[PATCH 0/6] drm/tegra: Fix some error handling paths

Christophe JAILLET posted 6 patches 2 years, 3 months ago
drivers/gpu/drm/tegra/dsi.c    | 55 ++++++++++++++++++++++------------
drivers/gpu/drm/tegra/hdmi.c   | 20 ++++++++-----
drivers/gpu/drm/tegra/output.c | 16 +++++++---
drivers/gpu/drm/tegra/rgb.c    | 18 +++++++----
4 files changed, 74 insertions(+), 35 deletions(-)
[PATCH 0/6] drm/tegra: Fix some error handling paths
Posted by Christophe JAILLET 2 years, 3 months ago
Most of the patches are retated to tegra_output_probe() and missing
tegra_output_remove(). Others are things spotted while writting the serie.


Patches 1, 3, 4 are verbose, but some functions called in the probe can
return -EPROBE_DEFER, so it is nice to correctly release resources.

Maybe moving the tegra_output_probe() call would minimize the changes, but I'm
always reluctant to move code, because of possible side-effects.


Christophe JAILLET (6):
  drm/tegra: dsi: Fix some error handling paths in tegra_dsi_probe()
  drm/tegra: dsi: Fix missing pm_runtime_disable() in the error handling
    path of tegra_dsi_probe()
  drm/tegra: dsi: Fix some error handling paths in tegra_hdmi_probe()
  drm/tegra: rgb: Fix some error handling paths in tegra_dc_rgb_probe()
  drm/tegra: rgb: Fix missing clk_put() in the error handling paths of
    tegra_dc_rgb_probe()
  drm/tegra: output: Fix missing i2c_put_adapter() in the error handling
    paths of tegra_output_probe()

 drivers/gpu/drm/tegra/dsi.c    | 55 ++++++++++++++++++++++------------
 drivers/gpu/drm/tegra/hdmi.c   | 20 ++++++++-----
 drivers/gpu/drm/tegra/output.c | 16 +++++++---
 drivers/gpu/drm/tegra/rgb.c    | 18 +++++++----
 4 files changed, 74 insertions(+), 35 deletions(-)

-- 
2.34.1
Re: [PATCH 0/6] drm/tegra: Fix some error handling paths
Posted by Thierry Reding 2 years ago
On Sat, Sep 02, 2023 at 05:22:07PM +0200, Christophe JAILLET wrote:
> Most of the patches are retated to tegra_output_probe() and missing
> tegra_output_remove(). Others are things spotted while writting the serie.
> 
> 
> Patches 1, 3, 4 are verbose, but some functions called in the probe can
> return -EPROBE_DEFER, so it is nice to correctly release resources.
> 
> Maybe moving the tegra_output_probe() call would minimize the changes, but I'm
> always reluctant to move code, because of possible side-effects.
> 
> 
> Christophe JAILLET (6):
>   drm/tegra: dsi: Fix some error handling paths in tegra_dsi_probe()
>   drm/tegra: dsi: Fix missing pm_runtime_disable() in the error handling
>     path of tegra_dsi_probe()
>   drm/tegra: dsi: Fix some error handling paths in tegra_hdmi_probe()
>   drm/tegra: rgb: Fix some error handling paths in tegra_dc_rgb_probe()
>   drm/tegra: rgb: Fix missing clk_put() in the error handling paths of
>     tegra_dc_rgb_probe()
>   drm/tegra: output: Fix missing i2c_put_adapter() in the error handling
>     paths of tegra_output_probe()
> 
>  drivers/gpu/drm/tegra/dsi.c    | 55 ++++++++++++++++++++++------------
>  drivers/gpu/drm/tegra/hdmi.c   | 20 ++++++++-----
>  drivers/gpu/drm/tegra/output.c | 16 +++++++---
>  drivers/gpu/drm/tegra/rgb.c    | 18 +++++++----
>  4 files changed, 74 insertions(+), 35 deletions(-)

Sorry, this fell through the cracks. Applied now, thanks.

Thierry