All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ramalingam C <ramalingam.c@intel.com>
To: dri-devel <dri-devel@lists.freedesktop.org>,
	intel-gfx <intel-gfx@lists.freedesktop.org>
Cc: Daniel Vetter <daniel@ffwll.ch>,
	Matthew Auld <matthew.auld@intel.com>,
	CQ Tang <cq.tang@intel.com>,
	Hellstrom Thomas <thomas.hellstrom@intel.com>,
	Ramalingam C <ramalingam.c@intel.com>
Subject: [PATCH 13/14] drm/i915/uapi: document behaviour for DG2 64K support
Date: Mon, 11 Oct 2021 21:41:54 +0530	[thread overview]
Message-ID: <20211011161155.6397-14-ramalingam.c@intel.com> (raw)
In-Reply-To: <20211011161155.6397-1-ramalingam.c@intel.com>

From: Matthew Auld <matthew.auld@intel.com>

On discrete platforms like DG2, we need to support a minimum page size
of 64K when dealing with device local-memory. This is quite tricky for
various reasons, so try to document the new implicit uapi for this.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
---
 include/uapi/drm/i915_drm.h | 61 ++++++++++++++++++++++++++++++++++---
 1 file changed, 56 insertions(+), 5 deletions(-)

diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index aa2a7eccfb94..d62e8b7ed8b6 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -1118,10 +1118,16 @@ struct drm_i915_gem_exec_object2 {
 	/**
 	 * When the EXEC_OBJECT_PINNED flag is specified this is populated by
 	 * the user with the GTT offset at which this object will be pinned.
+	 *
 	 * When the I915_EXEC_NO_RELOC flag is specified this must contain the
 	 * presumed_offset of the object.
+	 *
 	 * During execbuffer2 the kernel populates it with the value of the
 	 * current GTT offset of the object, for future presumed_offset writes.
+	 *
+	 * See struct drm_i915_gem_create_ext for the rules when dealing with
+	 * alignment restrictions with I915_MEMORY_CLASS_DEVICE, on devices with
+	 * minimum page sizes, like DG2.
 	 */
 	__u64 offset;
 
@@ -3001,11 +3007,56 @@ struct drm_i915_gem_create_ext {
 	 *
 	 * The (page-aligned) allocated size for the object will be returned.
 	 *
-	 * Note that for some devices we have might have further minimum
-	 * page-size restrictions(larger than 4K), like for device local-memory.
-	 * However in general the final size here should always reflect any
-	 * rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS
-	 * extension to place the object in device local-memory.
+	 * On discrete platforms, starting from DG2, we have to contend with GTT
+	 * page size restrictions when dealing with I915_MEMORY_CLASS_DEVICE
+	 * objects.  Specifically the hardware only supports 64K or larger GTT
+	 * page sizes for such memory. The kernel will already ensure that all
+	 * I915_MEMORY_CLASS_DEVICE memory is allocated using 64K or larger page
+	 * sizes underneath.
+	 *
+	 * Note that the returned size here will always reflect any required
+	 * rounding up done by the kernel, i.e 4K will now become 64K on devices
+	 * such as DG2. The GTT alignment will also need be at least 64K for
+	 * such objects.
+	 *
+	 * Note that due to how the hardware implements 64K GTT page support, we
+	 * have some further complications:
+	 *
+	 *   1.) The entire PDE(which covers a 2M virtual address range), must
+	 *   contain only 64K PTEs, i.e mixing 4K and 64K PTEs in the same
+	 *   PDE is forbidden by the hardware.
+	 *
+	 *   2.) We still need to support 4K PTEs for I915_MEMORY_CLASS_SYSTEM
+	 *   objects.
+	 *
+	 * To handle the above the kernel implements a memory coloring scheme to
+	 * prevent userspace from mixing I915_MEMORY_CLASS_DEVICE and
+	 * I915_MEMORY_CLASS_SYSTEM objects in the same PDE. If the kernel is
+	 * ever unable to evict the required pages for the given PDE(different
+	 * color) when inserting the object into the GTT then it will simply
+	 * fail the request.
+	 *
+	 * Since userspace needs to manage the GTT address space themselves,
+	 * special care is needed to ensure this doesn't happen. The simplest
+	 * scheme is to simply align and round up all I915_MEMORY_CLASS_DEVICE
+	 * objects to 2M, which avoids any issues here. At the very least this
+	 * is likely needed for objects that can be placed in both
+	 * I915_MEMORY_CLASS_DEVICE and I915_MEMORY_CLASS_SYSTEM, to avoid
+	 * potential issues when the kernel needs to migrate the object behind
+	 * the scenes, since that might also involve evicting other objects.
+	 *
+	 * To summarise the GTT rules, on platforms like DG2:
+	 *
+	 *   1.) All objects that can be placed in I915_MEMORY_CLASS_DEVICE must
+	 *   have 64K alignment. The kernel will reject this otherwise.
+	 *
+	 *   2.) All I915_MEMORY_CLASS_DEVICE objects must never be placed in
+	 *   the same PDE with other I915_MEMORY_CLASS_SYSTEM objects. The
+	 *   kernel will reject this otherwise.
+	 *
+	 *   3.) Objects that can be placed in both I915_MEMORY_CLASS_DEVICE and
+	 *   I915_MEMORY_CLASS_SYSTEM should probably be aligned and padded out
+	 *   to 2M.
 	 */
 	__u64 size;
 	/**
-- 
2.20.1


WARNING: multiple messages have this Message-ID (diff)
From: Ramalingam C <ramalingam.c@intel.com>
To: dri-devel <dri-devel@lists.freedesktop.org>,
	intel-gfx <intel-gfx@lists.freedesktop.org>
Cc: Daniel Vetter <daniel@ffwll.ch>,
	Matthew Auld <matthew.auld@intel.com>,
	CQ Tang <cq.tang@intel.com>,
	Hellstrom Thomas <thomas.hellstrom@intel.com>,
	Ramalingam C <ramalingam.c@intel.com>
Subject: [Intel-gfx] [PATCH 13/14] drm/i915/uapi: document behaviour for DG2 64K support
Date: Mon, 11 Oct 2021 21:41:54 +0530	[thread overview]
Message-ID: <20211011161155.6397-14-ramalingam.c@intel.com> (raw)
In-Reply-To: <20211011161155.6397-1-ramalingam.c@intel.com>

From: Matthew Auld <matthew.auld@intel.com>

On discrete platforms like DG2, we need to support a minimum page size
of 64K when dealing with device local-memory. This is quite tricky for
various reasons, so try to document the new implicit uapi for this.

Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
---
 include/uapi/drm/i915_drm.h | 61 ++++++++++++++++++++++++++++++++++---
 1 file changed, 56 insertions(+), 5 deletions(-)

diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index aa2a7eccfb94..d62e8b7ed8b6 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -1118,10 +1118,16 @@ struct drm_i915_gem_exec_object2 {
 	/**
 	 * When the EXEC_OBJECT_PINNED flag is specified this is populated by
 	 * the user with the GTT offset at which this object will be pinned.
+	 *
 	 * When the I915_EXEC_NO_RELOC flag is specified this must contain the
 	 * presumed_offset of the object.
+	 *
 	 * During execbuffer2 the kernel populates it with the value of the
 	 * current GTT offset of the object, for future presumed_offset writes.
+	 *
+	 * See struct drm_i915_gem_create_ext for the rules when dealing with
+	 * alignment restrictions with I915_MEMORY_CLASS_DEVICE, on devices with
+	 * minimum page sizes, like DG2.
 	 */
 	__u64 offset;
 
@@ -3001,11 +3007,56 @@ struct drm_i915_gem_create_ext {
 	 *
 	 * The (page-aligned) allocated size for the object will be returned.
 	 *
-	 * Note that for some devices we have might have further minimum
-	 * page-size restrictions(larger than 4K), like for device local-memory.
-	 * However in general the final size here should always reflect any
-	 * rounding up, if for example using the I915_GEM_CREATE_EXT_MEMORY_REGIONS
-	 * extension to place the object in device local-memory.
+	 * On discrete platforms, starting from DG2, we have to contend with GTT
+	 * page size restrictions when dealing with I915_MEMORY_CLASS_DEVICE
+	 * objects.  Specifically the hardware only supports 64K or larger GTT
+	 * page sizes for such memory. The kernel will already ensure that all
+	 * I915_MEMORY_CLASS_DEVICE memory is allocated using 64K or larger page
+	 * sizes underneath.
+	 *
+	 * Note that the returned size here will always reflect any required
+	 * rounding up done by the kernel, i.e 4K will now become 64K on devices
+	 * such as DG2. The GTT alignment will also need be at least 64K for
+	 * such objects.
+	 *
+	 * Note that due to how the hardware implements 64K GTT page support, we
+	 * have some further complications:
+	 *
+	 *   1.) The entire PDE(which covers a 2M virtual address range), must
+	 *   contain only 64K PTEs, i.e mixing 4K and 64K PTEs in the same
+	 *   PDE is forbidden by the hardware.
+	 *
+	 *   2.) We still need to support 4K PTEs for I915_MEMORY_CLASS_SYSTEM
+	 *   objects.
+	 *
+	 * To handle the above the kernel implements a memory coloring scheme to
+	 * prevent userspace from mixing I915_MEMORY_CLASS_DEVICE and
+	 * I915_MEMORY_CLASS_SYSTEM objects in the same PDE. If the kernel is
+	 * ever unable to evict the required pages for the given PDE(different
+	 * color) when inserting the object into the GTT then it will simply
+	 * fail the request.
+	 *
+	 * Since userspace needs to manage the GTT address space themselves,
+	 * special care is needed to ensure this doesn't happen. The simplest
+	 * scheme is to simply align and round up all I915_MEMORY_CLASS_DEVICE
+	 * objects to 2M, which avoids any issues here. At the very least this
+	 * is likely needed for objects that can be placed in both
+	 * I915_MEMORY_CLASS_DEVICE and I915_MEMORY_CLASS_SYSTEM, to avoid
+	 * potential issues when the kernel needs to migrate the object behind
+	 * the scenes, since that might also involve evicting other objects.
+	 *
+	 * To summarise the GTT rules, on platforms like DG2:
+	 *
+	 *   1.) All objects that can be placed in I915_MEMORY_CLASS_DEVICE must
+	 *   have 64K alignment. The kernel will reject this otherwise.
+	 *
+	 *   2.) All I915_MEMORY_CLASS_DEVICE objects must never be placed in
+	 *   the same PDE with other I915_MEMORY_CLASS_SYSTEM objects. The
+	 *   kernel will reject this otherwise.
+	 *
+	 *   3.) Objects that can be placed in both I915_MEMORY_CLASS_DEVICE and
+	 *   I915_MEMORY_CLASS_SYSTEM should probably be aligned and padded out
+	 *   to 2M.
 	 */
 	__u64 size;
 	/**
-- 
2.20.1


  parent reply	other threads:[~2021-10-11 16:10 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-11 16:11 [PATCH 00/14] drm/i915/dg2: Enabling 64k page size and flat ccs Ramalingam C
2021-10-11 16:11 ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 01/14] drm/i915: Add has_64k_pages flag Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 02/14] drm/i915/xehpsdv: set min page-size to 64K Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 03/14] drm/i915/xehpsdv: enforce min GTT alignment Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-13 13:38   ` Daniel Vetter
2021-10-13 13:38     ` [Intel-gfx] " Daniel Vetter
2021-10-13 14:13     ` Matthew Auld
2021-10-13 14:13       ` [Intel-gfx] " Matthew Auld
2021-10-14 13:33       ` Daniel Vetter
2021-10-14 13:33         ` [Intel-gfx] " Daniel Vetter
2021-10-14 14:21         ` Matthew Auld
2021-10-14 14:21           ` [Intel-gfx] " Matthew Auld
2021-10-11 16:11 ` [PATCH 04/14] drm/i915: enforce min page size for scratch Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 05/14] drm/i915/gtt/xehpsdv: move scratch page to system memory Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 06/14] drm/i915/xehpsdv: support 64K GTT pages Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 07/14] drm/i915: Add vm min alignment support Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 08/14] drm/i915/selftests: account for min_alignment in GTT selftests Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 09/14] drm/i915/xehpsdv: implement memory coloring Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 10/14] drm/i915/xehpsdv: Add has_flat_ccs to device info Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 11/14] drm/i915/lmem: Enable lmem for platforms with Flat CCS Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` [PATCH 12/14] drm/i915/gt: Clear compress metadata for Gen12.5 >= platforms Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 16:11 ` Ramalingam C [this message]
2021-10-11 16:11   ` [Intel-gfx] [PATCH 13/14] drm/i915/uapi: document behaviour for DG2 64K support Ramalingam C
2021-10-13 13:46   ` Daniel Vetter
2021-10-13 13:46     ` [Intel-gfx] " Daniel Vetter
2021-10-11 16:11 ` [PATCH 14/14] Doc/gpu/rfc/i915: i915 DG2 uAPI Ramalingam C
2021-10-11 16:11   ` [Intel-gfx] " Ramalingam C
2021-10-11 17:08   ` Tang, CQ
2021-10-11 17:08     ` [Intel-gfx] " Tang, CQ
2021-10-12  5:23   ` Lucas De Marchi
2021-10-12  5:23     ` [Intel-gfx] " Lucas De Marchi
2021-10-13 13:50   ` Daniel Vetter
2021-10-13 13:50     ` [Intel-gfx] " Daniel Vetter
2021-10-11 18:05 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915/dg2: Enabling 64k page size and flat ccs Patchwork
2021-10-11 18:06 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-10-11 18:34 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
2021-10-13 13:51 ` [PATCH 00/14] " Daniel Vetter
2021-10-13 13:51   ` [Intel-gfx] " Daniel Vetter

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=20211011161155.6397-14-ramalingam.c@intel.com \
    --to=ramalingam.c@intel.com \
    --cc=cq.tang@intel.com \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=matthew.auld@intel.com \
    --cc=thomas.hellstrom@intel.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.