| Message ID | 20190722182439.44844-1-dianders@chromium.org (mailing list archive) |
|---|---|
| Headers | show |
| Series | video: of: display_timing: Adjust err printing of of_get_display_timing() | expand |
Hi Dough. On Mon, Jul 22, 2019 at 11:24:35AM -0700, Douglas Anderson wrote: > As reported by Sam Ravnborg [1], after commit b8a2948fa2b3 > ("drm/panel: simple: Add ability to override typical timing") we now > see a pointless error message printed on every boot for many systems. > Let's fix that by adjusting who is responsible for printing error > messages when of_get_display_timing() is used. > > Most certainly we can bikeshed the topic about whether this is the > right fix or we should instead add logic to panel_simple_probe() to > avoid calling of_get_display_timing() in the case where there is no > "panel-timing" sub-node. If there is consensus that I should move the > fix to panel_simple_probe() I'm happy to spin this series. In that > case we probably should _remove_ the extra prints that were already > present in the other two callers of of_get_display_timing(). > > While at it, fix a missing of_node_put() found by code inspection. > > NOTE: amba-clcd and panel-lvds were only compile-tested. > > [1] https://lkml.kernel.org/r/20190721093815.GA4375@ravnborg.org > > > Douglas Anderson (4): > video: of: display_timing: Add of_node_put() in > of_get_display_timing() > video: of: display_timing: Don't yell if no timing node is present > drm: panel-lvds: Spout an error if of_get_display_timing() gives an > error > video: amba-clcd: Spout an error if of_get_display_timing() gives an > error Series looks good - thanks. Reviewed-by: Sam Ravnborg <sam@ravnborg.org> You could consider silencing display_timing as the last patch, but thats a very small detail. How do we apply these fixes - to drm-misc-next? Bartlomiej? No need to go in via drm-misc-fixes as the offending commit is only in drm-misc-next. Sam
Hi, On 7/23/19 10:38 AM, Sam Ravnborg wrote: > Hi Dough. > > On Mon, Jul 22, 2019 at 11:24:35AM -0700, Douglas Anderson wrote: >> As reported by Sam Ravnborg [1], after commit b8a2948fa2b3 >> ("drm/panel: simple: Add ability to override typical timing") we now >> see a pointless error message printed on every boot for many systems. >> Let's fix that by adjusting who is responsible for printing error >> messages when of_get_display_timing() is used. >> >> Most certainly we can bikeshed the topic about whether this is the >> right fix or we should instead add logic to panel_simple_probe() to >> avoid calling of_get_display_timing() in the case where there is no >> "panel-timing" sub-node. If there is consensus that I should move the >> fix to panel_simple_probe() I'm happy to spin this series. In that >> case we probably should _remove_ the extra prints that were already >> present in the other two callers of of_get_display_timing(). >> >> While at it, fix a missing of_node_put() found by code inspection. >> >> NOTE: amba-clcd and panel-lvds were only compile-tested. >> >> [1] https://lkml.kernel.org/r/20190721093815.GA4375@ravnborg.org >> >> >> Douglas Anderson (4): >> video: of: display_timing: Add of_node_put() in >> of_get_display_timing() >> video: of: display_timing: Don't yell if no timing node is present >> drm: panel-lvds: Spout an error if of_get_display_timing() gives an >> error >> video: amba-clcd: Spout an error if of_get_display_timing() gives an >> error > > Series looks good - thanks. > Reviewed-by: Sam Ravnborg <sam@ravnborg.org> > > You could consider silencing display_timing as the last patch, but thats > a very small detail. > > How do we apply these fixes - to drm-misc-next? Bartlomiej? > > No need to go in via drm-misc-fixes as the offending commit is only in > drm-misc-next. I've merged the whole series to drm-misc-next, thanks! Best regards, -- Bartlomiej Zolnierkiewicz Samsung R&D Institute Poland Samsung Electronics
As reported by Sam Ravnborg [1], after commit b8a2948fa2b3 ("drm/panel: simple: Add ability to override typical timing") we now see a pointless error message printed on every boot for many systems. Let's fix that by adjusting who is responsible for printing error messages when of_get_display_timing() is used. Most certainly we can bikeshed the topic about whether this is the right fix or we should instead add logic to panel_simple_probe() to avoid calling of_get_display_timing() in the case where there is no "panel-timing" sub-node. If there is consensus that I should move the fix to panel_simple_probe() I'm happy to spin this series. In that case we probably should _remove_ the extra prints that were already present in the other two callers of of_get_display_timing(). While at it, fix a missing of_node_put() found by code inspection. NOTE: amba-clcd and panel-lvds were only compile-tested. [1] https://lkml.kernel.org/r/20190721093815.GA4375@ravnborg.org Douglas Anderson (4): video: of: display_timing: Add of_node_put() in of_get_display_timing() video: of: display_timing: Don't yell if no timing node is present drm: panel-lvds: Spout an error if of_get_display_timing() gives an error video: amba-clcd: Spout an error if of_get_display_timing() gives an error drivers/gpu/drm/panel/panel-lvds.c | 5 ++++- drivers/video/fbdev/amba-clcd.c | 4 +++- drivers/video/of_display_timing.c | 11 +++++++---- 3 files changed, 14 insertions(+), 6 deletions(-)