igc: Fix wrong timestamp latency numbers
Subsystems (2)
PATCH 1/2 igc: Fix wrong timestamp latency numbers
The previous timestamping latency numbers were obtained by interpolating the i210 numbers with the i225 crystal clock value. That calculation was wrong.
Use the correct values from real measurements.
--- drivers/net/ethernet/intel/igc/igc.h | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-)
drivers/net/ethernet/intel/igc/igc.h +8 -12
@@ -303,18 +303,14 @@ extern char igc_driver_name[];#define IGC_RX_HDR_LEN IGC_RXBUFFER_256/* Transmit and receive latency (for PTP timestamps) */ -/* FIXME: These values were estimated using the ones that i225 has as -* basis, they seem to provide good numbers with ptp4l/phc2sys, but we -* need to confirm them. -*/ -#define IGC_I225_TX_LATENCY_10 9542 -#define IGC_I225_TX_LATENCY_100 1024 -#define IGC_I225_TX_LATENCY_1000 178 -#define IGC_I225_TX_LATENCY_2500 64 -#define IGC_I225_RX_LATENCY_10 20662 -#define IGC_I225_RX_LATENCY_100 2213 -#define IGC_I225_RX_LATENCY_1000 448 -#define IGC_I225_RX_LATENCY_2500 160 +#define IGC_I225_TX_LATENCY_10 240 +#define IGC_I225_TX_LATENCY_100 58 +#define IGC_I225_TX_LATENCY_1000 80 +#define IGC_I225_TX_LATENCY_2500 1325 +#define IGC_I225_RX_LATENCY_10 6450 +#define IGC_I225_RX_LATENCY_100 185 +#define IGC_I225_RX_LATENCY_1000 300 +#define IGC_I225_RX_LATENCY_2500 1485/* RX and TX descriptor control thresholds. * PTHRESH - MAC will consider prefetch if it has fewer than this number of
Signature · 1 line
2.28.0
PATCH 2/2 igc: Fix not considering the TX delay for timestamps
When timestamping a packet there's a delay between the start of the packet and the point where the hardware actually captures the timestamp. This difference needs to be considered if we want accurate timestamps.
This was done on the RX side, but not on the TX side.
--- drivers/net/ethernet/intel/igc/igc_ptp.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
drivers/net/ethernet/intel/igc/igc_ptp.c +19 -0
@@ -366,6 +366,7 @@ static void igc_ptp_tx_hwtstamp(struct igc_adapter *adapter)struct sk_buff *skb = adapter->ptp_tx_skb;struct skb_shared_hwtstamps shhwtstamps;struct igc_hw *hw = &adapter->hw; +int adjust =0 ;u64 regval;if (WARN_ON_ONCE (!skb))
@@ -375,6 +376,24 @@ static void igc_ptp_tx_hwtstamp(struct igc_adapter *adapter) regval |= (u64 )rd32 (IGC_TXSTMPH ) <<32 ;igc_ptp_systim_to_hwtstamp (adapter, &shhwtstamps, regval); +switch (adapter->link_speed) { +case SPEED_10 : + adjust =IGC_I225_TX_LATENCY_10 ; +break ; +case SPEED_100 : + adjust =IGC_I225_TX_LATENCY_100 ; +break ; +case SPEED_1000 : + adjust =IGC_I225_TX_LATENCY_1000 ; +break ; +case SPEED_2500 : + adjust =IGC_I225_TX_LATENCY_2500 ; +break ; + } + + shhwtstamps.hwtstamp = +ktime_add_ns (shhwtstamps.hwtstamp, adjust); +/* Clear the lock early before calling skb_tstamp_tx so that * applications are not woken up before the lock bit is clear. We use * a copy of the skb pointer to ensure other threads can't change it
Signature · 1 line
2.28.0
PATCH 2/2 igc: Fix not considering the TX delay for timestamps
Quoted 8 lines · Vinicius Costa Gomes
When timestamping a packet there's a delay between the start of the packet and the point where the hardware actually captures the timestamp. This difference needs to be considered if we want accurate timestamps.
This was done on the RX side, but not on the TX side.
Signed-off-by Vinicius Costa Gomes <[email protected]>
Please see my responses to the other patch.
Brown
PATCH 2/2 igc: Fix not considering the TX delay for timestamps
In reply to Vinicius Costa GomesQuoted 13 lines
From: Intel-wired-lan <[email protected]> On Behalf Of Vinicius Costa Gomes
Sent: Monday, August 17, 2020 4:13 PM
To: intel-wired-lan at lists.osuosl.org
Cc: Guedes, Andre <[email protected]>
Subject: [Intel-wired-lan] [PATCH 2/2] igc: Fix not considering the TX delay for timestampsWhen timestamping a packet there's a delay between the start of the packet and the point where the hardware actually captures the timestamp. This difference needs to be considered if we want accurate timestamps.
Quoted 7 lines
This was done on the RX side, but not on the TX side.
Signed-off-by Vinicius Costa Gomes <[email protected]>--- drivers/net/ethernet/intel/igc/igc_ptp.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
PATCH 1/2 igc: Fix wrong timestamp latency numbers
In reply to Vinicius Costa GomesQuoted 7 lines · Vinicius Costa Gomes
The previous timestamping latency numbers were obtained by interpolating the i210 numbers with the i225 crystal clock value. That calculation was wrong.
Use the correct values from real measurements.
Signed-off-by Vinicius Costa Gomes <[email protected]>
Is this targeted at net or net-next, and because it is a fix I recommend you have subject
[PATCH net 1/2]
and you should include a "Fixes:" tag since this driver has been released in previous kernels.
PATCH 1/2 igc: Fix wrong timestamp latency numbers
Quoted 6 lines · Jesse Brandeburg
Quoted 4 lines · Vinicius Costa Gomes
The previous timestamping latency numbers were obtained by interpolating the i210 numbers with the i225 crystal clock value. That calculation was wrong.
Quoted 7 lines
Quoted 3 lines
Use the correct values from real measurements.
Signed-off-by Vinicius Costa Gomes <[email protected]>Is this targeted at net or net-next, and because it is a fix I recommend you have subject
[PATCH net 1/2]
Oh, sorry about that. I used the wrong 'git format-patch' line.
Quoted 3 lines
and you should include a "Fixes:" tag since this driver has been released in previous kernels.
Yes. This should have been directed to 'net-queue', thanks for noticing.
Will send a v2.
Cheers,
Signature · 1 line
Vinicius
Brown
PATCH 1/2 igc: Fix wrong timestamp latency numbers
In reply to Vinicius Costa GomesQuoted 11 lines
From: Intel-wired-lan <[email protected]> On Behalf Of Vinicius Costa Gomes
Sent: Monday, August 17, 2020 4:13 PM
To: intel-wired-lan at lists.osuosl.org
Cc: Guedes, Andre <[email protected]>
Subject: [Intel-wired-lan] [PATCH 1/2] igc: Fix wrong timestamp latency numbersThe previous timestamping latency numbers were obtained by interpolating the i210 numbers with the i225 crystal clock value. That calculation was wrong.
Quoted 7 lines
Use the correct values from real measurements.
Signed-off-by Vinicius Costa Gomes <[email protected]>--- drivers/net/ethernet/intel/igc/igc.h | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-)