lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CDCA1F64.1E87D%matthew.vick@intel.com>
Date:	Tue, 28 May 2013 15:58:07 +0000
From:	"Vick, Matthew" <matthew.vick@...el.com>
To:	Richard Cochran <richardcochran@...il.com>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>
CC:	David Miller <davem@...emloft.net>,
	"e1000-devel@...ts.sourceforge.net" 
	<e1000-devel@...ts.sourceforge.net>,
	"Keller, Jacob E" <jacob.e.keller@...el.com>,
	"Kirsher, Jeffrey T" <jeffrey.t.kirsher@...el.com>
Subject: Re: [PATCH net-next 4/4] igb: enable auxiliary PHC functions for
 the i210.

On 5/27/13 2:21 AM, "Richard Cochran" <richardcochran@...il.com> wrote:

>The i210 device offers a number of special PTP Hardware Clock features on
>the Software Defined Pins (SDPs). This patch adds support for three of the
>possible functions, namely time stamping external events, a periodic
>output signal, and an internal PPS event for adjusting the kernel system
>time.
>
>A pair of module parameters lets the user choose which of the four SDPs
>will be used as the input signal and which as the output.
>
>Signed-off-by: Richard Cochran <richardcochran@...il.com>
>---
> drivers/net/ethernet/intel/igb/igb.h      |    2 +
> drivers/net/ethernet/intel/igb/igb_main.c |   30 ++++++
> drivers/net/ethernet/intel/igb/igb_ptp.c  |  157
>++++++++++++++++++++++++++++-
> 3 files changed, 188 insertions(+), 1 deletion(-)
>
>diff --git a/drivers/net/ethernet/intel/igb/igb.h
>b/drivers/net/ethernet/intel/igb/igb.h
>index 15ea8dc..802b79c 100644
>--- a/drivers/net/ethernet/intel/igb/igb.h
>+++ b/drivers/net/ethernet/intel/igb/igb.h
>@@ -435,6 +435,8 @@ struct igb_adapter {
> 	struct timecounter tc;
> 	u32 tx_hwtstamp_timeouts;
> 	u32 rx_hwtstamp_cleared;
>+	struct timespec start;
>+	struct timespec period;
> 
> 	char fw_version[32];
> #ifdef CONFIG_IGB_HWMON
>diff --git a/drivers/net/ethernet/intel/igb/igb_main.c
>b/drivers/net/ethernet/intel/igb/igb_main.c
>index d25a965..58120cd 100644
>--- a/drivers/net/ethernet/intel/igb/igb_main.c
>+++ b/drivers/net/ethernet/intel/igb/igb_main.c
>@@ -5038,12 +5038,42 @@ void igb_update_stats(struct igb_adapter *adapter,
> static void igb_tsync_interrupt(struct igb_adapter *adapter)
> {
> 	struct e1000_hw *hw = &adapter->hw;
>+	struct ptp_clock_event event;
> 	u32 tsicr = rd32(E1000_TSICR);
> 
>+	if (tsicr & TSINTR_SYS_WRAP) {
>+		event.type = PTP_CLOCK_PPS;
>+		ptp_clock_event(adapter->ptp_clock, &event);
>+	}
>+
> 	if (tsicr & E1000_TSICR_TXTS) {
> 		/* retrieve hardware timestamp */
> 		schedule_work(&adapter->ptp_tx_work);
> 	}
>+
>+	if (tsicr & TSINTR_TT0) {
>+		struct timespec ts;
>+		u32 tsauxc;
>+		spin_lock(&adapter->tmreg_lock);
>+		ts = timespec_add(adapter->start, adapter->period);
>+		wr32(E1000_TRGTTIML0, ts.tv_nsec);
>+		wr32(E1000_TRGTTIMH0, ts.tv_sec);
>+		tsauxc = rd32(E1000_TSAUXC);
>+		tsauxc |= TSAUXC_EN_TT0;
>+		wr32(E1000_TSAUXC, tsauxc);
>+		adapter->start = ts;
>+		spin_unlock(&adapter->tmreg_lock);
>+	}
>+
>+	if (tsicr & TSINTR_AUTT0) {
>+		u32 sec, nsec;
>+		nsec = rd32(E1000_AUXSTMPL0);
>+		sec  = rd32(E1000_AUXSTMPH0);
>+		event.type = PTP_CLOCK_EXTTS;
>+		event.index = 0;
>+		event.timestamp = sec * 1000000000ULL + nsec;
>+		ptp_clock_event(adapter->ptp_clock, &event);
>+	}
> }

I would prefer it if we did a MAC check before these two TSICR checks,
since we're making some assumptions about the hardware within the
interrupt cases. At the very least, a comment that these are only
applicable to I210/I211 would be nice.

> 
> static irqreturn_t igb_msix_other(int irq, void *data)
>diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c
>b/drivers/net/ethernet/intel/igb/igb_ptp.c
>index 5944de0..8cf4b8a 100644
>--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
>+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
>@@ -23,6 +23,15 @@
> 
> #include "igb.h"
> 
>+static int igb_input_sdp = 0;
>+static int igb_output_sdp = 1;
>+module_param(igb_input_sdp, int, 0444);
>+module_param(igb_output_sdp, int, 0444);
>+MODULE_PARM_DESC(igb_input_sdp,
>+		 "The SDP used as an input, to time stamp external events");
>+MODULE_PARM_DESC(igb_output_sdp,
>+		 "The SDP used to output the programmable periodic signal");
>+

Is there any other mechanism we could use to control this? I would imagine
not, but I know module parameters are generally frowned upon.

> #define INCVALUE_MASK		0x7fffffff
> #define ISGN			0x80000000
> 
>@@ -359,6 +368,86 @@ static int igb_ptp_settime_i210(struct
>ptp_clock_info *ptp,
> 	return 0;
> }
> 
>+static int igb_ptp_enable_i210(struct ptp_clock_info *ptp,
>+			       struct ptp_clock_request *rq, int on)
>+{
>+	struct igb_adapter *igb = container_of(ptp, struct igb_adapter,
>+					       ptp_caps);
>+	struct e1000_hw *hw = &igb->hw;
>+	unsigned long flags;
>+	struct timespec ts;
>+	u32 tsauxc, tsim;
>+	s64 ns;
>+
>+	switch (rq->type) {
>+	case PTP_CLK_REQ_EXTTS:
>+		if (rq->extts.index != 0)
>+			return -EINVAL;
>+		spin_lock_irqsave(&igb->tmreg_lock, flags);
>+		tsauxc = rd32(E1000_TSAUXC);
>+		tsim = rd32(E1000_TSIM);
>+		if (on) {
>+			tsauxc |= TSAUXC_EN_TS0;
>+			tsim |= TSINTR_AUTT0;
>+		} else {
>+			tsauxc &= ~TSAUXC_EN_TS0;
>+			tsim &= ~TSINTR_AUTT0;
>+		}
>+		wr32(E1000_TSAUXC, tsauxc);
>+		wr32(E1000_TSIM, tsim);
>+		spin_unlock_irqrestore(&igb->tmreg_lock, flags);
>+		return 0;
>+
>+	case PTP_CLK_REQ_PEROUT:
>+		if (rq->perout.index != 0)
>+			return -EINVAL;
>+
>+		ts.tv_sec = rq->perout.period.sec;
>+		ts.tv_nsec = rq->perout.period.nsec;
>+		ns = timespec_to_ns(&ts);
>+		ns = ns >> 1;
>+		if (on && ns < 500000LL) {
>+			/* 2k interrupts per second is an awful lot. */
>+			return -EINVAL;
>+		}
>+		ts = ns_to_timespec(ns);
>+
>+		spin_lock_irqsave(&igb->tmreg_lock, flags);
>+		tsauxc = rd32(E1000_TSAUXC);
>+		tsim = rd32(E1000_TSIM);
>+		if (on) {
>+			igb->start.tv_sec = rq->perout.start.sec;
>+			igb->start.tv_nsec = rq->perout.start.nsec;
>+			igb->period.tv_sec = ts.tv_sec;
>+			igb->period.tv_nsec = ts.tv_nsec;
>+			wr32(E1000_TRGTTIML0, rq->perout.start.sec);
>+			wr32(E1000_TRGTTIMH0, rq->perout.start.nsec);
>+			tsauxc |= TSAUXC_EN_TT0;
>+			tsim |= TSINTR_TT0;
>+		} else {
>+			tsauxc &= ~TSAUXC_EN_TT0;
>+			tsim &= ~TSINTR_TT0;
>+		}
>+		wr32(E1000_TSAUXC, tsauxc);
>+		wr32(E1000_TSIM, tsim);
>+		spin_unlock_irqrestore(&igb->tmreg_lock, flags);
>+		return 0;
>+
>+	case PTP_CLK_REQ_PPS:
>+		spin_lock_irqsave(&igb->tmreg_lock, flags);
>+		tsim = rd32(E1000_TSIM);
>+		if (on)
>+			tsim |= TSINTR_SYS_WRAP;
>+		else
>+			tsim &= ~TSINTR_SYS_WRAP;
>+		wr32(E1000_TSIM, tsim);
>+		spin_unlock_irqrestore(&igb->tmreg_lock, flags);
>+		return 0;
>+	}
>+
>+	return -EOPNOTSUPP;
>+}
>+
> static int igb_ptp_enable(struct ptp_clock_info *ptp,
> 			  struct ptp_clock_request *rq, int on)
> {
>@@ -711,6 +800,68 @@ int igb_ptp_hwtstamp_ioctl(struct net_device *netdev,
> 		-EFAULT : 0;
> }
> 
>+static int igb_sdp_init(struct igb_adapter *adapter)
>+{
>+	struct e1000_hw *hw = &adapter->hw;
>+	u32 ctrl, ctrl_ext, tssdp = 0;
>+
>+	if (igb_input_sdp == igb_output_sdp) {
>+		pr_err("SDP %d set as both input and output\n", igb_input_sdp);
>+		return -1;

Shouldn't this return -EINVAL?

>+	}
>+
>+	ctrl     = rd32(E1000_CTRL);
>+	ctrl_ext = rd32(E1000_CTRL_EXT);
>+
>+	switch (igb_input_sdp) {
>+	case 0:
>+		tssdp |= AUX0_TS_SDP_EN | AUX0_SEL_SDP0;
>+		ctrl  &= ~E1000_CTRL_SDP0_DIR;
>+		break;
>+	case 1:
>+		tssdp |= AUX0_TS_SDP_EN | AUX0_SEL_SDP1;
>+		ctrl  &= ~E1000_CTRL_SDP1_DIR;
>+		break;
>+	case 2:
>+		tssdp |= AUX0_TS_SDP_EN | AUX0_SEL_SDP2;
>+		ctrl_ext &= ~E1000_CTRL_EXT_SDP2_DIR;
>+		break;
>+	case 3:
>+		tssdp |= AUX0_TS_SDP_EN | AUX0_SEL_SDP3;
>+		ctrl_ext &= ~E1000_CTRL_EXT_SDP3_DIR;
>+		break;
>+	default:
>+		pr_err("Input SDP %d out of range\n", igb_input_sdp);

Shouldn't this return -EINVAL as well?

>+	}
>+
>+	switch (igb_output_sdp) {
>+	case 0:
>+		tssdp |= TS_SDP0_SEL_TT0 | TS_SDP0_EN;
>+		ctrl |= E1000_CTRL_SDP0_DIR;
>+		break;
>+	case 1:
>+		tssdp |= TS_SDP1_SEL_TT0 | TS_SDP1_EN;
>+		ctrl |= E1000_CTRL_SDP1_DIR;
>+		break;
>+	case 2:
>+		tssdp |= TS_SDP2_SEL_TT0 | TS_SDP2_EN;
>+		ctrl_ext |= E1000_CTRL_EXT_SDP2_DIR;
>+		break;
>+	case 3:
>+		tssdp |= TS_SDP3_SEL_TT0 | TS_SDP3_EN;
>+		ctrl_ext |= E1000_CTRL_EXT_SDP3_DIR;
>+		break;
>+	default:
>+		pr_err("Output SDP %d out of range\n", igb_output_sdp);

Same here?

>+	}
>+
>+	wr32(E1000_TSSDP, tssdp);
>+	wr32(E1000_CTRL, ctrl);
>+	wr32(E1000_CTRL_EXT, ctrl_ext);
>+
>+	return 0;
>+}
>+
> void igb_ptp_init(struct igb_adapter *adapter)
> {
> 	struct e1000_hw *hw = &adapter->hw;
>@@ -766,7 +917,7 @@ void igb_ptp_init(struct igb_adapter *adapter)
> 		adapter->ptp_caps.adjtime = igb_ptp_adjtime_i210;
> 		adapter->ptp_caps.gettime = igb_ptp_gettime_i210;
> 		adapter->ptp_caps.settime = igb_ptp_settime_i210;
>-		adapter->ptp_caps.enable = igb_ptp_enable;
>+		adapter->ptp_caps.enable = igb_ptp_enable_i210;
> 		/* Enable the timer functions by clearing bit 31. */
> 		wr32(E1000_TSAUXC, 0x0);
> 		break;
>@@ -785,6 +936,10 @@ void igb_ptp_init(struct igb_adapter *adapter)
> 		struct timespec ts = ktime_to_timespec(ktime_get_real());
> 
> 		igb_ptp_settime_i210(&adapter->ptp_caps, &ts);
>+		igb_sdp_init(adapter);

What if igb_sdp_init fails?

>+		adapter->ptp_caps.n_ext_ts = 1;
>+		adapter->ptp_caps.n_per_out = 1;
>+		adapter->ptp_caps.pps = 1;
> 	} else {
> 		timecounter_init(&adapter->tc, &adapter->cc,
> 				 ktime_to_ns(ktime_get_real()));
>-- 
>1.7.10.4
>

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ