From: Philip Yang <Philip.Yang@amd.com>
To: <amd-gfx@lists.freedesktop.org>
Cc: Philip Yang <Philip.Yang@amd.com>
Subject: [PATCH 2/2] drm/amdkfd: fix dma mapping leaking warning
Date: Thu, 16 Sep 2021 20:50:37 -0400 [thread overview]
Message-ID: <20210917005037.18526-2-Philip.Yang@amd.com> (raw)
In-Reply-To: <20210917005037.18526-1-Philip.Yang@amd.com>
For xnack off, restore work dma unmap previous system memory page, and
dma map the updated system memory page to update GPU mapping, this is
not dma mapping leaking, remove the WARN_ONCE for dma mapping leaking.
prange->dma_addr store the VRAM page pfn after the range migrated to
VRAM, should not dma unmap VRAM page when updating GPU mapping or
remove prange. Add helper svm_is_valid_dma_mapping_addr to check VRAM
page and error cases.
Keep SVM_RANGE_VRAM_DOMAIN flag in dma_addr to diff VRAM page and dma
map system page, amdgpu vm update will mask out this bit before adding
flags to update PTE.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
---
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 57318edc4020..763c74d60c8e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -118,6 +118,13 @@ static void svm_range_remove_notifier(struct svm_range *prange)
mmu_interval_notifier_remove(&prange->notifier);
}
+static bool
+svm_is_valid_dma_mapping_addr(struct device *dev, dma_addr_t dma_addr)
+{
+ return dma_addr && !dma_mapping_error(dev, dma_addr) &&
+ !(dma_addr & SVM_RANGE_VRAM_DOMAIN);
+}
+
static int
svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
unsigned long offset, unsigned long npages,
@@ -139,8 +146,7 @@ svm_range_dma_map_dev(struct amdgpu_device *adev, struct svm_range *prange,
addr += offset;
for (i = 0; i < npages; i++) {
- if (WARN_ONCE(addr[i] && !dma_mapping_error(dev, addr[i]),
- "leaking dma mapping\n"))
+ if (svm_is_valid_dma_mapping_addr(dev, addr[i]))
dma_unmap_page(dev, addr[i], PAGE_SIZE, dir);
page = hmm_pfn_to_page(hmm_pfns[i]);
@@ -209,7 +215,7 @@ void svm_range_dma_unmap(struct device *dev, dma_addr_t *dma_addr,
return;
for (i = offset; i < offset + npages; i++) {
- if (!dma_addr[i] || dma_mapping_error(dev, dma_addr[i]))
+ if (!svm_is_valid_dma_mapping_addr(dev, dma_addr[i]))
continue;
pr_debug("dma unmapping 0x%llx\n", dma_addr[i] >> PAGE_SHIFT);
dma_unmap_page(dev, dma_addr[i], PAGE_SIZE, dir);
@@ -1177,7 +1183,6 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm,
for (i = offset; i < offset + npages; i++) {
last_domain = dma_addr[i] & SVM_RANGE_VRAM_DOMAIN;
- dma_addr[i] &= ~SVM_RANGE_VRAM_DOMAIN;
/* Collect all pages in the same address range and memory domain
* that can be mapped with a single call to update mapping.
--
2.17.1
next prev parent reply other threads:[~2021-09-17 0:51 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-17 0:50 [PATCH 1/2] drm/amdgpu: mask out invalid PTE address bit Philip Yang
2021-09-17 0:50 ` Philip Yang [this message]
2021-09-17 6:23 ` Christian König
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=20210917005037.18526-2-Philip.Yang@amd.com \
--to=philip.yang@amd.com \
--cc=amd-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