Re: [PATCH net-next 1/5] ch_ktls: Use kmap_local_page() instead of kmap_atomic()

On 11/18/2022 12:18 PM, Fabio M. De Francesco wrote:

On venerdì 18 novembre 2022 19:27:56 CET Anirudh Venkataramanan wrote:

quoted

On 11/18/2022 12:14 AM, Fabio M. De Francesco wrote:

quoted

On giovedì 17 novembre 2022 23:25:53 CET Anirudh Venkataramanan wrote:

quoted

kmap_atomic() is being deprecated in favor of kmap_local_page().
Replace kmap_atomic() and kunmap_atomic() with kmap_local_page()
and kunmap_local() respectively.

Note that kmap_atomic() disables preemption and page-fault processing,
but kmap_local_page() doesn't. Converting the former to the latter is

safe

quoted

quoted

quoted

only if there isn't an implicit dependency on preemption and page-fault
handling being disabled, which does appear to be the case here.

Also note that the page being mapped is not allocated by the driver,
and so the driver doesn't know if the page is in normal memory. This is

the

quoted

quoted

quoted

reason kmap_local_page() is used as opposed to page_address().

I don't have hardware, so this change has only been compile tested.

Cc: Ayush Sawal <ayush.sawal@chelsio.com>
Cc: Vinay Kumar Yadav <redacted>
Cc: Rohit Maheshwari <redacted>
Cc: Ira Weiny <redacted>
Cc: Fabio M. De Francesco <redacted>
Signed-off-by: Anirudh Venkataramanan <redacted>
---

   .../ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c | 10 +++++-----
   1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/

chcr_ktls.c

quoted

quoted

quoted

b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c index
da9973b..d95f230 100644

--- a/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c
+++ b/drivers/net/ethernet/chelsio/inline_crypto/ch_ktls/chcr_ktls.c

@@ -1853,24 +1853,24 @@ static int chcr_short_record_handler(struct

chcr_ktls_info *tx_info, i++;

   			}
   			f = &record->frags[i];

-			vaddr = kmap_atomic(skb_frag_page(f));
+			vaddr = kmap_local_page(skb_frag_page(f));

   			data = vaddr + skb_frag_off(f)  + remaining;
   			frag_delta = skb_frag_size(f) - remaining;
   			
   			if (frag_delta >= prior_data_len) {
   			
   				memcpy(prior_data, data,

prior_data_len);

quoted

-				kunmap_atomic(vaddr);
+				kunmap_local(vaddr);

   			} else {
   			
   				memcpy(prior_data, data, frag_delta);

-				kunmap_atomic(vaddr);
+				kunmap_local(vaddr);

   				/* get the next page */
   				f = &record->frags[i + 1];

-				vaddr = kmap_atomic(skb_frag_page(f));
+				vaddr =

kmap_local_page(skb_frag_page(f));

quoted

   				data = vaddr + skb_frag_off(f);
   				memcpy(prior_data + frag_delta,
   				
   				       data, (prior_data_len -

frag_delta));

quoted

-				kunmap_atomic(vaddr);
+				kunmap_local(vaddr);

   			}
   			/* reset tcp_seq as per the prior_data_required

len */

quoted

   			tcp_seq -= prior_data_len;

--
2.37.2

The last conversion could have been done with memcpy_from_page(). However,
it's not that a big deal. Therefore...

Reviewed-by: Fabio M. De Francesco <redacted>

Yeah, using memcpy_from_page() is cleaner. I'll update this patch, and
probably 4/5 too.

Thanks!
Ani

Well, I didn't ask you for a second version. This is why you already see my
"Reviewed-by:" tag. I'm OK with your changes. I just warned you that
maintainers might ask, so I'd wait and see. However it's up to you.

I understand and appreciate your "Reviewed-by", but that doesn't mean further improvements aren't possible. I believe using memcpy_from_page() is better, and plan to do this in v2.

However, if you decide to send this patch with memcpy_from_page(), why you
are not sure about 4/5? Since you decided to send 1/5 again, what does prevent
you from updating also 4/5?

I hadn't seen patch 4/5 when I replied to you. Since then I have, and so I'll be updating 4/5 to use memcpy_from_page() as well.

Ani