diff options
| author |   Jason Gunthorpe <jgg@mellanox.com> | 2020-05-01 15:20:45 -0300 |
|---|---|---|
| committer | Jason Gunthorpe <jgg@mellanox.com> | 2020-05-11 10:47:29 -0300 |
| commit | be957c886d92aa9caf0f63aee2c77d1497217d93 (patch) | |
| tree | 945037adedcbcdbf4f3f6e309334d1c0eb583e3f /drivers/gpu/drm/nouveau/nouveau_svm.c | |
| parent | Linux 5.7-rc4 (diff) | |
| download | linux-dev-be957c886d92aa9caf0f63aee2c77d1497217d93.tar.xz linux-dev-be957c886d92aa9caf0f63aee2c77d1497217d93.zip | |
mm/hmm: make hmm_range_fault return 0 or -1
hmm_vma_walk->last is supposed to be updated after every write to the
pfns, so that it can be returned by hmm_range_fault(). However, this is
not done consistently. Fortunately nothing checks the return code of
hmm_range_fault() for anything other than error.
More importantly last must be set before returning -EBUSY as it is used to
prevent reading an output pfn as an input flags when the loop restarts.
For clarity and simplicity make hmm_range_fault() return 0 or -ERRNO. Only
set last when returning -EBUSY.
Link: https://lore.kernel.org/r/2-v2-b4e84f444c7d+24f57-hmm_no_flags_jgg@mellanox.com
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Tested-by: Ralph Campbell <rcampbell@nvidia.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Diffstat (limited to 'drivers/gpu/drm/nouveau/nouveau_svm.c')
| -rw-r--r-- | drivers/gpu/drm/nouveau/nouveau_svm.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index 645fedd77e21..c68e9317cf07 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -536,7 +536,7 @@ static int nouveau_range_fault(struct nouveau_svmm *svmm, .pfn_shift = NVIF_VMM_PFNMAP_V0_ADDR_SHIFT, }; struct mm_struct *mm = notifier->notifier.mm; - long ret; + int ret; while (true) { if (time_after(jiffies, timeout)) @@ -548,8 +548,8 @@ static int nouveau_range_fault(struct nouveau_svmm *svmm, down_read(&mm->mmap_sem); ret = hmm_range_fault(&range); up_read(&mm->mmap_sem); - if (ret <= 0) { - if (ret == 0 || ret == -EBUSY) + if (ret) { + if (ret == -EBUSY) continue; return ret; } |