Nouveau - Jul 2019 - hmm_range_fault related fixes and legacy API removal v2

Hi Jérôme, Ben and Jason,

below is a series against the hmm tree which fixes up the mmap_sem
locking in nouveau and while at it also removes leftover legacy HMM APIs
only used by nouveau.

Changes since v1:
 - don't return the valid state from hmm_range_unregister
 - additional nouveau cleanups

We should not have two different error codes for the same condition.  In
addition this really complicates the code due to the special handling of
EAGAIN that drops the mmap_sem due to the FAULT_FLAG_ALLOW_RETRY logic
in the core vm.

Signed-off-by: Christoph Hellwig <hch at lst.de>
Reviewed-by: Ralph Campbell <rcampbell at nvidia.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling at amd.com>
---
 Documentation/vm/hmm.rst |  2 +-
 mm/hmm.c                 | 10 ++++------
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/Documentation/vm/hmm.rst b/Documentation/vm/hmm.rst
index 7d90964abbb0..710ce1c701bf 100644
--- a/Documentation/vm/hmm.rst
+++ b/Documentation/vm/hmm.rst
@@ -237,7 +237,7 @@ The usage pattern is::
       ret = hmm_range_snapshot(&range);
       if (ret) {
           up_read(&mm->mmap_sem);
-          if (ret == -EAGAIN) {
+          if (ret == -EBUSY) {
             /*
              * No need to check hmm_range_wait_until_valid() return value
              * on retry we will get proper error with hmm_range_snapshot()
diff --git a/mm/hmm.c b/mm/hmm.c
index d48b9283725a..1d57c39c1d8b 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -946,7 +946,7 @@ EXPORT_SYMBOL(hmm_range_unregister);
  * @range: range
  * Return: -EINVAL if invalid argument, -ENOMEM out of memory, -EPERM invalid
  *          permission (for instance asking for write and range is read only),
- *          -EAGAIN if you need to retry, -EFAULT invalid (ie either no valid
+ *          -EBUSY if you need to retry, -EFAULT invalid (ie either no valid
  *          vma or it is illegal to access that range), number of valid pages
  *          in range->pfns[] (from range start address).
  *
@@ -967,7 +967,7 @@ long hmm_range_snapshot(struct hmm_range *range)
 	do {
 		/* If range is no longer valid force retry. */
 		if (!range->valid)
-			return -EAGAIN;
+			return -EBUSY;
 
 		vma = find_vma(hmm->mm, start);
 		if (vma == NULL || (vma->vm_flags & device_vma))
@@ -1062,10 +1062,8 @@ long hmm_range_fault(struct hmm_range *range, bool block)
 
 	do {
 		/* If range is no longer valid force retry. */
-		if (!range->valid) {
-			up_read(&hmm->mm->mmap_sem);
-			return -EAGAIN;
-		}
+		if (!range->valid)
+			return -EBUSY;
 
 		vma = find_vma(hmm->mm, start);
 		if (vma == NULL || (vma->vm_flags & device_vma))
-- 
2.20.1

These two functions are marked as a legacy APIs to get rid of, but seem
to suit the current nouveau flow.  Move it to the only user in
preparation for fixing a locking bug involving caller and callee.
All comments referring to the old API have been removed as this now
is a driver private helper.

Signed-off-by: Christoph Hellwig <hch at lst.de>
---
 drivers/gpu/drm/nouveau/nouveau_svm.c | 52 +++++++++++++++++++++-
 include/linux/hmm.h                   | 63 ---------------------------
 2 files changed, 50 insertions(+), 65 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c
b/drivers/gpu/drm/nouveau/nouveau_svm.c
index 8c92374afcf2..033a9241a14a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -36,6 +36,13 @@
 #include <linux/sort.h>
 #include <linux/hmm.h>
 
+/*
+ * When waiting for mmu notifiers we need some kind of time out otherwise we
+ * could potentialy wait for ever, 1000ms ie 1s sounds like a long time to
+ * wait already.
+ */
+#define NOUVEAU_RANGE_FAULT_TIMEOUT 1000
+
 struct nouveau_svm {
 	struct nouveau_drm *drm;
 	struct mutex mutex;
@@ -475,6 +482,47 @@ nouveau_svm_fault_cache(struct nouveau_svm *svm,
 		fault->inst, fault->addr, fault->access);
 }
 
+static inline bool nouveau_range_done(struct hmm_range *range)
+{
+	bool ret = hmm_range_valid(range);
+
+	hmm_range_unregister(range);
+	return ret;
+}
+
+static int
+nouveau_range_fault(struct hmm_mirror *mirror, struct hmm_range *range,
+		    bool block)
+{
+	long ret;
+
+	range->default_flags = 0;
+	range->pfn_flags_mask = -1UL;
+
+	ret = hmm_range_register(range, mirror,
+				 range->start, range->end,
+				 PAGE_SHIFT);
+	if (ret)
+		return (int)ret;
+
+	if (!hmm_range_wait_until_valid(range, NOUVEAU_RANGE_FAULT_TIMEOUT)) {
+		up_read(&range->vma->vm_mm->mmap_sem);
+		return -EAGAIN;
+	}
+
+	ret = hmm_range_fault(range, block);
+	if (ret <= 0) {
+		if (ret == -EBUSY || !ret) {
+			up_read(&range->vma->vm_mm->mmap_sem);
+			ret = -EBUSY;
+		} else if (ret == -EAGAIN)
+			ret = -EBUSY;
+		hmm_range_unregister(range);
+		return ret;
+	}
+	return 0;
+}
+
 static int
 nouveau_svm_fault(struct nvif_notify *notify)
 {
@@ -649,10 +697,10 @@ nouveau_svm_fault(struct nvif_notify *notify)
 		range.values = nouveau_svm_pfn_values;
 		range.pfn_shift = NVIF_VMM_PFNMAP_V0_ADDR_SHIFT;
 again:
-		ret = hmm_vma_fault(&svmm->mirror, &range, true);
+		ret = nouveau_range_fault(&svmm->mirror, &range, true);
 		if (ret == 0) {
 			mutex_lock(&svmm->mutex);
-			if (!hmm_vma_range_done(&range)) {
+			if (!nouveau_range_done(&range)) {
 				mutex_unlock(&svmm->mutex);
 				goto again;
 			}
diff --git a/include/linux/hmm.h b/include/linux/hmm.h
index b8a08b2a10ca..fa43a9f53833 100644
--- a/include/linux/hmm.h
+++ b/include/linux/hmm.h
@@ -475,69 +475,6 @@ long hmm_range_dma_unmap(struct hmm_range *range,
 			 dma_addr_t *daddrs,
 			 bool dirty);
 
-/*
- * HMM_RANGE_DEFAULT_TIMEOUT - default timeout (ms) when waiting for a range
- *
- * When waiting for mmu notifiers we need some kind of time out otherwise we
- * could potentialy wait for ever, 1000ms ie 1s sounds like a long time to
- * wait already.
- */
-#define HMM_RANGE_DEFAULT_TIMEOUT 1000
-
-/* This is a temporary helper to avoid merge conflict between trees. */
-static inline bool hmm_vma_range_done(struct hmm_range *range)
-{
-	bool ret = hmm_range_valid(range);
-
-	hmm_range_unregister(range);
-	return ret;
-}
-
-/* This is a temporary helper to avoid merge conflict between trees. */
-static inline int hmm_vma_fault(struct hmm_mirror *mirror,
-				struct hmm_range *range, bool block)
-{
-	long ret;
-
-	/*
-	 * With the old API the driver must set each individual entries with
-	 * the requested flags (valid, write, ...). So here we set the mask to
-	 * keep intact the entries provided by the driver and zero out the
-	 * default_flags.
-	 */
-	range->default_flags = 0;
-	range->pfn_flags_mask = -1UL;
-
-	ret = hmm_range_register(range, mirror,
-				 range->start, range->end,
-				 PAGE_SHIFT);
-	if (ret)
-		return (int)ret;
-
-	if (!hmm_range_wait_until_valid(range, HMM_RANGE_DEFAULT_TIMEOUT)) {
-		/*
-		 * The mmap_sem was taken by driver we release it here and
-		 * returns -EAGAIN which correspond to mmap_sem have been
-		 * drop in the old API.
-		 */
-		up_read(&range->vma->vm_mm->mmap_sem);
-		return -EAGAIN;
-	}
-
-	ret = hmm_range_fault(range, block);
-	if (ret <= 0) {
-		if (ret == -EBUSY || !ret) {
-			/* Same as above, drop mmap_sem to match old API. */
-			up_read(&range->vma->vm_mm->mmap_sem);
-			ret = -EBUSY;
-		} else if (ret == -EAGAIN)
-			ret = -EBUSY;
-		hmm_range_unregister(range);
-		return ret;
-	}
-	return 0;
-}
-
 /* Below are for HMM internal use only! Not to be used by device driver! */
 static inline void hmm_mm_init(struct mm_struct *mm)
 {
-- 
2.20.1

The parameter is always false, so remove it as well as the -EAGAIN
handling that can only happen for the non-blocking case.

Signed-off-by: Christoph Hellwig <hch at lst.de>
---
 drivers/gpu/drm/nouveau/nouveau_svm.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c
b/drivers/gpu/drm/nouveau/nouveau_svm.c
index 033a9241a14a..9a9f71e4be29 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -491,8 +491,7 @@ static inline bool nouveau_range_done(struct hmm_range
*range)
 }
 
 static int
-nouveau_range_fault(struct hmm_mirror *mirror, struct hmm_range *range,
-		    bool block)
+nouveau_range_fault(struct hmm_mirror *mirror, struct hmm_range *range)
 {
 	long ret;
 
@@ -510,7 +509,7 @@ nouveau_range_fault(struct hmm_mirror *mirror, struct
hmm_range *range,
 		return -EAGAIN;
 	}
 
-	ret = hmm_range_fault(range, block);
+	ret = hmm_range_fault(range, true);
 	if (ret <= 0) {
 		if (ret == -EBUSY || !ret) {
 			up_read(&range->vma->vm_mm->mmap_sem);
@@ -697,7 +696,7 @@ nouveau_svm_fault(struct nvif_notify *notify)
 		range.values = nouveau_svm_pfn_values;
 		range.pfn_shift = NVIF_VMM_PFNMAP_V0_ADDR_SHIFT;
 again:
-		ret = nouveau_range_fault(&svmm->mirror, &range, true);
+		ret = nouveau_range_fault(&svmm->mirror, &range);
 		if (ret == 0) {
 			mutex_lock(&svmm->mutex);
 			if (!nouveau_range_done(&range)) {
-- 
2.20.1

Currently nouveau_svm_fault expects nouveau_range_fault to never unlock
mmap_sem, but the latter unlocks it for a random selection of error
codes. Fix this up by always unlocking mmap_sem for non-zero return
values in nouveau_range_fault, and only unlocking it in the caller
for successful returns.

Signed-off-by: Christoph Hellwig <hch at lst.de>
---
 drivers/gpu/drm/nouveau/nouveau_svm.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c
b/drivers/gpu/drm/nouveau/nouveau_svm.c
index 9a9f71e4be29..d97d862e8b7d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -501,8 +501,10 @@ nouveau_range_fault(struct hmm_mirror *mirror, struct
hmm_range *range)
 	ret = hmm_range_register(range, mirror,
 				 range->start, range->end,
 				 PAGE_SHIFT);
-	if (ret)
+	if (ret) {
+		up_read(&range->vma->vm_mm->mmap_sem);
 		return (int)ret;
+	}
 
 	if (!hmm_range_wait_until_valid(range, NOUVEAU_RANGE_FAULT_TIMEOUT)) {
 		up_read(&range->vma->vm_mm->mmap_sem);
@@ -511,11 +513,9 @@ nouveau_range_fault(struct hmm_mirror *mirror, struct
hmm_range *range)
 
 	ret = hmm_range_fault(range, true);
 	if (ret <= 0) {
-		if (ret == -EBUSY || !ret) {
-			up_read(&range->vma->vm_mm->mmap_sem);
-			ret = -EBUSY;
-		} else if (ret == -EAGAIN)
+		if (ret == 0)
 			ret = -EBUSY;
+		up_read(&range->vma->vm_mm->mmap_sem);
 		hmm_range_unregister(range);
 		return ret;
 	}
@@ -713,8 +713,8 @@ nouveau_svm_fault(struct nvif_notify *notify)
 						NULL);
 			svmm->vmm->vmm.object.client->super = false;
 			mutex_unlock(&svmm->mutex);
+			up_read(&svmm->mm->mmap_sem);
 		}
-		up_read(&svmm->mm->mmap_sem);
 
 		/* Cancel any faults in the window whose pages didn't manage
 		 * to keep their valid bit, or stay writeable when required.
-- 
2.20.1

-EAGAIN has a magic meaning for non-blocking faults, so don't overload
it.  Given that the caller doesn't check for specific error codes this
change is purely cosmetic.

Signed-off-by: Christoph Hellwig <hch at lst.de>
---
 drivers/gpu/drm/nouveau/nouveau_svm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c
b/drivers/gpu/drm/nouveau/nouveau_svm.c
index d97d862e8b7d..26853f59b0f4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -508,7 +508,7 @@ nouveau_range_fault(struct hmm_mirror *mirror, struct
hmm_range *range)
 
 	if (!hmm_range_wait_until_valid(range, NOUVEAU_RANGE_FAULT_TIMEOUT)) {
 		up_read(&range->vma->vm_mm->mmap_sem);
-		return -EAGAIN;
+		return -EBUSY;
 	}
 
 	ret = hmm_range_fault(range, true);
-- 
2.20.1

Switch the one remaining user in nouveau over to its replacement,
and remove all the wrappers.

Signed-off-by: Christoph Hellwig <hch at lst.de>
Reviewed-by: Ralph Campbell <rcampbell at nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg at mellanox.com>
---
 drivers/gpu/drm/nouveau/nouveau_dmem.c |  2 +-
 include/linux/hmm.h                    | 34 --------------------------
 2 files changed, 1 insertion(+), 35 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c
b/drivers/gpu/drm/nouveau/nouveau_dmem.c
index 42c026010938..b9ced2e61667 100644
--- a/drivers/gpu/drm/nouveau/nouveau_dmem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c
@@ -844,7 +844,7 @@ nouveau_dmem_convert_pfn(struct nouveau_drm *drm,
 		struct page *page;
 		uint64_t addr;
 
-		page = hmm_pfn_to_page(range, range->pfns[i]);
+		page = hmm_device_entry_to_page(range, range->pfns[i]);
 		if (page == NULL)
 			continue;
 
diff --git a/include/linux/hmm.h b/include/linux/hmm.h
index fa43a9f53833..bf641bbecc7e 100644
--- a/include/linux/hmm.h
+++ b/include/linux/hmm.h
@@ -290,40 +290,6 @@ static inline uint64_t hmm_device_entry_from_pfn(const
struct hmm_range *range,
 		range->flags[HMM_PFN_VALID];
 }
 
-/*
- * Old API:
- * hmm_pfn_to_page()
- * hmm_pfn_to_pfn()
- * hmm_pfn_from_page()
- * hmm_pfn_from_pfn()
- *
- * This are the OLD API please use new API, it is here to avoid cross-tree
- * merge painfullness ie we convert things to new API in stages.
- */
-static inline struct page *hmm_pfn_to_page(const struct hmm_range *range,
-					   uint64_t pfn)
-{
-	return hmm_device_entry_to_page(range, pfn);
-}
-
-static inline unsigned long hmm_pfn_to_pfn(const struct hmm_range *range,
-					   uint64_t pfn)
-{
-	return hmm_device_entry_to_pfn(range, pfn);
-}
-
-static inline uint64_t hmm_pfn_from_page(const struct hmm_range *range,
-					 struct page *page)
-{
-	return hmm_device_entry_from_page(range, page);
-}
-
-static inline uint64_t hmm_pfn_from_pfn(const struct hmm_range *range,
-					unsigned long pfn)
-{
-	return hmm_device_entry_from_pfn(range, pfn);
-}
-
 /*
  * Mirroring: how to synchronize device page table with CPU page table.
  *
-- 
2.20.1

On Wed, Jul 03, 2019 at 03:02:08PM -0700, Christoph Hellwig
wrote:
> Hi Jérôme, Ben and Jason,
> 
> below is a series against the hmm tree which fixes up the mmap_sem
> locking in nouveau and while at it also removes leftover legacy HMM APIs
> only used by nouveau.
> 
> Changes since v1:
>  - don't return the valid state from hmm_range_unregister
>  - additional nouveau cleanups
Ralph, since most of this is nouveau could you contribute a
Tested-by? Thanks

Jason

On Wed, Jul 03, 2019 at 03:02:08PM -0700, Christoph Hellwig
wrote:
> Hi Jérôme, Ben and Jason,
> 
> below is a series against the hmm tree which fixes up the mmap_sem
> locking in nouveau and while at it also removes leftover legacy HMM APIs
> only used by nouveau.
As much as I like this series, it won't make it to this merge window,
sorry.

Let's revisit it in a few weeks at rc1.

Regards,
Jason

On 7/4/19 9:42 AM, Jason Gunthorpe wrote:
> On Wed, Jul 03, 2019 at 03:02:08PM -0700, Christoph Hellwig wrote:
>> Hi Jérôme, Ben and Jason,
>>
>> below is a series against the hmm tree which fixes up the mmap_sem
>> locking in nouveau and while at it also removes leftover legacy HMM
APIs
>> only used by nouveau.
>>
>> Changes since v1:
>>   - don't return the valid state from hmm_range_unregister
>>   - additional nouveau cleanups
> 
> Ralph, since most of this is nouveau could you contribute a
> Tested-by? Thanks
> 
> Jason
> 
I can test things fairly easily but with all the different patches,
conflicts, and personal git trees, can you specify the git tree
and branch with everything applied that you want me to test?

On Fri, Jul 05, 2019 at 09:33:36AM -0300, Jason Gunthorpe
wrote:
> On Wed, Jul 03, 2019 at 03:02:08PM -0700, Christoph Hellwig wrote:
> > Hi Jérôme, Ben and Jason,
> > 
> > below is a series against the hmm tree which fixes up the mmap_sem
> > locking in nouveau and while at it also removes leftover legacy HMM
APIs
> > only used by nouveau.
> 
> As much as I like this series, it won't make it to this merge window,
> sorry.
Note that patch 4 fixes a pretty severe locking bug, and 1-3 is just
preparation for that.