[Intel-gfx] [PATCH v3 4/4] drm/i915/fbc: Register per-crtc debugfs files

[Ville Syrjälä]

On Sat, Dec 18, 2021 at 06:00:47PM -0700, Nathan Chancellor wrote:
> Hi Ville,
> 
> On Mon, Dec 13, 2021 at 05:14:35PM +0200, Ville Syrjala wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > 
> > Expose FBC debugfs files for each crtc. These may or may not point
> > to the same FBC instance depending on the platform.
> > 
> > We leave the old global debugfs files in place until
> > igt catches up to the new per-crtc approach.
> > 
> > v2: Take a trip via intel_crtc_debugfs_add() (Jani)
> > 
> > Cc: Jani Nikula <jani.nikula at intel.com>
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 
> Apologies if this has already been reported and fixed, I have not seen
> anything on lore.kernel.org or drm-intel about it.
> 
> This patch as commit e74c6aa955ca ("drm/i915/fbc: Register per-crtc
> debugfs files") breaks the build when CONFIG_DEBUG_FS is disabled.
> 
> drivers/gpu/drm/i915/display/intel_fbc.c: In function ‘intel_fbc_crtc_debugfs_add’:
> drivers/gpu/drm/i915/display/intel_fbc.c:1817:61: error: ‘struct drm_crtc’ has no member named ‘debugfs_entry’
>  1817 |                 intel_fbc_debugfs_add(plane->fbc, crtc->base.debugfs_entry);
>       |                                                             ^

Doh. I guess I didn't actually build test the final version with
DEBUGFS=n.

Does this fix it for you?

diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 13eeba2a750a..4d01b4d89775 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -1135,14 +1135,12 @@ struct drm_crtc {
 	 */
 	spinlock_t commit_lock;
 
-#ifdef CONFIG_DEBUG_FS
 	/**
 	 * @debugfs_entry:
 	 *
 	 * Debugfs directory for this CRTC.
 	 */
 	struct dentry *debugfs_entry;
-#endif
 
 	/**
 	 * @crc:
-- 
2.32.0

-- 
Ville Syrjälä
Intel