Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Ben Widawsky <ben@bwidawsk.net>
To: Intel GFX <intel-gfx@lists.freedesktop.org>
Cc: Ben Widawsky <ben@bwidawsk.net>
Subject: [PATCH 09/12] drm/i915: Split context enabling from init
Date: Tue, 23 Apr 2013 23:15:37 -0700	[thread overview]
Message-ID: <1366784140-2670-10-git-send-email-ben@bwidawsk.net> (raw)
In-Reply-To: <1366784140-2670-1-git-send-email-ben@bwidawsk.net>

Context init should have only been called once, ever. The code was
hackish because it predated some of the init cleanups. This patch
refactors the code to leave the part which must be run more than once
(reset, and thaw) as a distinct function.

I need this for some of my future plans to more closely tie a context to
a PPGTT.

Since context_init comes before ring_init now, we also have to add the
call to fini() if things go badly.

Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
---
 drivers/gpu/drm/i915/i915_drv.h         |  1 +
 drivers/gpu/drm/i915/i915_gem.c         | 18 +++++++++++++++---
 drivers/gpu/drm/i915/i915_gem_context.c | 18 +++++++++---------
 3 files changed, 25 insertions(+), 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 618845e..11cc159 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1699,6 +1699,7 @@ struct dma_buf *i915_gem_prime_export(struct drm_device *dev,
 /* i915_gem_context.c */
 void i915_gem_context_init(struct drm_device *dev);
 void i915_gem_context_fini(struct drm_device *dev, bool reset);
+int i915_gem_context_enable(struct drm_i915_private *dev_priv);
 void i915_gem_context_close(struct drm_device *dev, struct drm_file *file);
 int i915_switch_context(struct intel_ring_buffer *ring,
 			struct drm_file *file, int to_id);
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 9c5eaf0..fbd289d 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -2151,6 +2151,8 @@ void i915_gem_reset(struct drm_device *dev)
 	for_each_ring(ring, dev_priv, i)
 		i915_gem_reset_ring_lists(dev_priv, ring);
 
+	i915_gem_context_fini(dev, false);
+
 	/* Move everything out of the GPU domains to ensure we do any
 	 * necessary invalidation upon reuse.
 	 */
@@ -4029,10 +4031,18 @@ i915_gem_init_hw(struct drm_device *dev)
 		return ret;
 
 	/*
-	 * XXX: There was some w/a described somewhere suggesting loading
-	 * contexts before PPGTT.
+	 * XXX: Contexts should only be initialized once. Doing a switch to the
+	 * default context switch however is something we'd like to do after
+	 * reset or thaw (the latter may not actually be necessary for HW, but
+	 * goes with our code better).  Context switching requires rings (for
+	 * the do_switch), but before enabling PPGTT. So don't move this.
 	 */
-	i915_gem_context_init(dev);
+	if (!dev_priv->hw_contexts_disabled &&
+	    i915_gem_context_enable(dev_priv)) {
+		i915_gem_context_fini(dev, false);
+		dev_priv->hw_contexts_disabled = true;
+	}
+
 	if (dev_priv->mm.aliasing_ppgtt) {
 		ret = dev_priv->mm.aliasing_ppgtt->enable(dev);
 		if (ret) {
@@ -4060,6 +4070,8 @@ int i915_gem_init(struct drm_device *dev)
 
 	i915_gem_init_global_gtt(dev);
 
+	i915_gem_context_init(dev);
+
 	ret = i915_gem_init_hw(dev);
 	mutex_unlock(&dev->struct_mutex);
 	if (ret) {
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 6030d83..b2deddd 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -216,15 +216,10 @@ static int create_default_context(struct drm_i915_private *dev_priv)
 	if (ret)
 		goto err_destroy;
 
-	ret = do_switch(ctx);
-	if (ret)
-		goto err_unpin;
+	DRM_DEBUG_DRIVER("Default HW context created\n");
 
-	DRM_DEBUG_DRIVER("Default HW context loaded\n");
 	return 0;
 
-err_unpin:
-	i915_gem_object_unpin(ctx->obj);
 err_destroy:
 	do_destroy(ctx);
 	return ret;
@@ -239,9 +234,8 @@ void i915_gem_context_init(struct drm_device *dev)
 		return;
 	}
 
-	/* If called from reset, or thaw... we've been here already */
-	if (dev_priv->hw_contexts_disabled ||
-	    dev_priv->ring[RCS].default_context)
+	if (WARN_ON(dev_priv->hw_contexts_disabled ||
+	    dev_priv->ring[RCS].default_context))
 		return;
 
 	dev_priv->hw_context_size = round_up(get_context_size(dev), 4096);
@@ -279,6 +273,12 @@ void i915_gem_context_fini(struct drm_device *dev, bool reset)
 	dev_priv->ring[RCS].default_context = NULL;
 }
 
+int i915_gem_context_enable(struct drm_i915_private *dev_priv)
+{
+	BUG_ON(!dev_priv->ring[RCS].default_context);
+	return do_switch(dev_priv->ring[RCS].default_context);
+}
+
 static int context_idr_cleanup(int id, void *p, void *data)
 {
 	struct i915_hw_context *ctx = p;
-- 
1.8.2.1

  parent reply	other threads:[~2013-04-24  6:13 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-24  6:15 [PATCH 00/12] [RFC] PPGTT prep patches part 1 Ben Widawsky
2013-04-24  6:15 ` [PATCH 01/12] drm/i915: Assert mutex_is_locked on context lookup Ben Widawsky
2013-05-02 20:27   ` Jesse Barnes
2013-05-06  9:40     ` Daniel Vetter
2013-05-06  9:44       ` Daniel Vetter
2013-05-06 17:59         ` Ben Widawsky
2013-05-06 18:35           ` Daniel Vetter
2013-04-24  6:15 ` [PATCH 02/12] drm/i915: BUG_ON bad PPGTT offset Ben Widawsky
2013-05-02 20:28   ` Jesse Barnes
2013-05-06  9:48     ` Daniel Vetter
2013-05-06 18:03       ` Ben Widawsky
2013-05-06 18:37         ` Daniel Vetter
2013-05-08 16:48           ` Ben Widawsky
2013-05-08 17:55             ` Daniel Vetter
2013-04-24  6:15 ` [PATCH 03/12] drm/i915: make PDE|PTE platform specific Ben Widawsky
2013-05-02 21:26   ` Jesse Barnes
2013-05-02 22:49     ` Ben Widawsky
2013-05-02 22:55       ` Jesse Barnes
2013-05-06  9:47     ` Daniel Vetter
2013-05-08 16:49       ` Ben Widawsky
2013-05-08 17:52         ` Daniel Vetter
2013-04-24  6:15 ` [PATCH 04/12] drm/i915: Extract PDE writes Ben Widawsky
2013-05-02 21:27   ` Jesse Barnes
2013-05-06  9:50     ` Daniel Vetter
2013-04-24  6:15 ` [PATCH 05/12] drm: Optionally create mm blocks from top-to-bottom Ben Widawsky
2013-04-24  6:15 ` [PATCH 06/12] drm/i915: Use drm_mm for PPGTT PDEs Ben Widawsky
2013-05-02 21:42   ` Jesse Barnes
2013-04-24  6:15 ` [PATCH 07/12] drm/i915: Use PDEs as the guard page Ben Widawsky
2013-04-24  6:15 ` [PATCH 08/12] drm/i915: Update context_fini Ben Widawsky
2013-04-24 15:11   ` Mika Kuoppala
2013-04-25  4:11     ` Ben Widawsky
2013-04-25  5:17       ` Ben Widawsky
2013-04-25 15:01       ` Mika Kuoppala
2013-04-25 17:22         ` Ben Widawsky
2013-04-24  6:15 ` Ben Widawsky [this message]
2013-04-24 10:04   ` [PATCH 09/12] drm/i915: Split context enabling from init Chris Wilson
2013-04-24 16:39     ` Ben Widawsky
2013-04-24  6:15 ` [PATCH 10/12] drm/i915: destroy i915_gem_init_global_gtt Ben Widawsky
2013-04-24  6:15 ` [PATCH 11/12] drm/i915: Embed PPGTT into the context Ben Widawsky
2013-04-24  6:15 ` [PATCH 12/12] drm/i915: No contexts without ppgtt Ben Widawsky
2013-04-24 10:06   ` Chris Wilson
2013-04-24 16:39     ` Ben Widawsky
2013-04-24  9:53 ` [PATCH 00/12] [RFC] PPGTT prep patches part 1 Chris Wilson
2013-04-24 19:58 ` Chris Wilson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1366784140-2670-10-git-send-email-ben@bwidawsk.net \
    --to=ben@bwidawsk.net \
    --cc=intel-gfx@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox