch to mipi_dsi_dcs_write_seq_multi()
This is a mechanical conversion of the novatek-nt36672e driver to use
the new mipi_dsi_dcs_write_seq_multi(). The new function is easier for
clients to understand and using it also causes smaller code to be
generated. Specifically:
$ scripts/bloat-o-meter \
...after/panel-novatek-nt36672e.ko \
...ctx/panel-novatek-nt36672e.ko
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-988 (-988)
Function old new delta
nt36672e_1080x2408_60hz_init 6236 5248 -988
Total: Before=10651, After=9663, chg -9.28%
The current mipi_dsi_*_write_seq() macros are non-intutitive because
they contain a hidden "return" statement that will return out of the
_caller_ of the macro. Let's mark them as deprecated and instead
introduce some new macros that are more intuitive.
These new macros are less optimal when an error occurs but should
behave more optimally when there is no error. Specifically these new
macros cause smaller code to get generated and the code size savings
(less to fetch from RAM, less cache space used, less RAM used) are
important. Since the error case isn't something we need to optimize
for and these new macros are easier to understand and more flexible,
they should be used.
After converting to use these new functions, one example shows some
nice savings while also being easier to understand.
$ scripts/bloat-o-meter \
...after/panel-novatek-nt36672e.ko \
...ctx/panel-novatek-nt36672e.ko
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-988 (-988)
Function old new delta
nt36672e_1080x2408_60hz_init 6236 5248 -988
Total: Before=10651, After=9663, chg -9.28%
Douglas Anderson [Tue, 14 May 2024 17:20:54 +0000 (10:20 -0700)]
drm/mipi-dsi: Reduce driver bloat of mipi_dsi_*_write_seq()
Through a cooperative effort between Hsin-Yi Wang and Dmitry
Baryshkov, we have realized the dev_err() in the
mipi_dsi_*_write_seq() macros was causing quite a bit of bloat to the
kernel. Let's hoist this call into drm_mipi_dsi.c by adding a "chatty"
version of the functions that includes the print. While doing this,
add a bit more comments to these macros making it clear that they
print errors and also that they return out of _the caller's_ function.
Without any changes to clients this gives a nice savings. Specifically
the macro was inlined and thus the error report call was inlined into
every call to mipi_dsi_dcs_write_seq() and
mipi_dsi_generic_write_seq(). By using a call to a "chatty" function,
the usage is reduced to one call in the chatty function and a function
call at the invoking site.
Building with my build system shows one example:
$ scripts/bloat-o-meter \
.../before/panel-novatek-nt36672e.ko \
.../after/panel-novatek-nt36672e.ko
add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-4404 (-4404)
Function old new delta
nt36672e_1080x2408_60hz_init 10640 6236 -4404
Total: Before=15055, After=10651, chg -29.25%
Note that given the change in location of the print it's harder to
include the "cmd" in the printout for mipi_dsi_dcs_write_seq() since,
theoretically, someone could call the new chatty function with a
zero-size array and it would be illegal to dereference data[0].
There's a printk format to print the whole buffer and this is probably
more useful for debugging anyway. Given that we're doing this for
mipi_dsi_dcs_write_seq(), let's also print the buffer for
mipi_dsi_generic_write_seq() in the error case.
It should be noted that the current consensus of DRM folks is that the
mipi_dsi_*_write_seq() should be deprecated due to the non-intuitive
return behavior. A future patch will formally mark them as deprecated
and provide an alternative.
Douglas Anderson [Tue, 14 May 2024 17:20:52 +0000 (10:20 -0700)]
drm/mipi-dsi: Fix theoretical int overflow in mipi_dsi_generic_write_seq()
The mipi_dsi_generic_write_seq() macro makes a call to
mipi_dsi_generic_write() which returns a type ssize_t. The macro then
stores it in an int and checks to see if it's negative. This could
theoretically be a problem if "ssize_t" is larger than "int".
To see the issue, imagine that "ssize_t" is 32-bits and "int" is
16-bits, you could see a problem if there was some code out there that
looked like:
mipi_dsi_generic_write_seq(dsi, <32768 bytes as arguments>);
...since we'd get back that 32768 bytes were transferred and 32768
stored in a 16-bit int would look negative.
Though there are no callsites where we'd actually hit this (even if
"int" was only 16-bit), it's cleaner to make the types match so let's
fix it.
Douglas Anderson [Tue, 14 May 2024 17:20:51 +0000 (10:20 -0700)]
drm/mipi-dsi: Fix theoretical int overflow in mipi_dsi_dcs_write_seq()
The mipi_dsi_dcs_write_seq() macro makes a call to
mipi_dsi_dcs_write_buffer() which returns a type ssize_t. The macro
then stores it in an int and checks to see if it's negative. This
could theoretically be a problem if "ssize_t" is larger than "int".
To see the issue, imagine that "ssize_t" is 32-bits and "int" is
16-bits, you could see a problem if there was some code out there that
looked like:
mipi_dsi_dcs_write_seq(dsi, cmd, <32767 bytes as arguments>);
...since we'd get back that 32768 bytes were transferred and 32768
stored in a 16-bit int would look negative.
Though there are no callsites where we'd actually hit this (even if
"int" was only 16-bit), it's cleaner to make the types match so let's
fix it.
Tomasz Rusinowicz [Mon, 13 May 2024 12:04:31 +0000 (14:04 +0200)]
accel/ivpu: Share NPU busy time in sysfs
The driver tracks the time spent by NPU executing jobs
and shares it through sysfs `npu_busy_time_us` file.
It can be then used by user space applications to monitor device
utilization.
NPU is considered 'busy' starting with a first job submitted
to firmware and ending when there is no more jobs pending/executing.
Jacek Lawrynowicz [Mon, 13 May 2024 12:04:30 +0000 (14:04 +0200)]
accel/ivpu: Increase reset counter when warm boot fails
Failed warm boot causes a cold boot that looses FW state and is
equivalent to a recovery or reset, so reset_counter should be
incremented in order for this failure to be detected by tests.
Wachowski, Karol [Mon, 13 May 2024 12:04:28 +0000 (14:04 +0200)]
accel/ivpu: Add force snoop module parameter
Add module parameter that enforces snooping for all NPU accesses,
both through MMU PTEs mappings and through TCU page table walk
override register bits for MMU page walks / configuration access.
Wachowski, Karol [Mon, 13 May 2024 12:04:26 +0000 (14:04 +0200)]
accel/ivpu: Add resume engine support
Create debugfs interface that triggers sending resume engine IPC
command to VPU. It is used to test engine resume functionality in
driver user space tests.
Wachowski, Karol [Mon, 13 May 2024 12:04:22 +0000 (14:04 +0200)]
accel/ivpu: Create priority based command queues
Create multiple command queues per engine with different priorities.
The cmdqs are created on-demand and they support 4 priority levels.
These priorities will later be used by the HWS (hardware scheduler).
Wachowski, Karol [Mon, 13 May 2024 12:04:21 +0000 (14:04 +0200)]
accel/ivpu: Add sched_mode module param
This param will be used to enable/disable HWS (hardware scheduler).
The HWS is a FW side feature and may not be available on all
HW generations and FW versions.
Sui Jingfeng [Mon, 13 May 2024 15:31:09 +0000 (23:31 +0800)]
drm/bridge: analogix: Remove redundant checks on existence of bridge->encoder
The checks on the existence of bridge->encoder in the implementation of
drm_bridge_funcs::attach() is not necessary, as it has already been checked
in the drm_bridge_attach() function call by previous bridge or KMS driver.
The drm_bridge_attach() will quit with a negative error code returned if
it fails for some reasons, hence, it is guaranteed that the .encoder member
of the drm_bridge instance is not NULL when various bridge attach functions
are called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:08 +0000 (23:31 +0800)]
drm/bridge: imx: Remove redundant checks on existence of bridge->encoder
The checks on the existence of bridge->encoder in the implementation of
drm_bridge_funcs::attach() is not necessary, as it has already been checked
in the drm_bridge_attach() function call by previous bridge or KMS driver.
The drm_bridge_attach() will quit with a negative error code returned if
it fails for some reasons, hence, it is guaranteed that the .encoder member
of the drm_bridge instance is not NULL when various i.MX specific bridge
attach functions are called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:07 +0000 (23:31 +0800)]
drm/bridge: lt9611uxc: Remove a redundant check on existence of bridge->encoder
In the lt9611uxc_connector_init() function, the check on the existence
of bridge->encoder is not necessary, as it has already been checked in
the drm_bridge_attach() function. And the check on the drm bridge core
happens before check in the implementation. Hence, it is guaranteed that
the .encoder member of the struct drm_bridge is not NULL when
lt9611uxc_connector_init() function get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:06 +0000 (23:31 +0800)]
drm/bridge: synopsys: dw-mipi-dsi: Remove a redundant check on existence of bridge->encoder
In the dw_mipi_dsi_bridge_attach() function, the check on the existence
of bridge->encoder is not necessary, as it has already been checked in
the drm_bridge_attach() function invocked by previous bridge or KMS driver.
The previous drm_bridge_attach() will quit with a negative error code
returned if it fails for some reasons, hence, it is guaranteed that the
.encoder member of the struct drm_bridge is not NULL when
dw_mipi_dsi_bridge_attach() function gets called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:05 +0000 (23:31 +0800)]
drm/bridge: megachips-stdpxxxx-ge-b850v3-fw: Remove a redundant check on existence of bridge->encoder
In the ge_b850v3_lvds_create_connector function, the check on the existence
of bridge->encoder is not necessary, as it has already been checked in the
drm_bridge_attach() function called by upstream bridge or driver. Hence, it
is guaranteed that the .encoder member of the drm_bridge instance is not
NULL when cdns_mhdp_connector_init() function get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:04 +0000 (23:31 +0800)]
drm/bridge: cdns-mhdp8546: Remove a redundant check on existence of bridge->encoder
In the cdns_mhdp_connector_init() function, the check on the existence
of bridge->encoder is not necessary, as it has already been checked in
the drm_bridge_attach() function. As the cdns_mhdp_connector_init() is
only called by cdns_mhdp_attach(), it is guaranteed that the .encoder
member of the struct drm_bridge is not NULL when cdns_mhdp_attach() gets
called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:03 +0000 (23:31 +0800)]
drm/bridge: adv7511: Remove a redundant check on existence of bridge->encoder
In the adv7511_connector_init() function, the check on the existence of
bridge->encoder is not necessary. As it has already been checked in the
drm_bridge_attach() which happens prior to the adv7511_bridge_attach()
get called. Also note that the adv7511_connector_init() is only called by
adv7511_bridge_attach(). Hence, it is guaranteed that the .encoder member
of the drm_bridge instance is not NULL when adv7511_connector_init() get
called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:02 +0000 (23:31 +0800)]
drm/bridge: it6505: Remove a redundant check on existence of bridge->encoder
In it6505_bridge_attach(), the check on the existence of 'bridge->encoder'
is not necessary, as it has already been checked in the drm_bridge_attach()
which happens prior to it6505_bridge_attach() get called. Note that the
it6505_bridge_attach() will only be called by .attach() of the previous
bridge or KMS driver. The previous drm_bridge_attach() will quit with a
negative error code returned if it fails for some reasons. Hence, it is
guaranteed that the .encoder member of the drm_bridge instance is not NULL
when it6505_bridge_attach() function get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:01 +0000 (23:31 +0800)]
drm/bridge: panel: Remove a redundant check on existence of bridge->encoder
Because the existence of 'bridge->encoder' has already been checked before
the panel_bridge_attach() function get called, and the drm_bridge_attach()
will quit with a negative error code returned if it fails for some reasons.
Hence, it is guaranteed that the .encoder member of the drm_bridge instance
is not NULL when panel_bridge_attach() function get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:31:00 +0000 (23:31 +0800)]
drm/bridge: nxp-ptn3460: Remove a redundant check on existence of bridge->encoder
Because the existence of 'bridge->encoder' has already been checked before
the ptn3460_bridge_attach() function get called, and drm_bridge_attach()
will quit with a negative error code returned if it fails for some reasons.
Hence, it is guaranteed that the .encoder member of the drm_bridge instance
is not NULL when the ptn3460_bridge_attach() function get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:30:59 +0000 (23:30 +0800)]
drm/bridge: tfp410: Remove a redundant check on existence of bridge->encoder
Because the existence of bridge->encoder has already been checked before
the simple_bridge_attach() function get called, And drm_bridge_attach()
will quit with a negative error code returned if it fails for some reasons.
Hence, it is guaranteed that the .encoder member of the drm_bridge instance
is not NULL when the tfp410_attach() function get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Sui Jingfeng [Mon, 13 May 2024 15:30:58 +0000 (23:30 +0800)]
drm/bridge: simple-bridge: Remove a redundant check on existence of bridge->encoder
Because the existence of 'bridge->encoder' has already been checked before
the simple_bridge_attach() function get called, and drm_bridge_attach()
will quit with a negative error code returned if it fails for some reasons.
Hence, it is guaranteed that the .encoder member of the drm_bridge instance
is not NULL when the simple_bridge_attach() get called.
Remove the redundant checking codes "if (!bridge->encoder) { ... }".
Thomas Zimmermann [Fri, 10 May 2024 15:47:12 +0000 (17:47 +0200)]
drm/udl: Remove struct udl_connector
Udl's struct udl_connector is an empty wrapper around struct
drm_connector. Remove it. Allocate the connector as part of struct
udl_device and inline the init function into its only caller.
v2:
- fix return value in udl_modeset_init() (Dan)
Thomas Zimmermann [Fri, 10 May 2024 15:47:11 +0000 (17:47 +0200)]
drm/udl: Untangle .get_modes() and .detect_ctx()
Provide separate implementations of .get_modes() and .detect_ctx()
from struct drm_connector. Switch to struct drm_edid.
Udl's .detect() helper used to fetch the EDID from the adapter and the
.get_modes() helper provided display modes from the data. But this
relied on the DRM helpers to call the functions in the correct order.
When no EDID could be retrieved, .detect() regularly printed a warning
to the kernel log.
Switching to the new helpers around struct drm_edid separates both from
each other. The .get_modes() helper now fetches the EDID by itself and
the .detect_ctx() helper only tests for its presence. The patch does a
number of things to implement this.
- Move udl_get_edid_block() to udl_edid.c and rename it to
udl_read_edid_block(). Then use the helper to implement probing in
udl_probe_edid() and reading in udl_edid_read(). The latter helper
is build on top of DRM helpers.
- Replace the existing code in .get_modes() and .detect() with udl's
new EDID helpers. The new code behaves like DRM's similar DDC-based
helpers. Instead of .detect(), udl now implements .detect_ctx().
- Remove the edid data from struct udl_connector. The field cached
the EDID data between calls to .detect() and .get_modes(), but is now
unused.
v3:
- implement udl_probe_edid() with memchr_inv() (Jani)
v2:
- implement udl_probe_edid() within udl
- reword commit description
Thomas Zimmermann [Fri, 10 May 2024 15:47:09 +0000 (17:47 +0200)]
drm/udl: Move drm_dev_{enter, exit}() into udl_get_edid_block()
Protect the code in udl_get_edid_block() with drm_dev_enter() and
drm_dev_exit(), so that all callers automatically invoke it. The
function uses hardware resources, which can be hot-unplugged at
any time. The other code in udl_connector_detect() does not use the
resources of the hardware device and therefore does not require
protection.
This change will allow to use udl_get_edid_block() in various
contexts easily.
Thomas Zimmermann [Fri, 10 May 2024 15:47:08 +0000 (17:47 +0200)]
drm/udl: Remove DRM_CONNECTOR_POLL_HPD
DisplayLink devices do not generate hotplug events. Remove the poll
flag DRM_CONNECTOR_POLL_HPD, as it may not be specified together with
DRM_CONNECTOR_POLL_CONNECT or DRM_CONNECTOR_POLL_DISCONNECT.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de> Fixes: afdfc4c6f55f ("drm/udl: Fixed problem with UDL adpater reconnection") Reviewed-by: Jani Nikula <jani.nikula@intel.com> Cc: Robert Tarasov <tutankhamen@chromium.org> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Dave Airlie <airlied@redhat.com> Cc: Sean Paul <sean@poorly.run> Cc: Thomas Zimmermann <tzimmermann@suse.de> Cc: dri-devel@lists.freedesktop.org Cc: <stable@vger.kernel.org> # v4.15+ Link: https://patchwork.freedesktop.org/patch/msgid/20240510154841.11370-2-tzimmermann@suse.de
Jani Nikula [Fri, 10 May 2024 09:09:51 +0000 (12:09 +0300)]
drm: use "0" instead of "" for deprecated driver date
libdrm does not like the empty string for driver date. Use "0" instead,
which has been used by virtio previously.
Reported-by: Steven Price <steven.price@arm.com> Closes: https://lore.kernel.org/r/9d0cff47-308e-4b11-a9f3-4157dc26b6fa@arm.com Fixes: 7fb8af6798e8 ("drm: deprecate driver date") Signed-off-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240510090951.3398882-1-jani.nikula@intel.com Reviewed-by: Javier Martinez Canillas <javierm@redhat.com> Tested-by: Steven Price <steven.price@arm.com> Reviewed-by: Simon Ser <contact@emersion.fr>
Amjad Ouled-Ameur [Mon, 19 Feb 2024 10:09:15 +0000 (15:39 +0530)]
drm/arm/komeda: update DMA mask to 40 bits
Each layer in the DPU has a 40-bit base address register, which indicates
start of frame buffer data for that layer. Komeda driver does not set
its DMA mask, which makes it 32-bit by default which does not use
the entire available possible supported by the DPU.
Update the DMA mask to align with DPU Architecture v1.0 spec.
Faiz Abbas [Mon, 19 Feb 2024 10:09:14 +0000 (15:39 +0530)]
drm/arm/komeda: Move pipeline prints to after the entire pipeline has been enabled
The komeda driver prints a pretty verbose log in komeda_pipeline_dump()
detailing the components of each of the two pipelines. This gets printed
multiple times during boot as komeda EPROBE_DEFERs waiting for the
remote bridge drivers to come up. Move this log to after this has
happened indicating that the printed pipeline is actually completely up.
Faiz Abbas [Mon, 19 Feb 2024 10:09:13 +0000 (15:39 +0530)]
drm/arm/komeda: Fix komeda probe failing if there are no links in the secondary pipeline
Since commit 4cfe5cc02e3f ("drm/arm/komeda: Remove component framework and
add a simple encoder"), the devm_drm_of_get_bridge() call happens
regardless of whether any remote nodes are available on the pipeline. Fix
this by moving the bridge attach to its own function and calling it
conditional on there being an output link.
Fixes: 4cfe5cc02e3f ("drm/arm/komeda: Remove component framework and add a simple encoder") Signed-off-by: Faiz Abbas <faiz.abbas@arm.com>
[Corrected Commit-id of the fixed patch to match mainline] Signed-off-by: Liviu Dudau <liviu.dudau@arm.com> Acked-by: Liviu Dudau <liviu.dudau@arm.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240219100915.192475-2-faiz.abbas@arm.com
Wolfram Sang [Tue, 7 May 2024 09:02:01 +0000 (11:02 +0200)]
drm/arm/komeda: don't use confusing 'timeout' variable name
There is a confusing pattern in the kernel to use a variable named 'timeout' to
store the result of wait_for_completion_timeout() causing patterns like:
timeout = wait_for_completion_timeout(...)
if (!timeout) return -ETIMEDOUT;
with all kinds of permutations. Check the return value directly to drop
'timeout' which also fixes its wrong type.
Ville Syrjälä [Mon, 8 Apr 2024 17:04:19 +0000 (20:04 +0300)]
drm/omap: Open code phys_to_page()
phys_to_page() is not available on most architectures.
Just open code it like msm does. Allows COMPILE_TEST=y
builds of omapdrm on other architectures.
Ville Syrjälä [Mon, 8 Apr 2024 17:04:16 +0000 (20:04 +0300)]
drm/hisilicon/kirin: Allow build with COMPILE_TEST=y
Allow kirin to be built with COMPILE_TEST=y for greater
coverage. Builds fine on x86/x86_64 at least.
Cc: Xinliang Liu <xinliang.liu@linaro.org> Cc: Tian Tao <tiantao6@hisilicon.com> Cc: Xinwei Kong <kong.kongxinwei@hisilicon.com> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Yongqin Liu <yongqin.liu@linaro.org> Cc: John Stultz <jstultz@google.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240408170426.9285-12-ville.syrjala@linux.intel.com Acked-by: John Stultz <jstultz@google.com>
Ville Syrjälä [Mon, 8 Apr 2024 17:04:15 +0000 (20:04 +0300)]
drm/hisilicon/kirin: Fix MASK(32) on 32bit architectures
BIT(32) is illegal when sizeof(long)==4. Use BIT_ULL(32)
instead.
Cc: Xinliang Liu <xinliang.liu@linaro.org> Cc: Tian Tao <tiantao6@hisilicon.com> Cc: Xinwei Kong <kong.kongxinwei@hisilicon.com> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Yongqin Liu <yongqin.liu@linaro.org> Cc: John Stultz <jstultz@google.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240408170426.9285-11-ville.syrjala@linux.intel.com Acked-by: John Stultz <jstultz@google.com>
Ville Syrjälä [Mon, 8 Apr 2024 17:04:14 +0000 (20:04 +0300)]
drm/hisilicon/kirin: Fix 64bit divisions
Use the appropriate 64bit division helpers to make the code
build on 32bit architectures.
Cc: Xinliang Liu <xinliang.liu@linaro.org> Cc: Tian Tao <tiantao6@hisilicon.com> Cc: Xinwei Kong <kong.kongxinwei@hisilicon.com> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Yongqin Liu <yongqin.liu@linaro.org> Cc: John Stultz <jstultz@google.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240408170426.9285-10-ville.syrjala@linux.intel.com Acked-by: John Stultz <jstultz@google.com>
Ville Syrjälä [Mon, 8 Apr 2024 17:04:13 +0000 (20:04 +0300)]
drm/hisilicon/kirin: Include linux/io.h for readl()/writel()
Include linux/io.h for readl()/writel().
When built on x86_64 w/ COMPILE_TEST=y:
../drivers/gpu/drm/hisilicon/kirin/dw_dsi_reg.h:93:16: error: implicit declaration of function ‘readl’ [-Werror=implicit-function-declaration]
93 | orig = readl(addr);
| ^~~~~
../drivers/gpu/drm/hisilicon/kirin/dw_dsi_reg.h:96:9: error: implicit declaration of function ‘writel’ [-Werror=implicit-function-declaration]
96 | writel(tmp, addr);
| ^~~~~~
Cc: Xinliang Liu <xinliang.liu@linaro.org> Cc: Tian Tao <tiantao6@hisilicon.com> Cc: Xinwei Kong <kong.kongxinwei@hisilicon.com> Cc: Sumit Semwal <sumit.semwal@linaro.org> Cc: Yongqin Liu <yongqin.liu@linaro.org> Cc: John Stultz <jstultz@google.com> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240408170426.9285-9-ville.syrjala@linux.intel.com Acked-by: John Stultz <jstultz@google.com>
Sui Jingfeng [Tue, 7 May 2024 18:00:00 +0000 (02:00 +0800)]
drm/drm-bridge: Drop conditionals around of_node pointers
Having conditional around the of_node pointer of the drm_bridge structure
is not necessary, since drm_bridge structure always has the of_node as its
member.
Let's drop the conditional to get a better looks, please also note that
this is following the already accepted commitments. see commit d8dfccde2709
("drm/bridge: Drop conditionals around of_node pointers") for reference.
Jani Nikula [Mon, 29 Apr 2024 16:43:36 +0000 (19:43 +0300)]
drm: deprecate driver date
The driver date serves no useful purpose, because it's hardly ever
updated. The information is misleading at best.
As described in Documentation/gpu/drm-internals.rst:
The driver date, formatted as YYYYMMDD, is meant to identify the date
of the latest modification to the driver. However, as most drivers
fail to update it, its value is mostly useless. The DRM core prints it
to the kernel log at initialization time and passes it to userspace
through the DRM_IOCTL_VERSION ioctl.
Stop printing the driver date at init, and start returning the empty
string "" as driver date through the DRM_IOCTL_VERSION ioctl.
The driver date initialization in drivers and the struct drm_driver date
member can be removed in follow-up.
Ville Syrjälä [Mon, 22 Apr 2024 08:58:57 +0000 (11:58 +0300)]
drm/uapi: Move drm_color_ctm_3x4 out from drm_mode.h
drm_color_ctm_3x4 is some undocumented amgdpu private
uapi and thus has no business being in drm_mode.h.
At least move it to some amdgpu specific header, albeit
with the wrong namespace as maybe something somewhere
is using this already?
Cc: Harry Wentland <harry.wentland@amd.com> Cc: Joshua Ashton <joshua@froggi.es> Cc: Alex Deucher <alexander.deucher@amd.com> Fixes: 6872a189be50 ("drm/amd/display: Add 3x4 CTM support for plane CTM") Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240422085857.17651-1-ville.syrjala@linux.intel.com Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Douglas Anderson [Thu, 2 May 2024 23:47:47 +0000 (16:47 -0700)]
drm/panel-edp: Add ID for KD KD116N09-30NH-A016
As evidenced by in-field reports, this panel shipped on pompom but we
never added the ID and thus we're stuck w/ conservative timings. The
panel was part of early patches but somehow got left off in the
end. :( Add it in now.
projects / users / jedix / linux-maple.git / log