From: Ben Widawsky <benjamin.widawsky@intel.com>
To: Intel GFX <intel-gfx@lists.freedesktop.org>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>,
Ben Widawsky <ben@bwidawsk.net>,
Ben Widawsky <benjamin.widawsky@intel.com>
Subject: [PATCH 21/62] drm/i915/bdw: Support BDW caching
Date: Sat, 2 Nov 2013 21:07:19 -0700 [thread overview]
Message-ID: <1383451680-11173-22-git-send-email-benjamin.widawsky@intel.com> (raw)
In-Reply-To: <1383451680-11173-1-git-send-email-benjamin.widawsky@intel.com>
BDW caching works differently than the previous generations. Instead of
having bits in the PTE which directly control how the page is cached,
the 3 PTE bits PWT PCD and PAT provide an index into a PAT defined by
register 0x40e0. This style of caching is functionally equivalent to how
it works on HSW and before.
v2: Tiny bikeshed as discussed on internal irc.
v3: Squash in patch from Ville to mirror the x86 PAT setup more like
in arch/x86/mm/pat.c. Primarily, the 0th index will be WB, and not
uncached.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net> (v1)
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
---
drivers/gpu/drm/i915/i915_gem_gtt.c | 42 +++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/i915/i915_reg.h | 1 +
2 files changed, 43 insertions(+)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index df992dc..02de12d 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -58,12 +58,21 @@ typedef uint64_t gen8_gtt_pte_t;
#define HSW_WB_ELLC_LLC_AGE0 HSW_CACHEABILITY_CONTROL(0xb)
#define HSW_WT_ELLC_LLC_AGE0 HSW_CACHEABILITY_CONTROL(0x6)
+#define PPAT_UNCACHED_INDEX (_PAGE_PWT | _PAGE_PCD)
+#define PPAT_CACHED_PDE_INDEX 0 /* WB LLC */
+#define PPAT_CACHED_INDEX _PAGE_PAT /* WB LLCeLLC */
+#define PPAT_DISPLAY_ELLC_INDEX _PAGE_PCD /* WT eLLC */
+
static inline gen8_gtt_pte_t gen8_pte_encode(dma_addr_t addr,
enum i915_cache_level level,
bool valid)
{
gen8_gtt_pte_t pte = valid ? _PAGE_PRESENT | _PAGE_RW : 0;
pte |= addr;
+ if (level != I915_CACHE_NONE)
+ pte |= PPAT_CACHED_INDEX;
+ else
+ pte |= PPAT_UNCACHED_INDEX;
return pte;
}
@@ -805,6 +814,7 @@ static void i915_gtt_color_adjust(struct drm_mm_node *node,
*end -= 4096;
}
}
+
void i915_gem_setup_global_gtt(struct drm_device *dev,
unsigned long start,
unsigned long mappable_end,
@@ -1002,6 +1012,36 @@ static int ggtt_probe_common(struct drm_device *dev,
return ret;
}
+/* The GGTT and PPGTT need a private PPAT setup in order to handle cacheability
+ * bits. When using advanced contexts each context stores its own PAT, but
+ * writing this data shouldn't be harmful even in those cases. */
+static void gen8_setup_private_ppat(struct drm_i915_private *dev_priv)
+{
+#define GEN8_PPAT_UC (0<<0)
+#define GEN8_PPAT_WC (1<<0)
+#define GEN8_PPAT_WT (2<<0)
+#define GEN8_PPAT_WB (3<<0)
+#define GEN8_PPAT_ELLC_OVERRIDE (0<<2)
+#define GEN8_PPAT_LLC (1<<2)
+#define GEN8_PPAT_LLCELLC (2<<2)
+#define GEN8_PPAT_LLCeLLC (3<<2) /* BSPEC mistake? */
+#define GEN8_PPAT_AGE(x) (x<<4)
+#define GEN8_PPAT(i, x) ((uint64_t) (x) << ((i) * 8))
+ uint64_t pat;
+
+ pat = GEN8_PPAT(0, GEN8_PPAT_WB | GEN8_PPAT_LLC) | /* for normal objects, no eLLC */
+ GEN8_PPAT(1, GEN8_PPAT_WC | GEN8_PPAT_LLCELLC) | /* for something pointing to ptes? */
+ GEN8_PPAT(2, GEN8_PPAT_WT | GEN8_PPAT_LLCELLC) | /* for scanout with eLLC */
+ GEN8_PPAT(3, GEN8_PPAT_UC) | /* Uncached objects, mostly for scanout */
+ GEN8_PPAT(4, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(0)) |
+ GEN8_PPAT(5, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(1)) |
+ GEN8_PPAT(6, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(2)) |
+ GEN8_PPAT(7, GEN8_PPAT_WB | GEN8_PPAT_LLCELLC | GEN8_PPAT_AGE(3));
+
+ I915_WRITE(GEN8_PRIVATE_PAT, pat);
+ I915_WRITE(GEN8_PRIVATE_PAT + 4, pat >> 32);
+}
+
static int gen8_gmch_probe(struct drm_device *dev,
size_t *gtt_total,
size_t *stolen,
@@ -1027,6 +1067,8 @@ static int gen8_gmch_probe(struct drm_device *dev,
gtt_size = gen8_get_total_gtt_size(snb_gmch_ctl);
*gtt_total = (gtt_size / sizeof(gen8_gtt_pte_t)) << PAGE_SHIFT;
+ gen8_setup_private_ppat(dev_priv);
+
ret = ggtt_probe_common(dev, gtt_size);
dev_priv->gtt.base.clear_range = gen8_ggtt_clear_range;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index b801b88..9929750 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -664,6 +664,7 @@
#define RING_FAULT_FAULT_TYPE(x) ((x >> 1) & 0x3)
#define RING_FAULT_VALID (1<<0)
#define DONE_REG 0x40b0
+#define GEN8_PRIVATE_PAT 0x40e0
#define BSD_HWS_PGA_GEN7 (0x04180)
#define BLT_HWS_PGA_GEN7 (0x04280)
#define VEBOX_HWS_PGA_GEN7 (0x04380)
--
1.8.4.2
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2013-11-03 4:08 UTC|newest]
Thread overview: 145+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-03 4:06 [PATCH 00/62] Broadwell kernel driver support Ben Widawsky
2013-11-03 4:06 ` [PATCH 01/62] drm/i915/bdw: IS_GEN8 definition Ben Widawsky
2013-11-03 4:07 ` [PATCH 02/62] drm/i915/bdw: Handle forcewake for writes on gen8 Ben Widawsky
2013-11-04 14:19 ` Chris Wilson
2013-11-05 9:24 ` Mika Kuoppala
2013-11-03 4:07 ` [PATCH 03/62] drm/i915/bdw: Disable PPGTT for now Ben Widawsky
2013-11-04 14:44 ` Chris Wilson
2013-11-03 4:07 ` [PATCH 04/62] drm/i915/bdw: Add device IDs Ben Widawsky
2013-11-03 21:58 ` Chris Wilson
2013-11-04 0:36 ` [PATCH 04/62] [v6] " Ben Widawsky
2013-11-04 14:49 ` Chris Wilson
2013-11-04 15:49 ` Daniel Vetter
2013-11-04 16:04 ` Chris Wilson
2013-11-04 16:56 ` Ben Widawsky
2013-11-04 0:43 ` [PATCH 04/62] " Ben Widawsky
2013-11-04 0:47 ` [PATCH 04/62] [v7] " Ben Widawsky
2013-11-05 14:45 ` Mika Kuoppala
2013-11-03 4:07 ` [PATCH 05/62] drm/i915/bdw: Fences on gen8 look just like gen7 Ben Widawsky
2013-11-03 4:07 ` [PATCH 06/62] drm/i915/bdw: Swizzling support Ben Widawsky
2013-11-05 9:59 ` Mika Kuoppala
2013-11-03 4:07 ` [PATCH 07/62] drm/i915/bdw: HW context support Ben Widawsky
2013-11-03 4:07 ` [PATCH 08/62] drm/i915/bdw: Clock gating init Ben Widawsky
2013-11-03 4:07 ` [PATCH 09/62] drm/i915/bdw: display stuff Ben Widawsky
2013-11-06 8:13 ` Daniel Vetter
2013-11-03 4:07 ` [PATCH 10/62] drm/i915/bdw: support GMS and GGMS changes Ben Widawsky
2013-11-04 0:53 ` [PATCH 10/62] [v5] " Ben Widawsky
2013-11-03 4:07 ` [PATCH 11/62] drm/i915/bdw: Implement interrupt changes Ben Widawsky
2013-11-06 8:39 ` Daniel Vetter
2013-11-03 4:07 ` [PATCH 12/62] drm/i915/bdw: Add interrupt info to debugfs Ben Widawsky
2013-11-03 4:07 ` [PATCH 13/62] drm/i915/bdw: Support 64b relocations Ben Widawsky
2013-11-03 4:07 ` [PATCH 14/62] drm/i915/bdw: dispatch updates (64b related) Ben Widawsky
2013-11-05 15:50 ` Paulo Zanoni
2013-11-03 4:07 ` [PATCH 15/62] drm/i915/bdw: Update MI_FLUSH_DW Ben Widawsky
2013-11-03 4:07 ` [PATCH 16/62] drm/i915/bdw: debugfs updates Ben Widawsky
2013-11-04 14:28 ` Chris Wilson
2013-11-05 3:03 ` Ben Widawsky
2013-11-05 16:40 ` Paulo Zanoni
2013-11-03 4:07 ` [PATCH 17/62] drm/i915/bdw: Update relevant error state Ben Widawsky
2013-11-05 17:03 ` Paulo Zanoni
2013-11-03 4:07 ` [PATCH 18/62] drm/i915/bdw: Make gen8_gmch_probe Ben Widawsky
2013-11-04 22:01 ` Imre Deak
2013-11-05 3:32 ` [PATCH 18/62] [v6] " Ben Widawsky
2013-11-03 4:07 ` [PATCH 19/62] drm/i915/bdw: Create gen8_gtt_pte_t Ben Widawsky
2013-11-04 14:36 ` Chris Wilson
2013-11-04 22:03 ` Imre Deak
2013-11-03 4:07 ` [PATCH 20/62] drm/i915/bdw: Add GTT functions Ben Widawsky
2013-11-04 22:22 ` Imre Deak
2013-11-06 8:28 ` Bloomfield, Jon
2013-11-03 4:07 ` Ben Widawsky [this message]
2013-11-04 14:39 ` [PATCH 21/62] drm/i915/bdw: Support BDW caching Chris Wilson
2013-11-05 3:56 ` [PATCH 21/62] [v4] " Ben Widawsky
2013-11-05 15:19 ` [PATCH 21/62] " Imre Deak
2013-11-03 4:07 ` [PATCH 22/62] drm/i915/bdw: Implement Full Force Miss disables Ben Widawsky
2013-11-05 15:41 ` Imre Deak
2013-11-05 16:17 ` Daniel Vetter
2013-11-06 9:33 ` Daniel Vetter
2013-11-03 4:07 ` [PATCH 23/62] drm/i915/bdw: PPGTT init & cleanup Ben Widawsky
2013-11-04 14:58 ` Imre Deak
2013-11-05 4:47 ` [PATCH] " Ben Widawsky
2013-11-03 4:07 ` [PATCH 24/62] drm/i915/bdw: Initialize the PDEs Ben Widawsky
2013-11-04 14:10 ` Damien Lespiau
2013-11-05 5:20 ` [PATCH 24/62] [v3] " Ben Widawsky
2013-11-03 4:07 ` [PATCH 25/62] drm/i915/bdw: Implement PPGTT clear range Ben Widawsky
2013-11-03 4:07 ` [PATCH 26/62] drm/i915/bdw: Implement PPGTT insert Ben Widawsky
2013-11-03 4:07 ` [PATCH 27/62] drm/i915/bdw: Implement PPGTT enable Ben Widawsky
2013-11-04 14:47 ` Damien Lespiau
2013-11-05 6:29 ` [PATCH 27/62] [v7] " Ben Widawsky
2013-11-03 4:07 ` [PATCH 28/62] drm/i915/bdw: unleash PPGTT Ben Widawsky
2013-11-03 4:07 ` [PATCH 29/62] drm/i915/bdw: Render ring flushing Ben Widawsky
2013-11-03 4:07 ` [PATCH 30/62] drm/i915/bdw: BSD init for gen8 also Ben Widawsky
2013-11-03 4:07 ` [PATCH 31/62] drm/i915/bdw: Don't muck with gtt_size on Gen8 when PPGTT setup fails Ben Widawsky
2013-11-03 4:07 ` [PATCH 32/62] drm/i915/bdw: ppgtt info in debugfs Ben Widawsky
2013-11-03 4:07 ` [PATCH 33/62] drm/i915/bdw: add IS_BROADWELL macro Ben Widawsky
2013-11-03 4:07 ` [PATCH 34/62] drm/i915/bdw: Broadwell has 3 pipes Ben Widawsky
2013-11-03 4:07 ` [PATCH 35/62] drm/i915/bdw: add Broadwell sprite/plane/cursor checks Ben Widawsky
2013-11-03 4:07 ` [PATCH 36/62] drm/i915/bdw: Broadwell also has the "power down well" Ben Widawsky
2013-11-03 11:05 ` Ville Syrjälä
2013-11-03 11:24 ` Daniel Vetter
2013-11-03 11:25 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 37/62] drm/i915/bdw: pretend we have LPT LP on Broadwell Ben Widawsky
2013-11-03 11:19 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 38/62] drm/i915/bdw: get the correct LCPLL frequency " Ben Widawsky
2013-11-03 11:07 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 39/62] drm/i915/bdw: on Broadwell, the panel fitter is on the pipe Ben Widawsky
2013-11-03 11:19 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 40/62] drm/i915/bdw: Broadwell has PIPEMISC Ben Widawsky
2013-11-03 11:11 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 41/62] drm/i915/bdw: Use pipe CSC on Broadwell Ben Widawsky
2013-11-03 4:07 ` [PATCH 42/62] drm/i915/bdw: Implement WaSwitchSolVfFArbitrationPriority Ben Widawsky
2013-11-03 11:07 ` Ville Syrjälä
2013-11-03 17:44 ` Ben Widawsky
2013-11-04 14:23 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 43/62] drm/i915/bdw: Add BDW DDI buffer translation values Ben Widawsky
2013-11-04 23:59 ` Ben Widawsky
2013-11-03 4:07 ` [PATCH 44/62] drm/i915/bdw: add BDW DDI buf translations for eDP Ben Widawsky
2013-11-05 0:09 ` Ben Widawsky
2013-11-03 4:07 ` [PATCH 45/62] drm/i915/bdw: add support for BDW DP voltage swings and pre-emphasis Ben Widawsky
2013-11-05 0:45 ` Ben Widawsky
2013-11-05 13:01 ` Paulo Zanoni
2013-11-06 3:15 ` Todd Previte
2013-11-03 4:07 ` [PATCH 46/62] drm/i915/bdw: BDW also has only 2 FDI lanes Ben Widawsky
2013-11-03 4:07 ` [PATCH 47/62] drm/i915/bdw: check DPD on port D when setting the DDI buffers Ben Widawsky
2013-11-05 0:46 ` Ben Widawsky
2013-11-03 4:07 ` [PATCH 48/62] drm/i915/bdw: Add Broadwell display FIFO limits Ben Widawsky
2013-11-04 9:39 ` Jani Nikula
2013-11-04 13:59 ` Ville Syrjälä
2013-11-03 4:07 ` [PATCH 49/62] drm/i915/bdw: Use The GT mailbox for IPS enable/disable Ben Widawsky
2013-11-04 10:15 ` Jani Nikula
2013-11-03 4:07 ` [PATCH 50/62] drm/i915/bdw: Support eDP PSR Ben Widawsky
2013-11-04 10:34 ` Jani Nikula
2013-11-05 6:45 ` [PATCH 50/62] [v5] " Ben Widawsky
2014-03-04 9:31 ` Kumar, Kiran S
2014-03-05 6:31 ` Ben Widawsky
2013-11-03 4:07 ` [PATCH 51/62] drm/i915/bdw: Use HSW formula for ring freq scaling Ben Widawsky
2013-11-06 13:34 ` Daniel Vetter
2013-11-03 4:07 ` [PATCH 52/62] drm/i915/bdw: Don't wait for c0 threads on forcewake Ben Widawsky
2013-11-04 13:47 ` Jani Nikula
2013-11-03 4:07 ` [PATCH 53/62] drm/i915/bdw: Broadwell has a max port clock of 300Mhz on HDMI Ben Widawsky
2013-11-04 13:33 ` Jani Nikula
2013-11-03 4:07 ` [PATCH 54/62] drm/i915/bdw: Create a separate BDW rps enable Ben Widawsky
2013-11-04 21:04 ` Jesse Barnes
2013-11-03 4:07 ` [PATCH 55/62] drm/i915/bdw: Disable semaphores Ben Widawsky
2013-11-04 18:18 ` Jesse Barnes
2013-11-05 3:45 ` [PATCH 55/62] [v2] " Ben Widawsky
2013-11-03 4:07 ` [PATCH 56/62] drm/i915/bdw: Implement edp PSR workarounds Ben Widawsky
2013-11-05 17:19 ` Jesse Barnes
2013-11-06 15:44 ` Daniel Vetter
2013-11-03 4:07 ` [PATCH 57/62] drm/i915/bdw: BWGTLB clock gate disable Ben Widawsky
2013-11-05 17:22 ` Jesse Barnes
2013-11-03 4:07 ` [PATCH 58/62] drm/i915/bdw: Disable centroid pixel perf optimization Ben Widawsky
2013-11-04 13:20 ` Paulo Zanoni
2013-11-05 6:52 ` [PATCH] " Ben Widawsky
2013-11-05 17:24 ` Jesse Barnes
2013-11-03 4:07 ` [PATCH 59/62] drm/i915/bdw: Sampler power bypass disable Ben Widawsky
2013-11-03 4:07 ` [PATCH 60/62] drm/i915/bdw: Limit SDE poly depth FIFO to 2 Ben Widawsky
2013-11-03 4:07 ` [PATCH 61/62] drm/i915/bdw: conservative SBE VUE cache mode Ben Widawsky
2013-11-03 4:08 ` [PATCH 62/62] drm/i915/bdw: WaSingleSubspanDispatchOnAALinesAndPoints Ben Widawsky
2013-11-03 8:45 ` [PATCH 00/62] Broadwell kernel driver support Daniel Vetter
2013-11-04 14:15 ` Jani Nikula
2013-11-04 15:04 ` Damien Lespiau
2013-11-05 15:14 ` Daniel Vetter
2013-11-05 15:54 ` Imre Deak
2013-11-03 11:47 ` [PATCH 63/62] drm/i915/bdw: Enable trickle feed on Broadwell ville.syrjala
2013-11-04 15:05 ` Damien Lespiau
2013-11-05 7:11 ` [PATCH 64/62] drm/i915/bdw: Change dp aux timeout to 600us on DDIA Ben Widawsky
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=1383451680-11173-22-git-send-email-benjamin.widawsky@intel.com \
--to=benjamin.widawsky@intel.com \
--cc=ben@bwidawsk.net \
--cc=daniel.vetter@ffwll.ch \
--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