| Message ID | 55e413d20678a1bb4c7cce889062bbb07b0df892.1697116581.git.lstoakes@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | permit write-sealed memfd read-only shared mappings | expand |
On Thu, Oct 12, 2023 at 06:04:30PM +0100, Lorenzo Stoakes wrote: > In order for a F_SEAL_WRITE sealed memfd mapping to have an opportunity to > clear VM_MAYWRITE, we must be able to invoke the appropriate vm_ops->mmap() > handler to do so. We would otherwise fail the mapping_map_writable() check > before we had the opportunity to avoid it. > > This patch moves this check after the call_mmap() invocation. Only memfd > actively denies write access causing a potential failure here (in > memfd_add_seals()), so there should be no impact on non-memfd cases. > > This patch makes the userland-visible change that MAP_SHARED, PROT_READ > mappings of an F_SEAL_WRITE sealed memfd mapping will now succeed. > > There is a delicate situation with cleanup paths assuming that a writable > mapping must have occurred in circumstances where it may now not have. In > order to ensure we do not accidentally mark a writable file unwritable by > mistake, we explicitly track whether we have a writable mapping and > unmap only if we do. > > Link: https://bugzilla.kernel.org/show_bug.cgi?id=217238 > Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com> > --- > mm/mmap.c | 23 ++++++++++++++--------- > 1 file changed, 14 insertions(+), 9 deletions(-) > [snip] Andrew, could you apply the following -fix patch to this? As a bug was detected in the implementation [0] - I was being over-zealous in setting the writable_file_mapping flag and had falsely assumed vma->vm_file == file in all instances of the cleanup. The fix is to only set it in one place. [0]: https://lore.kernel.org/all/CA+G9fYtL7wK-dE-Tnz4t-GWmQb50EPYa=TWGjpgYU2Z=oeAO_w@mail.gmail.com/ ----8<---- From 7feea6faada5b10a872c24755cc630220cba619a Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes <lstoakes@gmail.com> Date: Mon, 16 Oct 2023 17:17:13 +0100 Subject: [PATCH] mm: perform the mapping_map_writable() check after call_mmap() Do not set writable_file_mapping in an instance where it is not appropriate to do so. Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com> --- mm/mmap.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/mm/mmap.c b/mm/mmap.c index 7f45a08e7973..8b57e42fd980 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -2923,10 +2923,8 @@ unsigned long mmap_region(struct file *file, unsigned long addr, mm->map_count++; if (vma->vm_file) { i_mmap_lock_write(vma->vm_file->f_mapping); - if (vma_is_shared_maywrite(vma)) { + if (vma_is_shared_maywrite(vma)) mapping_allow_writable(vma->vm_file->f_mapping); - writable_file_mapping = true; - } flush_dcache_mmap_lock(vma->vm_file->f_mapping); vma_interval_tree_insert(vma, &vma->vm_file->f_mapping->i_mmap);
diff --git a/mm/mmap.c b/mm/mmap.c index 0041e3631f6c..7f45a08e7973 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -2752,6 +2752,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr, unsigned long charged = 0; unsigned long end = addr + len; unsigned long merge_start = addr, merge_end = end; + bool writable_file_mapping = false; pgoff_t vm_pgoff; int error; VMA_ITERATOR(vmi, mm, addr); @@ -2846,17 +2847,19 @@ unsigned long mmap_region(struct file *file, unsigned long addr, vma->vm_pgoff = pgoff; if (file) { - if (is_shared_maywrite(vm_flags)) { - error = mapping_map_writable(file->f_mapping); - if (error) - goto free_vma; - } - vma->vm_file = get_file(file); error = call_mmap(file, vma); if (error) goto unmap_and_free_vma; + if (vma_is_shared_maywrite(vma)) { + error = mapping_map_writable(file->f_mapping); + if (error) + goto close_and_free_vma; + + writable_file_mapping = true; + } + /* * Expansion is handled above, merging is handled below. * Drivers should not alter the address of the VMA. @@ -2920,8 +2923,10 @@ unsigned long mmap_region(struct file *file, unsigned long addr, mm->map_count++; if (vma->vm_file) { i_mmap_lock_write(vma->vm_file->f_mapping); - if (vma_is_shared_maywrite(vma)) + if (vma_is_shared_maywrite(vma)) { mapping_allow_writable(vma->vm_file->f_mapping); + writable_file_mapping = true; + } flush_dcache_mmap_lock(vma->vm_file->f_mapping); vma_interval_tree_insert(vma, &vma->vm_file->f_mapping->i_mmap); @@ -2937,7 +2942,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr, /* Once vma denies write, undo our temporary denial count */ unmap_writable: - if (file && is_shared_maywrite(vm_flags)) + if (writable_file_mapping) mapping_unmap_writable(file->f_mapping); file = vma->vm_file; ksm_add_vma(vma); @@ -2985,7 +2990,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr, unmap_region(mm, &vmi.mas, vma, prev, next, vma->vm_start, vma->vm_end, vma->vm_end, true); } - if (file && is_shared_maywrite(vm_flags)) + if (writable_file_mapping) mapping_unmap_writable(file->f_mapping); free_vma: vm_area_free(vma);
In order for a F_SEAL_WRITE sealed memfd mapping to have an opportunity to clear VM_MAYWRITE, we must be able to invoke the appropriate vm_ops->mmap() handler to do so. We would otherwise fail the mapping_map_writable() check before we had the opportunity to avoid it. This patch moves this check after the call_mmap() invocation. Only memfd actively denies write access causing a potential failure here (in memfd_add_seals()), so there should be no impact on non-memfd cases. This patch makes the userland-visible change that MAP_SHARED, PROT_READ mappings of an F_SEAL_WRITE sealed memfd mapping will now succeed. There is a delicate situation with cleanup paths assuming that a writable mapping must have occurred in circumstances where it may now not have. In order to ensure we do not accidentally mark a writable file unwritable by mistake, we explicitly track whether we have a writable mapping and unmap only if we do. Link: https://bugzilla.kernel.org/show_bug.cgi?id=217238 Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com> --- mm/mmap.c | 23 ++++++++++++++--------- 1 file changed, 14 insertions(+), 9 deletions(-)