October 2024 - Linux-stable-mirror

[merged mm-hotfixes-stable] mm-mremap-fix-move_normal_pmd-retract_page_tables-race.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: mm/mremap: fix move_normal_pmd/retract_page_tables race has been removed from the -mm tree. Its filename was mm-mremap-fix-move_normal_pmd-retract_page_tables-race.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Jann Horn <jannh(a)google.com> Subject: mm/mremap: fix move_normal_pmd/retract_page_tables race Date: Mon, 07 Oct 2024 23:42:04 +0200 In mremap(), move_page_tables() looks at the type of the PMD entry and the specified address range to figure out by which method the next chunk of page table entries should be moved. At that point, the mmap_lock is held in write mode, but no rmap locks are held yet. For PMD entries that point to page tables and are fully covered by the source address range, move_pgt_entry(NORMAL_PMD, ...) is called, which first takes rmap locks, then does move_normal_pmd(). move_normal_pmd() takes the necessary page table locks at source and destination, then moves an entire page table from the source to the destination. The problem is: The rmap locks, which protect against concurrent page table removal by retract_page_tables() in the THP code, are only taken after the PMD entry has been read and it has been decided how to move it. So we can race as follows (with two processes that have mappings of the same tmpfs file that is stored on a tmpfs mount with huge=advise); note that process A accesses page tables through the MM while process B does it through the file rmap: process A process B ========= ========= mremap mremap_to move_vma move_page_tables get_old_pmd alloc_new_pmd *** PREEMPT *** madvise(MADV_COLLAPSE) do_madvise madvise_walk_vmas madvise_vma_behavior madvise_collapse hpage_collapse_scan_file collapse_file retract_page_tables i_mmap_lock_read(mapping) pmdp_collapse_flush i_mmap_unlock_read(mapping) move_pgt_entry(NORMAL_PMD, ...) take_rmap_locks move_normal_pmd drop_rmap_locks When this happens, move_normal_pmd() can end up creating bogus PMD entries in the line `pmd_populate(mm, new_pmd, pmd_pgtable(pmd))`. The effect depends on arch-specific and machine-specific details; on x86, you can end up with physical page 0 mapped as a page table, which is likely exploitable for user->kernel privilege escalation. Fix the race by letting process B recheck that the PMD still points to a page table after the rmap locks have been taken. Otherwise, we bail and let the caller fall back to the PTE-level copying path, which will then bail immediately at the pmd_none() check. Bug reachability: Reaching this bug requires that you can create shmem/file THP mappings - anonymous THP uses different code that doesn't zap stuff under rmap locks. File THP is gated on an experimental config flag (CONFIG_READ_ONLY_THP_FOR_FS), so on normal distro kernels you need shmem THP to hit this bug. As far as I know, getting shmem THP normally requires that you can mount your own tmpfs with the right mount flags, which would require creating your own user+mount namespace; though I don't know if some distros maybe enable shmem THP by default or something like that. Bug impact: This issue can likely be used for user->kernel privilege escalation when it is reachable. Link: https://lkml.kernel.org/r/20241007-move_normal_pmd-vs-collapse-fix-2-v1-1-5… Fixes: 1d65b771bc08 ("mm/khugepaged: retract_page_tables() without mmap or vma lock") Signed-off-by: Jann Horn <jannh(a)google.com> Signed-off-by: David Hildenbrand <david(a)redhat.com> Co-developed-by: David Hildenbrand <david(a)redhat.com> Closes: https://project-zero.issues.chromium.org/371047675 Acked-by: Qi Zheng <zhengqi.arch(a)bytedance.com> Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes(a)oracle.com> Cc: Hugh Dickins <hughd(a)google.com> Cc: Joel Fernandes <joel(a)joelfernandes.org> Cc: Matthew Wilcox <willy(a)infradead.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- mm/mremap.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) --- a/mm/mremap.c~mm-mremap-fix-move_normal_pmd-retract_page_tables-race +++ a/mm/mremap.c @@ -238,6 +238,7 @@ static bool move_normal_pmd(struct vm_ar { spinlock_t *old_ptl, *new_ptl; struct mm_struct *mm = vma->vm_mm; + bool res = false; pmd_t pmd; if (!arch_supports_page_table_move()) @@ -277,19 +278,25 @@ static bool move_normal_pmd(struct vm_ar if (new_ptl != old_ptl) spin_lock_nested(new_ptl, SINGLE_DEPTH_NESTING); - /* Clear the pmd */ pmd = *old_pmd; + + /* Racing with collapse? */ + if (unlikely(!pmd_present(pmd) || pmd_leaf(pmd))) + goto out_unlock; + /* Clear the pmd */ pmd_clear(old_pmd); + res = true; VM_BUG_ON(!pmd_none(*new_pmd)); pmd_populate(mm, new_pmd, pmd_pgtable(pmd)); flush_tlb_range(vma, old_addr, old_addr + PMD_SIZE); +out_unlock: if (new_ptl != old_ptl) spin_unlock(new_ptl); spin_unlock(old_ptl); - return true; + return res; } #else static inline bool move_normal_pmd(struct vm_area_struct *vma, _ Patches currently in -mm which might be from jannh(a)google.com are mm-mark-mas-allocation-in-vms_abort_munmap_vmas-as-__gfp_nofail.patch

11 months

1
0
0 0

[merged mm-hotfixes-stable] selftests-mm-fix-deadlock-for-fork-after-pthread_create-on-arm.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: selftests/mm: fix deadlock for fork after pthread_create on ARM has been removed from the -mm tree. Its filename was selftests-mm-fix-deadlock-for-fork-after-pthread_create-on-arm.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Edward Liaw <edliaw(a)google.com> Subject: selftests/mm: fix deadlock for fork after pthread_create on ARM Date: Thu, 3 Oct 2024 21:17:11 +0000 On Android with arm, there is some synchronization needed to avoid a deadlock when forking after pthread_create. Link: https://lkml.kernel.org/r/20241003211716.371786-3-edliaw@google.com Fixes: cff294582798 ("selftests/mm: extend and rename uffd pagemap test") Signed-off-by: Edward Liaw <edliaw(a)google.com> Cc: Lokesh Gidra <lokeshgidra(a)google.com> Cc: Peter Xu <peterx(a)redhat.com> Cc: Shuah Khan <shuah(a)kernel.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- tools/testing/selftests/mm/uffd-unit-tests.c | 7 +++++++ 1 file changed, 7 insertions(+) --- a/tools/testing/selftests/mm/uffd-unit-tests.c~selftests-mm-fix-deadlock-for-fork-after-pthread_create-on-arm +++ a/tools/testing/selftests/mm/uffd-unit-tests.c @@ -241,6 +241,9 @@ static void *fork_event_consumer(void *d fork_event_args *args = data; struct uffd_msg msg = { 0 }; + /* Ready for parent thread to fork */ + pthread_barrier_wait(&ready_for_fork); + /* Read until a full msg received */ while (uffd_read_msg(args->parent_uffd, &msg)); @@ -308,8 +311,12 @@ static int pagemap_test_fork(int uffd, b /* Prepare a thread to resolve EVENT_FORK */ if (with_event) { + pthread_barrier_init(&ready_for_fork, NULL, 2); if (pthread_create(&thread, NULL, fork_event_consumer, &args)) err("pthread_create()"); + /* Wait for child thread to start before forking */ + pthread_barrier_wait(&ready_for_fork); + pthread_barrier_destroy(&ready_for_fork); } child = fork(); _ Patches currently in -mm which might be from edliaw(a)google.com are

11 months

1
0
0 0

[merged mm-hotfixes-stable] selftests-mm-replace-atomic_bool-with-pthread_barrier_t.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: selftests/mm: replace atomic_bool with pthread_barrier_t has been removed from the -mm tree. Its filename was selftests-mm-replace-atomic_bool-with-pthread_barrier_t.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Edward Liaw <edliaw(a)google.com> Subject: selftests/mm: replace atomic_bool with pthread_barrier_t Date: Thu, 3 Oct 2024 21:17:10 +0000 Patch series "selftests/mm: fix deadlock after pthread_create". On Android arm, pthread_create followed by a fork caused a deadlock in the case where the fork required work to be completed by the created thread. Update the synchronization primitive to use pthread_barrier instead of atomic_bool. Apply the same fix to the wp-fork-with-event test. This patch (of 2): Swap synchronization primitive with pthread_barrier, so that stdatomic.h does not need to be included. The synchronization is needed on Android ARM64; we see a deadlock with pthread_create when the parent thread races forward before the child has a chance to start doing work. Link: https://lkml.kernel.org/r/20241003211716.371786-1-edliaw@google.com Link: https://lkml.kernel.org/r/20241003211716.371786-2-edliaw@google.com Fixes: cff294582798 ("selftests/mm: extend and rename uffd pagemap test") Signed-off-by: Edward Liaw <edliaw(a)google.com> Cc: Lokesh Gidra <lokeshgidra(a)google.com> Cc: Peter Xu <peterx(a)redhat.com> Cc: Shuah Khan <shuah(a)kernel.org> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- tools/testing/selftests/mm/uffd-common.c | 5 +++-- tools/testing/selftests/mm/uffd-common.h | 3 +-- tools/testing/selftests/mm/uffd-unit-tests.c | 14 ++++++++------ 3 files changed, 12 insertions(+), 10 deletions(-) --- a/tools/testing/selftests/mm/uffd-common.c~selftests-mm-replace-atomic_bool-with-pthread_barrier_t +++ a/tools/testing/selftests/mm/uffd-common.c @@ -18,7 +18,7 @@ bool test_uffdio_wp = true; unsigned long long *count_verify; uffd_test_ops_t *uffd_test_ops; uffd_test_case_ops_t *uffd_test_case_ops; -atomic_bool ready_for_fork; +pthread_barrier_t ready_for_fork; static int uffd_mem_fd_create(off_t mem_size, bool hugetlb) { @@ -519,7 +519,8 @@ void *uffd_poll_thread(void *arg) pollfd[1].fd = pipefd[cpu*2]; pollfd[1].events = POLLIN; - ready_for_fork = true; + /* Ready for parent thread to fork */ + pthread_barrier_wait(&ready_for_fork); for (;;) { ret = poll(pollfd, 2, -1); --- a/tools/testing/selftests/mm/uffd-common.h~selftests-mm-replace-atomic_bool-with-pthread_barrier_t +++ a/tools/testing/selftests/mm/uffd-common.h @@ -33,7 +33,6 @@ #include <inttypes.h> #include <stdint.h> #include <sys/random.h> -#include <stdatomic.h> #include "../kselftest.h" #include "vm_util.h" @@ -105,7 +104,7 @@ extern bool map_shared; extern bool test_uffdio_wp; extern unsigned long long *count_verify; extern volatile bool test_uffdio_copy_eexist; -extern atomic_bool ready_for_fork; +extern pthread_barrier_t ready_for_fork; extern uffd_test_ops_t anon_uffd_test_ops; extern uffd_test_ops_t shmem_uffd_test_ops; --- a/tools/testing/selftests/mm/uffd-unit-tests.c~selftests-mm-replace-atomic_bool-with-pthread_barrier_t +++ a/tools/testing/selftests/mm/uffd-unit-tests.c @@ -774,7 +774,7 @@ static void uffd_sigbus_test_common(bool char c; struct uffd_args args = { 0 }; - ready_for_fork = false; + pthread_barrier_init(&ready_for_fork, NULL, 2); fcntl(uffd, F_SETFL, uffd_flags | O_NONBLOCK); @@ -791,8 +791,9 @@ static void uffd_sigbus_test_common(bool if (pthread_create(&uffd_mon, NULL, uffd_poll_thread, &args)) err("uffd_poll_thread create"); - while (!ready_for_fork) - ; /* Wait for the poll_thread to start executing before forking */ + /* Wait for child thread to start before forking */ + pthread_barrier_wait(&ready_for_fork); + pthread_barrier_destroy(&ready_for_fork); pid = fork(); if (pid < 0) @@ -833,7 +834,7 @@ static void uffd_events_test_common(bool char c; struct uffd_args args = { 0 }; - ready_for_fork = false; + pthread_barrier_init(&ready_for_fork, NULL, 2); fcntl(uffd, F_SETFL, uffd_flags | O_NONBLOCK); if (uffd_register(uffd, area_dst, nr_pages * page_size, @@ -844,8 +845,9 @@ static void uffd_events_test_common(bool if (pthread_create(&uffd_mon, NULL, uffd_poll_thread, &args)) err("uffd_poll_thread create"); - while (!ready_for_fork) - ; /* Wait for the poll_thread to start executing before forking */ + /* Wait for child thread to start before forking */ + pthread_barrier_wait(&ready_for_fork); + pthread_barrier_destroy(&ready_for_fork); pid = fork(); if (pid < 0) _ Patches currently in -mm which might be from edliaw(a)google.com are

11 months

1
0
0 0

[merged mm-hotfixes-stable] fat-fix-uninitialized-variable.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: fat: fix uninitialized variable has been removed from the -mm tree. Its filename was fat-fix-uninitialized-variable.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: OGAWA Hirofumi <hirofumi(a)mail.parknet.co.jp> Subject: fat: fix uninitialized variable Date: Fri, 04 Oct 2024 15:03:49 +0900 syszbot produced this with a corrupted fs image. In theory, however an IO error would trigger this also. This affects just an error report, so should not be a serious error. Link: https://lkml.kernel.org/r/87r08wjsnh.fsf@mail.parknet.co.jp Link: https://lkml.kernel.org/r/66ff2c95.050a0220.49194.03e9.GAE@google.com Signed-off-by: OGAWA Hirofumi <hirofumi(a)mail.parknet.co.jp> Reported-by: syzbot+ef0d7bc412553291aa86(a)syzkaller.appspotmail.com Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- fs/fat/namei_vfat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/fs/fat/namei_vfat.c~fat-fix-uninitialized-variable +++ a/fs/fat/namei_vfat.c @@ -1037,7 +1037,7 @@ error_inode: if (corrupt < 0) { fat_fs_error(new_dir->i_sb, "%s: Filesystem corrupted (i_pos %lld)", - __func__, sinfo.i_pos); + __func__, new_i_pos); } goto out; } _ Patches currently in -mm which might be from hirofumi(a)mail.parknet.co.jp are

11 months

1
0
0 0

[merged mm-hotfixes-stable] nilfs2-propagate-directory-read-errors-from-nilfs_find_entry.patch removed from -mm tree

by Andrew Morton

The quilt patch titled Subject: nilfs2: propagate directory read errors from nilfs_find_entry() has been removed from the -mm tree. Its filename was nilfs2-propagate-directory-read-errors-from-nilfs_find_entry.patch This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm ------------------------------------------------------ From: Ryusuke Konishi <konishi.ryusuke(a)gmail.com> Subject: nilfs2: propagate directory read errors from nilfs_find_entry() Date: Fri, 4 Oct 2024 12:35:31 +0900 Syzbot reported that a task hang occurs in vcs_open() during a fuzzing test for nilfs2. The root cause of this problem is that in nilfs_find_entry(), which searches for directory entries, ignores errors when loading a directory page/folio via nilfs_get_folio() fails. If the filesystem images is corrupted, and the i_size of the directory inode is large, and the directory page/folio is successfully read but fails the sanity check, for example when it is zero-filled, nilfs_check_folio() may continue to spit out error messages in bursts. Fix this issue by propagating the error to the callers when loading a page/folio fails in nilfs_find_entry(). The current interface of nilfs_find_entry() and its callers is outdated and cannot propagate error codes such as -EIO and -ENOMEM returned via nilfs_find_entry(), so fix it together. Link: https://lkml.kernel.org/r/20241004033640.6841-1-konishi.ryusuke@gmail.com Fixes: 2ba466d74ed7 ("nilfs2: directory entry operations") Signed-off-by: Ryusuke Konishi <konishi.ryusuke(a)gmail.com> Reported-by: Lizhi Xu <lizhi.xu(a)windriver.com> Closes: https://lkml.kernel.org/r/20240927013806.3577931-1-lizhi.xu@windriver.com Reported-by: syzbot+8a192e8d090fa9a31135(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=8a192e8d090fa9a31135 Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- fs/nilfs2/dir.c | 48 ++++++++++++++++++++++---------------------- fs/nilfs2/namei.c | 39 +++++++++++++++++++++++------------ fs/nilfs2/nilfs.h | 2 - 3 files changed, 52 insertions(+), 37 deletions(-) --- a/fs/nilfs2/dir.c~nilfs2-propagate-directory-read-errors-from-nilfs_find_entry +++ a/fs/nilfs2/dir.c @@ -289,7 +289,7 @@ static int nilfs_readdir(struct file *fi * The folio is mapped and unlocked. When the caller is finished with * the entry, it should call folio_release_kmap(). * - * On failure, returns NULL and the caller should ignore foliop. + * On failure, returns an error pointer and the caller should ignore foliop. */ struct nilfs_dir_entry *nilfs_find_entry(struct inode *dir, const struct qstr *qstr, struct folio **foliop) @@ -312,22 +312,24 @@ struct nilfs_dir_entry *nilfs_find_entry do { char *kaddr = nilfs_get_folio(dir, n, foliop); - if (!IS_ERR(kaddr)) { - de = (struct nilfs_dir_entry *)kaddr; - kaddr += nilfs_last_byte(dir, n) - reclen; - while ((char *) de <= kaddr) { - if (de->rec_len == 0) { - nilfs_error(dir->i_sb, - "zero-length directory entry"); - folio_release_kmap(*foliop, kaddr); - goto out; - } - if (nilfs_match(namelen, name, de)) - goto found; - de = nilfs_next_entry(de); + if (IS_ERR(kaddr)) + return ERR_CAST(kaddr); + + de = (struct nilfs_dir_entry *)kaddr; + kaddr += nilfs_last_byte(dir, n) - reclen; + while ((char *)de <= kaddr) { + if (de->rec_len == 0) { + nilfs_error(dir->i_sb, + "zero-length directory entry"); + folio_release_kmap(*foliop, kaddr); + goto out; } - folio_release_kmap(*foliop, kaddr); + if (nilfs_match(namelen, name, de)) + goto found; + de = nilfs_next_entry(de); } + folio_release_kmap(*foliop, kaddr); + if (++n >= npages) n = 0; /* next folio is past the blocks we've got */ @@ -340,7 +342,7 @@ struct nilfs_dir_entry *nilfs_find_entry } } while (n != start); out: - return NULL; + return ERR_PTR(-ENOENT); found: ei->i_dir_start_lookup = n; @@ -384,18 +386,18 @@ fail: return NULL; } -ino_t nilfs_inode_by_name(struct inode *dir, const struct qstr *qstr) +int nilfs_inode_by_name(struct inode *dir, const struct qstr *qstr, ino_t *ino) { - ino_t res = 0; struct nilfs_dir_entry *de; struct folio *folio; de = nilfs_find_entry(dir, qstr, &folio); - if (de) { - res = le64_to_cpu(de->inode); - folio_release_kmap(folio, de); - } - return res; + if (IS_ERR(de)) + return PTR_ERR(de); + + *ino = le64_to_cpu(de->inode); + folio_release_kmap(folio, de); + return 0; } void nilfs_set_link(struct inode *dir, struct nilfs_dir_entry *de, --- a/fs/nilfs2/namei.c~nilfs2-propagate-directory-read-errors-from-nilfs_find_entry +++ a/fs/nilfs2/namei.c @@ -55,12 +55,20 @@ nilfs_lookup(struct inode *dir, struct d { struct inode *inode; ino_t ino; + int res; if (dentry->d_name.len > NILFS_NAME_LEN) return ERR_PTR(-ENAMETOOLONG); - ino = nilfs_inode_by_name(dir, &dentry->d_name); - inode = ino ? nilfs_iget(dir->i_sb, NILFS_I(dir)->i_root, ino) : NULL; + res = nilfs_inode_by_name(dir, &dentry->d_name, &ino); + if (res) { + if (res != -ENOENT) + return ERR_PTR(res); + inode = NULL; + } else { + inode = nilfs_iget(dir->i_sb, NILFS_I(dir)->i_root, ino); + } + return d_splice_alias(inode, dentry); } @@ -263,10 +271,11 @@ static int nilfs_do_unlink(struct inode struct folio *folio; int err; - err = -ENOENT; de = nilfs_find_entry(dir, &dentry->d_name, &folio); - if (!de) + if (IS_ERR(de)) { + err = PTR_ERR(de); goto out; + } inode = d_inode(dentry); err = -EIO; @@ -362,10 +371,11 @@ static int nilfs_rename(struct mnt_idmap if (unlikely(err)) return err; - err = -ENOENT; old_de = nilfs_find_entry(old_dir, &old_dentry->d_name, &old_folio); - if (!old_de) + if (IS_ERR(old_de)) { + err = PTR_ERR(old_de); goto out; + } if (S_ISDIR(old_inode->i_mode)) { err = -EIO; @@ -382,10 +392,12 @@ static int nilfs_rename(struct mnt_idmap if (dir_de && !nilfs_empty_dir(new_inode)) goto out_dir; - err = -ENOENT; - new_de = nilfs_find_entry(new_dir, &new_dentry->d_name, &new_folio); - if (!new_de) + new_de = nilfs_find_entry(new_dir, &new_dentry->d_name, + &new_folio); + if (IS_ERR(new_de)) { + err = PTR_ERR(new_de); goto out_dir; + } nilfs_set_link(new_dir, new_de, new_folio, old_inode); folio_release_kmap(new_folio, new_de); nilfs_mark_inode_dirty(new_dir); @@ -440,12 +452,13 @@ out: */ static struct dentry *nilfs_get_parent(struct dentry *child) { - unsigned long ino; + ino_t ino; + int res; struct nilfs_root *root; - ino = nilfs_inode_by_name(d_inode(child), &dotdot_name); - if (!ino) - return ERR_PTR(-ENOENT); + res = nilfs_inode_by_name(d_inode(child), &dotdot_name, &ino); + if (res) + return ERR_PTR(res); root = NILFS_I(d_inode(child))->i_root; --- a/fs/nilfs2/nilfs.h~nilfs2-propagate-directory-read-errors-from-nilfs_find_entry +++ a/fs/nilfs2/nilfs.h @@ -254,7 +254,7 @@ static inline __u32 nilfs_mask_flags(umo /* dir.c */ int nilfs_add_link(struct dentry *, struct inode *); -ino_t nilfs_inode_by_name(struct inode *, const struct qstr *); +int nilfs_inode_by_name(struct inode *dir, const struct qstr *qstr, ino_t *ino); int nilfs_make_empty(struct inode *, struct inode *); struct nilfs_dir_entry *nilfs_find_entry(struct inode *, const struct qstr *, struct folio **); _ Patches currently in -mm which might be from konishi.ryusuke(a)gmail.com are nilfs2-fix-kernel-bug-due-to-missing-clearing-of-buffer-delay-flag.patch

11 months

1
0
0 0

[PATCH] ACPI: resource: Add LG 16T90SP to irq1_level_low_skip_override[]

by Christian Heusel

The LG Gram Pro 16 2-in-1 (2024) the 16T90SP has its keybopard IRQ (1) described as ActiveLow in the DSDT, which the kernel overrides to EdgeHigh which breaks the keyboard. Add the 16T90SP to the irq1_level_low_skip_override[] quirk table to fix this. Reported-by: Dirk Holten <dirk.holten(a)gmx.de> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=219382 Cc: stable(a)vger.kernel.org Suggested-by: Dirk Holten <dirk.holten(a)gmx.de> Signed-off-by: Christian Heusel <christian(a)heusel.eu> --- Note that I do not have the relevant hardware since I'm sending in this quirk at the request of someone else. Also does this change need a "Fixes: ..." tag? --- drivers/acpi/resource.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 129bceb1f4a27df93439bcefdb27fd9c91258028..dd6249fb76c24f08db4149883be4548130d0ef1e 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -502,6 +502,11 @@ static const struct dmi_system_id irq1_level_low_skip_override[] = { DMI_MATCH(DMI_SYS_VENDOR, "LG Electronics"), DMI_MATCH(DMI_BOARD_NAME, "17U70P"), }, + /* LG Electronics 16T90SP */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LG Electronics"), + DMI_MATCH(DMI_BOARD_NAME, "16T90SP"), + }, }, { } }; --- base-commit: 8e929cb546ee42c9a61d24fae60605e9e3192354 change-id: 20241016-lg-gram-pro-keyboard-9a9d8b9aa647 Best regards, -- Christian Heusel <christian(a)heusel.eu>

11 months

2
1
0 0

[PATCH] KVM: arm64: vgic-its: Do not call vgic_put_irq() within vgic_its_inject_cached_translation()

by WangYuli

There is a probability that the host machine will also restart when the virtual machine is restarting. Commit ad362fe07fec ("KVM: arm64: vgic-its: Avoid potential UAF in LPI translation cache") released the reference count of an IRQ when it shouldn't have. This led to a situation where, when the system finally released the IRQ, it found that the structure had already been freed, triggering a 'refcount_t: underflow; use-after-free' error. In fact, the function "vgic_put_irq" should be called by "vgic_its_inject_cached_translation" instead of "vgic_its_trigger_msi". Call trace: its_free_ite+0x90/0xa0 vgic_its_free_device+0x3c/0xa0 vgic_its_destroy+0x4c/0xb8 kvm_put_kvm+0x214/0x358 kvm_vcpu_release+0x24/0x38 __fput+0x84/0x278 ____fput+0x20/0x30 task_work_run+0xcc/0x190 do_exit+0x36c/0xa88 do_group_exit+0x4c/0xb8 __arm64_sys_exit_group+0x24/0x28 invoke_syscall+0x54/0x120 el0_svc_common.constprop.4+0x16c/0x1f0 do_el0_svc+0x34/0xb0 el0_svc+0x1c/0x28 el0_sync_handler+0x8c/0xb0 el0_sync+0x148/0x180 Fixes: ad362fe07fec ("KVM: arm64: vgic-its: Avoid potential UAF in LPI translation cache") Cc: stable(a)vger.kernel.org Signed-off-by: Wenyao Hai <haiwenyao(a)uniontech.com> Signed-off-by: WangYuli <wangyuli(a)uniontech.com> --- arch/arm64/kvm/vgic/vgic-its.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c index ba945ba78cc7..fb5f57cbab42 100644 --- a/arch/arm64/kvm/vgic/vgic-its.c +++ b/arch/arm64/kvm/vgic/vgic-its.c @@ -679,6 +679,7 @@ static int vgic_its_trigger_msi(struct kvm *kvm, struct vgic_its *its, raw_spin_lock_irqsave(&irq->irq_lock, flags); irq->pending_latch = true; vgic_queue_irq_unlock(kvm, irq, flags); + vgic_put_irq(kvm, irq); return 0; } @@ -697,7 +698,6 @@ int vgic_its_inject_cached_translation(struct kvm *kvm, struct kvm_msi *msi) raw_spin_lock_irqsave(&irq->irq_lock, flags); irq->pending_latch = true; vgic_queue_irq_unlock(kvm, irq, flags); - vgic_put_irq(kvm, irq); return 0; } -- 2.45.2

11 months

2
1
0 0

[PATCH v3 1/2] riscv: vdso: Prevent the compiler from inserting calls to memset()

by Alexandre Ghiti

The compiler is smart enough to insert a call to memset() in riscv_vdso_get_cpus(), which generates a dynamic relocation. So prevent this by using -fno-builtin option. Fixes: e2c0cdfba7f6 ("RISC-V: User-facing API") Cc: stable(a)vger.kernel.org Signed-off-by: Alexandre Ghiti <alexghiti(a)rivosinc.com> --- arch/riscv/kernel/vdso/Makefile | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile index 960feb1526ca..3f1c4b2d0b06 100644 --- a/arch/riscv/kernel/vdso/Makefile +++ b/arch/riscv/kernel/vdso/Makefile @@ -18,6 +18,7 @@ obj-vdso = $(patsubst %, %.o, $(vdso-syms)) note.o ccflags-y := -fno-stack-protector ccflags-y += -DDISABLE_BRANCH_PROFILING +ccflags-y += -fno-builtin ifneq ($(c-gettimeofday-y),) CFLAGS_vgettimeofday.o += -fPIC -include $(c-gettimeofday-y) -- 2.39.2

11 months

2
1
0 0

[PATCH 2/2] [tip: sched/core] sched: Move PLACE_LAG and RUN_TO_PARITY to sysctl

by Cristian Prundeanu

These two scheduler features have a high impact on performance for some database workloads. Move them to sysctl as they are likely to be modified and persisted across reboots. Cc: <stable(a)vger.kernel.org> # 6.6.x Fixes: 86bfbb7ce4f6 ("sched/fair: Add lag based placement") Fixes: 63304558ba5d ("sched/eevdf: Curb wakeup-preemption") Signed-off-by: Cristian Prundeanu <cpru(a)amazon.com> --- include/linux/sched/sysctl.h | 8 ++++++++ kernel/sched/core.c | 13 +++++++++++++ kernel/sched/fair.c | 5 +++-- kernel/sched/features.h | 10 ---------- kernel/sysctl.c | 20 ++++++++++++++++++++ 5 files changed, 44 insertions(+), 12 deletions(-) diff --git a/include/linux/sched/sysctl.h b/include/linux/sched/sysctl.h index 5a64582b086b..0258fba3896a 100644 --- a/include/linux/sched/sysctl.h +++ b/include/linux/sched/sysctl.h @@ -29,4 +29,12 @@ extern int sysctl_numa_balancing_mode; #define sysctl_numa_balancing_mode 0 #endif +#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL) +extern unsigned int sysctl_sched_place_lag_enabled; +extern unsigned int sysctl_sched_run_to_parity_enabled; +#else +#define sysctl_sched_place_lag_enabled 0 +#define sysctl_sched_run_to_parity_enabled 0 +#endif + #endif /* _LINUX_SCHED_SYSCTL_H */ diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 43e453ab7e20..c6bd1bda8c7e 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -134,6 +134,19 @@ const_debug unsigned int sysctl_sched_features = 0; #undef SCHED_FEAT +#ifdef CONFIG_SYSCTL +/* + * Using the avg_vruntime, do the right thing and preserve lag across + * sleep+wake cycles. EEVDF placement strategy #1, #2 if disabled. + */ +__read_mostly unsigned int sysctl_sched_place_lag_enabled = 0; +/* + * Inhibit (wakeup) preemption until the current task has either matched the + * 0-lag point or until is has exhausted it's slice. + */ +__read_mostly unsigned int sysctl_sched_run_to_parity_enabled = 0; +#endif + /* * Print a warning if need_resched is set for the given duration (if * LATENCY_WARN is enabled). diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 5a621210c9c1..c58b76233f59 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -925,7 +925,8 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq) * Once selected, run a task until it either becomes non-eligible or * until it gets a new slice. See the HACK in set_next_entity(). */ - if (sched_feat(RUN_TO_PARITY) && curr && curr->vlag == curr->deadline) + if (sysctl_sched_run_to_parity_enabled && + curr && curr->vlag == curr->deadline) return curr; /* Pick the leftmost entity if it's eligible */ @@ -5280,7 +5281,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * * EEVDF: placement strategy #1 / #2 */ - if (sched_feat(PLACE_LAG) && cfs_rq->nr_running && se->vlag) { + if (sysctl_sched_place_lag_enabled && cfs_rq->nr_running && se->vlag) { struct sched_entity *curr = cfs_rq->curr; unsigned long load; diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 8a5ca80665b3..b39a9dde0b54 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -1,10 +1,5 @@ /* SPDX-License-Identifier: GPL-2.0 */ -/* - * Using the avg_vruntime, do the right thing and preserve lag across - * sleep+wake cycles. EEVDF placement strategy #1, #2 if disabled. - */ -SCHED_FEAT(PLACE_LAG, false) /* * Give new tasks half a slice to ease into the competition. */ @@ -13,11 +8,6 @@ SCHED_FEAT(PLACE_DEADLINE_INITIAL, true) * Preserve relative virtual deadline on 'migration'. */ SCHED_FEAT(PLACE_REL_DEADLINE, true) -/* - * Inhibit (wakeup) preemption until the current task has either matched the - * 0-lag point or until is has exhausted it's slice. - */ -SCHED_FEAT(RUN_TO_PARITY, false) /* * Allow wakeup of tasks with a shorter slice to cancel RUN_TO_PARITY for * current. diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 79e6cb1d5c48..f435b741654a 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2029,6 +2029,26 @@ static struct ctl_table kern_table[] = { .extra2 = SYSCTL_INT_MAX, }, #endif +#ifdef CONFIG_SCHED_DEBUG + { + .procname = "sched_place_lag_enabled", + .data = &sysctl_sched_place_lag_enabled, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, + { + .procname = "sched_run_to_parity_enabled", + .data = &sysctl_sched_run_to_parity_enabled, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, +#endif }; static struct ctl_table vm_table[] = { -- 2.40.1

11 months

1
0
0 0

[PATCH 1/2] [tip: sched/core] sched: Disable PLACE_LAG and RUN_TO_PARITY

by Cristian Prundeanu

With these features are enabled, the EEVDF scheduler introduces a large performance degradation, observed in multiple database tests on kernel versions using EEVDF, across multiple architectures (x86, aarch64, amd64) and CPU generations. Disable the features to minimize default performance impact. Cc: <stable(a)vger.kernel.org> # 6.6.x Fixes: 86bfbb7ce4f6 ("sched/fair: Add lag based placement") Fixes: 63304558ba5d ("sched/eevdf: Curb wakeup-preemption") Signed-off-by: Cristian Prundeanu <cpru(a)amazon.com> --- kernel/sched/features.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/features.h b/kernel/sched/features.h index a3d331dd2d8f..8a5ca80665b3 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -4,7 +4,7 @@ * Using the avg_vruntime, do the right thing and preserve lag across * sleep+wake cycles. EEVDF placement strategy #1, #2 if disabled. */ -SCHED_FEAT(PLACE_LAG, true) +SCHED_FEAT(PLACE_LAG, false) /* * Give new tasks half a slice to ease into the competition. */ @@ -17,7 +17,7 @@ SCHED_FEAT(PLACE_REL_DEADLINE, true) * Inhibit (wakeup) preemption until the current task has either matched the * 0-lag point or until is has exhausted it's slice. */ -SCHED_FEAT(RUN_TO_PARITY, true) +SCHED_FEAT(RUN_TO_PARITY, false) /* * Allow wakeup of tasks with a shorter slice to cancel RUN_TO_PARITY for * current. -- 2.40.1

11 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror October 2024