| Message ID | 1505409850-6468-2-git-send-email-zhi.a.wang@intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Quoting Zhi Wang (2017-09-14 18:24:10)
> struct_mutext needs to be held before call these two APIs.
True for the moment. But raises a question of whether introducing a
ppat->mutex would be beneficial? All depends on whether the caller is
already forced to use struct_mutex and is likely to continue to need it
next year...
-Chris
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 37cd086..b43d3c9 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -2869,6 +2869,7 @@ intel_ppat_get(struct drm_i915_private *i915, u8 value) unsigned int scanned, best_score; int i; + lockdep_assert_held(&i915->drm.struct_mutex); GEM_BUG_ON(!ppat->max_entries); scanned = best_score = 0; @@ -2924,6 +2925,7 @@ void intel_ppat_put(const struct intel_ppat_entry *entry) struct intel_ppat *ppat = entry->ppat; unsigned int index = entry - ppat->entries; + lockdep_assert_held(&ppat->i915->drm.struct_mutex); GEM_BUG_ON(!ppat->max_entries); kref_put(&ppat->entries[index].ref, release_ppat);
struct_mutext needs to be held before call these two APIs. Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Ben Widawsky <benjamin.widawsky@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Zhi Wang <zhi.a.wang@intel.com> --- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 ++ 1 file changed, 2 insertions(+)