> -----Original Message-----
> From: lon...@linuxonhyperv.com <lon...@linuxonhyperv.com>
> Sent: Monday, August 5, 2024 7:38 PM
> To: KY Srinivasan <k...@microsoft.com>; Haiyang Zhang
> <haiya...@microsoft.com>; Wei Liu <wei....@kernel.org>; Dexuan Cui
> <de...@microsoft.com>; David S. Miller <da...@davemloft.net>; Eric Dumazet
> <eduma...@google.com>; Jakub Kicinski <k...@kernel.org>; Paolo Abeni
> <pab...@redhat.com>; Shradha Gupta <shradhagu...@linux.microsoft.com>;
> Simon Horman <ho...@kernel.org>; Konstantin Taranov
> <kotara...@microsoft.com>; Souradeep Chakrabarti
> <schakraba...@linux.microsoft.com>; Erick Archer
> <erick.arc...@outlook.com>; linux-hyperv@vger.kernel.org;
> net...@vger.kernel.org; linux-ker...@vger.kernel.org; linux-
> r...@vger.kernel.org
> Cc: Long Li <lon...@microsoft.com>; sta...@vger.kernel.org
> Subject: [PATCH net] net: mana: Fix doorbell out of order violation and
> avoid unnecessary doorbell rings
> 
> From: Long Li <lon...@microsoft.com>
> 
> After napi_complete_done() is called, another NAPI may be running on
> another CPU and ring the doorbell before the current CPU does. When
> combined with unnecessary rings when there is no need to ARM the CQ, this
> triggers error paths in the hardware.
> 
> Fix this by always ring the doorbell in sequence and avoid unnecessary
> rings.
> 
> Cc: sta...@vger.kernel.org
> Fixes: e1b5683ff62e ("net: mana: Move NAPI from EQ to CQ")
> Signed-off-by: Long Li <lon...@microsoft.com>

Reviewed-by: Haiyang Zhang <haiya...@microsoft.com>

Thank you.

Reply via email to