[1/1] io_uring: don't vmap single page regions

Commit Message

Pavel Begunkov Dec. 8, 2024, 9:46 p.m. UTC

When io_check_coalesce_buffer() meets a single page buffer it bails out
and tells that it can be coalesced. That's fine for registered buffers
as io_coalesce_buffer() wouldn't change anything, but the region code
now uses the function to decided on whether to vmap the buffer or not.

Report that a single page buffer is trivially coalescable and let
io_sqe_buffer_register() to filter them.

Fixes: cc2f1a864c27a ("io_uring/memmap: optimise single folio regions")
Signed-off-by: Pavel Begunkov <[e-mail]>
---
 io_uring/rsrc.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

Comments

Jens Axboe Dec. 9, 2024, 3:17 p.m. UTC | #1

On Sun, 08 Dec 2024 21:46:01 +0000, Pavel Begunkov wrote:
> When io_check_coalesce_buffer() meets a single page buffer it bails out
> and tells that it can be coalesced. That's fine for registered buffers
> as io_coalesce_buffer() wouldn't change anything, but the region code
> now uses the function to decided on whether to vmap the buffer or not.
> 
> Report that a single page buffer is trivially coalescable and let
> io_sqe_buffer_register() to filter them.
> 
> [...]

Applied, thanks!

[1/1] io_uring: don't vmap single page regions
      commit: 21e3947c0cd2ecdfc043089288973d40979d3a81

Best regards,

Patch

diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c
index 9208cf77c41e..0f1cd0a1d880 100644
--- a/io_uring/rsrc.c
+++ b/io_uring/rsrc.c
@@ -675,14 +675,9 @@  bool io_check_coalesce_buffer(struct page **page_array, int nr_pages,
 	unsigned int count = 1, nr_folios = 1;
 	int i;
 
-	if (nr_pages <= 1)
-		return false;
-
 	data->nr_pages_mid = folio_nr_pages(folio);
-	if (data->nr_pages_mid == 1)
-		return false;
-
 	data->folio_shift = folio_shift(folio);
+
 	/*
 	 * Check if pages are contiguous inside a folio, and all folios have
 	 * the same page count except for the head and tail.
@@ -750,8 +745,10 @@  static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 	}
 
 	/* If it's huge page(s), try to coalesce them into fewer bvec entries */
-	if (io_check_coalesce_buffer(pages, nr_pages, &data))
-		coalesced = io_coalesce_buffer(&pages, &nr_pages, &data);
+	if (nr_pages > 1 && io_check_coalesce_buffer(pages, nr_pages, &data)) {
+		if (data.nr_pages_mid != 1)
+			coalesced = io_coalesce_buffer(&pages, &nr_pages, &data);
+	}
 
 	imu = kvmalloc(struct_size(imu, bvec, nr_pages), GFP_KERNEL);
 	if (!imu)