intel-xe.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>
To: intel-xe@lists.freedesktop.org
Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
	dri-devel@lists.freedesktop.org, himal.prasad.ghimiray@intel.com,
	apopple@nvidia.com, airlied@gmail.com,
	"Simona Vetter" <simona.vetter@ffwll.ch>,
	felix.kuehling@amd.com, "Matthew Brost" <matthew.brost@intel.com>,
	"Christian König" <christian.koenig@amd.com>,
	dakr@kernel.org, "Mrozek, Michal" <michal.mrozek@intel.com>,
	"Joonas Lahtinen" <joonas.lahtinen@linux.intel.com>
Subject: [PATCH v2 14/17] drm/xe: Support pcie p2p dma as a fast interconnect
Date: Tue, 11 Nov 2025 17:44:04 +0100	[thread overview]
Message-ID: <20251111164408.113070-15-thomas.hellstrom@linux.intel.com> (raw)
In-Reply-To: <20251111164408.113070-1-thomas.hellstrom@linux.intel.com>

Mimic the dma-buf method using dma_[map|unmap]_resource to map
for pcie-p2p dma.

There's an ongoing area of work upstream to sort out how this best
should be done. One method proposed is to add an additional
pci_p2p_dma_pagemap aliasing the device_private pagemap and use
the corresponding pci_p2p_dma_pagemap page as input for
dma_map_page(). However, that would incur double the amount of
memory and latency to set up the drm_pagemap and given the huge
amount of memory present on modern GPUs, that would really not work.
Hence the simple approach used in this patch.

v2:
- Simplify xe_page_to_pcie(). (Matt Brost)

Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
---
 drivers/gpu/drm/xe/xe_svm.c | 34 +++++++++++++++++++++++++++++++---
 drivers/gpu/drm/xe/xe_svm.h |  1 +
 2 files changed, 32 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
index c0b17b548a00..86d7b0882b60 100644
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@ -3,6 +3,8 @@
  * Copyright © 2024 Intel Corporation
  */
 
+#include <linux/pci-p2pdma.h>
+
 #include <drm/drm_drv.h>
 #include <drm/drm_managed.h>
 #include <drm/drm_pagemap.h>
@@ -441,6 +443,14 @@ static u64 xe_page_to_dpa(struct page *page)
 	return dpa;
 }
 
+static u64 xe_page_to_pcie(struct page *page)
+{
+	struct xe_pagemap *xpagemap = xe_page_to_pagemap(page);
+	struct xe_vram_region *vr = xe_pagemap_to_vr(xpagemap);
+
+	return xe_page_to_dpa(page) - vr->dpa_base + vr->io_start;
+}
+
 enum xe_svm_copy_dir {
 	XE_SVM_COPY_TO_VRAM,
 	XE_SVM_COPY_TO_SRAM,
@@ -792,7 +802,10 @@ static bool xe_has_interconnect(struct drm_pagemap_peer *peer1,
 	struct device *dev1 = xe_peer_to_dev(peer1);
 	struct device *dev2 = xe_peer_to_dev(peer2);
 
-	return dev1 == dev2;
+	if (dev1 == dev2)
+		return true;
+
+	return pci_p2pdma_distance(to_pci_dev(dev1), dev2, true) >= 0;
 }
 
 static DRM_PAGEMAP_OWNER_LIST_DEFINE(xe_owner_list);
@@ -1552,13 +1565,27 @@ xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,
 		addr = xe_page_to_dpa(page);
 		prot = XE_INTERCONNECT_VRAM;
 	} else {
-		addr = DMA_MAPPING_ERROR;
-		prot = 0;
+		addr = dma_map_resource(dev,
+					xe_page_to_pcie(page),
+					PAGE_SIZE << order, dir,
+					DMA_ATTR_SKIP_CPU_SYNC);
+		prot = XE_INTERCONNECT_P2P;
 	}
 
 	return drm_pagemap_addr_encode(addr, prot, order, dir);
 }
 
+static void xe_drm_pagemap_device_unmap(struct drm_pagemap *dpagemap,
+					struct device *dev,
+					struct drm_pagemap_addr addr)
+{
+	if (addr.proto != XE_INTERCONNECT_P2P)
+		return;
+
+	dma_unmap_resource(dev, addr.addr, PAGE_SIZE << addr.order,
+			   addr.dir, DMA_ATTR_SKIP_CPU_SYNC);
+}
+
 static void xe_pagemap_destroy_work(struct work_struct *work)
 {
 	struct xe_pagemap *xpagemap = container_of(work, typeof(*xpagemap), destroy_work);
@@ -1595,6 +1622,7 @@ static void xe_pagemap_destroy(struct drm_pagemap *dpagemap, bool from_atomic_or
 
 static const struct drm_pagemap_ops xe_drm_pagemap_ops = {
 	.device_map = xe_drm_pagemap_device_map,
+	.device_unmap = xe_drm_pagemap_device_unmap,
 	.populate_mm = xe_drm_pagemap_populate_mm,
 	.destroy = xe_pagemap_destroy,
 };
diff --git a/drivers/gpu/drm/xe/xe_svm.h b/drivers/gpu/drm/xe/xe_svm.h
index 60eae01a4220..64971c9b2a1a 100644
--- a/drivers/gpu/drm/xe/xe_svm.h
+++ b/drivers/gpu/drm/xe/xe_svm.h
@@ -13,6 +13,7 @@
 #include <drm/drm_pagemap_util.h>
 
 #define XE_INTERCONNECT_VRAM DRM_INTERCONNECT_DRIVER
+#define XE_INTERCONNECT_P2P (XE_INTERCONNECT_VRAM + 1)
 
 struct drm_device;
 struct drm_file;
-- 
2.51.1


  parent reply	other threads:[~2025-11-11 16:45 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-11 16:43 [PATCH v2 00/17] Dynamic drm_pagemaps and Initial multi-device SVM Thomas Hellström
2025-11-11 16:43 ` [PATCH v2 01/17] drm/xe/svm: Fix a debug printout Thomas Hellström
2025-11-12  4:29   ` Ghimiray, Himal Prasad
2025-11-11 16:43 ` [PATCH v2 02/17] drm/pagemap, drm/xe: Add refcounting to struct drm_pagemap Thomas Hellström
2025-11-12  6:07   ` Ghimiray, Himal Prasad
2025-11-21 10:19     ` Thomas Hellström
2025-11-11 16:43 ` [PATCH v2 03/17] drm/pagemap: Add a refcounted drm_pagemap backpointer to struct drm_pagemap_zdd Thomas Hellström
2025-11-11 16:43 ` [PATCH v2 04/17] drm/pagemap, drm/xe: Manage drm_pagemap provider lifetimes Thomas Hellström
2025-11-18  0:44   ` Matthew Brost
2025-11-11 16:43 ` [PATCH v2 05/17] drm/pagemap: Add a drm_pagemap cache and shrinker Thomas Hellström
2025-11-19 19:28   ` Matthew Brost
2025-11-11 16:43 ` [PATCH v2 06/17] drm/xe: Use the " Thomas Hellström
2025-11-11 16:43 ` [PATCH v2 07/17] drm/pagemap: Remove the drm_pagemap_create() interface Thomas Hellström
2025-11-11 16:43 ` [PATCH v2 08/17] drm/pagemap_util: Add a utility to assign an owner to a set of interconnected gpus Thomas Hellström
2025-11-11 16:43 ` [PATCH v2 09/17] drm/xe: Use the drm_pagemap_util helper to get a svm pagemap owner Thomas Hellström
2025-11-11 16:44 ` [PATCH v2 10/17] drm/xe: Pass a drm_pagemap pointer around with the memory advise attributes Thomas Hellström
2025-11-11 16:44 ` [PATCH v2 11/17] drm/xe: Use the vma attibute drm_pagemap to select where to migrate Thomas Hellström
2025-11-12  5:22   ` kernel test robot
2025-11-12  7:16   ` kernel test robot
2025-11-13  4:51   ` kernel test robot
2025-11-11 16:44 ` [PATCH v2 12/17] drm/xe: Simplify madvise_preferred_mem_loc() Thomas Hellström
2025-11-11 16:44 ` [PATCH v2 13/17] drm/xe/uapi: Extend the madvise functionality to support foreign pagemap placement for svm Thomas Hellström
2025-11-11 16:44 ` Thomas Hellström [this message]
2025-11-11 16:44 ` [PATCH v2 15/17] drm/xe/vm: Add a couple of VM debug printouts Thomas Hellström
2025-11-11 16:44 ` [PATCH v2 16/17] drm/pagemap, drm/xe: Support migration over interconnect Thomas Hellström
2025-11-11 16:44 ` [PATCH v2 17/17] drm/xe/svm: Document how xe keeps drm_pagemap references Thomas Hellström
2025-11-18  0:49   ` Matthew Brost
2025-11-11 17:07 ` ✗ CI.checkpatch: warning for Dynamic drm_pagemaps and Initial multi-device SVM (rev2) Patchwork
2025-11-11 17:08 ` ✓ CI.KUnit: success " Patchwork
2025-11-11 17:45 ` ✓ Xe.CI.BAT: " Patchwork
2025-11-12  2:53 ` ✗ Xe.CI.Full: failure " Patchwork
2025-11-18  6:15 ` [PATCH v2 00/17] Dynamic drm_pagemaps and Initial multi-device SVM Alistair Popple
2025-11-18  9:31   ` Thomas Hellström

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=20251111164408.113070-15-thomas.hellstrom@linux.intel.com \
    --to=thomas.hellstrom@linux.intel.com \
    --cc=airlied@gmail.com \
    --cc=apopple@nvidia.com \
    --cc=christian.koenig@amd.com \
    --cc=dakr@kernel.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=felix.kuehling@amd.com \
    --cc=himal.prasad.ghimiray@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=joonas.lahtinen@linux.intel.com \
    --cc=matthew.brost@intel.com \
    --cc=michal.mrozek@intel.com \
    --cc=simona.vetter@ffwll.ch \
    /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;
as well as URLs for NNTP newsgroup(s).