All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: keithp@keithp.com
Cc: intel-gfx@lists.freedesktop.org
Subject: [PATCH 01/16] drm/i915: Cache GT fifo count for SandyBridge
Date: Thu, 12 May 2011 22:17:09 +0100	[thread overview]
Message-ID: <1305235044-9159-2-git-send-email-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <1305235044-9159-1-git-send-email-chris@chris-wilson.co.uk>

The read back of the available FIFO entries is vital for system
stability, but extremely costly. However, we only need a guide so as to
avoid eating into the reserved entries and since we are the only
consumer we can cache the read of the count from the last write.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
---
 drivers/gpu/drm/i915/i915_drv.c |   15 ++++++++++-----
 drivers/gpu/drm/i915/i915_drv.h |    1 +
 drivers/gpu/drm/i915/i915_reg.h |    1 +
 3 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 52e52ce..50c6065 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -313,12 +313,17 @@ void gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
 
 void __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv)
 {
-	int loop = 500;
-	u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
-	while (fifo < 20 && loop--) {
-		udelay(10);
-		fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+	if (dev_priv->gt_fifo_count < GT_FIFO_NUM_RESERVED_ENTRIES ) {
+		int loop = 500;
+		u32 fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+		while (fifo <= GT_FIFO_NUM_RESERVED_ENTRIES && loop--) {
+			udelay(10);
+			fifo = I915_READ_NOTRACE(GT_FIFO_FREE_ENTRIES);
+		}
+		WARN_ON(loop < 0 && fifo <= GT_FIFO_NUM_RESERVED_ENTRIES);
+		dev_priv->gt_fifo_count = fifo;
 	}
+	dev_priv->gt_fifo_count--;
 }
 
 static int i915_drm_freeze(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index e9d8243..d029476 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -268,6 +268,7 @@ typedef struct drm_i915_private {
 	int relative_constants_mode;
 
 	void __iomem *regs;
+	u32 gt_fifo_count;
 
 	struct intel_gmbus {
 		struct i2c_adapter adapter;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 289adaa..ef13f39 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -3329,6 +3329,7 @@
 #define  FORCEWAKE_ACK				0x130090
 
 #define  GT_FIFO_FREE_ENTRIES			0x120008
+#define    GT_FIFO_NUM_RESERVED_ENTRIES		20
 
 #define GEN6_RPNSWREQ				0xA008
 #define   GEN6_TURBO_DISABLE			(1<<31)
-- 
1.7.5.1

  reply	other threads:[~2011-05-12 21:17 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-12 21:17 Select patches for 2.6.40 Chris Wilson
2011-05-12 21:17 ` Chris Wilson [this message]
2011-05-12 21:17 ` [PATCH 02/16] drm/i915: Refactor pwrite/pread to use single copy of get_user_pages Chris Wilson
2011-05-13  0:21   ` Keith Packard
2011-05-15  8:00     ` Chris Wilson
2011-05-15 15:36       ` Keith Packard
2011-05-12 21:17 ` [PATCH 03/16] drm/i915: s/addr & ~PAGE_MASK/offset_in_page(addr)/ Chris Wilson
2011-05-13  0:26   ` Keith Packard
2011-05-12 21:17 ` [PATCH 04/16] drm/i915: Only print out the actual number of fences for i915_error_state Chris Wilson
2011-05-13  0:28   ` Keith Packard
2011-05-12 21:17 ` [PATCH 05/16] drm/i915/tv: Use a direct pointer for tv_mode Chris Wilson
2011-05-13  0:30   ` Keith Packard
2011-05-12 21:17 ` [PATCH 06/16] drm/i915: Replace ironlake_compute_wm0 with g4x_compute_wm0 Chris Wilson
2011-05-13  0:34   ` Keith Packard
2011-05-13  9:19     ` Chris Wilson
2011-05-13 15:00       ` Keith Packard
2011-05-12 21:17 ` [PATCH 07/16] drm/i915/crt: Explicitly return false if connected to a digital monitor Chris Wilson
2011-05-12 21:17 ` [PATCH 08/16] drm/i915/i2c: Convert from using GMBUS1 + reg_offset idiom to reg + 0 Chris Wilson
2011-05-13  0:36   ` Keith Packard
2011-05-13  0:40   ` Keith Packard
2011-05-13  9:28     ` Chris Wilson
2011-05-13 15:00       ` Keith Packard
2011-06-03 20:55         ` Keith Packard
2011-06-03 23:09           ` Chris Wilson
2011-06-03 23:48             ` Keith Packard
2011-07-13 18:33             ` Keith Packard
2011-07-13 19:22               ` Chris Wilson
2011-05-12 21:17 ` [PATCH 09/16] drm/i915/gmbus: Reset the controller on initialisation Chris Wilson
2011-05-13  0:41   ` Keith Packard
2011-05-13  9:32     ` Chris Wilson
2011-05-13 15:01       ` Keith Packard
2011-05-13 15:53         ` Chris Wilson
2011-06-03 20:56   ` Keith Packard
2011-05-12 21:17 ` [PATCH 10/16] drm/i915: Retire requests before disabling pagefaults Chris Wilson
2011-05-13  1:08   ` Keith Packard
2011-05-12 21:17 ` [PATCH 11/16] drm/i915: not finding a fence is a non-recoverable condition Chris Wilson
2011-05-13  1:10   ` Keith Packard
2011-05-15 20:38     ` Daniel Vetter
2011-05-12 21:17 ` [PATCH 12/16] drm/915: fix relaxed tiling on gen2: tile height Chris Wilson
2011-05-13  1:13   ` Keith Packard
2011-05-15 20:43     ` Daniel Vetter
2011-05-15 21:58       ` Keith Packard
2011-05-16 17:48         ` Daniel Vetter
2011-05-12 21:17 ` [PATCH 13/16] drm/i915: Remove unused enum "chip_family" Chris Wilson
2011-05-13  1:13   ` Keith Packard
2011-05-12 21:17 ` [PATCH 14/16] drm/i915: Use PCI-ID to identify Broadwater and Crestline Chris Wilson
2011-05-13  1:16   ` Keith Packard
2011-05-13  9:39     ` Chris Wilson
2011-05-15 20:49     ` Daniel Vetter
2011-05-15 22:01       ` Keith Packard
2011-06-03 20:59   ` Keith Packard
2011-05-12 21:17 ` [PATCH 15/16] drm/i915: Convert partial to full CPU read domain if we touch every page Chris Wilson
2011-05-13  1:17   ` Keith Packard
2011-05-12 21:17 ` [PATCH 16/16] drm/i915: Share the common force-audio property between connectors Chris Wilson
2011-05-13  1:21   ` Keith Packard
2011-05-13  3:39 ` Select patches for 2.6.40 Zou, Nanhai

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=1305235044-9159-2-git-send-email-chris@chris-wilson.co.uk \
    --to=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=keithp@keithp.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.