aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/i915/i915_vma.c
diff options
context:
space:
mode:
authorMaarten Lankhorst <maarten.lankhorst@linux.intel.com>2022-01-17 08:56:04 +0100
committerMaarten Lankhorst <maarten.lankhorst@linux.intel.com>2022-01-18 12:00:30 +0100
commit6945c53bc712cf4a28a46fe46c2bd8526ea261d1 (patch)
tree7b958600148f5b806cd6170e66ceb11b1930cc12 /drivers/gpu/drm/i915/i915_vma.c
parentdrm/i915: Call i915_gem_evict_vm in vm_fault_gtt to prevent new ENOSPC errors, v2. (diff)
downloadlinux-dev-6945c53bc712cf4a28a46fe46c2bd8526ea261d1.tar.xz
linux-dev-6945c53bc712cf4a28a46fe46c2bd8526ea261d1.zip
drm/i915: Add locking to i915_gem_evict_vm(), v3.
i915_gem_evict_vm will need to be able to evict objects that are locked by the current ctx. By testing if the current context already locked the object, we can do this correctly. This allows us to evict the entire vm even if we already hold some objects' locks. Previously, this was spread over several commits, but it makes more sense to commit the changes to i915_gem_evict_vm separately from the changes to i915_gem_evict_something() and i915_gem_evict_for_node(). Changes since v1: - Handle evicting dead objects better. Changes since v2: - Use for_i915_gem_ww in igt_evict_vm. (Thomas) Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com> [mlankhorst: Fix up doc warning.] Link: https://patchwork.freedesktop.org/patch/msgid/20220117075604.131477-1-maarten.lankhorst@linux.intel.com
Diffstat (limited to 'drivers/gpu/drm/i915/i915_vma.c')
-rw-r--r--drivers/gpu/drm/i915/i915_vma.c7
1 files changed, 6 insertions, 1 deletions
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index 9d859b0a3fbe..8bbc08a0e88f 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -1479,7 +1479,12 @@ static int __i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
/* Unlike i915_vma_pin, we don't take no for an answer! */
flush_idle_contexts(vm->gt);
if (mutex_lock_interruptible(&vm->mutex) == 0) {
- i915_gem_evict_vm(vm);
+ /*
+ * We pass NULL ww here, as we don't want to unbind
+ * locked objects when called from execbuf when pinning
+ * is removed. This would probably regress badly.
+ */
+ i915_gem_evict_vm(vm, NULL);
mutex_unlock(&vm->mutex);
}
} while (1);