[Intel-gfx] [PATCH 3/9] drm/i915: split intel_modeset_driver_remove() to pre/post irq uninstall

[Ville Syrjälä]

On Tue, Feb 04, 2020 at 03:42:22PM +0200, Jani Nikula wrote:
> Split intel_modeset_driver_remove() to two, the part with working irqs
> before irq uninstall, and the part after irq uninstall. Move
> irq_unintall() closer to the layer it belongs.
> 
> The error path in i915_driver_modeset_probe() looks obviously weird
> after this, but remains as good or broken as it ever was. No functional
> changes.
> 
> Cc: Ville Syrjälä <ville.syrjala at linux.intel.com>
> Signed-off-by: Jani Nikula <jani.nikula at intel.com>
> ---
>  drivers/gpu/drm/i915/display/intel_display.c | 12 +++++-------
>  drivers/gpu/drm/i915/display/intel_display.h |  1 +
>  drivers/gpu/drm/i915/i915_drv.c              |  5 +++++
>  3 files changed, 11 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
> index b0af37fb6d4a..69d6fe626a19 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.c
> +++ b/drivers/gpu/drm/i915/display/intel_display.c
> @@ -18826,6 +18826,7 @@ static void intel_hpd_poll_fini(struct drm_i915_private *i915)
>  	drm_connector_list_iter_end(&conn_iter);
>  }
>  
> +/* part #1: call before irq uninstall */
>  void intel_modeset_driver_remove(struct drm_i915_private *i915)
>  {
>  	flush_workqueue(i915->flip_wq);
> @@ -18833,14 +18834,11 @@ void intel_modeset_driver_remove(struct drm_i915_private *i915)
>  
>  	flush_work(&i915->atomic_helper.free_work);
>  	WARN_ON(!llist_empty(&i915->atomic_helper.free_list));
> +}
>  
> -	/*
> -	 * Interrupts and polling as the first thing to avoid creating havoc.
> -	 * Too much stuff here (turning of connectors, ...) would
> -	 * experience fancy races otherwise.
> -	 */

Maybe the comment should stay at the start of the _noirq() function?
Or maybe it's obvious now. Shrug.

> -	intel_irq_uninstall(i915);
> -
> +/* part #2: call after irq uninstall */
> +void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915)
> +{
>  	/*
>  	 * Due to the hpd irq storm handling the hotplug work can re-arm the
>  	 * poll handlers. Hence disable polling after hpd handling is shut down.
> diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
> index 75438a136d58..f92efbbec838 100644
> --- a/drivers/gpu/drm/i915/display/intel_display.h
> +++ b/drivers/gpu/drm/i915/display/intel_display.h
> @@ -616,6 +616,7 @@ intel_format_info_is_yuv_semiplanar(const struct drm_format_info *info,
>  void intel_modeset_init_hw(struct drm_i915_private *i915);
>  int intel_modeset_init(struct drm_i915_private *i915);
>  void intel_modeset_driver_remove(struct drm_i915_private *i915);
> +void intel_modeset_driver_remove_noirq(struct drm_i915_private *i915);
>  void intel_display_resume(struct drm_device *dev);
>  void intel_init_pch_refclk(struct drm_i915_private *dev_priv);
>  
> diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
> index 6ab2de82b5f7..5330a0f10e97 100644
> --- a/drivers/gpu/drm/i915/i915_drv.c
> +++ b/drivers/gpu/drm/i915/i915_drv.c
> @@ -332,6 +332,9 @@ static int i915_driver_modeset_probe(struct drm_i915_private *i915)
>  	i915_gem_driver_release(i915);
>  cleanup_modeset:
>  	intel_modeset_driver_remove(i915);
> +	intel_irq_uninstall(i915);
> +	intel_modeset_driver_remove_noirq(i915);
> +	goto cleanup_csr;
>  cleanup_irq:
>  	intel_irq_uninstall(i915);
>  cleanup_csr:
> @@ -348,6 +351,8 @@ static void i915_driver_modeset_remove(struct drm_i915_private *i915)

A bit funny with the naming of driver_modeset_remove() vs.
modeset_driver_remove(). But that's already in there.

Reviewed-by: Ville Syrjälä <ville.syrjala at linux.intel.com>

>  
>  	intel_irq_uninstall(i915);
>  
> +	intel_modeset_driver_remove_noirq(i915);
> +
>  	intel_bios_driver_remove(i915);
>  
>  	intel_vga_unregister(i915);
> -- 
> 2.20.1

-- 
Ville Syrjälä
Intel