All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: "Felix Kuehling" <felix.kuehling@amd.com>,
	"Hao Zhou" <hao.zhou@amd.com>,
	"Christian König" <christian.koenig@amd.com>,
	"Alex Deucher" <alexander.deucher@amd.com>,
	"Sasha Levin" <sashal@kernel.org>,
	airlied@gmail.com, simona@ffwll.ch, Yunxiang.Li@amd.com,
	tvrtko.ursulin@igalia.com, matthew.auld@intel.com,
	amd-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Subject: [PATCH AUTOSEL 6.6 18/21] drm/amdgpu: Allow P2P access through XGMI
Date: Tue, 29 Apr 2025 19:52:30 -0400	[thread overview]
Message-ID: <20250429235233.537828-18-sashal@kernel.org> (raw)
In-Reply-To: <20250429235233.537828-1-sashal@kernel.org>

From: Felix Kuehling <felix.kuehling@amd.com>

[ Upstream commit a92741e72f91b904c1d8c3d409ed8dbe9c1f2b26 ]

If peer memory is accessible through XGMI, allow leaving it in VRAM
rather than forcing its migration to GTT on DMABuf attachment.

Signed-off-by: Felix Kuehling <felix.kuehling@amd.com>
Tested-by: Hao (Claire) Zhou <hao.zhou@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 372c8d72c3680fdea3fbb2d6b089f76b4a6d596a)
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 30 ++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
index be4cc4868a748..493e18bcea069 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c
@@ -43,6 +43,29 @@
 #include <linux/pci-p2pdma.h>
 #include <linux/pm_runtime.h>
 
+static const struct dma_buf_attach_ops amdgpu_dma_buf_attach_ops;
+
+/**
+ * dma_buf_attach_adev - Helper to get adev of an attachment
+ *
+ * @attach: attachment
+ *
+ * Returns:
+ * A struct amdgpu_device * if the attaching device is an amdgpu device or
+ * partition, NULL otherwise.
+ */
+static struct amdgpu_device *dma_buf_attach_adev(struct dma_buf_attachment *attach)
+{
+	if (attach->importer_ops == &amdgpu_dma_buf_attach_ops) {
+		struct drm_gem_object *obj = attach->importer_priv;
+		struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+
+		return amdgpu_ttm_adev(bo->tbo.bdev);
+	}
+
+	return NULL;
+}
+
 /**
  * amdgpu_dma_buf_attach - &dma_buf_ops.attach implementation
  *
@@ -54,12 +77,14 @@
 static int amdgpu_dma_buf_attach(struct dma_buf *dmabuf,
 				 struct dma_buf_attachment *attach)
 {
+	struct amdgpu_device *attach_adev = dma_buf_attach_adev(attach);
 	struct drm_gem_object *obj = dmabuf->priv;
 	struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
 	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	int r;
 
-	if (pci_p2pdma_distance(adev->pdev, attach->dev, false) < 0)
+	if (!amdgpu_dmabuf_is_xgmi_accessible(attach_adev, bo) &&
+	    pci_p2pdma_distance(adev->pdev, attach->dev, false) < 0)
 		attach->peer2peer = false;
 
 	r = pm_runtime_get_sync(adev_to_drm(adev)->dev);
@@ -482,6 +507,9 @@ bool amdgpu_dmabuf_is_xgmi_accessible(struct amdgpu_device *adev,
 	struct drm_gem_object *obj = &bo->tbo.base;
 	struct drm_gem_object *gobj;
 
+	if (!adev)
+		return false;
+
 	if (obj->import_attach) {
 		struct dma_buf *dma_buf = obj->import_attach->dmabuf;
 
-- 
2.39.5


  parent reply	other threads:[~2025-04-29 23:53 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-29 23:52 [PATCH AUTOSEL 6.6 01/21] cpufreq: Add SM8650 to cpufreq-dt-platdev blocklist Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 02/21] nvmem: rockchip-otp: Move read-offset into variant-data Sasha Levin
2025-04-29 23:52   ` Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 03/21] nvmem: rockchip-otp: add rk3576 variant data Sasha Levin
2025-04-29 23:52   ` Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 04/21] nvmem: core: verify cell's raw_len Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 05/21] nvmem: core: update raw_len if the bit reading is required Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 06/21] nvmem: qfprom: switch to 4-byte aligned reads Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 07/21] scsi: target: iscsi: Fix timeout on deleted connection Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 08/21] scsi: ufs: Introduce quirk to extend PA_HIBERN8TIME for UFS devices Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 09/21] virtio_ring: Fix data race by tagging event_triggered as racy for KCSAN Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 10/21] intel_th: avoid using deprecated page->mapping, index fields Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 11/21] dma-mapping: avoid potential unused data compilation warning Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 12/21] cgroup: Fix compilation issue due to cgroup_mutex not being exported Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 13/21] vhost_task: fix vhost_task_create() documentation Sasha Levin
2025-04-30  9:10   ` Stefano Garzarella
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 14/21] vhost-scsi: protect vq->log_used with vq->mutex Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 15/21] scsi: mpi3mr: Add level check to control event logging Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 16/21] net: enetc: refactor bulk flipping of RX buffers to separate function Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 17/21] ima: process_measurement() needlessly takes inode_lock() on MAY_READ Sasha Levin
2025-04-29 23:52 ` Sasha Levin [this message]
2025-04-30 12:57   ` [PATCH AUTOSEL 6.6 18/21] drm/amdgpu: Allow P2P access through XGMI Alex Deucher
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 19/21] selftests/bpf: Mitigate sockmap_ktls disconnect_after_delete failure Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 20/21] bpf: fix possible endless loop in BPF map iteration Sasha Levin
2025-04-29 23:52 ` [PATCH AUTOSEL 6.6 21/21] samples/bpf: Fix compilation failure for samples/bpf on LoongArch Fedora Sasha Levin

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=20250429235233.537828-18-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=Yunxiang.Li@amd.com \
    --cc=airlied@gmail.com \
    --cc=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=christian.koenig@amd.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=felix.kuehling@amd.com \
    --cc=hao.zhou@amd.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matthew.auld@intel.com \
    --cc=simona@ffwll.ch \
    --cc=stable@vger.kernel.org \
    --cc=tvrtko.ursulin@igalia.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.