net: reject PTP periodic output requests with unsupported flags
Subsystems (11)
Richard Cochran
maintainerPATCH 2/13 net: reject PTP periodic output requests with unsupported flags
From: Jacob Keller <[email protected]>
Commit 823eb2a3c4c7 ("PTP: add support for one-shot output") introduced a new flag for the PTP periodic output request ioctl. This flag is not currently supported by any driver.
Fix all drivers which implement the periodic output request ioctl to explicitly reject any request with flags they do not understand. This ensures that the driver does not accidentally misinterpret the PTP_PEROUT_ONE_SHOT flag, or any new flag introduced in the future.
This is important for forward compatibility: if a new flag is introduced, the driver should reject requests to enable the flag until the driver has actually been modified to support the flag in question.
--- drivers/net/ethernet/broadcom/tg3.c | 4 ++++ drivers/net/ethernet/intel/igb/igb_ptp.c | 4 ++++ drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c | 4 ++++ drivers/net/ethernet/microchip/lan743x_ptp.c | 4 ++++ drivers/net/ethernet/renesas/ravb_ptp.c | 4 ++++ drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c | 4 ++++ drivers/net/phy/dp83640.c | 3 +++ 7 files changed, 27 insertions(+)
drivers/net/ethernet/broadcom/tg3.c +4 -0
@@ -6280,6 +6280,10 @@ static int tg3_ptp_enable(struct ptp_clock_info *ptp,switch (rq->type) {case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; +if (rq->perout.index !=0 )return -EINVAL ;
drivers/net/ethernet/intel/igb/igb_ptp.c +4 -0
@@ -551,6 +551,10 @@ static int igb_ptp_feature_enable_i210(struct ptp_clock_info *ptp,return 0 ;case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; +if (on) { pin =ptp_find_pin (igb->ptp_clock,PTP_PF_PEROUT , rq->perout.index);
drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +4 -0
@@ -290,6 +290,10 @@ static int mlx5_perout_configure(struct ptp_clock_info *ptp,if (!MLX5_PPS_CAP (mdev))return -EOPNOTSUPP ; +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; +if (rq->perout.index >= clock->ptp_info.n_pins)return -EINVAL ;
drivers/net/ethernet/microchip/lan743x_ptp.c +4 -0
@@ -429,6 +429,10 @@ static int lan743x_ptp_perout(struct lan743x_adapter *adapter, int on,int pulse_width =0 ;int perout_bit =0 ; +/* Reject requests with unsupported flags */ +if (perout->flags) +return -EOPNOTSUPP ; +if (!on) {lan743x_ptp_perout_off (adapter);return 0 ;
drivers/net/ethernet/renesas/ravb_ptp.c +4 -0
@@ -211,6 +211,10 @@ static int ravb_ptp_perout(struct ptp_clock_info *ptp,unsigned long flags;int error =0 ; +/* Reject requests with unsupported flags */ +if (req->flags) +return -EOPNOTSUPP ; +if (req->index)return -EINVAL ;
drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c +4 -0
@@ -140,6 +140,10 @@ static int stmmac_enable(struct ptp_clock_info *ptp,switch (rq->type) {case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; + cfg = &priv->pps[rq->perout.index]; cfg->start.tv_sec = rq->perout.start.sec;
drivers/net/phy/dp83640.c +3 -0
@@ -491,6 +491,9 @@ static int ptp_dp83640_enable(struct ptp_clock_info *ptp,return 0 ;case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ;if (rq->perout.index >=N_PER_OUT )return -EINVAL ;return periodic_output (clock, rq, on, rq->perout.index);
Signature · 1 line
2.20.1
Quoted 13 lines · Richard Cochran
From: Jacob Keller <[email protected]>
Commit
823eb2a3c4c7("PTP: add support for one-shot output") introduced
a new flag for the PTP periodic output request ioctl. This flag is not
currently supported by any driver.Fix all drivers which implement the periodic output request ioctl to explicitly reject any request with flags they do not understand. This ensures that the driver does not accidentally misinterpret the
PTP_PEROUT_ONE_SHOTflag, or any new flag introduced in the future.
Quoted 5 lines
This is important for forward compatibility: if a new flag is introduced, the driver should reject requests to enable the flag until
the driver has actually been modified to support the flag in question.
LGTM, just there might be a problem with old tools that didn't clear the flags upon request and expected PTP periodic .. they will stop to work with new kernel, am i am not sure if such tools do exist.
But the fact now that we have PTP_PEROUT_ONE_SHOT, we need to trust both driver and tools to do the right thing.
What are the tools to test PTP_PEROUT_ONE_SHOT ? to support this in mlx5 it is just a matter of a flipping a bit.
Quoted 136 lines
Cc Felipe Balbi <[email protected]>Cc David S. Miller <[email protected]>Cc Christopher Hall <[email protected]>Signed-off-by Jacob Keller <[email protected]>Signed-off-by Richard Cochran <[email protected]>Tested-by Aaron Brown <[email protected]>--- drivers/net/ethernet/broadcom/tg3.c | 4 ++++ drivers/net/ethernet/intel/igb/igb_ptp.c | 4 ++++ drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c | 4 ++++ drivers/net/ethernet/microchip/lan743x_ptp.c | 4 ++++ drivers/net/ethernet/renesas/ravb_ptp.c | 4 ++++ drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c | 4 ++++ drivers/net/phy/dp83640.c | 3 +++ 7 files changed, 27 insertions(+)drivers/net/ethernet/broadcom/tg3.c +4
-0@@ -6280,6 +6280,10 @@ static int tg3_ptp_enable(struct ptp_clock_info *ptp, switch (rq->type) {case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; +if (rq->perout.index !=0 )return -EINVAL ;drivers/net/ethernet/intel/igb/igb_ptp.c +4
-0@@ -551,6 +551,10 @@ static int igb_ptp_feature_enable_i210(structptp_clock_info *ptp, return 0;case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; +if (on) { pin = ptp_find_pin(igb->ptp_clock,
PTP_PF_PEROUT,drivers/net/ethernet/intel/igb/igb_ptp.c +0
-0rq->perout.index);drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +4
-0@@ -290,6 +290,10 @@ static int mlx5_perout_configure(struct ptp_clock_info *ptp, if (!MLX5_PPS_CAP(mdev)) return -EOPNOTSUPP; +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; +if (rq->perout.index >= clock->ptp_info.n_pins)return -EINVAL ;drivers/net/ethernet/microchip/lan743x_ptp.c +4
-0@@ -429,6 +429,10 @@ static int lan743x_ptp_perout(struct lan743x_adapter *adapter, int on, int pulse_width = 0;int perout_bit =0 ; +/* Reject requests with unsupported flags */ +if (perout->flags) +return -EOPNOTSUPP ; +if (!on) {lan743x_ptp_perout_off (adapter);drivers/net/ethernet/microchip/lan743x_ptp.c +0
-0return 0 ;drivers/net/ethernet/renesas/ravb_ptp.c +4
-0@@ -211,6 +211,10 @@ static int ravb_ptp_perout(struct ptp_clock_info *ptp, unsigned long flags;int error =0 ; +/* Reject requests with unsupported flags */ +if (req->flags) +return -EOPNOTSUPP ; +if (req->index)return -EINVAL ;drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c +4
-0@@ -140,6 +140,10 @@ static int stmmac_enable(struct ptp_clock_info *ptp, switch (rq->type) {case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ; + cfg = &priv->pps[rq->perout.index];drivers/net/ethernet/stmicro/stmmac/stmmac_ptp.c +0
-0cfg->start.tv_sec = rq->perout.start.sec;drivers/net/phy/dp83640.c +3
-0@@ -491,6 +491,9 @@ static int ptp_dp83640_enable(structptp_clock_info *ptp, return 0;case PTP_CLK_REQ_PEROUT : +/* Reject requests with unsupported flags */ +if (rq->perout.flags) +return -EOPNOTSUPP ;if (rq->perout.index >=N_PER_OUT )return -EINVAL ;return periodic_output(clock, rq, on, rq-
Quoted 1 line
perout.index);
Quoted 14 lines
-----Original Message-----
From: Saeed Mahameed <[email protected]>
Sent: Thursday, November 14, 2019 3:58 PM
To: richardcochran at gmail.com; netdev at vger.kernel.org
Cc: Hall, Christopher S <[email protected]>; Eugenia Emantayev <[email protected]>; davem at davemloft.net;
sergei.shtylyov at cogentembedded.com; Feras Daoud <[email protected]>; stefan.sorensen at spectralink.com; Brown, Aaron F <[email protected]>; brandon.streiff at ni.com; Keller, Jacob E <[email protected]>; Kirsher, Jeffrey T <[email protected]>; intel-wired-lan at lists.osuosl.org; felipe.balbi at linux.intel.com
Subject: Re: [PATCH net 02/13] net: reject PTP periodic output requests with unsupported flags
Quoted 7 lines
Quoted 2 lines · Richard Cochran
From: Jacob Keller <[email protected]>
LGTM, just there might be a problem with old tools that didn't clear the flags upon request and expected PTP periodic .. they will stop to work with new kernel, am i am not sure if such tools do exist.
Not quite. This is why there is a V2 ioctl now. The V1 ioctl now explicitly disables any bits set beyond the originally defined ones, so that callers of v1 will continue to work, even if they send junk data.
Thus, old tools should continue working with the old interface, while tools that want new functionality must be upgraded to use the v2 ioctls.
Thanks,
Jake
Quoted 8 lines
But the fact now that we have
PTP_PEROUT_ONE_SHOT, we need to trust both driver and tools to do the right thing.What are the tools to test
PTP_PEROUT_ONE_SHOT? to support this in mlx5 it is just a matter of a flipping a bit.Reviewed-by Saeed Mahameed <[email protected]>