kmap_atomic() is deprecated and should be replaced with kmap_local_page()
[1][2]. kmap_local_page() is faster in kernels with HIGHMEM enabled, can
take page faults, and allows preemption.

According to [2], this replacement is safe as long as the code between
kmap_atomic() and kunmap_atomic() does not implicitly depend on disabling
page faults or preemption. In this patch, the only thing happening between
mapping and unmapping the page is a memcpy, and I don't suspect it depends
on disabling page faults or preemption.

[1] https://lwn.net/Articles/836144/
[2] https://docs.kernel.org/mm/highmem.html#temporary-virtual-mappings

Signed-off-by: David Reaver <me@davidreaver.com>
---
 drivers/char/virtio_console.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index c62b208b42f1..24442485e73e 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -883,9 +883,9 @@ static int pipe_to_sg(struct pipe_inode_info *pipe, struct pipe_buffer *buf,
 		if (len + offset > PAGE_SIZE)
 			len = PAGE_SIZE - offset;
 
-		src = kmap_atomic(buf->page);
+		src = kmap_local_page(buf->page);
 		memcpy(page_address(page) + offset, src + buf->offset, len);
-		kunmap_atomic(src);
+		kunmap_local(src);
 
 		sg_set_page(&(sgl->sg[sgl->n]), page, len, offset);
 	}

David Reaver wrote:
> kmap_atomic() is deprecated and should be replaced with kmap_local_page()
> [1][2]. kmap_local_page() is faster in kernels with HIGHMEM enabled, can
> take page faults, and allows preemption.

Thanks for taking these on!  It is good to see some progress here.

> 
> According to [2], this replacement is safe as long as the code between
> kmap_atomic() and kunmap_atomic() does not implicitly depend on disabling
> page faults or preemption. In this patch, the only thing happening between
> mapping and unmapping the page is a memcpy, and I don't suspect it depends
> on disabling page faults or preemption.
> 
> [1] https://lwn.net/Articles/836144/
> [2] https://docs.kernel.org/mm/highmem.html#temporary-virtual-mappings
> 
> Signed-off-by: David Reaver <me@davidreaver.com>
> ---
>  drivers/char/virtio_console.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
> index c62b208b42f1..24442485e73e 100644
> --- a/drivers/char/virtio_console.c
> +++ b/drivers/char/virtio_console.c
> @@ -883,9 +883,9 @@ static int pipe_to_sg(struct pipe_inode_info *pipe, struct pipe_buffer *buf,
>  		if (len + offset > PAGE_SIZE)
>  			len = PAGE_SIZE - offset;
>  
> -		src = kmap_atomic(buf->page);
> +		src = kmap_local_page(buf->page);
>  		memcpy(page_address(page) + offset, src + buf->offset, len);
> -		kunmap_atomic(src);
> +		kunmap_local(src);

memcpy_to_page()

Ira

[snip]

On Wed, 2025-01-08 at 19:59 -0800, David Reaver wrote:
> kmap_atomic() is deprecated and should be replaced with
> kmap_local_page()
> [1][2]. kmap_local_page() is faster in kernels with HIGHMEM enabled,
> can
> take page faults, and allows preemption.
> 
> According to [2], this replacement is safe as long as the code
> between
> kmap_atomic() and kunmap_atomic() does not implicitly depend on
> disabling
> page faults or preemption. In this patch, the only thing happening
> between
> mapping and unmapping the page is a memcpy, and I don't suspect it
> depends
> on disabling page faults or preemption.
> 
> [1] https://lwn.net/Articles/836144/
> [2]
> https://docs.kernel.org/mm/highmem.html#temporary-virtual-mappings
> 
> Signed-off-by: David Reaver <me@davidreaver.com>

Reviewed-by: Amit Shah <amit@kernel.org>