i40e: sync next_to_clean and next_to_process for programming status desc
authorTirthendu Sarkar <tirthendu.sarkar@intel.com>
Thu, 19 Oct 2023 20:38:52 +0000 (13:38 -0700)
committerJakub Kicinski <kuba@kernel.org>
Sat, 21 Oct 2023 01:17:11 +0000 (18:17 -0700)
When a programming status desc is encountered on the rx_ring,
next_to_process is bumped along with cleaned_count but next_to_clean is
not. This causes I40E_DESC_UNUSED() macro to misbehave resulting in
overwriting whole ring with new buffers.

Update next_to_clean to point to next_to_process on seeing a programming
status desc if not in the middle of handling a multi-frag packet. Also,
bump cleaned_count only for such case as otherwise next_to_clean buffer
may be returned to hardware on reaching clean_threshold.

Fixes: e9031f2da1ae ("i40e: introduce next_to_process to i40e_ring")
Suggested-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Reported-by: hq.dev+kernel@msdfc.xyz
Reported by: Solomon Peachy <pizza@shaftnet.org>
Closes: https://bugzilla.kernel.org/show_bug.cgi?id=217678
Tested-by: hq.dev+kernel@msdfc.xyz
Tested by: Indrek Järve <incx@dustbite.net>
Signed-off-by: Tirthendu Sarkar <tirthendu.sarkar@intel.com>
Tested-by: Arpana Arland <arpanax.arland@intel.com> (A Contingent worker at Intel)
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20231019203852.3663665-1-jacob.e.keller@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
drivers/net/ethernet/intel/i40e/i40e_txrx.c

index 0b3a27f..50c70a8 100644 (file)
@@ -2544,7 +2544,14 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget,
                        rx_buffer = i40e_rx_bi(rx_ring, ntp);
                        i40e_inc_ntp(rx_ring);
                        i40e_reuse_rx_page(rx_ring, rx_buffer);
-                       cleaned_count++;
+                       /* Update ntc and bump cleaned count if not in the
+                        * middle of mb packet.
+                        */
+                       if (rx_ring->next_to_clean == ntp) {
+                               rx_ring->next_to_clean =
+                                       rx_ring->next_to_process;
+                               cleaned_count++;
+                       }
                        continue;
                }