The patch titled Subject: mm: defer second attempt at merge on mmap() has been added to the -mm mm-unstable branch. Its filename is mm-defer-second-attempt-at-merge-on-mmap.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-defer-second-attempt-at-merge-on-mmap.patch This patch will later appear in the mm-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days ------------------------------------------------------ From: Lorenzo Stoakes <lorenzo.stoakes@xxxxxxxxxx> Subject: mm: defer second attempt at merge on mmap() Date: Wed, 23 Oct 2024 21:38:33 +0100 Rather than trying to merge again when ostensibly allocating a new VMA, instead defer until the VMA is added and attempt to merge the existing range. This way we have no complicated unwinding logic midway through the process of mapping the VMA. In addition this removes limitations on the VMA not being able to be the first in the virtual memory address space which was previously implicitly required. It also performs this merge after the final flag adjustments are performed, something that was not done previously and thus might have prevented possibly valid merges in the past. In theory, for this very same reason, we should unconditionally attempt merge here, however this is likely to have a performance impact so it is better to avoid this given the unlikely outcome of a merge. The vmg state will already have been reset by the first attempt at a merge so we only need to reset the iterator, set the vma and flags and try again. Link: https://lkml.kernel.org/r/e063d9d31a4dbd83c2d77a3f1e0114e6a0788724.1729715266.git.lorenzo.stoakes@xxxxxxxxxx Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@xxxxxxxxxx> Cc: Jann Horn <jannh@xxxxxxxxxx> Cc: Liam R. Howlett <Liam.Howlett@xxxxxxxxxx> Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Cc: Vlastimil Babka <vbabka@xxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- mm/vma.c | 75 ++++++++++++++++++----------------------------------- 1 file changed, 26 insertions(+), 49 deletions(-) --- a/mm/vma.c~mm-defer-second-attempt-at-merge-on-mmap +++ a/mm/vma.c @@ -19,6 +19,7 @@ struct mmap_state { struct file *file; unsigned long charged; + bool retry_merge; struct vm_area_struct *prev; struct vm_area_struct *next; @@ -2280,9 +2281,9 @@ static int __mmap_prepare(struct mmap_st return 0; } + static int __mmap_new_file_vma(struct mmap_state *map, - struct vma_merge_struct *vmg, - struct vm_area_struct **vmap, bool *mergedp) + struct vm_area_struct **vmap) { struct vma_iterator *vmi = map->vmi; struct vm_area_struct *vma = *vmap; @@ -2311,37 +2312,11 @@ static int __mmap_new_file_vma(struct mm !(map->flags & VM_MAYWRITE) && (vma->vm_flags & VM_MAYWRITE)); - vma_iter_config(vmi, map->addr, map->end); - /* - * If flags changed after mmap_file(), we should try merge - * vma again as we may succeed this time. - */ - if (unlikely(map->flags != vma->vm_flags && map->prev)) { - struct vm_area_struct *merge; - - vmg->flags = vma->vm_flags; - /* If this fails, state is reset ready for a reattempt. */ - merge = vma_merge_new_range(vmg); - - if (merge) { - /* - * ->mmap() can change vma->vm_file and fput - * the original file. So fput the vma->vm_file - * here or we would add an extra fput for file - * and cause general protection fault - * ultimately. - */ - fput(vma->vm_file); - vm_area_free(vma); - vma = merge; - *mergedp = true; - } else { - vma_iter_config(vmi, map->addr, map->end); - } - } + /* If the flags change (and are mergeable), let's retry later. */ + map->retry_merge = vma->vm_flags != map->flags && !(vma->vm_flags & VM_SPECIAL); + vma_iter_config(vmi, map->addr, map->end); map->flags = vma->vm_flags; - *vmap = vma; return 0; } @@ -2349,22 +2324,15 @@ static int __mmap_new_file_vma(struct mm * __mmap_new_vma() - Allocate a new VMA for the region, as merging was not * possible. * - * An exception to this is if the mapping is file-backed, and the underlying - * driver changes the VMA flags, permitting a subsequent merge of the VMA, in - * which case the returned VMA is one that was merged on a second attempt. - * * @map: Mapping state. - * @vmg: VMA merge state. * @vmap: Output pointer for the new VMA. * * Returns: Zero on success, or an error. */ -static int __mmap_new_vma(struct mmap_state *map, struct vma_merge_struct *vmg, - struct vm_area_struct **vmap) +static int __mmap_new_vma(struct mmap_state *map, struct vm_area_struct **vmap) { struct vma_iterator *vmi = map->vmi; int error = 0; - bool merged = false; struct vm_area_struct *vma; /* @@ -2387,7 +2355,7 @@ static int __mmap_new_vma(struct mmap_st } if (map->file) - error = __mmap_new_file_vma(map, vmg, &vma, &merged); + error = __mmap_new_file_vma(map, &vma); else if (map->flags & VM_SHARED) error = shmem_zero_setup(vma); else @@ -2396,9 +2364,6 @@ static int __mmap_new_vma(struct mmap_st if (error) goto free_iter_vma; - if (merged) - goto file_expanded; - #ifdef CONFIG_SPARC64 /* TODO: Fix SPARC ADI! */ WARN_ON_ONCE(!arch_validate_flags(map->flags)); @@ -2415,8 +2380,6 @@ static int __mmap_new_vma(struct mmap_st * call covers the non-merge case. */ khugepaged_enter_vma(vma, map->flags); - -file_expanded: ksm_add_vma(vma); *vmap = vma; return 0; @@ -2430,13 +2393,17 @@ free_vma: /* * __mmap_complete() - Unmap any VMAs we overlap, account memory mapping - * statistics, handle locking and finalise the VMA. + * statistics, handle locking and finalise the VMA, + * attempt a final merge if required. * * @map: Mapping state. * @vma: Merged or newly allocated VMA for the mmap()'d region. + * @vmg: VMA merge state. */ -static void __mmap_complete(struct mmap_state *map, struct vm_area_struct *vma) +static void __mmap_complete(struct mmap_state *map, struct vm_area_struct *vma, + struct vma_merge_struct *vmg) { + struct mm_struct *mm = map->mm; unsigned long vm_flags = vma->vm_flags; @@ -2468,6 +2435,16 @@ static void __mmap_complete(struct mmap_ vm_flags_set(vma, VM_SOFTDIRTY); vma_set_page_prot(vma); + + /* OK VMA flags changed in __mmap_new_vma(), try a merge again. */ + if (map->retry_merge) { + vma_iter_config(map->vmi, map->addr, map->end); + vmg->vma = vma; + vmg->flags = map->flags; + vmg->next = NULL; /* Must be set by merge logic. */ + + vma_merge_existing_range(vmg); + } } unsigned long __mmap_region(struct file *file, unsigned long addr, @@ -2490,12 +2467,12 @@ unsigned long __mmap_region(struct file vma = vma_merge_new_range(&vmg); if (!vma) { /* ...but if we can't, allocate a new VMA. */ - error = __mmap_new_vma(&map, &vmg, &vma); + error = __mmap_new_vma(&map, &vma); if (error) goto unacct_error; } - __mmap_complete(&map, vma); + __mmap_complete(&map, vma, &vmg); return addr; _ Patches currently in -mm which might be from lorenzo.stoakes@xxxxxxxxxx are fork-do-not-invoke-uffd-on-fork-if-error-occurs.patch fork-only-invoke-khugepaged-ksm-hooks-if-no-error.patch mm-vma-add-expand-only-vma-merge-mode-and-optimise-do_brk_flags.patch tools-testing-add-expand-only-mode-vma-test.patch mm-avoid-unsafe-vma-hook-invocation-when-error-arises-on-mmap-hook.patch mm-unconditionally-close-vmas-on-error.patch mm-refactor-map_deny_write_exec.patch mm-resolve-faulty-mmap_region-error-path-behaviour.patch selftests-mm-add-pkey_sighandler_xx-hugetlb_dio-to-gitignore.patch mm-refactor-mm_access-to-not-return-null.patch mm-refactor-mm_access-to-not-return-null-fix.patch mm-madvise-unrestrict-process_madvise-for-current-process.patch maple_tree-do-not-hash-pointers-on-dump-in-debug-mode.patch tools-testing-fix-phys_addr_t-size-on-64-bit-systems.patch tools-testing-add-additional-vma_internalh-stubs.patch mm-isolate-mmap-internal-logic-to-mm-vmac.patch mm-refactor-__mmap_region.patch mm-defer-second-attempt-at-merge-on-mmap.patch