From: Matthew Auld <matthew.auld@intel.com>
To: intel-gfx@lists.freedesktop.org
Subject: [PATCH 10/23] drm/i915: enable IPS bit for 64K pages
Date: Mon, 21 Aug 2017 19:34:50 +0100 [thread overview]
Message-ID: <20170821183503.12246-11-matthew.auld@intel.com> (raw)
In-Reply-To: <20170821183503.12246-1-matthew.auld@intel.com>
Before we can enable 64K pages through the IPS bit, we must first enable
it through MMIO, otherwise the page-walker will simply ignore it.
v2: add comment mentioning that 64K is BDW+
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
---
drivers/gpu/drm/i915/i915_gem.c | 16 ++++++++++++++++
drivers/gpu/drm/i915/i915_reg.h | 3 +++
2 files changed, 19 insertions(+)
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index af3b9cfcc1fe..12c365c69edf 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -4773,6 +4773,22 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv)
}
}
+ /* To support 64K PTEs we need to first enable the use of the
+ * Intermediate-Page-Size(IPS) bit of the PDE field via some magical
+ * mmio, otherwise the page-walker will simply ignore the IPS bit. This
+ * shouldn't be needed after GEN10.
+ *
+ * 64K pages were first introduced from BDW+, although technically they
+ * only *work* from gen9+. For pre-BDW we instead have the option for
+ * 32K pages, but we don't currently have any support for it in our
+ * driver.
+ */
+ if (HAS_PAGE_SIZE(dev_priv, I915_GTT_PAGE_SIZE_64K) &&
+ INTEL_GEN(dev_priv) <= 10)
+ I915_WRITE(GEN8_GAMW_ECO_DEV_RW_IA,
+ I915_READ(GEN8_GAMW_ECO_DEV_RW_IA) |
+ GAMW_ECO_ENABLE_64K_IPS_FIELD);
+
i915_gem_init_swizzling(dev_priv);
/*
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index d4ecb1905ad8..767952aef2ab 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2371,6 +2371,9 @@ enum i915_power_well_id {
#define GEN9_GAMT_ECO_REG_RW_IA _MMIO(0x4ab0)
#define GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS (1<<18)
+#define GEN8_GAMW_ECO_DEV_RW_IA _MMIO(0x4080)
+#define GAMW_ECO_ENABLE_64K_IPS_FIELD 0xF
+
#define GAMT_CHKN_BIT_REG _MMIO(0x4ab8)
#define GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING (1<<28)
--
2.13.5
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2017-08-21 18:35 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-21 18:34 [PATCH 00/23] huge gtt pages Matthew Auld
2017-08-21 18:34 ` [PATCH 01/23] mm/shmem: introduce shmem_file_setup_with_mnt Matthew Auld
2017-08-23 9:31 ` Joonas Lahtinen
2017-08-23 22:34 ` Andrew Morton
2017-08-24 12:04 ` Matthew Auld
2017-08-25 20:49 ` Andrew Morton
2017-08-29 14:09 ` Joonas Lahtinen
2017-08-21 18:34 ` [PATCH 02/23] drm/i915: introduce simple gemfs Matthew Auld
2017-08-29 14:33 ` Joonas Lahtinen
2017-08-21 18:34 ` [PATCH 03/23] drm/i915/gemfs: enable THP Matthew Auld
2017-08-29 14:49 ` Joonas Lahtinen
2017-09-04 12:09 ` Matthew Auld
2017-08-21 18:34 ` [PATCH 04/23] drm/i915: introduce page_size_mask to dev_info Matthew Auld
2017-08-21 18:34 ` [PATCH 05/23] drm/i915: push set_pages down to the callers Matthew Auld
2017-08-29 14:44 ` Joonas Lahtinen
2017-08-21 18:34 ` [PATCH 06/23] drm/i915: introduce page_size members Matthew Auld
2017-09-05 9:25 ` Joonas Lahtinen
2017-08-21 18:34 ` [PATCH 07/23] drm/i915: introduce vm set_pages/clear_pages Matthew Auld
2017-08-21 18:34 ` [PATCH 08/23] drm/i915: align the vma start to the largest gtt page size Matthew Auld
2017-08-21 18:34 ` [PATCH 09/23] drm/i915: align 64K objects to 2M Matthew Auld
2017-08-21 18:34 ` Matthew Auld [this message]
2017-08-21 18:34 ` [PATCH 11/23] drm/i915: disable GTT cache for 2M/1G pages Matthew Auld
2017-08-21 18:34 ` [PATCH 12/23] drm/i915: support 1G pages for the 48b PPGTT Matthew Auld
2017-08-21 18:34 ` [PATCH 13/23] drm/i915: support 2M " Matthew Auld
2017-08-21 18:34 ` [PATCH 14/23] drm/i915: add support for 64K scratch page Matthew Auld
2017-08-21 18:34 ` [PATCH 15/23] drm/i915: support 64K pages for the 48b PPGTT Matthew Auld
2017-08-21 18:34 ` [PATCH 16/23] drm/i915: accurate page size tracking for the ppgtt Matthew Auld
2017-08-21 18:34 ` [PATCH 17/23] drm/i915/debugfs: include some gtt page size metrics Matthew Auld
2017-08-21 18:34 ` [PATCH 18/23] drm/i915/selftests: huge page tests Matthew Auld
2017-08-24 17:56 ` kbuild test robot
2017-08-28 14:36 ` Chris Wilson
2017-08-21 18:34 ` [PATCH 19/23] drm/i915/selftests: mix huge pages Matthew Auld
2017-08-21 18:35 ` [PATCH 20/23] drm/i915: disable platform support for vGPU huge gtt pages Matthew Auld
2017-08-21 18:35 ` [PATCH 21/23] drm/i915: enable platform support for 64K pages Matthew Auld
2017-08-21 18:35 ` [PATCH 22/23] drm/i915: enable platform support for 2M pages Matthew Auld
2017-08-21 18:35 ` [PATCH 23/23] drm/i915: enable platform support for 1G pages Matthew Auld
2017-08-21 18:54 ` ✓ Fi.CI.BAT: success for huge gtt pages (rev7) Patchwork
2017-08-22 14:21 ` [PATCH 00/23] huge gtt pages Chris Wilson
2017-08-22 14:23 ` Chris Wilson
2017-08-22 15:23 ` Matthew Auld
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=20170821183503.12246-11-matthew.auld@intel.com \
--to=matthew.auld@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox