Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking updates from Davic Miller:

 1) Support busy polling generically, for all NAPI drivers.  From Eric
    Dumazet.

 2) Add byte/packet counter support to nft_ct, from Floriani Westphal.

 3) Add RSS/XPS support to mvneta driver, from Gregory Clement.

 4) Implement IPV6_HDRINCL socket option for raw sockets, from Hannes
    Frederic Sowa.

 5) Add support for T6 adapter to cxgb4 driver, from Hariprasad Shenai.

 6) Add support for VLAN device bridging to mlxsw switch driver, from
    Ido Schimmel.

 7) Add driver for Netronome NFP4000/NFP6000, from Jakub Kicinski.

 8) Provide hwmon interface to mlxsw switch driver, from Jiri Pirko.

 9) Reorganize wireless drivers into per-vendor directories just like we
    do for ethernet drivers.  From Kalle Valo.

10) Provide a way for administrators "destroy" connected sockets via the
    SOCK_DESTROY socket netlink diag operation.  From Lorenzo Colitti.

11) Add support to add/remove multicast routes via netlink, from Nikolay
    Aleksandrov.

12) Make TCP keepalive settings per-namespace, from Nikolay Borisov.

13) Add forwarding and packet duplication facilities to nf_tables, from
    Pablo Neira Ayuso.

14) Dead route support in MPLS, from Roopa Prabhu.

15) TSO support for thunderx chips, from Sunil Goutham.

16) Add driver for IBM's System i/p VNIC protocol, from Thomas Falcon.

17) Rationalize, consolidate, and more completely document the checksum
    offloading facilities in the networking stack.  From Tom Herbert.

18) Support aborting an ongoing scan in mac80211/cfg80211, from
    Vidyullatha Kanchanapally.

19) Use per-bucket spinlock for bpf hash facility, from Tom Leiming.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1375 commits)
  net: bnxt: always return values from _bnxt_get_max_rings
  net: bpf: reject invalid shifts
  phonet: properly unshare skbs in phonet_rcv()
  dwc_eth_qos: Fix dma address for multi-fragment skbs
  phy: remove an unneeded condition
  mdio: remove an unneed condition
  mdio_bus: NULL dereference on allocation error
  net: Fix typo in netdev_intersect_features
  net: freescale: mac-fec: Fix build error from phy_device API change
  net: freescale: ucc_geth: Fix build error from phy_device API change
  bonding: Prevent IPv6 link local address on enslaved devices
  IB/mlx5: Add flow steering support
  net/mlx5_core: Export flow steering API
  net/mlx5_core: Make ipv4/ipv6 location more clear
  net/mlx5_core: Enable flow steering support for the IB driver
  net/mlx5_core: Initialize namespaces only when supported by device
  net/mlx5_core: Set priority attributes
  net/mlx5_core: Connect flow tables
  net/mlx5_core: Introduce modify flow table command
  net/mlx5_core: Managing root flow table
  ...

1 files changed, 35 insertions, 5 deletions

diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index 75991979f667..4286f3618bd0 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -193,7 +193,7 @@ struct jit_context {
 	 32 /* space for rbx, r13, r14, r15 */ + \
 	 8 /* space for skb_copy_bits() buffer */)
 
-#define PROLOGUE_SIZE 51
+#define PROLOGUE_SIZE 48
 
 /* emit x64 prologue code for BPF program and check it's size.
  * bpf_tail_call helper will skip it while jumping into another program
@@ -229,11 +229,15 @@ static void emit_prologue(u8 **pprog)
 	/* mov qword ptr [rbp-X],r15 */
 	EMIT3_off32(0x4C, 0x89, 0xBD, -STACKSIZE + 24);
 
-	/* clear A and X registers */
-	EMIT2(0x31, 0xc0); /* xor eax, eax */
-	EMIT3(0x4D, 0x31, 0xED); /* xor r13, r13 */
+	/* Clear the tail call counter (tail_call_cnt): for eBPF tail calls
+	 * we need to reset the counter to 0. It's done in two instructions,
+	 * resetting rax register to 0 (xor on eax gets 0 extended), and
+	 * moving it to the counter location.
+	 */
 
-	/* clear tail_cnt: mov qword ptr [rbp-X], rax */
+	/* xor eax, eax */
+	EMIT2(0x31, 0xc0);
+	/* mov qword ptr [rbp-X], rax */
 	EMIT3_off32(0x48, 0x89, 0x85, -STACKSIZE + 32);
 
 	BUILD_BUG_ON(cnt != PROLOGUE_SIZE);
@@ -455,6 +459,18 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
 			}
 
 		case BPF_ALU | BPF_MOV | BPF_K:
+			/* optimization: if imm32 is zero, use 'xor <dst>,<dst>'
+			 * to save 3 bytes.
+			 */
+			if (imm32 == 0) {
+				if (is_ereg(dst_reg))
+					EMIT1(add_2mod(0x40, dst_reg, dst_reg));
+				b2 = 0x31; /* xor */
+				b3 = 0xC0;
+				EMIT2(b2, add_2reg(b3, dst_reg, dst_reg));
+				break;
+			}
+
 			/* mov %eax, imm32 */
 			if (is_ereg(dst_reg))
 				EMIT1(add_1mod(0x40, dst_reg));
@@ -469,6 +485,20 @@ static int do_jit(struct bpf_prog *bpf_prog, int *addrs, u8 *image,
 				return -EINVAL;
 			}
 
+			/* optimization: if imm64 is zero, use 'xor <dst>,<dst>'
+			 * to save 7 bytes.
+			 */
+			if (insn[0].imm == 0 && insn[1].imm == 0) {
+				b1 = add_2mod(0x48, dst_reg, dst_reg);
+				b2 = 0x31; /* xor */
+				b3 = 0xC0;
+				EMIT3(b1, b2, add_2reg(b3, dst_reg, dst_reg));
+
+				insn++;
+				i++;
+				break;
+			}
+
 			/* movabsq %rax, imm64 */
 			EMIT2(add_1mod(0x48, dst_reg), add_1reg(0xB8, dst_reg));
 			EMIT(insn[0].imm, 4);