| Message ID | 0f864bee4e43c1fd7dbc2ab6917c99001510073c.1553086417.git.leonard.crestez@nxp.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
| Series | ARM: imx_v6_v7_defconfig: Enable CONFIG_PWM_IMX27 | expand |
Hello, On Wed, Mar 20, 2019 at 01:01:26PM +0000, Leonard Crestez wrote: > Commit d80f8206905c ("pwm: imx: Split into two drivers") also adds a new > CONFIG_PWM_IMX27 for the PWM block on recent IMX chips and we should > enable it by default for imx. > > Restoring the PWM driver fixes an infinite probe loop in 5.1-rc1 on > various imx6qdl-sabresd boards. > > Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> > Reported-by: Abel Vesa <abel.vesa@nxp.com> This is prior art: https://patchwork.ozlabs.org/project/linux-pwm/list/?series=85452&state=%2A&archive=both unfortunately this didn't get an Ack by Shawn and Thierry (added to recipients) marked it as "Not Applicable". Given that the driver change is in Linus Torvald's tree now (since v5.1-rc1~38) it doesn't matter much via which tree this goes in and Shawn's tree is the easier now. @Shawn: Would you please apply my patches? If you don't have them any more, I can bounce them to you. See git show v5.1-rc1~38^2~17 -- drivers/pwm/Kconfig for their justification. > --- > arch/arm/configs/imx_v6_v7_defconfig | 1 + > 1 file changed, 1 insertion(+) > > Probe loop repeats following lines: > > [ 3.625031] pwm-backlight backlight-lvds: backlight-lvds supply power not found, using dummy regulator > [ 3.635868] panel-simple panel: panel supply power not found, using dummy regulator > [ 3.644844] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). > [ 3.651478] [drm] No driver support for vblank timestamp query. > [ 3.657660] imx-drm display-subsystem: bound imx-ipuv3-crtc.2 (ops ipu_crtc_ops) > [ 3.665240] imx-drm display-subsystem: bound imx-ipuv3-crtc.3 (ops ipu_crtc_ops) > [ 3.672819] imx-drm display-subsystem: bound imx-ipuv3-crtc.6 (ops ipu_crtc_ops) > [ 3.680393] imx-drm display-subsystem: bound imx-ipuv3-crtc.7 (ops ipu_crtc_ops) > [ 3.688312] dwhdmi-imx 120000.hdmi: Detected HDMI TX controller v1.30a with HDCP (DWC HDMI 3D TX PHY) > [ 3.699743] imx-drm display-subsystem: bound 120000.hdmi (ops dw_hdmi_imx_ops) > [ 3.707343] imx-drm display-subsystem: failed to bind ldb (ops imx_ldb_ops): -517 > [ 3.716921] imx-drm display-subsystem: master bind failed: -517 > > Maybe it's an imx-drm bug which got exposed by accident? IMHO this should be debugged independent of this patch. > diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig > index 5586a5074a96..2fa5074f5244 100644 > --- a/arch/arm/configs/imx_v6_v7_defconfig > +++ b/arch/arm/configs/imx_v6_v7_defconfig > @@ -397,10 +397,11 @@ CONFIG_SENSORS_ISL29018=y > CONFIG_MAG3110=y > CONFIG_MPL3115=y > CONFIG_PWM=y > CONFIG_PWM_FSL_FTM=y > CONFIG_PWM_IMX=y > +CONFIG_PWM_IMX27=y PWM_IMX is gone, so this can be dropped (but see my patch referenced above). > CONFIG_NVMEM_IMX_OCOTP=y > CONFIG_NVMEM_VF610_OCOTP=y > CONFIG_TEE=y > CONFIG_OPTEE=y > CONFIG_MUX_MMIO=y Best regards Uwe
On 3/21/2019 11:49 AM, Uwe Kleine-König wrote: > On Wed, Mar 20, 2019 at 01:01:26PM +0000, Leonard Crestez wrote: >> Commit d80f8206905c ("pwm: imx: Split into two drivers") also adds a new >> CONFIG_PWM_IMX27 for the PWM block on recent IMX chips and we should >> enable it by default for imx. >> >> Restoring the PWM driver fixes an infinite probe loop in 5.1-rc1 on >> various imx6qdl-sabresd boards. > This is prior art: > > @Shawn: Would you please apply my patches? If you don't have them any > more, I can bounce them to you. +1 from me because it fixes boot. >> --- a/arch/arm/configs/imx_v6_v7_defconfig >> +++ b/arch/arm/configs/imx_v6_v7_defconfig >> @@ -397,10 +397,11 @@ CONFIG_SENSORS_ISL29018=y >> CONFIG_MAG3110=y >> CONFIG_MPL3115=y >> CONFIG_PWM=y >> CONFIG_PWM_FSL_FTM=y >> CONFIG_PWM_IMX=y >> +CONFIG_PWM_IMX27=y > > PWM_IMX is gone, so this can be dropped (but see my patch referenced > above). Noticed that after sending; your patch deals with this correctly and also covers imx_v4_v5 -- Regards, Leonard
On Thu, Mar 21, 2019 at 10:49:03AM +0100, Uwe Kleine-König wrote: > Hello, > > On Wed, Mar 20, 2019 at 01:01:26PM +0000, Leonard Crestez wrote: > > Commit d80f8206905c ("pwm: imx: Split into two drivers") also adds a new > > CONFIG_PWM_IMX27 for the PWM block on recent IMX chips and we should > > enable it by default for imx. > > > > Restoring the PWM driver fixes an infinite probe loop in 5.1-rc1 on > > various imx6qdl-sabresd boards. > > > > Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> > > Reported-by: Abel Vesa <abel.vesa@nxp.com> > > This is prior art: > > https://patchwork.ozlabs.org/project/linux-pwm/list/?series=85452&state=%2A&archive=both > > unfortunately this didn't get an Ack by Shawn and Thierry (added to > recipients) marked it as "Not Applicable". > > Given that the driver change is in Linus Torvald's tree now (since > v5.1-rc1~38) it doesn't matter much via which tree this goes in and > Shawn's tree is the easier now. > > @Shawn: Would you please apply my patches? If you don't have them any > more, I can bounce them to you. See Yeah, just applied. Sorry for being sluggish. Shawn
On Thu, Mar 21, 2019 at 10:49:03AM +0100, Uwe Kleine-König wrote: > Hello, > > On Wed, Mar 20, 2019 at 01:01:26PM +0000, Leonard Crestez wrote: > > Commit d80f8206905c ("pwm: imx: Split into two drivers") also adds a new > > CONFIG_PWM_IMX27 for the PWM block on recent IMX chips and we should > > enable it by default for imx. > > > > Restoring the PWM driver fixes an infinite probe loop in 5.1-rc1 on > > various imx6qdl-sabresd boards. > > > > Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> > > Reported-by: Abel Vesa <abel.vesa@nxp.com> > > This is prior art: > > https://patchwork.ozlabs.org/project/linux-pwm/list/?series=85452&state=%2A&archive=both > > unfortunately this didn't get an Ack by Shawn and Thierry (added to > recipients) marked it as "Not Applicable". If I mark patches as "not applicable" it generally means that I don't intend to apply them to the PWM tree. Thierry > Given that the driver change is in Linus Torvald's tree now (since > v5.1-rc1~38) it doesn't matter much via which tree this goes in and > Shawn's tree is the easier now. > > @Shawn: Would you please apply my patches? If you don't have them any > more, I can bounce them to you. See > > git show v5.1-rc1~38^2~17 -- drivers/pwm/Kconfig > > for their justification. > > > --- > > arch/arm/configs/imx_v6_v7_defconfig | 1 + > > 1 file changed, 1 insertion(+) > > > > Probe loop repeats following lines: > > > > [ 3.625031] pwm-backlight backlight-lvds: backlight-lvds supply power not found, using dummy regulator > > [ 3.635868] panel-simple panel: panel supply power not found, using dummy regulator > > [ 3.644844] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). > > [ 3.651478] [drm] No driver support for vblank timestamp query. > > [ 3.657660] imx-drm display-subsystem: bound imx-ipuv3-crtc.2 (ops ipu_crtc_ops) > > [ 3.665240] imx-drm display-subsystem: bound imx-ipuv3-crtc.3 (ops ipu_crtc_ops) > > [ 3.672819] imx-drm display-subsystem: bound imx-ipuv3-crtc.6 (ops ipu_crtc_ops) > > [ 3.680393] imx-drm display-subsystem: bound imx-ipuv3-crtc.7 (ops ipu_crtc_ops) > > [ 3.688312] dwhdmi-imx 120000.hdmi: Detected HDMI TX controller v1.30a with HDCP (DWC HDMI 3D TX PHY) > > [ 3.699743] imx-drm display-subsystem: bound 120000.hdmi (ops dw_hdmi_imx_ops) > > [ 3.707343] imx-drm display-subsystem: failed to bind ldb (ops imx_ldb_ops): -517 > > [ 3.716921] imx-drm display-subsystem: master bind failed: -517 > > > > Maybe it's an imx-drm bug which got exposed by accident? > > IMHO this should be debugged independent of this patch. > > > diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig > > index 5586a5074a96..2fa5074f5244 100644 > > --- a/arch/arm/configs/imx_v6_v7_defconfig > > +++ b/arch/arm/configs/imx_v6_v7_defconfig > > @@ -397,10 +397,11 @@ CONFIG_SENSORS_ISL29018=y > > CONFIG_MAG3110=y > > CONFIG_MPL3115=y > > CONFIG_PWM=y > > CONFIG_PWM_FSL_FTM=y > > CONFIG_PWM_IMX=y > > +CONFIG_PWM_IMX27=y > > PWM_IMX is gone, so this can be dropped (but see my patch referenced > above). > > > CONFIG_NVMEM_IMX_OCOTP=y > > CONFIG_NVMEM_VF610_OCOTP=y > > CONFIG_TEE=y > > CONFIG_OPTEE=y > > CONFIG_MUX_MMIO=y > > Best regards > Uwe > > -- > Pengutronix e.K. | Uwe Kleine-König | > Industrial Linux Solutions | http://www.pengutronix.de/ |
On Fri, Mar 22, 2019 at 10:47:00AM +0100, Thierry Reding wrote: > On Thu, Mar 21, 2019 at 10:49:03AM +0100, Uwe Kleine-König wrote: > > Hello, > > > > On Wed, Mar 20, 2019 at 01:01:26PM +0000, Leonard Crestez wrote: > > > Commit d80f8206905c ("pwm: imx: Split into two drivers") also adds a new > > > CONFIG_PWM_IMX27 for the PWM block on recent IMX chips and we should > > > enable it by default for imx. > > > > > > Restoring the PWM driver fixes an infinite probe loop in 5.1-rc1 on > > > various imx6qdl-sabresd boards. > > > > > > Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> > > > Reported-by: Abel Vesa <abel.vesa@nxp.com> > > > > This is prior art: > > > > https://patchwork.ozlabs.org/project/linux-pwm/list/?series=85452&state=%2A&archive=both > > > > unfortunately this didn't get an Ack by Shawn and Thierry (added to > > recipients) marked it as "Not Applicable". > > If I mark patches as "not applicable" it generally means that I don't > intend to apply them to the PWM tree. I understood that. In this case however it would in my eyes have made sense to take these patches together with the patch that does the driver split via the pwm tree to make it less likely that people using the respective defconfigs don't get the PWM driver enabled. That's the situation now on 5.1-rc1 and people actuall hit this problem. :-| That's why I wrote: I think the easiest handling would be to let them go via the pwm tree with Shawn's Ack to get the update near to the actual split into the mainline. in the cover letter. Best regards Uwe
diff --git a/arch/arm/configs/imx_v6_v7_defconfig b/arch/arm/configs/imx_v6_v7_defconfig index 5586a5074a96..2fa5074f5244 100644 --- a/arch/arm/configs/imx_v6_v7_defconfig +++ b/arch/arm/configs/imx_v6_v7_defconfig @@ -397,10 +397,11 @@ CONFIG_SENSORS_ISL29018=y CONFIG_MAG3110=y CONFIG_MPL3115=y CONFIG_PWM=y CONFIG_PWM_FSL_FTM=y CONFIG_PWM_IMX=y +CONFIG_PWM_IMX27=y CONFIG_NVMEM_IMX_OCOTP=y CONFIG_NVMEM_VF610_OCOTP=y CONFIG_TEE=y CONFIG_OPTEE=y CONFIG_MUX_MMIO=y
Commit d80f8206905c ("pwm: imx: Split into two drivers") also adds a new CONFIG_PWM_IMX27 for the PWM block on recent IMX chips and we should enable it by default for imx. Restoring the PWM driver fixes an infinite probe loop in 5.1-rc1 on various imx6qdl-sabresd boards. Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com> Reported-by: Abel Vesa <abel.vesa@nxp.com> --- arch/arm/configs/imx_v6_v7_defconfig | 1 + 1 file changed, 1 insertion(+) Probe loop repeats following lines: [ 3.625031] pwm-backlight backlight-lvds: backlight-lvds supply power not found, using dummy regulator [ 3.635868] panel-simple panel: panel supply power not found, using dummy regulator [ 3.644844] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 3.651478] [drm] No driver support for vblank timestamp query. [ 3.657660] imx-drm display-subsystem: bound imx-ipuv3-crtc.2 (ops ipu_crtc_ops) [ 3.665240] imx-drm display-subsystem: bound imx-ipuv3-crtc.3 (ops ipu_crtc_ops) [ 3.672819] imx-drm display-subsystem: bound imx-ipuv3-crtc.6 (ops ipu_crtc_ops) [ 3.680393] imx-drm display-subsystem: bound imx-ipuv3-crtc.7 (ops ipu_crtc_ops) [ 3.688312] dwhdmi-imx 120000.hdmi: Detected HDMI TX controller v1.30a with HDCP (DWC HDMI 3D TX PHY) [ 3.699743] imx-drm display-subsystem: bound 120000.hdmi (ops dw_hdmi_imx_ops) [ 3.707343] imx-drm display-subsystem: failed to bind ldb (ops imx_ldb_ops): -517 [ 3.716921] imx-drm display-subsystem: master bind failed: -517 Maybe it's an imx-drm bug which got exposed by accident?