i915: fix backlight configuration issue
Subsystems (3)
PATCH 1/1 i915: fix backlight configuration issue
The i915 driver can use the backlight subsystem as an option, and usually selects it when CONFIG_ACPI is set. However it is possible to configure a kernel with modular backlight classdev support and a built-in i915 driver, which leads to a linker error:
drivers/gpu/drm/i915/display/intel_panel.o: In function `intel_backlight_device_register':
intel_panel.c:(.text+0x2f58): undefined reference to backlight_device_register' drivers/gpu/drm/i915/display/intel_panel.o: In function intel_backlight_device_unregister':
intel_panel.c:(.text+0x2fe4): undefined reference to `backlight_device_unregister'
Add another Kconfig option to ensure the driver only tries to use the backlight support when it can in fact be linked that way. The new option is on by default to keep the existing behavior.
This is roughly what other drivers like nouveau do as well.
---
I've had this one lying around for a long time, it is still needed but I am not sure which solution is best here. This version is probably the least invasive, but it does not solve the bigger problem around too many 'select' statements in drm
--- drivers/gpu/drm/i915/Kconfig | 11 ++++++++++- drivers/gpu/drm/i915/display/intel_panel.c | 4 ++-- drivers/gpu/drm/i915/display/intel_panel.h | 6 +++--- 3 files changed, 15 insertions(+), 6 deletions(-)
drivers/gpu/drm/i915/Kconfig +10 -1
@@ -16,7 +16,7 @@ config DRM_I915 select IRQ_WORK # i915 depends on ACPI_VIDEO when ACPI is enabled # but for select to work, need to select ACPI_VIDEO's dependencies, ick - select BACKLIGHT_CLASS_DEVICE if ACPI + select DRM_I915_BACKLIGHT if ACPI select INPUT if ACPI select ACPI_VIDEO if ACPI select ACPI_BUTTON if ACPI
@@ -68,6 +68,15 @@ config DRM_I915_FORCE_PROBE Use "*" to force probe the driver for all known devices. +config DRM_I915_BACKLIGHT + tristate "Control backlight support" + depends on DRM_I915 + default DRM_I915 + select BACKLIGHT_CLASS_DEVICE + help + Say Y here if you want to control the backlight of your display + (e.g. a laptop panel). + config DRM_I915_CAPTURE_ERROR bool "Enable capturing GPU state following a hang" depends on DRM_I915
drivers/gpu/drm/i915/display/intel_panel.c +2 -2
@@ -1203,7 +1203,7 @@ void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state,mutex_unlock (&dev_priv->backlight_lock); } -#if IS_ENABLED (CONFIG_BACKLIGHT_CLASS_DEVICE ) +#if IS_ENABLED (CONFIG_DRM_I915_BACKLIGHT )static u32 intel_panel_get_backlight (struct intel_connector *connector) {struct drm_i915_private *dev_priv =to_i915 (connector->base.dev);
@@ -1370,7 +1370,7 @@ void intel_backlight_device_unregister(struct intel_connector *connector) panel->backlight.device = NULL; } } -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#endif /* CONFIG_DRM_I915_BACKLIGHT */ /* * CNP: PWM clock frequency is 19.2 MHz or 24 MHz.
drivers/gpu/drm/i915/display/intel_panel.h +3 -3
@@ -49,10 +49,10 @@ intel_panel_edid_fixed_mode(struct intel_connector *connector);struct drm_display_mode *intel_panel_vbt_fixed_mode (struct intel_connector *connector); -#if IS_ENABLED (CONFIG_BACKLIGHT_CLASS_DEVICE ) +#if IS_ENABLED (CONFIG_DRM_I915_BACKLIGHT )int intel_backlight_device_register (struct intel_connector *connector);void intel_backlight_device_unregister (struct intel_connector *connector); -#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#else /* CONFIG_DRM_I915_BACKLIGHT */ static inline int intel_backlight_device_register (struct intel_connector *connector) {return 0 ;
@@ -60,6 +60,6 @@ static inline int intel_backlight_device_register(struct intel_connector *connecstatic inline void intel_backlight_device_unregister (struct intel_connector *connector) { } -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#endif /* CONFIG_DRM_I915_BACKLIGHT */ #endif /* __INTEL_PANEL_H__ */
Signature · 6 lines
2.20.0 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Patchwork
CI[Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for i915: fix backlight configuration issue
== Series Details ==
== Series Details ==
Series: i915: fix backlight configuration issue
URL : https://patchwork.freedesktop.org/series/71766/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip 3237c6dfef7c i915: fix backlight configuration issue -:12: WARNING:COMMIT_LOG_LONG_LINE: Possible unwrapped commit description (prefer a maximum 75 chars per line) #12: intel_panel.c:(.text+0x2f58): undefined reference to `backlight_device_register'
-:41: WARNING:CONFIG_DESCRIPTION: please write a paragraph that describes the config symbol fully #41: FILE: drivers/gpu/drm/i915/Kconfig:71:
+config DRM_I915_BACKLIGHT
total: 0 errors, 2 warnings, 0 checks, 58 lines checked
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI[Intel-gfx] ✓ Fi.CI.BAT: success for i915: fix backlight configuration issue
In reply to Arnd Bergmann== Series Details ==
== Series Details ==
Series: i915: fix backlight configuration issue
URL : https://patchwork.freedesktop.org/series/71766/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_7703 -> Patchwork_16026
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/index.html
Known issues
------------
Here are the changes found in Patchwork_16026 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@gem_close_race@basic-threads:
- fi-byt-j1900: [PASS][1] -> [TIMEOUT][2] ([i915#816])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-byt-j1900/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-byt-j1900/igt@[email protected]
* igt@i915_module_load@reload-with-fault-injection:
- fi-kbl-x1275: [PASS][3] -> [INCOMPLETE][4] ([i915#879])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-kbl-x1275/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-kbl-x1275/igt@[email protected]
#### Possible fixes ####
* igt@i915_module_load@reload-with-fault-injection:
- fi-skl-6700k2: [INCOMPLETE][5] ([i915#671]) -> [PASS][6]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-skl-6700k2/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-skl-6700k2/igt@[email protected]
* igt@i915_pm_rpm@module-reload:
- fi-skl-6770hq: [FAIL][7] ([i915#178]) -> [PASS][8]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-skl-6770hq/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-skl-6770hq/igt@[email protected]
* igt@i915_selftest@live_execlists:
- fi-kbl-soraka: [DMESG-FAIL][9] ([i915#656]) -> [PASS][10]
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-kbl-soraka/igt@i915_selftest@live_execlists.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-kbl-soraka/igt@i915_selftest@live_execlists.html
* igt@i915_selftest@live_gem_contexts:
- fi-cfl-8700k: [INCOMPLETE][11] ([i915#424]) -> [PASS][12]
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-cfl-8700k/igt@i915_selftest@live_gem_contexts.html
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-cfl-8700k/igt@i915_selftest@live_gem_contexts.html
#### Warnings ####
* igt@i915_selftest@live_blt:
- fi-ivb-3770: [DMESG-FAIL][13] ([i915#725]) -> [DMESG-FAIL][14] ([i915#770])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7703/fi-ivb-3770/igt@i915_selftest@live_blt.html
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/fi-ivb-3770/igt@i915_selftest@live_blt.html
[i915#178]: https://gitlab.freedesktop.org/drm/intel/issues/178
[i915#424]: https://gitlab.freedesktop.org/drm/intel/issues/424
[i915#656]: https://gitlab.freedesktop.org/drm/intel/issues/656
[i915#671]: https://gitlab.freedesktop.org/drm/intel/issues/671
[i915#725]: https://gitlab.freedesktop.org/drm/intel/issues/725
[i915#770]: https://gitlab.freedesktop.org/drm/intel/issues/770
[i915#816]: https://gitlab.freedesktop.org/drm/intel/issues/816
[i915#879]: https://gitlab.freedesktop.org/drm/intel/issues/879
Participating hosts (51 -> 43)
------------------------------
Additional (1): fi-tgl-y
Missing (9): fi-hsw-4200u fi-byt-squawks fi-bsw-cyan fi-kbl-7500u fi-ctg-p8600 fi-skl-lmem fi-blb-e6850 fi-byt-clapper fi-bdw-samus
Build changes
-------------
* CI: CI-20190529 -> None
* Linux: CI_DRM_7703 -> Patchwork_16026
CI-20190529: 20190529
CI_DRM_7703: 17da00c0d18968f69a34991f54804c7eaa8a3312 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_5358: c6fc013f414b806175dc4143c58ab445e5235ea5 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_16026: 3237c6dfef7c2ce97a999c38725250b4dc5f4577 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
3237c6dfef7c i915: fix backlight configuration issue
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_16026/index.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Jani Nikula
maintainerQuoted 14 lines · Arnd Bergmann
The i915 driver can use the backlight subsystem as an option, and usually selects it when
CONFIG_ACPIis set. However it is possible to configure a kernel with modular backlight classdev support and a built-in i915 driver, which leads to a linker error:
drivers/gpu/drm/i915/display/intel_panel.o: In function `intel_backlight_device_register':
intel_panel.c:(.text+0x2f58): undefined reference tobacklight_device_register' drivers/gpu/drm/i915/display/intel_panel.o: In functionintel_backlight_device_unregister':
intel_panel.c:(.text+0x2fe4): undefined reference to `backlight_device_unregister'Add another
Kconfigoption to ensure the driver only tries to use the backlight support when it can in fact be linked that way. The new option is on by default to keep the existing behavior.
Quoted 8 lines
This is roughly what other drivers like nouveau do as well.
Signed-off-by Arnd Bergmann <[email protected]>---
I've had this one lying around for a long time, it is still needed but I am not sure which solution is best here. This version is probably the least invasive, but it does not solve the bigger problem around too many 'select' statements in drm
This is just another hack that's only required because backlight is selected instead of depended on throughout the kernel. (*)
i915 (and most drm drivers, with some variations) could easily handle this with:
depends on (ACPI && ACPI_VIDEO) || ACPI=n depends on BACKLIGHT_CLASS_DEVICE || BACKLIGHT_CLASS_DEVICE=n
Those two lines express the allowed configurations. It's just that we can't do that in i915 *alone*. The combinations of depends and selects lead to impossible configurations. It's all or nothing.
I am not amused by adding more hacks, and I am really *not* interested in adding another useless i915 config option to "solve" this issue.
So thanks, but no thanks. I'm not taking this patch.
BR,
Jani.
(*) The deeper issue is that people as well as the kconfig tools ignore the warnings in Documentation/kbuild/kconfig-language.rst:
select should be used with care. select will force a symbol to a value without visiting the dependencies. By abusing select you are able to select a symbol FOO even if FOO depends on BAR that is not set.
In general use select only for non-visible symbols (no prompts anywhere) and for symbols with no dependencies. That will limit the usefulness but on the other hand avoid the illegal configurations all over.
I don't think we can, uh, fix the people, but it might be possible to warn about selecting visible symbols or symbols with dependencies.
Quoted 82 lines
--- drivers/gpu/drm/i915/Kconfig | 11 ++++++++++- drivers/gpu/drm/i915/display/intel_panel.c | 4 ++-- drivers/gpu/drm/i915/display/intel_panel.h | 6 +++--- 3 files changed, 15 insertions(+), 6 deletions(-)drivers/gpu/drm/i915/Kconfig +10
-1@@ -16,7 +16,7 @@ config DRM_I915 select IRQ_WORK # i915 depends on ACPI_VIDEO when ACPI is enabled # but for select to work, need to select ACPI_VIDEO's dependencies, ick - select BACKLIGHT_CLASS_DEVICE if ACPI + select DRM_I915_BACKLIGHT if ACPI select INPUT if ACPI select ACPI_VIDEO if ACPI select ACPI_BUTTON if ACPI@@ -68,6 +68,15 @@ config DRM_I915_FORCE_PROBE Use "*" to force probe the driver for all known devices. +config DRM_I915_BACKLIGHT + tristate "Control backlight support" + depends on DRM_I915 + default DRM_I915 + select BACKLIGHT_CLASS_DEVICE + help + Say Y here if you want to control the backlight of your display + (e.g. a laptop panel). + config DRM_I915_CAPTURE_ERROR bool "Enable capturing GPU state following a hang" depends on DRM_I915drivers/gpu/drm/i915/display/intel_panel.c +2
-2@@ -1203,7 +1203,7 @@ void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state,mutex_unlock (&dev_priv->backlight_lock); } -#if IS_ENABLED (CONFIG_BACKLIGHT_CLASS_DEVICE ) +#if IS_ENABLED (CONFIG_DRM_I915_BACKLIGHT )static u32 intel_panel_get_backlight (struct intel_connector *connector) {struct drm_i915_private *dev_priv =to_i915 (connector->base.dev);@@ -1370,7 +1370,7 @@ void intel_backlight_device_unregister(struct intel_connector *connector) panel->backlight.device = NULL; } } -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#endif /* CONFIG_DRM_I915_BACKLIGHT */ /* * CNP: PWM clock frequency is 19.2 MHz or 24 MHz. drivers/gpu/drm/i915/display/intel_panel.h +3
-3@@ -49,10 +49,10 @@ intel_panel_edid_fixed_mode(struct intel_connector *connector);struct drm_display_mode *intel_panel_vbt_fixed_mode (struct intel_connector *connector); -#if IS_ENABLED (CONFIG_BACKLIGHT_CLASS_DEVICE ) +#if IS_ENABLED (CONFIG_DRM_I915_BACKLIGHT )int intel_backlight_device_register (struct intel_connector *connector);void intel_backlight_device_unregister (struct intel_connector *connector); -#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#else /* CONFIG_DRM_I915_BACKLIGHT */ static inline int intel_backlight_device_register (struct intel_connector *connector) {return 0 ;@@ -60,6 +60,6 @@ static inline int intel_backlight_device_register(struct intel_connector *connecstatic inline void intel_backlight_device_unregister (struct intel_connector *connector) { } -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#endif /* CONFIG_DRM_I915_BACKLIGHT */ #endif /* __INTEL_PANEL_H__ */
Signature · 1 line
Jani Nikula, Intel Open Source Graphics Center
Quoted 33 lines · Jani Nikula
Quoted 14 lines · Arnd Bergmann
The i915 driver can use the backlight subsystem as an option, and usually selects it when
CONFIG_ACPIis set. However it is possible to configure a kernel with modular backlight classdev support and a built-in i915 driver, which leads to a linker error:
drivers/gpu/drm/i915/display/intel_panel.o: In function `intel_backlight_device_register':
intel_panel.c:(.text+0x2f58): undefined reference tobacklight_device_register' drivers/gpu/drm/i915/display/intel_panel.o: In functionintel_backlight_device_unregister':
intel_panel.c:(.text+0x2fe4): undefined reference to `backlight_device_unregister'Add another
Kconfigoption to ensure the driver only tries to use the backlight support when it can in fact be linked that way. The new option is on by default to keep the existing behavior.Quoted 8 lines
This is roughly what other drivers like nouveau do as well.
Signed-off-by Arnd Bergmann <[email protected]>---
I've had this one lying around for a long time, it is still needed but I am not sure which solution is best here. This version is probably the least invasive, but it does not solve the bigger problem around too many 'select' statements in drmThis is just another hack that's only required because backlight is selected instead of depended on throughout the kernel. (*)
i915 (and most drm drivers, with some variations) could easily handle this with:
depends on (ACPI && ACPI_VIDEO) || ACPI=n depends on BACKLIGHT_CLASS_DEVICE || BACKLIGHT_CLASS_DEVICE=n
Quoted 8 lines
Those two lines express the allowed configurations. It's just that we can't do that in i915 *alone*. The combinations of depends and selects lead to impossible configurations. It's all or nothing.
I am not amused by adding more hacks, and I am really *not* interested in adding another useless i915 config option to "solve" this issue.
So thanks, but no thanks. I'm not taking this patch.
Yeah I'm also leaning towards that the real fix here is to convert backlight over to be a depends on symbol, not a select symbol. It's clearly not a simple stand-alone helper. Or someone makes select recursive and adds a SAT solver to Kconfig :-)
-Daniel
Quoted 106 lines
BR,
Jani.(*) The deeper issue is that people as well as the kconfig tools ignore the warnings in
Documentation/kbuild/kconfig-language.rst:select should be used with care. select will force a symbol to a value without visiting the dependencies. By abusing select you are able to select a symbol FOO even if FOO depends on BAR that is not set.
In general use select only for non-visible symbols (no prompts anywhere) and for symbols with no dependencies. That will limit the usefulness but on the other hand avoid the illegal configurations all over.I don't think we can, uh, fix the people, but it might be possible to warn about selecting visible symbols or symbols with dependencies.
Quoted 82 lines
--- drivers/gpu/drm/i915/Kconfig | 11 ++++++++++- drivers/gpu/drm/i915/display/intel_panel.c | 4 ++-- drivers/gpu/drm/i915/display/intel_panel.h | 6 +++--- 3 files changed, 15 insertions(+), 6 deletions(-)drivers/gpu/drm/i915/Kconfig +10
-1@@ -16,7 +16,7 @@ config DRM_I915 select IRQ_WORK # i915 depends on ACPI_VIDEO when ACPI is enabled # but for select to work, need to select ACPI_VIDEO's dependencies, ick - select BACKLIGHT_CLASS_DEVICE if ACPI + select DRM_I915_BACKLIGHT if ACPI select INPUT if ACPI select ACPI_VIDEO if ACPI select ACPI_BUTTON if ACPI@@ -68,6 +68,15 @@ config DRM_I915_FORCE_PROBE Use "*" to force probe the driver for all known devices. +config DRM_I915_BACKLIGHT + tristate "Control backlight support" + depends on DRM_I915 + default DRM_I915 + select BACKLIGHT_CLASS_DEVICE + help + Say Y here if you want to control the backlight of your display + (e.g. a laptop panel). + config DRM_I915_CAPTURE_ERROR bool "Enable capturing GPU state following a hang" depends on DRM_I915drivers/gpu/drm/i915/display/intel_panel.c +2
-2@@ -1203,7 +1203,7 @@ void intel_panel_enable_backlight(const struct intel_crtc_state *crtc_state,mutex_unlock (&dev_priv->backlight_lock); } -#if IS_ENABLED (CONFIG_BACKLIGHT_CLASS_DEVICE ) +#if IS_ENABLED (CONFIG_DRM_I915_BACKLIGHT )static u32 intel_panel_get_backlight (struct intel_connector *connector) {struct drm_i915_private *dev_priv =to_i915 (connector->base.dev);@@ -1370,7 +1370,7 @@ void intel_backlight_device_unregister(struct intel_connector *connector) panel->backlight.device = NULL; } } -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#endif /* CONFIG_DRM_I915_BACKLIGHT */ /* * CNP: PWM clock frequency is 19.2 MHz or 24 MHz. drivers/gpu/drm/i915/display/intel_panel.h +3
-3@@ -49,10 +49,10 @@ intel_panel_edid_fixed_mode(struct intel_connector *connector);struct drm_display_mode *intel_panel_vbt_fixed_mode (struct intel_connector *connector); -#if IS_ENABLED (CONFIG_BACKLIGHT_CLASS_DEVICE ) +#if IS_ENABLED (CONFIG_DRM_I915_BACKLIGHT )int intel_backlight_device_register (struct intel_connector *connector);void intel_backlight_device_unregister (struct intel_connector *connector); -#else /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#else /* CONFIG_DRM_I915_BACKLIGHT */ static inline int intel_backlight_device_register (struct intel_connector *connector) {return 0 ;@@ -60,6 +60,6 @@ static inline int intel_backlight_device_register(struct intel_connector *connecstatic inline void intel_backlight_device_unregister (struct intel_connector *connector) { } -#endif /* CONFIG_BACKLIGHT_CLASS_DEVICE */ +#endif /* CONFIG_DRM_I915_BACKLIGHT */ #endif /* __INTEL_PANEL_H__ */ Signature · 1 line
Jani Nikula, Intel Open Source Graphics Center
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel