net: ptp: move PTP classifier in its own file

This commit fixes a build error reported by Fengguang, that is
triggered when CONFIG_NETWORK_PHY_TIMESTAMPING is not set:

  ERROR: "ptp_classify_raw" [drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe.ko] undefined!

The fix is to introduce its own file for the PTP BPF classifier,
so that PTP_1588_CLOCK and/or NETWORK_PHY_TIMESTAMPING can select
it independently from each other. IXP4xx driver on ARM needs to
select it as well since it does not seem to select PTP_1588_CLOCK
or similar that would pull it in automatically.

This also allows for hiding all of the internals of the BPF PTP
program inside that file, and only exporting relevant API bits
to drivers.

This patch also adds a kdoc documentation of ptp_classify_raw()
API to make it clear that it can return PTP_CLASS_* defines. Also,
the BPF program has been translated into bpf_asm code, so that it
can be more easily read and altered (extensively documented in [1]).

In the kernel tree under tools/net/ we have bpf_asm and bpf_dbg
tools, so the commented program can simply be translated via
`./bpf_asm -c prog` where prog is a file that contains the
commented code. This makes it easily readable/verifiable and when
there's a need to change something, jump offsets etc do not need
to be replaced manually which can be very error prone. Instead,
a newly translated version via bpf_asm can simply replace the old
code. I have checked opcode diffs before/after and it's the very
same filter.

  [1] Documentation/networking/filter.txt

Fixes: 164d8c666521 ("net: ptp: do not reimplement PTP/BPF classifier")
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Jiri Benc <jbenc@redhat.com>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

10 files changed, 173 insertions, 96 deletions

diff --git a/drivers/net/ethernet/xscale/Kconfig b/drivers/net/ethernet/xscale/Kconfig
index 3f43101..b81bc9f 100644
--- a/drivers/net/ethernet/xscale/Kconfig
+++ b/drivers/net/ethernet/xscale/Kconfig
@@ -23,6 +23,7 @@ config IXP4XX_ETH
 	tristate "Intel IXP4xx Ethernet support"
 	depends on ARM && ARCH_IXP4XX && IXP4XX_NPE && IXP4XX_QMGR
 	select PHYLIB
+	select NET_PTP_CLASSIFY
 	---help---
 	  Say Y here if you want to use built-in Ethernet ports
 	  on IXP4xx processor.
diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index 352c5e4..6a999e6 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -27,6 +27,7 @@
 #include <linux/module.h>
 #include <linux/net_tstamp.h>
 #include <linux/netdevice.h>
+#include <linux/if_vlan.h>
 #include <linux/phy.h>
 #include <linux/ptp_classify.h>
 #include <linux/ptp_clock_kernel.h>
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index 5a7910e..6963bdf 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -7,6 +7,7 @@ menu "PTP clock support"
 config PTP_1588_CLOCK
 	tristate "PTP clock support"
 	select PPS
+	select NET_PTP_CLASSIFY
 	help
 	  The IEEE 1588 standard defines a method to precisely
 	  synchronize distributed clocks over Ethernet networks. The
diff --git a/include/linux/ptp_classify.h b/include/linux/ptp_classify.h
index 6d3b0a2..7dfed71 100644
--- a/include/linux/ptp_classify.h
+++ b/include/linux/ptp_classify.h
@@ -23,11 +23,8 @@
 #ifndef _PTP_CLASSIFY_H_
 #define _PTP_CLASSIFY_H_
 
-#include <linux/if_ether.h>
-#include <linux/if_vlan.h>
 #include <linux/ip.h>
-#include <linux/filter.h>
-#include <linux/in.h>
+#include <linux/skbuff.h>
 
 #define PTP_CLASS_NONE  0x00 /* not a PTP event message */
 #define PTP_CLASS_V1    0x01 /* protocol version 1 */
@@ -40,7 +37,7 @@
 #define PTP_CLASS_PMASK 0xf0 /* mask for the packet type field */
 
 #define PTP_CLASS_V1_IPV4 (PTP_CLASS_V1 | PTP_CLASS_IPV4)
-#define PTP_CLASS_V1_IPV6 (PTP_CLASS_V1 | PTP_CLASS_IPV6) /*probably DNE*/
+#define PTP_CLASS_V1_IPV6 (PTP_CLASS_V1 | PTP_CLASS_IPV6) /* probably DNE */
 #define PTP_CLASS_V2_IPV4 (PTP_CLASS_V2 | PTP_CLASS_IPV4)
 #define PTP_CLASS_V2_IPV6 (PTP_CLASS_V2 | PTP_CLASS_IPV6)
 #define PTP_CLASS_V2_L2   (PTP_CLASS_V2 | PTP_CLASS_L2)
@@ -49,82 +46,34 @@
 #define PTP_EV_PORT 319
 #define PTP_GEN_BIT 0x08 /* indicates general message, if set in message type */
 
-#define OFF_ETYPE	12
-#define OFF_IHL		14
-#define OFF_FRAG	20
-#define OFF_PROTO4	23
-#define OFF_NEXT	6
-#define OFF_UDP_DST	2
-
 #define OFF_PTP_SOURCE_UUID	22 /* PTPv1 only */
 #define OFF_PTP_SEQUENCE_ID	30
 #define OFF_PTP_CONTROL		32 /* PTPv1 only */
 
-#define IPV4_HLEN(data) (((struct iphdr *)(data + OFF_IHL))->ihl << 2)
-
+/* Below defines should actually be removed at some point in time. */
 #define IP6_HLEN	40
 #define UDP_HLEN	8
-
-#define RELOFF_DST4	(ETH_HLEN + OFF_UDP_DST)
-#define OFF_DST6	(ETH_HLEN + IP6_HLEN + OFF_UDP_DST)
+#define OFF_IHL		14
 #define OFF_PTP6	(ETH_HLEN + IP6_HLEN + UDP_HLEN)
+#define IPV4_HLEN(data) (((struct iphdr *)(data + OFF_IHL))->ihl << 2)
 
-#define OP_AND	(BPF_ALU | BPF_AND  | BPF_K)
-#define OP_JEQ	(BPF_JMP | BPF_JEQ  | BPF_K)
-#define OP_JSET	(BPF_JMP | BPF_JSET | BPF_K)
-#define OP_LDB	(BPF_LD  | BPF_B    | BPF_ABS)
-#define OP_LDH	(BPF_LD  | BPF_H    | BPF_ABS)
-#define OP_LDHI	(BPF_LD  | BPF_H    | BPF_IND)
-#define OP_LDX	(BPF_LDX | BPF_B    | BPF_MSH)
-#define OP_OR	(BPF_ALU | BPF_OR   | BPF_K)
-#define OP_RETA	(BPF_RET | BPF_A)
-#define OP_RETK	(BPF_RET | BPF_K)
-
-#define PTP_FILTER \
-	{OP_LDH,	0,   0, OFF_ETYPE		}, /*              */ \
-	{OP_JEQ,	0,  12, ETH_P_IP		}, /* f goto L20   */ \
-	{OP_LDB,	0,   0, OFF_PROTO4		}, /*              */ \
-	{OP_JEQ,	0,   9, IPPROTO_UDP		}, /* f goto L10   */ \
-	{OP_LDH,	0,   0, OFF_FRAG		}, /*              */ \
-	{OP_JSET,	7,   0, 0x1fff			}, /* t goto L11   */ \
-	{OP_LDX,	0,   0, OFF_IHL			}, /*              */ \
-	{OP_LDHI,	0,   0, RELOFF_DST4		}, /*              */ \
-	{OP_JEQ,	0,   4, PTP_EV_PORT		}, /* f goto L12   */ \
-	{OP_LDHI,	0,   0, ETH_HLEN + UDP_HLEN	}, /*              */ \
-	{OP_AND,	0,   0, PTP_CLASS_VMASK		}, /*              */ \
-	{OP_OR,		0,   0, PTP_CLASS_IPV4		}, /*              */ \
-	{OP_RETA,	0,   0, 0			}, /*              */ \
-/*L1x*/	{OP_RETK,	0,   0, PTP_CLASS_NONE		}, /*              */ \
-/*L20*/	{OP_JEQ,	0,   9, ETH_P_IPV6		}, /* f goto L40   */ \
-	{OP_LDB,	0,   0, ETH_HLEN + OFF_NEXT	}, /*              */ \
-	{OP_JEQ,	0,   6, IPPROTO_UDP		}, /* f goto L30   */ \
-	{OP_LDH,	0,   0, OFF_DST6		}, /*              */ \
-	{OP_JEQ,	0,   4, PTP_EV_PORT		}, /* f goto L31   */ \
-	{OP_LDH,	0,   0, OFF_PTP6		}, /*              */ \
-	{OP_AND,	0,   0, PTP_CLASS_VMASK		}, /*              */ \
-	{OP_OR,		0,   0, PTP_CLASS_IPV6		}, /*              */ \
-	{OP_RETA,	0,   0, 0			}, /*              */ \
-/*L3x*/	{OP_RETK,	0,   0, PTP_CLASS_NONE		}, /*              */ \
-/*L40*/	{OP_JEQ,	0,   9, ETH_P_8021Q		}, /* f goto L50   */ \
-	{OP_LDH,	0,   0, OFF_ETYPE + 4		}, /*              */ \
-	{OP_JEQ,	0,  15, ETH_P_1588		}, /* f goto L60   */ \
-	{OP_LDB,	0,   0, ETH_HLEN + VLAN_HLEN	}, /*              */ \
-	{OP_AND,	0,   0, PTP_GEN_BIT		}, /*              */ \
-	{OP_JEQ,	0,  12, 0			}, /* f goto L6x   */ \
-	{OP_LDH,	0,   0, ETH_HLEN + VLAN_HLEN	}, /*              */ \
-	{OP_AND,	0,   0, PTP_CLASS_VMASK		}, /*              */ \
-	{OP_OR,		0,   0, PTP_CLASS_VLAN		}, /*              */ \
-	{OP_RETA,	0,   0, 0			}, /*              */ \
-/*L50*/	{OP_JEQ,	0,   7, ETH_P_1588		}, /* f goto L61   */ \
-	{OP_LDB,	0,   0, ETH_HLEN		}, /*              */ \
-	{OP_AND,	0,   0, PTP_GEN_BIT		}, /*              */ \
-	{OP_JEQ,	0,   4, 0			}, /* f goto L6x   */ \
-	{OP_LDH,	0,   0, ETH_HLEN		}, /*              */ \
-	{OP_AND,	0,   0, PTP_CLASS_VMASK		}, /*              */ \
-	{OP_OR,		0,   0, PTP_CLASS_L2		}, /*              */ \
-	{OP_RETA,	0,   0, 0			}, /*              */ \
-/*L6x*/	{OP_RETK,	0,   0, PTP_CLASS_NONE		},
-
+#if defined(CONFIG_NET_PTP_CLASSIFY)
+/**
+ * ptp_classify_raw - classify a PTP packet
+ * @skb: buffer
+ *
+ * Runs a minimal BPF dissector to classify a network packet to
+ * determine the PTP class. In case the skb does not contain any
+ * PTP protocol data, PTP_CLASS_NONE will be returned, otherwise
+ * PTP_CLASS_V1_IPV{4,6}, PTP_CLASS_V2_IPV{4,6} or
+ * PTP_CLASS_V2_{L2,VLAN}, depending on the packet content.
+ */
 unsigned int ptp_classify_raw(const struct sk_buff *skb);
 
+void __init ptp_classifier_init(void);
+#else
+static inline void ptp_classifier_init(void)
+{
+}
 #endif
+#endif /* _PTP_CLASSIFY_H_ */
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 18ef022..31edf63 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2630,8 +2630,6 @@ static inline ktime_t net_invalid_timestamp(void)
 	return ktime_set(0, 0);
 }
 
-void skb_timestamping_init(void);
-
 #ifdef CONFIG_NETWORK_PHY_TIMESTAMPING
 
 void skb_clone_tx_timestamp(struct sk_buff *skb);
diff --git a/net/Kconfig b/net/Kconfig
index e411046..d1f6f96 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -89,8 +89,12 @@ config NETWORK_SECMARK
 	  to nfmark, but designated for security purposes.
 	  If you are unsure how to answer this question, answer N.
 
+config NET_PTP_CLASSIFY
+	def_bool n
+
 config NETWORK_PHY_TIMESTAMPING
 	bool "Timestamping in PHY devices"
+	select NET_PTP_CLASSIFY
 	help
 	  This allows timestamping of network packets by PHYs with
 	  hardware timestamping capabilities. This option adds some
diff --git a/net/core/Makefile b/net/core/Makefile
index 9628c20..826b925 100644
--- a/net/core/Makefile
+++ b/net/core/Makefile
@@ -21,5 +21,6 @@ obj-$(CONFIG_FIB_RULES) += fib_rules.o
 obj-$(CONFIG_TRACEPOINTS) += net-traces.o
 obj-$(CONFIG_NET_DROP_MONITOR) += drop_monitor.o
 obj-$(CONFIG_NETWORK_PHY_TIMESTAMPING) += timestamping.o
+obj-$(CONFIG_NET_PTP_CLASSIFY) += ptp_classifier.o
 obj-$(CONFIG_CGROUP_NET_PRIO) += netprio_cgroup.o
 obj-$(CONFIG_CGROUP_NET_CLASSID) += netclassid_cgroup.o
diff --git a/net/core/ptp_classifier.c b/net/core/ptp_classifier.c
new file mode 100644
index 0000000..eaba0f6
--- /dev/null
+++ b/net/core/ptp_classifier.c
@@ -0,0 +1,141 @@
+/* PTP classifier
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+
+/* The below program is the bpf_asm (tools/net/) representation of
+ * the opcode array in the ptp_filter structure.
+ *
+ * For convenience, this can easily be altered and reviewed with
+ * bpf_asm and bpf_dbg, e.g. `./bpf_asm -c prog` where prog is a
+ * simple file containing the below program:
+ *
+ * ldh [12]                        ; load ethertype
+ *
+ * ; PTP over UDP over IPv4 over Ethernet
+ * test_ipv4:
+ *   jneq #0x800, test_ipv6        ; ETH_P_IP ?
+ *   ldb [23]                      ; load proto
+ *   jneq #17, drop_ipv4           ; IPPROTO_UDP ?
+ *   ldh [20]                      ; load frag offset field
+ *   jset #0x1fff, drop_ipv4       ; don't allow fragments
+ *   ldxb 4*([14]&0xf)             ; load IP header len
+ *   ldh [x + 16]                  ; load UDP dst port
+ *   jneq #319, drop_ipv4          ; is port PTP_EV_PORT ?
+ *   ldh [x + 22]                  ; load payload
+ *   and #0xf                      ; mask PTP_CLASS_VMASK
+ *   or #0x10                      ; PTP_CLASS_IPV4
+ *   ret a                         ; return PTP class
+ *   drop_ipv4: ret #0x0           ; PTP_CLASS_NONE
+ *
+ * ; PTP over UDP over IPv6 over Ethernet
+ * test_ipv6:
+ *   jneq #0x86dd, test_8021q      ; ETH_P_IPV6 ?
+ *   ldb [20]                      ; load proto
+ *   jneq #17, drop_ipv6           ; IPPROTO_UDP ?
+ *   ldh [56]                      ; load UDP dst port
+ *   jneq #319, drop_ipv6          ; is port PTP_EV_PORT ?
+ *   ldh [62]                      ; load payload
+ *   and #0xf                      ; mask PTP_CLASS_VMASK
+ *   or #0x20                      ; PTP_CLASS_IPV6
+ *   ret a                         ; return PTP class
+ *   drop_ipv6: ret #0x0           ; PTP_CLASS_NONE
+ *
+ * ; PTP over 802.1Q over Ethernet
+ * test_8021q:
+ *   jneq #0x8100, test_ieee1588   ; ETH_P_8021Q ?
+ *   ldh [16]                      ; load inner type
+ *   jneq #0x88f7, drop_ieee1588   ; ETH_P_1588 ?
+ *   ldb [18]                      ; load payload
+ *   and #0x8                      ; as we don't have ports here, test
+ *   jneq #0x0, drop_ieee1588      ; for PTP_GEN_BIT and drop these
+ *   ldh [18]                      ; reload payload
+ *   and #0xf                      ; mask PTP_CLASS_VMASK
+ *   or #0x40                      ; PTP_CLASS_V2_VLAN
+ *   ret a                         ; return PTP class
+ *
+ * ; PTP over Ethernet
+ * test_ieee1588:
+ *   jneq #0x88f7, drop_ieee1588   ; ETH_P_1588 ?
+ *   ldb [14]                      ; load payload
+ *   and #0x8                      ; as we don't have ports here, test
+ *   jneq #0x0, drop_ieee1588      ; for PTP_GEN_BIT and drop these
+ *   ldh [14]                      ; reload payload
+ *   and #0xf                      ; mask PTP_CLASS_VMASK
+ *   or #0x30                      ; PTP_CLASS_L2
+ *   ret a                         ; return PTP class
+ *   drop_ieee1588: ret #0x0       ; PTP_CLASS_NONE
+ */
+
+#include <linux/skbuff.h>
+#include <linux/filter.h>
+#include <linux/ptp_classify.h>
+
+static struct sk_filter *ptp_insns __read_mostly;
+
+unsigned int ptp_classify_raw(const struct sk_buff *skb)
+{
+	return SK_RUN_FILTER(ptp_insns, skb);
+}
+EXPORT_SYMBOL_GPL(ptp_classify_raw);
+
+void __init ptp_classifier_init(void)
+{
+	static struct sock_filter ptp_filter[] = {
+		{ 0x28,  0,  0, 0x0000000c },
+		{ 0x15,  0, 12, 0x00000800 },
+		{ 0x30,  0,  0, 0x00000017 },
+		{ 0x15,  0,  9, 0x00000011 },
+		{ 0x28,  0,  0, 0x00000014 },
+		{ 0x45,  7,  0, 0x00001fff },
+		{ 0xb1,  0,  0, 0x0000000e },
+		{ 0x48,  0,  0, 0x00000010 },
+		{ 0x15,  0,  4, 0x0000013f },
+		{ 0x48,  0,  0, 0x00000016 },
+		{ 0x54,  0,  0, 0x0000000f },
+		{ 0x44,  0,  0, 0x00000010 },
+		{ 0x16,  0,  0, 0x00000000 },
+		{ 0x06,  0,  0, 0x00000000 },
+		{ 0x15,  0,  9, 0x000086dd },
+		{ 0x30,  0,  0, 0x00000014 },
+		{ 0x15,  0,  6, 0x00000011 },
+		{ 0x28,  0,  0, 0x00000038 },
+		{ 0x15,  0,  4, 0x0000013f },
+		{ 0x28,  0,  0, 0x0000003e },
+		{ 0x54,  0,  0, 0x0000000f },
+		{ 0x44,  0,  0, 0x00000020 },
+		{ 0x16,  0,  0, 0x00000000 },
+		{ 0x06,  0,  0, 0x00000000 },
+		{ 0x15,  0,  9, 0x00008100 },
+		{ 0x28,  0,  0, 0x00000010 },
+		{ 0x15,  0, 15, 0x000088f7 },
+		{ 0x30,  0,  0, 0x00000012 },
+		{ 0x54,  0,  0, 0x00000008 },
+		{ 0x15,  0, 12, 0x00000000 },
+		{ 0x28,  0,  0, 0x00000012 },
+		{ 0x54,  0,  0, 0x0000000f },
+		{ 0x44,  0,  0, 0x00000040 },
+		{ 0x16,  0,  0, 0x00000000 },
+		{ 0x15,  0,  7, 0x000088f7 },
+		{ 0x30,  0,  0, 0x0000000e },
+		{ 0x54,  0,  0, 0x00000008 },
+		{ 0x15,  0,  4, 0x00000000 },
+		{ 0x28,  0,  0, 0x0000000e },
+		{ 0x54,  0,  0, 0x0000000f },
+		{ 0x44,  0,  0, 0x00000030 },
+		{ 0x16,  0,  0, 0x00000000 },
+		{ 0x06,  0,  0, 0x00000000 },
+	};
+	struct sock_fprog ptp_prog = {
+		.len = ARRAY_SIZE(ptp_filter), .filter = ptp_filter,
+	};
+
+	BUG_ON(sk_unattached_filter_create(&ptp_insns, &ptp_prog));
+}
diff --git a/net/core/timestamping.c b/net/core/timestamping.c
index 9ff26b3..6521dfd 100644
--- a/net/core/timestamping.c
+++ b/net/core/timestamping.c
@@ -23,14 +23,6 @@
 #include <linux/skbuff.h>
 #include <linux/export.h>
 
-static struct sk_filter *ptp_insns __read_mostly;
-
-unsigned int ptp_classify_raw(const struct sk_buff *skb)
-{
-	return SK_RUN_FILTER(ptp_insns, skb);
-}
-EXPORT_SYMBOL_GPL(ptp_classify_raw);
-
 static unsigned int classify(const struct sk_buff *skb)
 {
 	if (likely(skb->dev && skb->dev->phydev &&
@@ -140,13 +132,3 @@ bool skb_defer_rx_timestamp(struct sk_buff *skb)
 	return false;
 }
 EXPORT_SYMBOL_GPL(skb_defer_rx_timestamp);
-
-void __init skb_timestamping_init(void)
-{
-	static struct sock_filter ptp_filter[] = { PTP_FILTER };
-	struct sock_fprog ptp_prog = {
-		.len = ARRAY_SIZE(ptp_filter), .filter = ptp_filter,
-	};
-
-	BUG_ON(sk_unattached_filter_create(&ptp_insns, &ptp_prog));
-}
diff --git a/net/socket.c b/net/socket.c
index f25eaa3..1b1e7e6 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -72,6 +72,7 @@
 #include <linux/if_bridge.h>
 #include <linux/if_frad.h>
 #include <linux/if_vlan.h>
+#include <linux/ptp_classify.h>
 #include <linux/init.h>
 #include <linux/poll.h>
 #include <linux/cache.h>
@@ -2685,9 +2686,7 @@ static int __init sock_init(void)
 		goto out;
 #endif
 
-#ifdef CONFIG_NETWORK_PHY_TIMESTAMPING
-	skb_timestamping_init();
-#endif
+	ptp_classifier_init();
 
 out:
 	return err;