All the mail mirrored from lore.kernel.org
 help / color / mirror / Atom feed
From: Ville Syrjala <ville.syrjala@linux.intel.com>
To: intel-gfx@lists.freedesktop.org
Subject: [Intel-gfx] [PATCH 8/9] drm/i915/fbc: Make the cfb allocation loop a bit more legible
Date: Thu, 10 Jun 2021 21:32:36 +0300	[thread overview]
Message-ID: <20210610183237.3920-9-ville.syrjala@linux.intel.com> (raw)
In-Reply-To: <20210610183237.3920-1-ville.syrjala@linux.intel.com>

From: Ville Syrjälä <ville.syrjala@linux.intel.com>

Write the cfb allocation loop as an actual loop instead of some
hard to read goto thing.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
 drivers/gpu/drm/i915/display/intel_fbc.c | 51 ++++++++++++------------
 1 file changed, 26 insertions(+), 25 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_fbc.c b/drivers/gpu/drm/i915/display/intel_fbc.c
index 6415f2cfd1ac..b6bfb4439a8b 100644
--- a/drivers/gpu/drm/i915/display/intel_fbc.c
+++ b/drivers/gpu/drm/i915/display/intel_fbc.c
@@ -440,44 +440,45 @@ static u64 intel_fbc_stolen_end(struct drm_i915_private *dev_priv)
 	return min(end, intel_fbc_cfb_base_max(dev_priv));
 }
 
+static int intel_fbc_max_limit(struct drm_i915_private *dev_priv, int fb_cpp)
+{
+	/*
+	 * FIXME: FBC1 can have arbitrary cfb stride,
+	 * so we could support different compression ratios.
+	 */
+	if (DISPLAY_VER(dev_priv) < 5 && !IS_G4X(dev_priv))
+		return 1;
+
+	/* WaFbcOnly1to1Ratio:ctg */
+	if (IS_G4X(dev_priv))
+		return 1;
+
+	/* FBC2 can only do 1:1, 1:2, 1:4 */
+	return fb_cpp == 2 ? 2 : 4;
+}
+
 static int find_compression_limit(struct drm_i915_private *dev_priv,
 				  unsigned int size,
 				  unsigned int fb_cpp)
 {
 	struct intel_fbc *fbc = &dev_priv->fbc;
 	u64 end = intel_fbc_stolen_end(dev_priv);
-	int compression_limit = 1;
-	int ret;
-
-	/* HACK: This code depends on what we will do in *_enable_fbc. If that
-	 * code changes, this code needs to change as well.
-	 *
-	 * The enable_fbc code will attempt to use one of our 2 compression
-	 * limits, therefore, in that case, we only have 1 resort.
-	 */
+	int ret, limit = 1;
 
 	/* Try to over-allocate to reduce reallocations and fragmentation. */
 	ret = i915_gem_stolen_insert_node_in_range(dev_priv, &fbc->compressed_fb,
 						   size <<= 1, 4096, 0, end);
 	if (ret == 0)
-		return compression_limit;
+		return limit;
 
-again:
-	/* HW's ability to limit the CFB is 1:4 */
-	if (compression_limit > 4 ||
-	    (fb_cpp == 2 && compression_limit == 2))
-		return 0;
-
-	ret = i915_gem_stolen_insert_node_in_range(dev_priv, &fbc->compressed_fb,
-						   size >>= 1, 4096, 0, end);
-	if (ret && DISPLAY_VER(dev_priv) <= 4) {
-		return 0;
-	} else if (ret) {
-		compression_limit <<= 1;
-		goto again;
-	} else {
-		return compression_limit;
+	for (; limit <= intel_fbc_max_limit(dev_priv, fb_cpp); limit <<= 1) {
+		ret = i915_gem_stolen_insert_node_in_range(dev_priv, &fbc->compressed_fb,
+							   size >>= 1, 4096, 0, end);
+		if (ret == 0)
+			return limit;
 	}
+
+	return 0;
 }
 
 static int intel_fbc_alloc_cfb(struct drm_i915_private *dev_priv,
-- 
2.31.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2021-06-10 18:33 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-10 18:32 [Intel-gfx] [PATCH 0/9] drm/i915/fbc: Clean up cfb allocation code Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 1/9] drm/i915/fbc: s/threshold/limit/ Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 2/9] drm/i915/fbc: Extract intel_fbc_program_cfb() Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 3/9] drm/i915/fbc: Embed the compressed_llb node Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 4/9] drm/i915/fbc: Don't pass around the mm node Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 5/9] drm/i915/fbc: Handle 16bpp compression limit better Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 6/9] drm/i915/fbc: Introduce g4x_dpfc_ctl_limit() Ville Syrjala
2021-06-10 18:32 ` [Intel-gfx] [PATCH 7/9] drm/i915/fbc: Extract intel_fbc_stolen_end() Ville Syrjala
2021-06-10 18:32 ` Ville Syrjala [this message]
2021-06-10 18:32 ` [Intel-gfx] [PATCH 9/9] drm/i915/fbc: Allocate llb before cfb Ville Syrjala
2021-06-18 21:07   ` Souza, Jose
2021-06-10 18:50 ` [Intel-gfx] ✗ Fi.CI.SPARSE: warning for drm/i915/fbc: Clean up cfb allocation code Patchwork
2021-06-10 19:20 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-06-10 22:15 ` [Intel-gfx] ✗ Fi.CI.IGT: failure " Patchwork

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=20210610183237.3920-9-ville.syrjala@linux.intel.com \
    --to=ville.syrjala@linux.intel.com \
    --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 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.