| Message ID | 20230823221537.816541-1-vadim.fedorenko@linux.dev |
|---|---|
| State | Accepted |
| Delegated to: | Anthony Nguyen |
| Headers | show |
| Series | [net] ixgbe: fix timestamp configuration code | expand |
On Wed, Aug 23, 2023 at 11:15:37PM +0100, Vadim Fedorenko wrote: > The commit in fixes introduced flags to control the status of hardware > configuration while processing packets. At the same time another structure > is used to provide configuration of timestamper to user-space applications. > The way it was coded makes this structures go out of sync easily. The > repro is easy for 82599 chips: > > [root@hostname ~]# hwstamp_ctl -i eth0 -r 12 -t 1 > current settings: > tx_type 0 > rx_filter 0 > new settings: > tx_type 1 > rx_filter 12 > > The eth0 device is properly configured to timestamp any PTPv2 events. > > [root@hostname ~]# hwstamp_ctl -i eth0 -r 1 -t 1 > current settings: > tx_type 1 > rx_filter 12 > SIOCSHWTSTAMP failed: Numerical result out of range > The requested time stamping mode is not supported by the hardware. > > The error is properly returned because HW doesn't support all packets > timestamping. But the adapter->flags is cleared of timestamp flags > even though no HW configuration was done. From that point no RX timestamps > are received by user-space application. But configuration shows good > values: > > [root@hostname ~]# hwstamp_ctl -i eth0 > current settings: > tx_type 1 > rx_filter 12 > > Fix the issue by applying new flags only when the HW was actually > configured. > > Fixes: a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x devices") > Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> Reviewed-by: Simon Horman <horms@kernel.org>
> -----Original Message----- > From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Vadim Fedorenko > Sent: Thursday, August 24, 2023 3:46 AM > To: Brandeburg, Jesse <jesse.brandeburg@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Jakub Kicinski <kuba@kernel.org>; Alexander Duyck <alexander.duyck@gmail.com>; Rustad, Mark D <mark.d.rustad@intel.com>; Darin Miller <darin.j.miller@intel.com>; Jeff Kirsher <jeffrey.t.kirsher@intel.com>; Richard Cochran <richardcochran@gmail.com> > Cc: netdev@vger.kernel.org; intel-wired-lan@lists.osuosl.org; Vadim Fedorenko <vadim.fedorenko@linux.dev> > Subject: [Intel-wired-lan] [PATCH net] ixgbe: fix timestamp configuration code > > The commit in fixes introduced flags to control the status of hardware > configuration while processing packets. At the same time another structure > is used to provide configuration of timestamper to user-space applications. > The way it was coded makes this structures go out of sync easily. The > repro is easy for 82599 chips: > > [root@hostname ~]# hwstamp_ctl -i eth0 -r 12 -t 1 > current settings: > tx_type 0 > rx_filter 0 > new settings: > tx_type 1 > rx_filter 12 > > The eth0 device is properly configured to timestamp any PTPv2 events. > > [root@hostname ~]# hwstamp_ctl -i eth0 -r 1 -t 1 > current settings: > tx_type 1 > rx_filter 12 > SIOCSHWTSTAMP failed: Numerical result out of range > The requested time stamping mode is not supported by the hardware. > > The error is properly returned because HW doesn't support all packets > timestamping. But the adapter->flags is cleared of timestamp flags > even though no HW configuration was done. From that point no RX timestamps > are received by user-space application. But configuration shows good > values: > > [root@hostname ~]# hwstamp_ctl -i eth0 > current settings: > tx_type 1 > rx_filter 12 > > Fix the issue by applying new flags only when the HW was actually > configured. > > Fixes: a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x devices") > Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> > --- > drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c | 28 +++++++++++--------- > 1 file changed, 15 insertions(+), 13 deletions(-) > Hi, With patch also we are observing same issue. # ./hwstamp_ctl -i eth10 current settings: tx_type 1 rx_filter 12 # ./hwstamp_ctl -i eth10 -r 1 -t 1 current settings: tx_type 1 rx_filter 12 SIOCSHWTSTAMP failed: Numerical result out of range The requested time stamping mode is not supported by the hardware. Adapter details: Niantic (Spring Fountain) SUT info: H/W: Manufacturer: Intel Corporation Product Name: S2600STQ RAM: [62G/8G/49G] CPU: Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz [112/112] PF bus-info: 0000:d8:00.1 0x8086:0x10fb 0x8086 0x000c (0x01) S/W: OS: "Red Hat Enterprise Linux 8.6 (Ootpa)" 6.5.0-rc7_next-queue_28-Aug-2023-01755-g938672aefaeb CMD: BOOT_IMAGE=(hd0,msdos2)/vmlinuz-6.5.0-rc7_next-queue_28-Aug-2023-01755-g938672aefaeb root=/dev/mapper/rhel_os--delivery-root ro crashkernel=1G-4G:192M,4G-64G:256M,64G-:512M resume=/dev/mapper/rhel_os--delivery-swap rd.lvm.lv=rhel_os-delivery/root rd.lvm.lv=rhel_os-delivery/swap selinux=0 biosdevname=0 net.ifnames=0 rhgb quiet FW firmware-version: 0x000161bf PF version: 6.5.0-rc7_next-queue_28-Aug-202
On 31/08/2023 06:18, Pucha, HimasekharX Reddy wrote: >> -----Original Message----- >> From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Vadim Fedorenko >> Sent: Thursday, August 24, 2023 3:46 AM >> To: Brandeburg, Jesse <jesse.brandeburg@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Jakub Kicinski <kuba@kernel.org>; Alexander Duyck <alexander.duyck@gmail.com>; Rustad, Mark D <mark.d.rustad@intel.com>; Darin Miller <darin.j.miller@intel.com>; Jeff Kirsher <jeffrey.t.kirsher@intel.com>; Richard Cochran <richardcochran@gmail.com> >> Cc: netdev@vger.kernel.org; intel-wired-lan@lists.osuosl.org; Vadim Fedorenko <vadim.fedorenko@linux.dev> >> Subject: [Intel-wired-lan] [PATCH net] ixgbe: fix timestamp configuration code >> >> The commit in fixes introduced flags to control the status of hardware >> configuration while processing packets. At the same time another structure >> is used to provide configuration of timestamper to user-space applications. >> The way it was coded makes this structures go out of sync easily. The >> repro is easy for 82599 chips: >> >> [root@hostname ~]# hwstamp_ctl -i eth0 -r 12 -t 1 >> current settings: >> tx_type 0 >> rx_filter 0 >> new settings: >> tx_type 1 >> rx_filter 12 >> >> The eth0 device is properly configured to timestamp any PTPv2 events. >> >> [root@hostname ~]# hwstamp_ctl -i eth0 -r 1 -t 1 >> current settings: >> tx_type 1 >> rx_filter 12 >> SIOCSHWTSTAMP failed: Numerical result out of range >> The requested time stamping mode is not supported by the hardware. >> >> The error is properly returned because HW doesn't support all packets >> timestamping. But the adapter->flags is cleared of timestamp flags >> even though no HW configuration was done. From that point no RX timestamps >> are received by user-space application. But configuration shows good >> values: >> >> [root@hostname ~]# hwstamp_ctl -i eth0 >> current settings: >> tx_type 1 >> rx_filter 12 >> >> Fix the issue by applying new flags only when the HW was actually >> configured. >> >> Fixes: a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x devices") >> Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> >> --- >> drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c | 28 +++++++++++--------- >> 1 file changed, 15 insertions(+), 13 deletions(-) >> > > Hi, > With patch also we are observing same issue. Hi, What kind of issue do you observe? The hardware doesn't support timestamping of all packets. The issue this patch fixes is that after failed attempt to setup the timestamping of all RX packets, the driver stops reading timestamps at all even though it reports that timestamping of PTP RX packets is enabled. > > # ./hwstamp_ctl -i eth10 > current settings: > tx_type 1 > rx_filter 12 > # ./hwstamp_ctl -i eth10 -r 1 -t 1 > current settings: > tx_type 1 > rx_filter 12 > SIOCSHWTSTAMP failed: Numerical result out of range > The requested time stamping mode is not supported by the hardware. > > Adapter details: Niantic (Spring Fountain) > > SUT info: > H/W: > Manufacturer: Intel Corporation > Product Name: S2600STQ > RAM: [62G/8G/49G] > CPU: Intel(R) Xeon(R) Platinum 8180 CPU @ 2.50GHz [112/112] > PF bus-info: 0000:d8:00.1 0x8086:0x10fb 0x8086 0x000c (0x01) > S/W: > OS: "Red Hat Enterprise Linux 8.6 (Ootpa)" 6.5.0-rc7_next-queue_28-Aug-2023-01755-g938672aefaeb > CMD: BOOT_IMAGE=(hd0,msdos2)/vmlinuz-6.5.0-rc7_next-queue_28-Aug-2023-01755-g938672aefaeb root=/dev/mapper/rhel_os--delivery-root ro crashkernel=1G-4G:192M,4G-64G:256M,64G-:512M resume=/dev/mapper/rhel_os--delivery-swap rd.lvm.lv=rhel_os-delivery/root rd.lvm.lv=rhel_os-delivery/swap selinux=0 biosdevname=0 net.ifnames=0 rhgb quiet > FW firmware-version: 0x000161bf > PF version: 6.5.0-rc7_next-queue_28-Aug-202 > >
> -----Original Message----- > From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of Vadim Fedorenko > Sent: Thursday, August 24, 2023 3:46 AM > To: Brandeburg, Jesse <jesse.brandeburg@intel.com>; Nguyen, Anthony L <anthony.l.nguyen@intel.com>; Jakub Kicinski <kuba@kernel.org>; Alexander Duyck <alexander.duyck@gmail.com>; Rustad, Mark D <mark.d.rustad@intel.com>; Darin Miller <darin.j.miller@intel.com>; Jeff Kirsher <jeffrey.t.kirsher@intel.com>; Richard Cochran <richardcochran@gmail.com> > Cc: netdev@vger.kernel.org; intel-wired-lan@lists.osuosl.org; Vadim Fedorenko <vadim.fedorenko@linux.dev> > Subject: [Intel-wired-lan] [PATCH net] ixgbe: fix timestamp configuration code > > The commit in fixes introduced flags to control the status of hardware > configuration while processing packets. At the same time another structure > is used to provide configuration of timestamper to user-space applications. > The way it was coded makes this structures go out of sync easily. The > repro is easy for 82599 chips: > > [root@hostname ~]# hwstamp_ctl -i eth0 -r 12 -t 1 > current settings: > tx_type 0 > rx_filter 0 > new settings: > tx_type 1 > rx_filter 12 > > The eth0 device is properly configured to timestamp any PTPv2 events. > > [root@hostname ~]# hwstamp_ctl -i eth0 -r 1 -t 1 > current settings: > tx_type 1 > rx_filter 12 > SIOCSHWTSTAMP failed: Numerical result out of range > The requested time stamping mode is not supported by the hardware. > > The error is properly returned because HW doesn't support all packets > timestamping. But the adapter->flags is cleared of timestamp flags > even though no HW configuration was done. From that point no RX timestamps > are received by user-space application. But configuration shows good > values: > > [root@hostname ~]# hwstamp_ctl -i eth0 > current settings: > tx_type 1 > rx_filter 12 > > Fix the issue by applying new flags only when the HW was actually > configured. > > Fixes: a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x devices") > Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> > --- > drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c | 28 +++++++++++--------- > 1 file changed, 15 insertions(+), 13 deletions(-) > Tested-by: Pucha Himasekhar Reddy <himasekharx.reddy.pucha@intel.com> (A Contingent worker at Intel)
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c index 0310af851086..9339edbd9082 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c @@ -979,6 +979,7 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, u32 tsync_tx_ctl = IXGBE_TSYNCTXCTL_ENABLED; u32 tsync_rx_ctl = IXGBE_TSYNCRXCTL_ENABLED; u32 tsync_rx_mtrl = PTP_EV_PORT << 16; + u32 aflags = adapter->flags; bool is_l2 = false; u32 regval; @@ -996,20 +997,20 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, case HWTSTAMP_FILTER_NONE: tsync_rx_ctl = 0; tsync_rx_mtrl = 0; - adapter->flags &= ~(IXGBE_FLAG_RX_HWTSTAMP_ENABLED | - IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); + aflags &= ~(IXGBE_FLAG_RX_HWTSTAMP_ENABLED | + IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); break; case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_L4_V1; tsync_rx_mtrl |= IXGBE_RXMTRL_V1_SYNC_MSG; - adapter->flags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED | - IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); + aflags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED | + IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); break; case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_L4_V1; tsync_rx_mtrl |= IXGBE_RXMTRL_V1_DELAY_REQ_MSG; - adapter->flags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED | - IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); + aflags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED | + IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); break; case HWTSTAMP_FILTER_PTP_V2_EVENT: case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: @@ -1023,8 +1024,8 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_EVENT_V2; is_l2 = true; config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; - adapter->flags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED | - IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); + aflags |= (IXGBE_FLAG_RX_HWTSTAMP_ENABLED | + IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); break; case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: case HWTSTAMP_FILTER_NTP_ALL: @@ -1035,7 +1036,7 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, if (hw->mac.type >= ixgbe_mac_X550) { tsync_rx_ctl |= IXGBE_TSYNCRXCTL_TYPE_ALL; config->rx_filter = HWTSTAMP_FILTER_ALL; - adapter->flags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED; + aflags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED; break; } fallthrough; @@ -1046,8 +1047,6 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, * Delay_Req messages and hardware does not support * timestamping all packets => return error */ - adapter->flags &= ~(IXGBE_FLAG_RX_HWTSTAMP_ENABLED | - IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER); config->rx_filter = HWTSTAMP_FILTER_NONE; return -ERANGE; } @@ -1079,8 +1078,8 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, IXGBE_TSYNCRXCTL_TYPE_ALL | IXGBE_TSYNCRXCTL_TSIP_UT_EN; config->rx_filter = HWTSTAMP_FILTER_ALL; - adapter->flags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED; - adapter->flags &= ~IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER; + aflags |= IXGBE_FLAG_RX_HWTSTAMP_ENABLED; + aflags &= ~IXGBE_FLAG_RX_HWTSTAMP_IN_REGISTER; is_l2 = true; break; default: @@ -1113,6 +1112,9 @@ static int ixgbe_ptp_set_timestamp_mode(struct ixgbe_adapter *adapter, IXGBE_WRITE_FLUSH(hw); + /* configure adapter flags only when HW is actually configured */ + adapter->flags = aflags; + /* clear TX/RX time stamp registers, just to be sure */ ixgbe_ptp_clear_tx_timestamp(adapter); IXGBE_READ_REG(hw, IXGBE_RXSTMPH);
The commit in fixes introduced flags to control the status of hardware configuration while processing packets. At the same time another structure is used to provide configuration of timestamper to user-space applications. The way it was coded makes this structures go out of sync easily. The repro is easy for 82599 chips: [root@hostname ~]# hwstamp_ctl -i eth0 -r 12 -t 1 current settings: tx_type 0 rx_filter 0 new settings: tx_type 1 rx_filter 12 The eth0 device is properly configured to timestamp any PTPv2 events. [root@hostname ~]# hwstamp_ctl -i eth0 -r 1 -t 1 current settings: tx_type 1 rx_filter 12 SIOCSHWTSTAMP failed: Numerical result out of range The requested time stamping mode is not supported by the hardware. The error is properly returned because HW doesn't support all packets timestamping. But the adapter->flags is cleared of timestamp flags even though no HW configuration was done. From that point no RX timestamps are received by user-space application. But configuration shows good values: [root@hostname ~]# hwstamp_ctl -i eth0 current settings: tx_type 1 rx_filter 12 Fix the issue by applying new flags only when the HW was actually configured. Fixes: a9763f3cb54c ("ixgbe: Update PTP to support X550EM_x devices") Signed-off-by: Vadim Fedorenko <vadim.fedorenko@linux.dev> --- drivers/net/ethernet/intel/ixgbe/ixgbe_ptp.c | 28 +++++++++++--------- 1 file changed, 15 insertions(+), 13 deletions(-)