On 22/09/2024 20:55, Salvatore Bonaccorso wrote:
Hi Salvatore,Control: tags -1 + moreinfo Hi Eric,Since you seem to easily have the effect reproducible with your hardware, and can determine if a version between 6.1.99 and 6.1.106 upstream would be affected, can you bisect the changes to identify which is the commit introducing the issue?Indeed, reproducing is easy. I have begun bisecting, will report back when it is finished (the box taks quite a bit of time for each kernel build).
The bisect is complete and yields this result
15115033f056cbd7649b8e1806287f71bdb7ce5c is the first bad commit
commit 15115033f056cbd7649b8e1806287f71bdb7ce5c
Author: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Date: Fri Jul 26 20:17:10 2024 +0200
ice: don't busy wait for Rx queue disable in ice_qp_dis()
[ Upstream commit 1ff72a2f67791cd4ddad19ed830445f57b30e992 ]
When ice driver is spammed with multiple xdpsock instances and
flow
control is enabled, there are cases when Rx queue gets stuck
and unable
to reflect the disable state in QRX_CTRL register. Similar
issue has
previously been addressed in commit 13a6233b033f ("ice: Add
support to
enable/disable all Rx queues before waiting").
To workaround this, let us simply not wait for a disabled
state as later
patch will make sure that regardless of the encountered error
in the
process of disabling a queue pair, the Rx queue will be
enabled.
Fixes: 2d4238f55697 ("ice: Add support for AF_XDP")
Reviewed-by: Shannon Nelson <shannon.nelson@amd.com>
Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com>
(A Contingent Worker at Intel)
Signed-off-by: Maciej Fijalkowski
<maciej.fijalkowski@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Regards, SalvatoreRegards
Eric