Skip to contentSkip to conversation

mm/hmm: A fix and a selftest

PATCH Superseded Dev Jain 5 participants 13 messages May 30 – Jun 1

Subsystems (8)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Dev Jain

author
#0 lore raw

v2 cover mm/hmm: A fix and a selftest

Patch 1 fixes a stale warning present from the time when only migration softleaf entries were supported at the PMD level.

Patch 2 adds some code into hmm-tests.c which exercises the pagemap path for PMD device-private entries.

---
Applies on mm-unstable (404fb4f38e8f).

Dev Jain (2):
fs/proc/task_mmu: do not warn on seeing non-migration pmd entry selftests/mm/hmm-tests: test pagemap reads of PMD device-private
entries

 fs/proc/task_mmu.c                     |  1 -
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 2 files changed, 29 insertions(+), 1 deletion(-)
Signature · 1 line
2.43.0
author
#1 lore raw

PATCH v2 1/2 fs/proc/task_mmu: do not warn on seeing non-migration pmd entry

pagemap_pmd_range_thp() warns if a non-present PMD is not a migration entry. This became false once device-private entries at the PMD level were added.

Therefore, remove the stale migration-only assertion.

Fixes a30b48bf1b24 ("mm/migrate_device: implement THP migration of zone device pages")
---
 fs/proc/task_mmu.c | 1 -
 1 file changed, 1 deletion(-)
fs/proc/task_mmu.c +0 -1
index 1e3a15bf46f4e..58938e62154d9 100644
@@ -2129,7 +2129,6 @@ static int pagemap_pmd_range_thp(pmd_t *pmdp, unsigned long addr,
 			flags |= PM_SOFT_DIRTY;
 		if (pmd_swp_uffd_wp(pmd))
 			flags |= PM_UFFD_WP;
-		VM_WARN_ON_ONCE(!pmd_is_migration_entry(pmd));
 		page = softleaf_to_page(entry);
 	}
 
Signature · 1 line
2.43.0
Reviewed-by on v2 1/2
#2 lore raw

Re: [PATCH v2 1/2] fs/proc/task_mmu: do not warn on seeing non-migration pmd entry

Quoted 18 lines · Dev Jain

pagemap_pmd_range_thp() warns if a non-present PMD is not a migration entry. This became false once device-private entries at the PMD level were added.

Therefore, remove the stale migration-only assertion.

Fixes a30b48bf1b24 ("mm/migrate_device: implement THP migration of zone device pages")
---
 fs/proc/task_mmu.c | 1 -
 1 file changed, 1 deletion(-)
fs/proc/task_mmu.c +0 -0
index 1e3a15bf46f4e..58938e62154d9 100644

fs/proc/task_mmu.c static int pagemap_pmd_range_thp(pmd_t *pmdp, unsigned long addr, · line 2134

Quoted 7 lines
fs/proc/task_mmu.c +0 -1
@@ -2129,7 +2129,6 @@ static int pagemap_pmd_range_thp(pmd_t *pmdp, unsigned long addr,
 			flags |= PM_SOFT_DIRTY;
 		if (pmd_swp_uffd_wp(pmd))
 			flags |= PM_UFFD_WP;
-		VM_WARN_ON_ONCE(!pmd_is_migration_entry(pmd));
 		page = softleaf_to_page(entry);
 	}
 

Thanks!

Balbir

maintainer Reviewed-by Tested-by on v2 1/2
#3 lore raw

Re: [PATCH v2 1/2] fs/proc/task_mmu: do not warn on seeing non-migration pmd entry

In reply to Dev Jain
Quoted 9 lines · Dev Jain

pagemap_pmd_range_thp() warns if a non-present PMD is not a migration entry. This became false once device-private entries at the PMD level were added.

Therefore, remove the stale migration-only assertion.

Fixes a30b48bf1b24 ("mm/migrate_device: implement THP migration of zone device pages")

LGTM so:

I also verified that I could trigger this locally with your test, so feel free to add:

Cheers, Lorenzo

Quoted 19 lines
---
 fs/proc/task_mmu.c | 1 -
 1 file changed, 1 deletion(-)
fs/proc/task_mmu.c +0 -2
index 1e3a15bf46f4e..58938e62154d9 100644
@@ -2129,7 +2129,6 @@ static int pagemap_pmd_range_thp(pmd_t *pmdp, unsigned long addr,
 			flags |= PM_SOFT_DIRTY;
 		if (pmd_swp_uffd_wp(pmd))
 			flags |= PM_UFFD_WP;
-		VM_WARN_ON_ONCE(!pmd_is_migration_entry(pmd));
 		page = softleaf_to_page(entry);
 	}
 
--

2.43.0

author
#4 lore raw

PATCH v2 2/2 selftests/mm/hmm-tests: test pagemap reads of PMD device-private entries

In reply to Dev Jain

To cover pagemap paths scanning PMD entries, add assertions to check whether a device-private PMD entry has the correct pagemap information - the PM_SWAP bit must be on in the pagemap entry. Before that, we must assert through HMM_DMIRROR_SNAPSHOT snapshot that the leaf entry is at PMD level and not PTE level.

---
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
tools/testing/selftests/mm/hmm-tests.c +29 -0
index e1c8a679a4cf3..d09d4a9081de1 100644
@@ -2276,8 +2276,11 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	unsigned long npages;
 	unsigned long size;
 	unsigned long i;
+	unsigned char *m;
+	uint64_t entry;
 	void *old_ptr;
 	void *map;
+	int pagemap_fd;
 	int *ptr;
 	int ret;
 
@@ -2318,6 +2321,32 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);
 
+	if (!hmm_is_coherent_type(variant->device_number)) {
+		ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT,
+				      buffer, npages);
+		ASSERT_EQ(ret, 0);
+		ASSERT_EQ(buffer->cpages, npages);
+
+		m = buffer->mirror;
+		for (i = 0; i < npages; ++i)
+			ASSERT_EQ(m[i], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL |
+					HMM_DMIRROR_PROT_WRITE |
+					HMM_DMIRROR_PROT_PMD);
+
+		pagemap_fd = open("/proc/self/pagemap", O_RDONLY);
+		ASSERT_GE(pagemap_fd, 0);
+
+		for (i = 0; i < npages; ++i) {
+			entry = pagemap_get_entry(pagemap_fd,
+					(char *)buffer->ptr + i * self->page_size);
+
+			ASSERT_NE(entry & PM_SWAP, 0);
+			ASSERT_EQ(entry & PM_PRESENT, 0);
+		}
+
+		close(pagemap_fd);
+	}
+
 	/* Fault pages back to system memory and check them. */
 	for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);
Signature · 1 line
2.43.0
#5 lore raw

Re: [PATCH v2 2/2] selftests/mm/hmm-tests: test pagemap reads of PMD device-private entries

Quoted 32 lines · Dev Jain

To cover pagemap paths scanning PMD entries, add assertions to check whether a device-private PMD entry has the correct pagemap information - the PM_SWAP bit must be on in the pagemap entry. Before that, we must assert through HMM_DMIRROR_SNAPSHOT snapshot that the leaf entry is at PMD level and not PTE level.

---
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
tools/testing/selftests/mm/hmm-tests.c +4 -0
index e1c8a679a4cf3..d09d4a9081de1 100644
@@ -2276,8 +2276,11 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	unsigned long npages;
 	unsigned long size;
 	unsigned long i;
+	unsigned char *m;
+	uint64_t entry;
 	void *old_ptr;
 	void *map;
+	int pagemap_fd;
 	int *ptr;
 	int ret;
 
@@ -2318,6 +2321,32 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);
 
+	if (!hmm_is_coherent_type(variant->device_number)) {

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2334

Quoted 10 lines
tools/testing/selftests/mm/hmm-tests.c +10 -0
+		ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT,
+				      buffer, npages);
+		ASSERT_EQ(ret, 0);
+		ASSERT_EQ(buffer->cpages, npages);
+
+		m = buffer->mirror;
+		for (i = 0; i < npages; ++i)
+			ASSERT_EQ(m[i], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL |
+					HMM_DMIRROR_PROT_WRITE |
+					HMM_DMIRROR_PROT_PMD);

madvise(..., MADV_HUGEPAGE) is not sufficient to guarantee that the allocation was indeed converted to THP. Might be worth using the kpageflags interface (but that requires elevated privileges) and then KPF_THP? Otherwise the HMM_DMIRROR_PROT_PMD can be a miss from time to time. One other option is not to assert, but to check and inform?

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2342

Quoted 8 lines
tools/testing/selftests/mm/hmm-tests.c +8 -0
+
+		pagemap_fd = open("/proc/self/pagemap", O_RDONLY);
+		ASSERT_GE(pagemap_fd, 0);
+
+		for (i = 0; i < npages; ++i) {
+			entry = pagemap_get_entry(pagemap_fd,
+					(char *)buffer->ptr + i * self->page_size);
+

If this is a THP entry, do we have valid pagemap entries for offset of i * page_size?

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2344

Quoted 2 lines
tools/testing/selftests/mm/hmm-tests.c +2 -0
+			ASSERT_NE(entry & PM_SWAP, 0);
+			ASSERT_EQ(entry & PM_PRESENT, 0);

Nit: You can use PAGEMAP_PRESENT()

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2352

Quoted 8 lines
tools/testing/selftests/mm/hmm-tests.c +5 -0
+		}
+
+		close(pagemap_fd);
+	}
+
 	/* Fault pages back to system memory and check them. */
 	for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);

Balbir Singh

author
#6 lore raw

Re: [PATCH v2 2/2] selftests/mm/hmm-tests: test pagemap reads of PMD device-private entries

Quoted 41 lines · Balbir Singh
Quoted 40 lines · Dev Jain

To cover pagemap paths scanning PMD entries, add assertions to check whether a device-private PMD entry has the correct pagemap information - the PM_SWAP bit must be on in the pagemap entry. Before that, we must assert through HMM_DMIRROR_SNAPSHOT snapshot that the leaf entry is at PMD level and not PTE level.

---
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
tools/testing/selftests/mm/hmm-tests.c +12 -0
index e1c8a679a4cf3..d09d4a9081de1 100644
@@ -2276,8 +2276,11 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	unsigned long npages;
 	unsigned long size;
 	unsigned long i;
+	unsigned char *m;
+	uint64_t entry;
 	void *old_ptr;
 	void *map;
+	int pagemap_fd;
 	int *ptr;
 	int ret;
 
@@ -2318,6 +2321,32 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);
 
+	if (!hmm_is_coherent_type(variant->device_number)) {
+		ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT,
+				      buffer, npages);
+		ASSERT_EQ(ret, 0);
+		ASSERT_EQ(buffer->cpages, npages);
+
+		m = buffer->mirror;
+		for (i = 0; i < npages; ++i)
+			ASSERT_EQ(m[i], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL |
Quoted 8 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2334

Quoted 2 lines
tools/testing/selftests/mm/hmm-tests.c +2 -0
+					HMM_DMIRROR_PROT_WRITE |
+					HMM_DMIRROR_PROT_PMD);

madvise(..., MADV_HUGEPAGE) is not sufficient to guarantee that the allocation was indeed converted to THP. Might be worth using the kpageflags interface (but that requires elevated privileges) and then KPF_THP? Otherwise the HMM_DMIRROR_PROT_PMD can be a miss from time to time. One other option is not to assert, but to check and inform?

I'll then use the existing check_huge_anon() to assert that a PMD THP got allocated.

Quoted 11 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2342

Quoted 9 lines
tools/testing/selftests/mm/hmm-tests.c +8 -0
+
+		pagemap_fd = open("/proc/self/pagemap", O_RDONLY);
+		ASSERT_GE(pagemap_fd, 0);
+
+		for (i = 0; i < npages; ++i) {
+			entry = pagemap_get_entry(pagemap_fd,
+					(char *)buffer->ptr + i * self->page_size);
+

If this is a THP entry, do we have valid pagemap entries for offset of i * page_size?

Yep we do, see the populate_pagemap label in pagemap_pmd_range_thp.

Quoted 5 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2344

Quoted 3 lines
tools/testing/selftests/mm/hmm-tests.c +2 -0
+			ASSERT_NE(entry & PM_SWAP, 0);
+			ASSERT_EQ(entry & PM_PRESENT, 0);

Nit: You can use PAGEMAP_PRESENT()

Okay.

Quoted 11 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2352

Quoted 9 lines
tools/testing/selftests/mm/hmm-tests.c +5 -0
+		}
+
+		close(pagemap_fd);
+	}
+
 	/* Fault pages back to system memory and check them. */
 	for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);

Balbir Singh

maintainer
#7 lore raw

Re: [PATCH v2 2/2] selftests/mm/hmm-tests: test pagemap reads of PMD device-private entries

Quoted 53 lines · Dev Jain
Quoted 41 lines · Balbir Singh
Quoted 40 lines · Dev Jain

To cover pagemap paths scanning PMD entries, add assertions to check whether a device-private PMD entry has the correct pagemap information - the PM_SWAP bit must be on in the pagemap entry. Before that, we must assert through HMM_DMIRROR_SNAPSHOT snapshot that the leaf entry is at PMD level and not PTE level.

---
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
tools/testing/selftests/mm/hmm-tests.c +12 -0
index e1c8a679a4cf3..d09d4a9081de1 100644
@@ -2276,8 +2276,11 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	unsigned long npages;
 	unsigned long size;
 	unsigned long i;
+	unsigned char *m;
+	uint64_t entry;
 	void *old_ptr;
 	void *map;
+	int pagemap_fd;
 	int *ptr;
 	int ret;
 
@@ -2318,6 +2321,32 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);
 
+	if (!hmm_is_coherent_type(variant->device_number)) {
+		ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT,
+				      buffer, npages);
+		ASSERT_EQ(ret, 0);
+		ASSERT_EQ(buffer->cpages, npages);
+
+		m = buffer->mirror;
+		for (i = 0; i < npages; ++i)
+			ASSERT_EQ(m[i], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL |
Quoted 8 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2334

Quoted 2 lines
tools/testing/selftests/mm/hmm-tests.c +2 -0
+					HMM_DMIRROR_PROT_WRITE |
+					HMM_DMIRROR_PROT_PMD);

madvise(..., MADV_HUGEPAGE) is not sufficient to guarantee that the allocation was indeed converted to THP. Might be worth using the kpageflags interface (but that requires elevated privileges) and then KPF_THP? Otherwise the HMM_DMIRROR_PROT_PMD can be a miss from time to time. One other option is not to assert, but to check and inform?

Quoted 1 line

I'll then use the existing check_huge_anon() to assert that a PMD THP got allocated.

Would MADV_COLLAPSE work here?

Quoted 16 lines
Quoted 12 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2342

Quoted 10 lines
tools/testing/selftests/mm/hmm-tests.c +8 -0
+
+		pagemap_fd = open("/proc/self/pagemap", O_RDONLY);
+		ASSERT_GE(pagemap_fd, 0);
+
+		for (i = 0; i < npages; ++i) {
+			entry = pagemap_get_entry(pagemap_fd,
+					(char *)buffer->ptr + i * self->page_size);
+

If this is a THP entry, do we have valid pagemap entries for offset of i * page_size?

Yep we do, see the populate_pagemap label in pagemap_pmd_range_thp.

Quoted 1 line
Quoted 19 lines
Quoted 4 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2344

Quoted 2 lines
tools/testing/selftests/mm/hmm-tests.c +2 -0
+			ASSERT_NE(entry & PM_SWAP, 0);
+			ASSERT_EQ(entry & PM_PRESENT, 0);

Nit: You can use PAGEMAP_PRESENT()

Okay.

Quoted 11 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2352

Quoted 9 lines
tools/testing/selftests/mm/hmm-tests.c +5 -0
+		}
+
+		close(pagemap_fd);
+	}
+
 	/* Fault pages back to system memory and check them. */
 	for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);

Balbir Singh

Cheers, Lorenzo

author
#8 lore raw

Re: [PATCH v2 2/2] selftests/mm/hmm-tests: test pagemap reads of PMD device-private entries

Quoted 54 lines · Lorenzo Stoakes
Quoted 53 lines · Dev Jain
Quoted 49 lines · Balbir Singh
Quoted 32 lines · Dev Jain

To cover pagemap paths scanning PMD entries, add assertions to check whether a device-private PMD entry has the correct pagemap information - the PM_SWAP bit must be on in the pagemap entry. Before that, we must assert through HMM_DMIRROR_SNAPSHOT snapshot that the leaf entry is at PMD level and not PTE level.

---
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
tools/testing/selftests/mm/hmm-tests.c +4 -0
index e1c8a679a4cf3..d09d4a9081de1 100644
@@ -2276,8 +2276,11 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	unsigned long npages;
 	unsigned long size;
 	unsigned long i;
+	unsigned char *m;
+	uint64_t entry;
 	void *old_ptr;
 	void *map;
+	int pagemap_fd;
 	int *ptr;
 	int ret;
 
@@ -2318,6 +2321,32 @@ TEST_F(hmm, migrate_anon_huge_fault)
 	for (i = 0, ptr = buffer->mirror; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);
 
+	if (!hmm_is_coherent_type(variant->device_number)) {

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2334

Quoted 10 lines
tools/testing/selftests/mm/hmm-tests.c +10 -0
+		ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_SNAPSHOT,
+				      buffer, npages);
+		ASSERT_EQ(ret, 0);
+		ASSERT_EQ(buffer->cpages, npages);
+
+		m = buffer->mirror;
+		for (i = 0; i < npages; ++i)
+			ASSERT_EQ(m[i], HMM_DMIRROR_PROT_DEV_PRIVATE_LOCAL |
+					HMM_DMIRROR_PROT_WRITE |
+					HMM_DMIRROR_PROT_PMD);

madvise(..., MADV_HUGEPAGE) is not sufficient to guarantee that the allocation was indeed converted to THP. Might be worth using the kpageflags interface (but that requires elevated privileges) and then KPF_THP? Otherwise the HMM_DMIRROR_PROT_PMD can be a miss from time to time. One other option is not to assert, but to check and inform?

Quoted 3 lines
Quoted 1 line

I'll then use the existing check_huge_anon() to assert that a PMD THP got allocated.

Would MADV_COLLAPSE work here?

So MADV_COLLAPSE requires at least one page to be faulted in (or precisely, HPAGE_PMD_NR - khugepaged_max_ptes_none). So I rejected that, but now I noticed that we are initializing the buffer first.

So yes I will use this, thanks.

Quoted 38 lines
Quoted 17 lines
Quoted 13 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2342

Quoted 11 lines
tools/testing/selftests/mm/hmm-tests.c +8 -0
+
+		pagemap_fd = open("/proc/self/pagemap", O_RDONLY);
+		ASSERT_GE(pagemap_fd, 0);
+
+		for (i = 0; i < npages; ++i) {
+			entry = pagemap_get_entry(pagemap_fd,
+					(char *)buffer->ptr + i * self->page_size);
+

If this is a THP entry, do we have valid pagemap entries for offset of i * page_size?

Yep we do, see the populate_pagemap label in pagemap_pmd_range_thp.

Quoted 1 line
Quoted 19 lines
Quoted 4 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2344

Quoted 2 lines
tools/testing/selftests/mm/hmm-tests.c +2 -0
+			ASSERT_NE(entry & PM_SWAP, 0);
+			ASSERT_EQ(entry & PM_PRESENT, 0);

Nit: You can use PAGEMAP_PRESENT()

Okay.

Quoted 11 lines

tools/testing/selftests/mm/hmm-tests.c TEST_F(hmm, migrate_anon_huge_fault) · line 2352

Quoted 9 lines
tools/testing/selftests/mm/hmm-tests.c +5 -0
+		}
+
+		close(pagemap_fd);
+	}
+
 	/* Fault pages back to system memory and check them. */
 	for (i = 0, ptr = buffer->ptr; i < size / sizeof(*ptr); ++i)
 		ASSERT_EQ(ptr[i], i);

Balbir Singh

Cheers, Lorenzo

maintainer
#9 lore raw
In reply to Dev Jain
Quoted 5 lines · On Sat, 30 May 2026 08:54:10 +0000 Dev Jain <[email protected]> wrote:

Patch 1 fixes a stale warning present from the time when only migration softleaf entries were supported at the PMD level.

Patch 2 adds some code into hmm-tests.c which exercises the pagemap path for PMD device-private entries.

David, I'll retain your ack on patch 1.

Sashiko has a concern about the selftest:

	https://sashiko.dev/#/patchset/[email protected]
#10 lore raw
Quoted 9 lines · Andrew Morton
Quoted 5 lines · On Sat, 30 May 2026 08:54:10 +0000 Dev Jain <[email protected]> wrote:

Patch 1 fixes a stale warning present from the time when only migration softleaf entries were supported at the PMD level.

Patch 2 adds some code into hmm-tests.c which exercises the pagemap path for PMD device-private entries.

David, I'll retain your ack on patch 1.

Thanks!

Signature · 3 lines
Cheers,

David
author
#11 lore raw
In reply to Andrew Morton
Quoted 5 lines · Andrew Morton
Quoted 3 lines · On Sat, 30 May 2026 08:54:10 +0000 Dev Jain <[email protected]> wrote:

Patch 1 fixes a stale warning present from the time when only migration softleaf entries were supported at the PMD level.

Quoted 7 lines
Quoted 2 lines

Patch 2 adds some code into hmm-tests.c which exercises the pagemap path for PMD device-private entries.

David, I'll retain your ack on patch 1.

Sashiko has a concern about the selftest:

	https://sashiko.dev/#/patchset/[email protected]

Same concern as raised by Balbir.

maintainer
#12 lore raw
In reply to Dev Jain
Quoted 5 lines · Dev Jain

Patch 1 fixes a stale warning present from the time when only migration softleaf entries were supported at the PMD level.

Patch 2 adds some code into hmm-tests.c which exercises the pagemap path for PMD device-private entries.

Thanks!

FYI running the HMM tests locally I ran into a nasty issue that I sent a patch for [0]!

[0]: https://lore.kernel.org/linux-mm/[email protected]/

Cheers, Lorenzo

Quoted 16 lines

---
Applies on mm-unstable (404fb4f38e8f).

Dev Jain (2):
fs/proc/task_mmu: do not warn on seeing non-migration pmd entry selftests/mm/hmm-tests: test pagemap reads of PMD device-private
entries

 fs/proc/task_mmu.c                     |  1 -
 tools/testing/selftests/mm/hmm-tests.c | 29 ++++++++++++++++++++++++++
 2 files changed, 29 insertions(+), 1 deletion(-)

--
2.43.0