| Message ID | 20210427225244.4326-7-axelrasmussen@google.com |
|---|---|
| State | New |
| Headers | show |
| Series | userfaultfd: add minor fault handling for shmem | expand |
On Tue, 27 Apr 2021, Axel Rasmussen wrote: > In a previous commit, we added the mcopy_atomic_install_pte() helper. > This helper does the job of setting up PTEs for an existing page, to map > it into a given VMA. It deals with both the anon and shmem cases, as > well as the shared and private cases. > > In other words, shmem_mcopy_atomic_pte() duplicates a case it already > handles. So, expose it, and let shmem_mcopy_atomic_pte() use it > directly, to reduce code duplication. > > This requires that we refactor shmem_mcopy_atomic_pte() a bit: > > Instead of doing accounting (shmem_recalc_inode() et al) part-way > through the PTE setup, do it afterward. This frees up > mcopy_atomic_install_pte() from having to care about this accounting, > and means we don't need to e.g. shmem_uncharge() in the error path. > > A side effect is this switches shmem_mcopy_atomic_pte() to use > lru_cache_add_inactive_or_unevictable() instead of just lru_cache_add(). > This wrapper does some extra accounting in an exceptional case, if > appropriate, so it's actually the more correct thing to use. > > Signed-off-by: Axel Rasmussen <axelrasmussen@google.com> Not quite. Two things. One, in this version, delete_from_page_cache(page) has vanished from the particular error path which needs it. Two, and I think this predates your changes (so needs a separate fix patch first, for backport to stable? a user with bad intentions might be able to trigger the BUG), in pondering the new error paths and that /* don't free the page */ one in particular, isn't it the case that the shmem_inode_acct_block() on entry might succeed the first time, but atomic copy fail so -ENOENT, then something else fill up the tmpfs before the retry comes in, so that retry then fail with -ENOMEM, and hit the BUG_ON(page) in __mcopy_atomic()? (As I understand it, the shmem_inode_unacct_blocks() has to be done before returning, because the caller may be unable to retry.) What the right fix is rather depends on other uses of __mcopy_atomic(): if they obviously cannot hit that BUG_ON(page), you may prefer to leave it in, and fix it here where shmem_inode_acct_block() fails. Or you may prefer instead to delete that "else BUG_ON(page);" - looks as if that would end up doing the right thing. Peter may have a preference. (Or, we could consider doing the shmem_inode_acct_block() only after the page has been copied in: its current placing reflects how shmem.c does it elsewhere, and there's reason for that, but it doesn't always work out right. Don't be surprised if I change the ordering in future, but it's probably best not to mess with that ordering now.) Sorry, if this is a pre-existing issue, then we are taking advantage of you, in asking you to fix it: but I hope that while you're in there, it will make sense to do so. Thanks, Hugh > --- > include/linux/userfaultfd_k.h | 5 ++++ > mm/shmem.c | 48 +++++------------------------------ > mm/userfaultfd.c | 17 +++++-------- > 3 files changed, 18 insertions(+), 52 deletions(-)
On Tue, Apr 27, 2021 at 05:58:16PM -0700, Hugh Dickins wrote: > On Tue, 27 Apr 2021, Axel Rasmussen wrote: > > > In a previous commit, we added the mcopy_atomic_install_pte() helper. > > This helper does the job of setting up PTEs for an existing page, to map > > it into a given VMA. It deals with both the anon and shmem cases, as > > well as the shared and private cases. > > > > In other words, shmem_mcopy_atomic_pte() duplicates a case it already > > handles. So, expose it, and let shmem_mcopy_atomic_pte() use it > > directly, to reduce code duplication. > > > > This requires that we refactor shmem_mcopy_atomic_pte() a bit: > > > > Instead of doing accounting (shmem_recalc_inode() et al) part-way > > through the PTE setup, do it afterward. This frees up > > mcopy_atomic_install_pte() from having to care about this accounting, > > and means we don't need to e.g. shmem_uncharge() in the error path. > > > > A side effect is this switches shmem_mcopy_atomic_pte() to use > > lru_cache_add_inactive_or_unevictable() instead of just lru_cache_add(). > > This wrapper does some extra accounting in an exceptional case, if > > appropriate, so it's actually the more correct thing to use. > > > > Signed-off-by: Axel Rasmussen <axelrasmussen@google.com> > > Not quite. Two things. > > One, in this version, delete_from_page_cache(page) has vanished > from the particular error path which needs it. Agreed. I also spotted that the set_page_dirty() seems to have been overlooked when reusing mcopy_atomic_install_pte(), which afaiu should be move into the helper. > > Two, and I think this predates your changes (so needs a separate > fix patch first, for backport to stable? a user with bad intentions > might be able to trigger the BUG), in pondering the new error paths > and that /* don't free the page */ one in particular, isn't it the > case that the shmem_inode_acct_block() on entry might succeed the > first time, but atomic copy fail so -ENOENT, then something else > fill up the tmpfs before the retry comes in, so that retry then > fail with -ENOMEM, and hit the BUG_ON(page) in __mcopy_atomic()? > > (As I understand it, the shmem_inode_unacct_blocks() has to be > done before returning, because the caller may be unable to retry.) > > What the right fix is rather depends on other uses of __mcopy_atomic(): > if they obviously cannot hit that BUG_ON(page), you may prefer to leave > it in, and fix it here where shmem_inode_acct_block() fails. Or you may > prefer instead to delete that "else BUG_ON(page);" - looks as if that > would end up doing the right thing. Peter may have a preference. To me, the BUG_ON(page) wanted to guarantee mfill_atomic_pte() should have consumed the page properly when possible. Removing the BUG_ON() looks good already, it will just stop covering the case when e.g. ret==0. So maybe slightly better to release the page when shmem_inode_acct_block() fails (so as to still keep some guard on the page)? Thanks, -- Peter Xu
On Wed, Apr 28, 2021 at 8:56 AM Peter Xu <peterx@redhat.com> wrote: > > On Tue, Apr 27, 2021 at 05:58:16PM -0700, Hugh Dickins wrote: > > On Tue, 27 Apr 2021, Axel Rasmussen wrote: > > > > > In a previous commit, we added the mcopy_atomic_install_pte() helper. > > > This helper does the job of setting up PTEs for an existing page, to map > > > it into a given VMA. It deals with both the anon and shmem cases, as > > > well as the shared and private cases. > > > > > > In other words, shmem_mcopy_atomic_pte() duplicates a case it already > > > handles. So, expose it, and let shmem_mcopy_atomic_pte() use it > > > directly, to reduce code duplication. > > > > > > This requires that we refactor shmem_mcopy_atomic_pte() a bit: > > > > > > Instead of doing accounting (shmem_recalc_inode() et al) part-way > > > through the PTE setup, do it afterward. This frees up > > > mcopy_atomic_install_pte() from having to care about this accounting, > > > and means we don't need to e.g. shmem_uncharge() in the error path. > > > > > > A side effect is this switches shmem_mcopy_atomic_pte() to use > > > lru_cache_add_inactive_or_unevictable() instead of just lru_cache_add(). > > > This wrapper does some extra accounting in an exceptional case, if > > > appropriate, so it's actually the more correct thing to use. > > > > > > Signed-off-by: Axel Rasmussen <axelrasmussen@google.com> > > > > Not quite. Two things. > > > > One, in this version, delete_from_page_cache(page) has vanished > > from the particular error path which needs it. > > Agreed. I also spotted that the set_page_dirty() seems to have been overlooked > when reusing mcopy_atomic_install_pte(), which afaiu should be move into the > helper. I think this is covered: we explicitly call SetPageDirty() just before returning in shmem_mcopy_atomic_pte(). If I remember correctly from a couple of revisions ago, we consciously put it here instead of in the helper because it resulted in simpler code (error handling in particular, I think?), and not all callers of the new helper need it. > > > > > Two, and I think this predates your changes (so needs a separate > > fix patch first, for backport to stable? a user with bad intentions > > might be able to trigger the BUG), in pondering the new error paths > > and that /* don't free the page */ one in particular, isn't it the > > case that the shmem_inode_acct_block() on entry might succeed the > > first time, but atomic copy fail so -ENOENT, then something else > > fill up the tmpfs before the retry comes in, so that retry then > > fail with -ENOMEM, and hit the BUG_ON(page) in __mcopy_atomic()? > > > > (As I understand it, the shmem_inode_unacct_blocks() has to be > > done before returning, because the caller may be unable to retry.) > > > > What the right fix is rather depends on other uses of __mcopy_atomic(): > > if they obviously cannot hit that BUG_ON(page), you may prefer to leave > > it in, and fix it here where shmem_inode_acct_block() fails. Or you may > > prefer instead to delete that "else BUG_ON(page);" - looks as if that > > would end up doing the right thing. Peter may have a preference. > > To me, the BUG_ON(page) wanted to guarantee mfill_atomic_pte() should have > consumed the page properly when possible. Removing the BUG_ON() looks good > already, it will just stop covering the case when e.g. ret==0. > > So maybe slightly better to release the page when shmem_inode_acct_block() > fails (so as to still keep some guard on the page)? This second issue, I will take some more time to investigate. :) > > Thanks, > > -- > Peter Xu >
On Wed, Apr 28, 2021 at 08:59:53AM -0700, Axel Rasmussen wrote: > On Wed, Apr 28, 2021 at 8:56 AM Peter Xu <peterx@redhat.com> wrote: > > > > On Tue, Apr 27, 2021 at 05:58:16PM -0700, Hugh Dickins wrote: > > > On Tue, 27 Apr 2021, Axel Rasmussen wrote: > > > > > > > In a previous commit, we added the mcopy_atomic_install_pte() helper. > > > > This helper does the job of setting up PTEs for an existing page, to map > > > > it into a given VMA. It deals with both the anon and shmem cases, as > > > > well as the shared and private cases. > > > > > > > > In other words, shmem_mcopy_atomic_pte() duplicates a case it already > > > > handles. So, expose it, and let shmem_mcopy_atomic_pte() use it > > > > directly, to reduce code duplication. > > > > > > > > This requires that we refactor shmem_mcopy_atomic_pte() a bit: > > > > > > > > Instead of doing accounting (shmem_recalc_inode() et al) part-way > > > > through the PTE setup, do it afterward. This frees up > > > > mcopy_atomic_install_pte() from having to care about this accounting, > > > > and means we don't need to e.g. shmem_uncharge() in the error path. > > > > > > > > A side effect is this switches shmem_mcopy_atomic_pte() to use > > > > lru_cache_add_inactive_or_unevictable() instead of just lru_cache_add(). > > > > This wrapper does some extra accounting in an exceptional case, if > > > > appropriate, so it's actually the more correct thing to use. > > > > > > > > Signed-off-by: Axel Rasmussen <axelrasmussen@google.com> > > > > > > Not quite. Two things. > > > > > > One, in this version, delete_from_page_cache(page) has vanished > > > from the particular error path which needs it. > > > > Agreed. I also spotted that the set_page_dirty() seems to have been overlooked > > when reusing mcopy_atomic_install_pte(), which afaiu should be move into the > > helper. > > I think this is covered: we explicitly call SetPageDirty() just before > returning in shmem_mcopy_atomic_pte(). If I remember correctly from a > couple of revisions ago, we consciously put it here instead of in the > helper because it resulted in simpler code (error handling in > particular, I think?), and not all callers of the new helper need it. Indeed, yes that looks okay. > > > > > > > > > Two, and I think this predates your changes (so needs a separate > > > fix patch first, for backport to stable? a user with bad intentions > > > might be able to trigger the BUG), in pondering the new error paths > > > and that /* don't free the page */ one in particular, isn't it the > > > case that the shmem_inode_acct_block() on entry might succeed the > > > first time, but atomic copy fail so -ENOENT, then something else > > > fill up the tmpfs before the retry comes in, so that retry then > > > fail with -ENOMEM, and hit the BUG_ON(page) in __mcopy_atomic()? > > > > > > (As I understand it, the shmem_inode_unacct_blocks() has to be > > > done before returning, because the caller may be unable to retry.) > > > > > > What the right fix is rather depends on other uses of __mcopy_atomic(): > > > if they obviously cannot hit that BUG_ON(page), you may prefer to leave > > > it in, and fix it here where shmem_inode_acct_block() fails. Or you may > > > prefer instead to delete that "else BUG_ON(page);" - looks as if that > > > would end up doing the right thing. Peter may have a preference. > > > > To me, the BUG_ON(page) wanted to guarantee mfill_atomic_pte() should have > > consumed the page properly when possible. Removing the BUG_ON() looks good > > already, it will just stop covering the case when e.g. ret==0. > > > > So maybe slightly better to release the page when shmem_inode_acct_block() > > fails (so as to still keep some guard on the page)? > > This second issue, I will take some more time to investigate. :) No worry - take your time. :) -- Peter Xu
diff --git a/include/linux/userfaultfd_k.h b/include/linux/userfaultfd_k.h index 794d1538b8ba..39c094cc6641 100644 --- a/include/linux/userfaultfd_k.h +++ b/include/linux/userfaultfd_k.h @@ -53,6 +53,11 @@ enum mcopy_atomic_mode { MCOPY_ATOMIC_CONTINUE, }; +extern int mcopy_atomic_install_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd, + struct vm_area_struct *dst_vma, + unsigned long dst_addr, struct page *page, + bool newly_allocated, bool wp_copy); + extern ssize_t mcopy_atomic(struct mm_struct *dst_mm, unsigned long dst_start, unsigned long src_start, unsigned long len, bool *mmap_changing, __u64 mode); diff --git a/mm/shmem.c b/mm/shmem.c index 30c0bb501dc9..37db52f45cb5 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -2378,10 +2378,8 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, struct address_space *mapping = inode->i_mapping; gfp_t gfp = mapping_gfp_mask(mapping); pgoff_t pgoff = linear_page_index(dst_vma, dst_addr); - spinlock_t *ptl; void *page_kaddr; struct page *page; - pte_t _dst_pte, *dst_pte; int ret; pgoff_t max_off; @@ -2404,9 +2402,9 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, /* fallback to copy_from_user outside mmap_lock */ if (unlikely(ret)) { *pagep = page; - shmem_inode_unacct_blocks(inode, 1); + ret = -ENOENT; /* don't free the page */ - return -ENOENT; + goto out_unacct_blocks; } } else { /* ZEROPAGE */ clear_highpage(page); @@ -2432,32 +2430,10 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, if (ret) goto out_release; - _dst_pte = mk_pte(page, dst_vma->vm_page_prot); - if (dst_vma->vm_flags & VM_WRITE) - _dst_pte = pte_mkwrite(pte_mkdirty(_dst_pte)); - else { - /* - * We don't set the pte dirty if the vma has no - * VM_WRITE permission, so mark the page dirty or it - * could be freed from under us. We could do it - * unconditionally before unlock_page(), but doing it - * only if VM_WRITE is not set is faster. - */ - set_page_dirty(page); - } - - dst_pte = pte_offset_map_lock(dst_mm, dst_pmd, dst_addr, &ptl); - - ret = -EFAULT; - max_off = DIV_ROUND_UP(i_size_read(inode), PAGE_SIZE); - if (unlikely(pgoff >= max_off)) - goto out_release_unlock; - - ret = -EEXIST; - if (!pte_none(*dst_pte)) - goto out_release_unlock; - - lru_cache_add(page); + ret = mcopy_atomic_install_pte(dst_mm, dst_pmd, dst_vma, dst_addr, + page, true, false); + if (ret) + goto out_release; spin_lock_irq(&info->lock); info->alloced++; @@ -2465,21 +2441,11 @@ int shmem_mcopy_atomic_pte(struct mm_struct *dst_mm, shmem_recalc_inode(inode); spin_unlock_irq(&info->lock); - inc_mm_counter(dst_mm, mm_counter_file(page)); - page_add_file_rmap(page, false); - set_pte_at(dst_mm, dst_addr, dst_pte, _dst_pte); - - /* No need to invalidate - it was non-present before */ - update_mmu_cache(dst_vma, dst_addr, dst_pte); - pte_unmap_unlock(dst_pte, ptl); + SetPageDirty(page); unlock_page(page); ret = 0; out: return ret; -out_release_unlock: - pte_unmap_unlock(dst_pte, ptl); - ClearPageDirty(page); - delete_from_page_cache(page); out_release: unlock_page(page); put_page(page); diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index 51d8c0127161..3a9ddbb2dbbd 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -51,18 +51,13 @@ struct vm_area_struct *find_dst_vma(struct mm_struct *dst_mm, /* * Install PTEs, to map dst_addr (within dst_vma) to page. * - * This function handles MCOPY_ATOMIC_CONTINUE (which is always file-backed), - * whether or not dst_vma is VM_SHARED. It also handles the more general - * MCOPY_ATOMIC_NORMAL case, when dst_vma is *not* VM_SHARED (it may be file - * backed, or not). - * - * Note that MCOPY_ATOMIC_NORMAL for a VM_SHARED dst_vma is handled by - * shmem_mcopy_atomic_pte instead. + * This function handles both MCOPY_ATOMIC_NORMAL and _CONTINUE for both shmem + * and anon, and for both shared and private VMAs. */ -static int mcopy_atomic_install_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd, - struct vm_area_struct *dst_vma, - unsigned long dst_addr, struct page *page, - bool newly_allocated, bool wp_copy) +int mcopy_atomic_install_pte(struct mm_struct *dst_mm, pmd_t *dst_pmd, + struct vm_area_struct *dst_vma, + unsigned long dst_addr, struct page *page, + bool newly_allocated, bool wp_copy) { int ret; pte_t _dst_pte, *dst_pte;
In a previous commit, we added the mcopy_atomic_install_pte() helper. This helper does the job of setting up PTEs for an existing page, to map it into a given VMA. It deals with both the anon and shmem cases, as well as the shared and private cases. In other words, shmem_mcopy_atomic_pte() duplicates a case it already handles. So, expose it, and let shmem_mcopy_atomic_pte() use it directly, to reduce code duplication. This requires that we refactor shmem_mcopy_atomic_pte() a bit: Instead of doing accounting (shmem_recalc_inode() et al) part-way through the PTE setup, do it afterward. This frees up mcopy_atomic_install_pte() from having to care about this accounting, and means we don't need to e.g. shmem_uncharge() in the error path. A side effect is this switches shmem_mcopy_atomic_pte() to use lru_cache_add_inactive_or_unevictable() instead of just lru_cache_add(). This wrapper does some extra accounting in an exceptional case, if appropriate, so it's actually the more correct thing to use. Signed-off-by: Axel Rasmussen <axelrasmussen@google.com> --- include/linux/userfaultfd_k.h | 5 ++++ mm/shmem.c | 48 +++++------------------------------ mm/userfaultfd.c | 17 +++++-------- 3 files changed, 18 insertions(+), 52 deletions(-)