mbox series

[net-next,0/3] ice: restore timestamp config after reset

Message ID 20231028002322.2224777-1-jacob.e.keller@intel.com (mailing list archive)
Headers show
Series ice: restore timestamp config after reset | expand

Message

Jacob Keller Oct. 28, 2023, 12:23 a.m. UTC
We recently discovered during internal validation that the ice driver has
not been properly restoring Tx timestamp configuration after a device reset,
which resulted in application failures after a device reset.

It turned out this problem is two-fold. Since the introduction of the PTP
support, the driver has been clobbering the user space configuration
settings during reset. Thus after a reset, the driver will not restore
timestamps, and will report timestamps as disabled if SIOCGHWTSTAMP ioctl is
issued.

In addition, the recently merged auxiliary bus support code missed that
PFINT_TSYN_MSK must be reprogrammed on the clock owner for E822 devices.
Failure to restore this register configuration results in the driver no
longer responding to interrupts from other ports. Depending on the traffic
pattern, this can either result in increased latency responding to
timestamps on the non-owner ports, or it can result in the driver never
reporting any timestamps.

This series fixes both issues, as well as removes a redundant Tx ring field
since we can rely on the skb flag as the primary detector for a Tx timestamp
request.

I opted to send this to net-next, because my primary focus was on fixing the
E822 issue which was not introduced until the auxiliary bus which isn't in
the net tree. I do not know if the fix for the overall timestamp
configuration issue is applicable to net, since we have had a lot of
refactor going into the driver to support auxiliary bus as well as in
preparation for new hardware support.

I'd like to see this merged so that the timestamping issues are fixed in
6.6.

Jacob Keller (3):
  ice: remove ptp_tx ring parameter flag
  ice: unify logic for programming PFINT_TSYN_MSK
  ice: restore timestamp configuration after device reset

 drivers/net/ethernet/intel/ice/ice_main.c |  12 +-
 drivers/net/ethernet/intel/ice/ice_ptp.c  | 146 ++++++++++++----------
 drivers/net/ethernet/intel/ice/ice_ptp.h  |   5 +-
 drivers/net/ethernet/intel/ice/ice_txrx.c |   3 -
 drivers/net/ethernet/intel/ice/ice_txrx.h |   1 -
 5 files changed, 84 insertions(+), 83 deletions(-)


base-commit: 3a04927f8d4b7a4f008f04af41e31173002eb1ea

Comments

Jesse Brandeburg Oct. 28, 2023, 12:30 a.m. UTC | #1
On 10/27/2023 5:23 PM, Jacob Keller wrote:
> We recently discovered during internal validation that the ice driver has
> not been properly restoring Tx timestamp configuration after a device reset,
> which resulted in application failures after a device reset.
> 
> It turned out this problem is two-fold. Since the introduction of the PTP
> support, the driver has been clobbering the user space configuration
> settings during reset. Thus after a reset, the driver will not restore
> timestamps, and will report timestamps as disabled if SIOCGHWTSTAMP ioctl is
> issued.
> 
> In addition, the recently merged auxiliary bus support code missed that
> PFINT_TSYN_MSK must be reprogrammed on the clock owner for E822 devices.
> Failure to restore this register configuration results in the driver no
> longer responding to interrupts from other ports. Depending on the traffic
> pattern, this can either result in increased latency responding to
> timestamps on the non-owner ports, or it can result in the driver never
> reporting any timestamps.
> 
> This series fixes both issues, as well as removes a redundant Tx ring field
> since we can rely on the skb flag as the primary detector for a Tx timestamp
> request.
> 
> I opted to send this to net-next, because my primary focus was on fixing the
> E822 issue which was not introduced until the auxiliary bus which isn't in
> the net tree. I do not know if the fix for the overall timestamp
> configuration issue is applicable to net, since we have had a lot of
> refactor going into the driver to support auxiliary bus as well as in
> preparation for new hardware support.
> 
> I'd like to see this merged so that the timestamping issues are fixed in
> 6.6.
> 
> Jacob Keller (3):
>   ice: remove ptp_tx ring parameter flag
>   ice: unify logic for programming PFINT_TSYN_MSK
>   ice: restore timestamp configuration after device reset

For the series:

Reviewed-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Jacob Keller Oct. 30, 2023, 9:38 p.m. UTC | #2
On 10/27/2023 5:23 PM, Jacob Keller wrote:
> We recently discovered during internal validation that the ice driver has
> not been properly restoring Tx timestamp configuration after a device reset,
> which resulted in application failures after a device reset.
> 
> It turned out this problem is two-fold. Since the introduction of the PTP
> support, the driver has been clobbering the user space configuration
> settings during reset. Thus after a reset, the driver will not restore
> timestamps, and will report timestamps as disabled if SIOCGHWTSTAMP ioctl is
> issued.
> 
> In addition, the recently merged auxiliary bus support code missed that
> PFINT_TSYN_MSK must be reprogrammed on the clock owner for E822 devices.
> Failure to restore this register configuration results in the driver no
> longer responding to interrupts from other ports. Depending on the traffic
> pattern, this can either result in increased latency responding to
> timestamps on the non-owner ports, or it can result in the driver never
> reporting any timestamps.
> 
> This series fixes both issues, as well as removes a redundant Tx ring field
> since we can rely on the skb flag as the primary detector for a Tx timestamp
> request.
> 
> I opted to send this to net-next, because my primary focus was on fixing the
> E822 issue which was not introduced until the auxiliary bus which isn't in
> the net tree. I do not know if the fix for the overall timestamp
> configuration issue is applicable to net, since we have had a lot of
> refactor going into the driver to support auxiliary bus as well as in
> preparation for new hardware support.
> 
> I'd like to see this merged so that the timestamping issues are fixed in
> 6.6.
> 

Heh, re-reading this now, I meant 6.7 here. The commit with the PTP
auxiliary busy support is in the code headed for 6.7, not in 6.6

Thanks,
Jake

> Jacob Keller (3):
>   ice: remove ptp_tx ring parameter flag
>   ice: unify logic for programming PFINT_TSYN_MSK
>   ice: restore timestamp configuration after device reset
> 
>  drivers/net/ethernet/intel/ice/ice_main.c |  12 +-
>  drivers/net/ethernet/intel/ice/ice_ptp.c  | 146 ++++++++++++----------
>  drivers/net/ethernet/intel/ice/ice_ptp.h  |   5 +-
>  drivers/net/ethernet/intel/ice/ice_txrx.c |   3 -
>  drivers/net/ethernet/intel/ice/ice_txrx.h |   1 -
>  5 files changed, 84 insertions(+), 83 deletions(-)
> 
> 
> base-commit: 3a04927f8d4b7a4f008f04af41e31173002eb1ea