The patch below does not apply to the 5.4-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
Possible dependencies:
73bdf65ea748 ("migrate: hugetlb: check for hugetlb shared PMD in node migration")
7ce82f4c3f3e ("mm/migration: return errno when isolate_huge_page failed")
1b7f7e58decc ("mm/gup: Convert check_and_migrate_movable_pages() to use a folio")
f9f38f78c5d5 ("mm: refactor check_and_migrate_movable_pages")
5ac95884a784 ("mm/migrate: enable returning precise migrate_pages() success count")
c5b5a3dd2c1f ("mm: thp: refactor NUMA fault handling")
5db4f15c4fd7 ("mm: memory: add orig_pmd to struct vm_fault")
8f34f1eac382 ("mm/userfaultfd: fix uffd-wp special cases for fork()")
25182f05ffed ("mm,hwpoison: fix race with hugetlb page allocation")
f68749ec342b ("mm/gup: longterm pin migration cleanup")
d1e153fea2a8 ("mm/gup: migrate pinned pages out of movable zone")
1a08ae36cf8b ("mm cma: rename PF_MEMALLOC_NOCMA to PF_MEMALLOC_PIN")
6e7f34ebb8d2 ("mm/gup: check for isolation errors")
f0f4463837da ("mm/gup: return an error on migration failure")
83c02c23d074 ("mm/gup: check every subpage of a compound page during isolation")
c991ffef7bce ("mm/gup: don't pin migrated cma pages in movable zone")
7ee820ee7238 ("Revert "mm: migrate: skip shared exec THP for NUMA balancing"")
ae37c7ff79f1 ("mm: make alloc_contig_range handle in-use hugetlb pages")
369fa227c219 ("mm: make alloc_contig_range handle free hugetlb pages")
c2ad7a1ffeaf ("mm,compaction: let isolate_migratepages_{range,block} return error codes")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 73bdf65ea74857d7fb2ec3067a3cec0e261b1462 Mon Sep 17 00:00:00 2001
From: Mike Kravetz <mike.kravetz(a)oracle.com>
Date: Thu, 26 Jan 2023 14:27:21 -0800
Subject: [PATCH] migrate: hugetlb: check for hugetlb shared PMD in node
migration
migrate_pages/mempolicy semantics state that CAP_SYS_NICE is required to
move pages shared with another process to a different node. page_mapcount
> 1 is being used to determine if a hugetlb page is shared. However, a
hugetlb page will have a mapcount of 1 if mapped by multiple processes via
a shared PMD. As a result, hugetlb pages shared by multiple processes and
mapped with a shared PMD can be moved by a process without CAP_SYS_NICE.
To fix, check for a shared PMD if mapcount is 1. If a shared PMD is found
consider the page shared.
Link: https://lkml.kernel.org/r/20230126222721.222195-3-mike.kravetz@oracle.com
Fixes: e2d8cf405525 ("migrate: add hugepage migration code to migrate_pages()")
Signed-off-by: Mike Kravetz <mike.kravetz(a)oracle.com>
Acked-by: Peter Xu <peterx(a)redhat.com>
Acked-by: David Hildenbrand <david(a)redhat.com>
Cc: James Houghton <jthoughton(a)google.com>
Cc: Matthew Wilcox <willy(a)infradead.org>
Cc: Michal Hocko <mhocko(a)suse.com>
Cc: Muchun Song <songmuchun(a)bytedance.com>
Cc: Naoya Horiguchi <naoya.horiguchi(a)linux.dev>
Cc: Vishal Moola (Oracle) <vishal.moola(a)gmail.com>
Cc: Yang Shi <shy828301(a)gmail.com>
Cc: <stable(a)vger.kernel.org>
Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org>
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 02c8a712282f..f940395667c8 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -600,7 +600,8 @@ static int queue_pages_hugetlb(pte_t *pte, unsigned long hmask,
/* With MPOL_MF_MOVE, we migrate only unshared hugepage. */
if (flags & (MPOL_MF_MOVE_ALL) ||
- (flags & MPOL_MF_MOVE && page_mapcount(page) == 1)) {
+ (flags & MPOL_MF_MOVE && page_mapcount(page) == 1 &&
+ !hugetlb_pmd_shared(pte))) {
if (isolate_hugetlb(page, qp->pagelist) &&
(flags & MPOL_MF_STRICT))
/*
From: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com>
It's unusual that we have enumeration by class in the middle of the table.
It might potentially be problematic in the future if we add another entry
after it.
So, move class matching entry to be the last in the ID table.
[ Upstream commit 0b85f59d30b91bd2b93ea7ef0816a4b7e7039e8c ]
Without this change, quirks set in driver_data added after the catch-all
are ignored.
Cc: <stable(a)vger.kernel.org>
Signed-off-by: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com>
Reviewed-by: Keith Busch <kbusch(a)kernel.org>
Reviewed-by: Sagi Grimberg <sagi(a)grimberg.me>
Reviewed-by: Chaitanya Kulkarni <chaitanya.kulkarni(a)wdc.com>
Signed-off-by: Christoph Hellwig <hch(a)lst.de>
Signed-off-by: Gwendal Grignou <gwendal(a)chromium.org>
---
drivers/nvme/host/pci.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 5d62d1042c0e6..a58711c488509 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -3199,7 +3199,6 @@ static const struct pci_device_id nvme_id_table[] = {
NVME_QUIRK_IGNORE_DEV_SUBNQN, },
{ PCI_DEVICE(0x1c5c, 0x1504), /* SK Hynix PC400 */
.driver_data = NVME_QUIRK_DISABLE_WRITE_ZEROES, },
- { PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) },
{ PCI_DEVICE(0x2646, 0x2263), /* KINGSTON A2000 NVMe SSD */
.driver_data = NVME_QUIRK_NO_DEEPEST_PS, },
{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, 0x2001),
@@ -3209,6 +3208,8 @@ static const struct pci_device_id nvme_id_table[] = {
.driver_data = NVME_QUIRK_SINGLE_VECTOR |
NVME_QUIRK_128_BYTES_SQES |
NVME_QUIRK_SHARED_TAGS },
+
+ { PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) },
{ 0, }
};
MODULE_DEVICE_TABLE(pci, nvme_id_table);
--
2.39.1.519.gcb327c4b5f-goog
Mark the Tiger Lake UP{3,4} AHCI controller as "low_power". This enables
S0ix to work out of the box. Otherwise this isn't working unless the
user manually sets /sys/class/scsi_host/*/link_power_management_policy.
Intel lists a total of 4 SATA controller IDs in [1] for those mobile
PCHs. This commit just adds the "AHCI" variant since I only tested
those.
[1]: https://cdrdv2.intel.com/v1/dl/getContent/631119
Signed-off-by: Simon Gaiser <simon(a)invisiblethingslab.com>
CC: stable(a)vger.kernel.org
---
As noted above this doesn't include the other PCI IDs listed by Intel
for those PCHs (RAID modes). Also the same is probably needed for newer
generations. But for both I don't have hardware to test handy right now,
so only included what I have actually tested.
Added stable to CC, since on systems using S0ix this prevents S0ix
residency and therefore leads to such high power consumption that
suspend is effectively broken.
drivers/ata/ahci.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 14a1c0d14916..3bb9bb483fe3 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -421,6 +421,7 @@ static const struct pci_device_id ahci_pci_tbl[] = {
{ PCI_VDEVICE(INTEL, 0x34d3), board_ahci_low_power }, /* Ice Lake LP AHCI */
{ PCI_VDEVICE(INTEL, 0x02d3), board_ahci_low_power }, /* Comet Lake PCH-U AHCI */
{ PCI_VDEVICE(INTEL, 0x02d7), board_ahci_low_power }, /* Comet Lake PCH RAID */
+ { PCI_VDEVICE(INTEL, 0xa0d3), board_ahci_low_power }, /* Tiger Lake UP{3,4} AHCI */
/* JMicron 360/1/3/5/6, match class to avoid IDE function */
{ PCI_VENDOR_ID_JMICRON, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
--
2.39.1
A hugetlb page will have a mapcount of 1 if mapped by multiple processes
via a shared PMD. This is because only the first process increases the
map count, and subsequent processes just add the shared PMD page to
their page table.
page_mapcount is being used to decide if a hugetlb page is shared or
private in /proc/PID/smaps. Pages referenced via a shared PMD were
incorrectly being counted as private.
To fix, check for a shared PMD if mapcount is 1. If a shared PMD is
found count the hugetlb page as shared. A new helper to check for a
shared PMD is added.
Fixes: 25ee01a2fca0 ("mm: hugetlb: proc: add hugetlb-related fields to /proc/PID/smaps")
Cc: stable(a)vger.kernel.org
Signed-off-by: Mike Kravetz <mike.kravetz(a)oracle.com>
---
fs/proc/task_mmu.c | 10 ++++++++--
include/linux/hugetlb.h | 12 ++++++++++++
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index e35a0398db63..cb9539879402 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -749,8 +749,14 @@ static int smaps_hugetlb_range(pte_t *pte, unsigned long hmask,
if (mapcount >= 2)
mss->shared_hugetlb += huge_page_size(hstate_vma(vma));
- else
- mss->private_hugetlb += huge_page_size(hstate_vma(vma));
+ else {
+ if (hugetlb_pmd_shared(pte))
+ mss->shared_hugetlb +=
+ huge_page_size(hstate_vma(vma));
+ else
+ mss->private_hugetlb +=
+ huge_page_size(hstate_vma(vma));
+ }
}
return 0;
}
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index e3aa336df900..8e65920e4363 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -1225,6 +1225,18 @@ static inline __init void hugetlb_cma_reserve(int order)
}
#endif
+#ifdef CONFIG_ARCH_WANT_HUGE_PMD_SHARE
+static inline bool hugetlb_pmd_shared(pte_t *pte)
+{
+ return page_count(virt_to_page(pte)) > 1;
+}
+#else
+static inline bool hugetlb_pmd_shared(pte_t *pte)
+{
+ return false;
+}
+#endif
+
bool want_pmd_share(struct vm_area_struct *vma, unsigned long addr);
#ifndef __HAVE_ARCH_FLUSH_HUGETLB_TLB_RANGE
--
2.39.1
The patch below does not apply to the 4.14-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
Possible dependencies:
73bdf65ea748 ("migrate: hugetlb: check for hugetlb shared PMD in node migration")
7ce82f4c3f3e ("mm/migration: return errno when isolate_huge_page failed")
1b7f7e58decc ("mm/gup: Convert check_and_migrate_movable_pages() to use a folio")
f9f38f78c5d5 ("mm: refactor check_and_migrate_movable_pages")
5ac95884a784 ("mm/migrate: enable returning precise migrate_pages() success count")
c5b5a3dd2c1f ("mm: thp: refactor NUMA fault handling")
5db4f15c4fd7 ("mm: memory: add orig_pmd to struct vm_fault")
8f34f1eac382 ("mm/userfaultfd: fix uffd-wp special cases for fork()")
25182f05ffed ("mm,hwpoison: fix race with hugetlb page allocation")
f68749ec342b ("mm/gup: longterm pin migration cleanup")
d1e153fea2a8 ("mm/gup: migrate pinned pages out of movable zone")
1a08ae36cf8b ("mm cma: rename PF_MEMALLOC_NOCMA to PF_MEMALLOC_PIN")
6e7f34ebb8d2 ("mm/gup: check for isolation errors")
f0f4463837da ("mm/gup: return an error on migration failure")
83c02c23d074 ("mm/gup: check every subpage of a compound page during isolation")
c991ffef7bce ("mm/gup: don't pin migrated cma pages in movable zone")
7ee820ee7238 ("Revert "mm: migrate: skip shared exec THP for NUMA balancing"")
ae37c7ff79f1 ("mm: make alloc_contig_range handle in-use hugetlb pages")
369fa227c219 ("mm: make alloc_contig_range handle free hugetlb pages")
c2ad7a1ffeaf ("mm,compaction: let isolate_migratepages_{range,block} return error codes")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 73bdf65ea74857d7fb2ec3067a3cec0e261b1462 Mon Sep 17 00:00:00 2001
From: Mike Kravetz <mike.kravetz(a)oracle.com>
Date: Thu, 26 Jan 2023 14:27:21 -0800
Subject: [PATCH] migrate: hugetlb: check for hugetlb shared PMD in node
migration
migrate_pages/mempolicy semantics state that CAP_SYS_NICE is required to
move pages shared with another process to a different node. page_mapcount
> 1 is being used to determine if a hugetlb page is shared. However, a
hugetlb page will have a mapcount of 1 if mapped by multiple processes via
a shared PMD. As a result, hugetlb pages shared by multiple processes and
mapped with a shared PMD can be moved by a process without CAP_SYS_NICE.
To fix, check for a shared PMD if mapcount is 1. If a shared PMD is found
consider the page shared.
Link: https://lkml.kernel.org/r/20230126222721.222195-3-mike.kravetz@oracle.com
Fixes: e2d8cf405525 ("migrate: add hugepage migration code to migrate_pages()")
Signed-off-by: Mike Kravetz <mike.kravetz(a)oracle.com>
Acked-by: Peter Xu <peterx(a)redhat.com>
Acked-by: David Hildenbrand <david(a)redhat.com>
Cc: James Houghton <jthoughton(a)google.com>
Cc: Matthew Wilcox <willy(a)infradead.org>
Cc: Michal Hocko <mhocko(a)suse.com>
Cc: Muchun Song <songmuchun(a)bytedance.com>
Cc: Naoya Horiguchi <naoya.horiguchi(a)linux.dev>
Cc: Vishal Moola (Oracle) <vishal.moola(a)gmail.com>
Cc: Yang Shi <shy828301(a)gmail.com>
Cc: <stable(a)vger.kernel.org>
Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org>
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 02c8a712282f..f940395667c8 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -600,7 +600,8 @@ static int queue_pages_hugetlb(pte_t *pte, unsigned long hmask,
/* With MPOL_MF_MOVE, we migrate only unshared hugepage. */
if (flags & (MPOL_MF_MOVE_ALL) ||
- (flags & MPOL_MF_MOVE && page_mapcount(page) == 1)) {
+ (flags & MPOL_MF_MOVE && page_mapcount(page) == 1 &&
+ !hugetlb_pmd_shared(pte))) {
if (isolate_hugetlb(page, qp->pagelist) &&
(flags & MPOL_MF_STRICT))
/*