| Message ID | 642128d50f5423b3331e3108f8faf6b8ac0d957e.1684097002.git.lstoakes@gmail.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | None | expand |
Looks good:
Reviewed-by: Christoph Hellwig <hch@lst.de>
On 5/14/23 3:26 PM, Lorenzo Stoakes wrote: > Now that the GUP explicitly checks FOLL_LONGTERM pin_user_pages() for > broken file-backed mappings in "mm/gup: disallow FOLL_LONGTERM GUP-nonfast > writing to file-backed mappings", there is no need to explicitly check VMAs > for this condition, so simply remove this logic from io_uring altogether. Don't have the prerequisite patch handy (not in mainline yet), but if it just moves the check, then: Reviewed-by: Jens Axboe <axboe@kernel.dk>
On 15.05.23 21:55, Jens Axboe wrote: > On 5/14/23 3:26 PM, Lorenzo Stoakes wrote: >> Now that the GUP explicitly checks FOLL_LONGTERM pin_user_pages() for >> broken file-backed mappings in "mm/gup: disallow FOLL_LONGTERM GUP-nonfast >> writing to file-backed mappings", there is no need to explicitly check VMAs >> for this condition, so simply remove this logic from io_uring altogether. > > Don't have the prerequisite patch handy (not in mainline yet), but if it > just moves the check, then: > > Reviewed-by: Jens Axboe <axboe@kernel.dk> > Jens, please see my note regarding iouring: https://lore.kernel.org/bpf/6e96358e-bcb5-cc36-18c3-ec5153867b9a@redhat.com/ With this patch, MAP_PRIVATE will work as expected (2), but there will be a change in return code handling (1) that we might have to document in the man page.
On 14.05.23 23:26, Lorenzo Stoakes wrote: > Now that the GUP explicitly checks FOLL_LONGTERM pin_user_pages() for > broken file-backed mappings in "mm/gup: disallow FOLL_LONGTERM GUP-nonfast > writing to file-backed mappings", there is no need to explicitly check VMAs > for this condition, so simply remove this logic from io_uring altogether. > Worth adding "Note that this change will make iouring fixed buffers work on MAP_PRIVATE file mappings." I'll run my test cases with this series and expect no surprises :) Reviewed-by: David Hildenbrand <david@redhat.com>
On 5/16/23 2:25?AM, David Hildenbrand wrote: > On 15.05.23 21:55, Jens Axboe wrote: >> On 5/14/23 3:26?PM, Lorenzo Stoakes wrote: >>> Now that the GUP explicitly checks FOLL_LONGTERM pin_user_pages() for >>> broken file-backed mappings in "mm/gup: disallow FOLL_LONGTERM GUP-nonfast >>> writing to file-backed mappings", there is no need to explicitly check VMAs >>> for this condition, so simply remove this logic from io_uring altogether. >> >> Don't have the prerequisite patch handy (not in mainline yet), but if it >> just moves the check, then: >> >> Reviewed-by: Jens Axboe <axboe@kernel.dk> >> > > Jens, please see my note regarding iouring: > > https://lore.kernel.org/bpf/6e96358e-bcb5-cc36-18c3-ec5153867b9a@redhat.com/ > > With this patch, MAP_PRIVATE will work as expected (2), but there will > be a change in return code handling (1) that we might have to document > in the man page. I think documenting that newer kernels will return -EFAULT rather than -EOPNOTSUPP should be fine. It's not a new failure case, just a different error value for an already failing case. Should be fine with just a doc update. Will do that now.
diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index d46f72a5ef73..b6451f8bc5d5 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -1030,9 +1030,8 @@ static int io_buffer_account_pin(struct io_ring_ctx *ctx, struct page **pages, struct page **io_pin_pages(unsigned long ubuf, unsigned long len, int *npages) { unsigned long start, end, nr_pages; - struct vm_area_struct **vmas = NULL; struct page **pages = NULL; - int i, pret, ret = -ENOMEM; + int pret, ret = -ENOMEM; end = (ubuf + len + PAGE_SIZE - 1) >> PAGE_SHIFT; start = ubuf >> PAGE_SHIFT; @@ -1042,45 +1041,24 @@ struct page **io_pin_pages(unsigned long ubuf, unsigned long len, int *npages) if (!pages) goto done; - vmas = kvmalloc_array(nr_pages, sizeof(struct vm_area_struct *), - GFP_KERNEL); - if (!vmas) - goto done; - ret = 0; mmap_read_lock(current->mm); pret = pin_user_pages(ubuf, nr_pages, FOLL_WRITE | FOLL_LONGTERM, - pages, vmas); - if (pret == nr_pages) { - /* don't support file backed memory */ - for (i = 0; i < nr_pages; i++) { - struct vm_area_struct *vma = vmas[i]; - - if (vma_is_shmem(vma)) - continue; - if (vma->vm_file && - !is_file_hugepages(vma->vm_file)) { - ret = -EOPNOTSUPP; - break; - } - } + pages, NULL); + if (pret == nr_pages) *npages = nr_pages; - } else { + else ret = pret < 0 ? pret : -EFAULT; - } + mmap_read_unlock(current->mm); if (ret) { - /* - * if we did partial map, or found file backed vmas, - * release any pages we did get - */ + /* if we did partial map, release any pages we did get */ if (pret > 0) unpin_user_pages(pages, pret); goto done; } ret = 0; done: - kvfree(vmas); if (ret < 0) { kvfree(pages); pages = ERR_PTR(ret);
Now that the GUP explicitly checks FOLL_LONGTERM pin_user_pages() for broken file-backed mappings in "mm/gup: disallow FOLL_LONGTERM GUP-nonfast writing to file-backed mappings", there is no need to explicitly check VMAs for this condition, so simply remove this logic from io_uring altogether. Signed-off-by: Lorenzo Stoakes <lstoakes@gmail.com> --- io_uring/rsrc.c | 34 ++++++---------------------------- 1 file changed, 6 insertions(+), 28 deletions(-)