mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Mukul Joshi <mukul.joshi@amd.com>,
	Felix Kuehling <felix.kuehling@amd.com>,
	Alex Deucher <alexander.deucher@amd.com>,
	Sasha Levin <sashal@kernel.org>,
	Felix.Kuehling@amd.com, christian.koenig@amd.com,
	Xinhui.Pan@amd.com, airlied@gmail.com, daniel@ffwll.ch,
	amd-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Subject: [PATCH AUTOSEL 6.6 10/19] drm/amdkfd: Add VRAM accounting for SVM migration
Date: Tue,  7 May 2024 18:58:32 -0400	[thread overview]
Message-ID: <20240507225910.390914-10-sashal@kernel.org> (raw)
In-Reply-To: <20240507225910.390914-1-sashal@kernel.org>

From: Mukul Joshi <mukul.joshi@amd.com>

[ Upstream commit 1e214f7faaf5d842754cd5cfcd76308bfedab3b5 ]

Do VRAM accounting when doing migrations to vram to make sure
there is enough available VRAM and migrating to VRAM doesn't evict
other possible non-unified memory BOs. If migrating to VRAM fails,
driver can fall back to using system memory seamlessly.

Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/amd/amdkfd/kfd_migrate.c | 16 +++++++++++++++-
 drivers/gpu/drm/amd/amdkfd/kfd_svm.c     |  2 +-
 2 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
index 659313648b200..3263b5fa182d2 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c
@@ -516,10 +516,19 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 	start = prange->start << PAGE_SHIFT;
 	end = (prange->last + 1) << PAGE_SHIFT;
 
+	r = amdgpu_amdkfd_reserve_mem_limit(node->adev,
+					prange->npages * PAGE_SIZE,
+					KFD_IOC_ALLOC_MEM_FLAGS_VRAM,
+					node->xcp ? node->xcp->id : 0);
+	if (r) {
+		dev_dbg(node->adev->dev, "failed to reserve VRAM, r: %ld\n", r);
+		return -ENOSPC;
+	}
+
 	r = svm_range_vram_node_new(node, prange, true);
 	if (r) {
 		dev_dbg(node->adev->dev, "fail %ld to alloc vram\n", r);
-		return r;
+		goto out;
 	}
 	ttm_res_offset = prange->offset << PAGE_SHIFT;
 
@@ -549,6 +558,11 @@ svm_migrate_ram_to_vram(struct svm_range *prange, uint32_t best_loc,
 		svm_range_vram_node_free(prange);
 	}
 
+out:
+	amdgpu_amdkfd_unreserve_mem_limit(node->adev,
+					prange->npages * PAGE_SIZE,
+					KFD_IOC_ALLOC_MEM_FLAGS_VRAM,
+					node->xcp ? node->xcp->id : 0);
 	return r < 0 ? r : 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 87e9ca65e58e0..ce76d45549984 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -3416,7 +3416,7 @@ svm_range_trigger_migration(struct mm_struct *mm, struct svm_range *prange,
 	r = svm_migrate_to_vram(prange, best_loc, mm, KFD_MIGRATE_TRIGGER_PREFETCH);
 	*migrated = !r;
 
-	return r;
+	return 0;
 }
 
 int svm_range_schedule_evict_svm_bo(struct amdgpu_amdkfd_fence *fence)
-- 
2.43.0


  parent reply	other threads:[~2024-05-07 22:59 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-05-07 22:58 [PATCH AUTOSEL 6.6 01/19] wifi: mac80211: don't use rate mask for scanning Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 02/19] wifi: mac80211: ensure beacon is non-S1G prior to extracting the beacon timestamp field Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 03/19] wifi: cfg80211: fix the order of arguments for trace events of the tx_rx_evt class Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 04/19] dt-bindings: rockchip: grf: Add missing type to 'pcie-phy' node Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 05/19] HID: mcp-2221: cancel delayed_work only when CONFIG_IIO is enabled Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 06/19] net: usb: qmi_wwan: add Telit FN920C04 compositions Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 07/19] drm/amd/display: Set color_mgmt_changed to true on unsuspend Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 08/19] drm/amdgpu: Update BO eviction priorities Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 09/19] drm/amd/pm: Restore config space after reset Sasha Levin
2024-05-07 22:58 ` Sasha Levin [this message]
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 11/19] drm/amdgpu: Fix the ring buffer size for queue VM flush Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 12/19] drm/amdgpu/mes: fix use-after-free issue Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 13/19] Revert "net: txgbe: fix i2c dev name cannot match clkdev" Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 14/19] Revert "net: txgbe: fix clk_name exceed MAX_DEV_ID limits" Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 15/19] cpu: Ignore "mitigations" kernel parameter if CPU_MITIGATIONS=n Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 16/19] LoongArch: Lately init pmu after smp is online Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 17/19] drm/etnaviv: fix tx clock gating on some GC7000 variants Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 18/19] selftests: sud_test: return correct emulated syscall value on RISC-V Sasha Levin
2024-05-07 22:58 ` [PATCH AUTOSEL 6.6 19/19] sched/isolation: Fix boot crash when maxcpus < first housekeeping CPU 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=20240507225910.390914-10-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=Xinhui.Pan@amd.com \
    --cc=airlied@gmail.com \
    --cc=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=christian.koenig@amd.com \
    --cc=daniel@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=felix.kuehling@amd.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mukul.joshi@amd.com \
    --cc=stable@vger.kernel.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

all inboxes | Powered by JetHome®