Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1360956
| From | Alexey Brodkin <Alexey.Brodkin@synopsys.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH 1/3] drm: introduce drm_connector_plug_all() helper |
| Date | 2016-03-18 23:00 +0100 |
| Message-ID | <reaB4-1dO-11@gated-at.bofh.it> (permalink) |
| References | <rdZvX-4at-3@gated-at.bofh.it> <rdZvZ-4at-25@gated-at.bofh.it> <re70u-49j-11@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
Hi Daniel, On Fri, 2016-03-18 at 19:06 +-0100, Daniel Vetter wrote: +AD4- On Fri, Mar 18, 2016 at 01:01:42PM +-0300, Alexey Brodkin wrote: +AD4- +AD4- +AD4- +AD4- As a pair to already existing drm+AF8-connector+AF8-unplug+AF8-all() we're adding +AD4- +AD4- generic implementation of what is already done in some drivers. +AD4- +AD4- +AD4- +AD4- Once this helper is implemented we'll be ready to switch existing +AD4- +AD4- driver-specific implementations with generic one. +AD4- +AD4- +AD4- +AD4- Signed-off-by: Alexey Brodkin +ADw-abrodkin+AEA-synopsys.com+AD4- +AD4- +AD4- Cc: Daniel Vetter +ADw-daniel+AEA-ffwll.ch+AD4- +AD4- +AD4- Cc: David Airlie +ADw-airlied+AEA-linux.ie+AD4- +AD4- +AD4- --- +AD4- +AD4- +AKA-drivers/gpu/drm/drm+AF8-crtc.c +AHw- 44 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-- +AD4- +AD4- +AKA-drivers/gpu/drm/drm+AF8-drv.c+AKAAoAB8AKAAoA-3 +-+-- +AD4- +AD4- +AKA-include/drm/drm+AF8-crtc.h+AKAAoACgAKAAoAB8AKAAoA-3 +-+-- +AD4- +AD4- +AKA-3 files changed, 47 insertions(+-), 3 deletions(-) +AD4- +AD4- +AD4- +AD4- diff --git a/drivers/gpu/drm/drm+AF8-crtc.c b/drivers/gpu/drm/drm+AF8-crtc.c +AD4- +AD4- index 65258ac..ce27420 100644 +AD4- +AD4- --- a/drivers/gpu/drm/drm+AF8-crtc.c +AD4- +AD4- +-+-+- b/drivers/gpu/drm/drm+AF8-crtc.c +AD4- +AD4- +AEAAQA- -1080,6 +-1080,46 +AEAAQA- void drm+AF8-connector+AF8-unregister(struct drm+AF8-connector +ACo-connector) +AD4- +AD4- +AKAAfQ- +AD4- +AD4- +AKA-EXPORT+AF8-SYMBOL(drm+AF8-connector+AF8-unregister)+ADs- +AD4- +AD4- +AKA- +AD4- +AD4- +-/+ACoAKg- +AD4- +AD4- +- +ACo- drm+AF8-connector+AF8-plug+AF8-all - register connector userspace interfaces +AD4- +AD4- +- +ACo- +AEA-dev: drm device +AD4- +AD4- +- +ACo- +AD4- +AD4- +- +ACo- This function registers all connector userspace interfaces in sysfs. Should +AD4- +AD4- +- +ACo- be call when the device is disconnected, e.g. from an usb driver's +AD4- Still talks about disconnect +ADs--) Please also mention that this just calls +AD4- drm+AF8-connector+AF8-register() exactly like this including () to generate a +AD4- kerneldoc hyperlink. Well I intentionally left in description of drm+AF8-connector+AF8-register+AF8-all(): +ACI-Should be call when the device is disconnected, e.g. from an usb driver's +AKA--+AD4-connect callback.+ACI- I did mean it. Or is this statement is incorrect and example of invocation of drm+AF8-connector+AF8-register+AF8-all() should be different? Which one works better then? +AD4- +AD4- +AD4- +AD4- +- +ACo- -+AD4-connect callback. +AD4- Returns: section is missing, specifying how this can fail. Just copy the +AD4- one from connector+AF8-register(). Yeah, correct. Blind copy-paste doesn't work equally good always :( +AD4- +AD4- +AD4- +AD4- +- +ACo-/ +AD4- +AD4- +-int drm+AF8-connector+AF8-plug+AF8-all(struct drm+AF8-device +ACo-dev) +AD4- +AD4- +-+AHs- +AD4- +AD4- +- struct drm+AF8-connector +ACo-connector, +ACo-failed+ADs- +AD4- +AD4- +- int ret+ADs- +AD4- +AD4- +- +AD4- +AD4- +- mutex+AF8-lock(+ACY-dev-+AD4-mode+AF8-config.mutex)+ADs- +AD4- +AD4- +- +AD4- +AD4- +- list+AF8-for+AF8-each+AF8-entry(connector, +ACY-dev-+AD4-mode+AF8-config.connector+AF8-list, head) +AHs- +AD4- for+AF8-each+AF8-connector here please. And the s/plug/register/ naming discussion +AD4- we've had. Ok. +AD4- +AD4- +AD4- +AD4- +- ret +AD0- drm+AF8-connector+AF8-register(connector)+ADs- +AD4- +AD4- +- if (ret) +AHs- +AD4- +AD4- +- failed +AD0- connector+ADs- +AD4- +AD4- +- goto err+ADs- +AD4- +AD4- +- +AH0- +AD4- +AD4- +- +AH0- +AD4- +AD4- +- +AD4- +AD4- +- mutex+AF8-unlock(+ACY-dev-+AD4-mode+AF8-config.mutex)+ADs- +AD4- +AD4- +- +AD4- +AD4- +- return 0+ADs- +AD4- +AD4- +- +AD4- +AD4- +-err: +AD4- +AD4- +- list+AF8-for+AF8-each+AF8-entry(connector, +ACY-dev-+AD4-mode+AF8-config.connector+AF8-list, head) +AHs- +AD4- +AD4- +- if (failed +AD0APQ- connector) +AD4- +AD4- +- break+ADs- +AD4- +AD4- +- +AD4- +AD4- +- drm+AF8-connector+AF8-unregister(connector)+ADs- +AD4- +AD4- +- +AH0- +AD4- +AD4- +- +AD4- +AD4- +- mutex+AF8-unlock(+ACY-dev-+AD4-mode+AF8-config.mutex)+ADs- +AD4- +AD4- +- +AD4- +AD4- +- return ret+ADs- +AD4- +AD4- +-+AH0- +AD4- +AD4- +-EXPORT+AF8-SYMBOL(drm+AF8-connector+AF8-plug+AF8-all)+ADs- +AD4- +AD4- +AKA- +AD4- +AD4- +AKA-/+ACoAKg- +AD4- +AD4- +AKA- +ACo- drm+AF8-connector+AF8-unplug+AF8-all - unregister connector userspace interfaces +AD4- +AD4- +AEAAQA- -1093,10 +-1133,12 +AEAAQA- void drm+AF8-connector+AF8-unplug+AF8-all(struct drm+AF8-device +ACo-dev) +AD4- +AD4- +AKAAew- +AD4- +AD4- +AKA- struct drm+AF8-connector +ACo-connector+ADs- +AD4- +AD4- +AKA- +AD4- +AD4- - /+ACo- FIXME: taking the mode config mutex ends up in a clash with sysfs +ACo-/ +AD4- +AD4- +- mutex+AF8-lock(+ACY-dev-+AD4-mode+AF8-config.mutex)+ADs- +AD4- You can't drop that FIXME, the bug is still there. That's clear given your explanation in the previous email. +AD4- +AD4- +AD4- +AD4- +- +AD4- +AD4- +AKA- list+AF8-for+AF8-each+AF8-entry(connector, +ACY-dev-+AD4-mode+AF8-config.connector+AF8-list, head) +AD4- +AD4- +AKA- drm+AF8-connector+AF8-unregister(connector)+ADs- +AD4- +AD4- +AKA- +AD4- +AD4- +- mutex+AF8-unlock(+ACY-dev-+AD4-mode+AF8-config.mutex)+ADs- +AD4- +AD4- +AKAAfQ- +AD4- +AD4- +AKA-EXPORT+AF8-SYMBOL(drm+AF8-connector+AF8-unplug+AF8-all)+ADs- +AD4- +AD4- +AKA- +AD4- +AD4- diff --git a/drivers/gpu/drm/drm+AF8-drv.c b/drivers/gpu/drm/drm+AF8-drv.c +AD4- +AD4- index 167c8d3..4a559c6 100644 +AD4- +AD4- --- a/drivers/gpu/drm/drm+AF8-drv.c +AD4- +AD4- +-+-+- b/drivers/gpu/drm/drm+AF8-drv.c +AD4- +AD4- +AEAAQA- -715,7 +-715,8 +AEAAQA- EXPORT+AF8-SYMBOL(drm+AF8-dev+AF8-unref)+ADs- +AD4- +AD4- +AKA- +ACo- +AD4- +AD4- +AKA- +ACo- Register the DRM device +AEA-dev with the system, advertise device to user-space +AD4- +AD4- +AKA- +ACo- and start normal device operation. +AEA-dev must be allocated via drm+AF8-dev+AF8-alloc() +AD4- +AD4- - +ACo- previously. +AD4- +AD4- +- +ACo- previously and right after drm+AF8-dev+AF8-register() driver should call +AD4- It'd do 2 sentences here for simplicity, not connect them with and. Also +AD4- +ACI-... +AF8-the+AF8- driver should ...+ACI- Ok. +AD4- +AD4- +AD4- +AD4- +- +ACo- drm+AF8-connector+AF8-plug+AF8-all() to register all connectors in sysfs. +AD4- Maybe mention why this is separate: +ACI-This is a separate call for backwards +AD4- compatibility with drivers still using the deprecated -+AD4-load() callback, +AD4- where connectors are registered from within the -+AD4-load() callback.+ACI- Ok. -Alexey
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH 0/3] drm: introduce drm_connector_plug_all() helper Alexey Brodkin <Alexey.Brodkin@synopsys.com> - 2016-03-18 11:10 +0100
[PATCH 1/3] drm: introduce drm_connector_plug_all() helper Alexey Brodkin <Alexey.Brodkin@synopsys.com> - 2016-03-18 11:10 +0100
Re: [PATCH 1/3] drm: introduce drm_connector_plug_all() helper Daniel Vetter <daniel@ffwll.ch> - 2016-03-18 19:10 +0100
Re: [PATCH 1/3] drm: introduce drm_connector_plug_all() helper Alexey Brodkin <Alexey.Brodkin@synopsys.com> - 2016-03-18 23:00 +0100
Re: [PATCH 1/3] drm: introduce drm_connector_plug_all() helper Daniel Vetter <daniel@ffwll.ch> - 2016-03-19 11:10 +0100
Re: [PATCH 1/3] drm: introduce drm_connector_plug_all() helper Alexey Brodkin <Alexey.Brodkin@synopsys.com> - 2016-03-21 12:10 +0100
Re: [PATCH 1/3] drm: introduce drm_connector_plug_all() helper Daniel Vetter <daniel@ffwll.ch> - 2016-03-21 18:20 +0100
csiph-web