Skip to content

Commit

Permalink
MIPS: BPF: Introduce BPF ASM helpers
Browse files Browse the repository at this point in the history 
This commit introduces BPF ASM helpers for MIPS and MIPS64 kernels.
The purpose of this patch is to twofold:

1) We are now able to handle negative offsets instead of either
falling back to the interpreter or to simply not do anything and
bail out.

2) Optimize reads from the packet header instead of calling the C
helpers

Because of this patch, we are now able to get rid of quite a bit of
code in the JIT generation process by using MIPS optimized assembly
code. The new assembly code makes the test_bpf testsuite happy with
all 60 test passing successfully compared to the previous
implementation where 2 tests were failing.
Doing some basic analysis in the results between the old
implementation and the new one we can obtain the following
summary running current mainline on an ER8 board (+/- 30us delta is
ignored to prevent noise from kernel scheduling or IRQ latencies):

Summary: 22 tests are faster, 7 are slower and 47 saw no improvement

with the most notable improvement being the tcpdump tests. The 7 tests
that seem to be a bit slower is because they all follow the slow path
(bpf_internal_load_pointer_neg_helper) which is meant to be slow so
that's not a problem.

Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: netdev@vger.kernel.org
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Alexei Starovoitov <ast@plumgrid.com>
Cc: Daniel Borkmann <dborkman@redhat.com>
Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Cc: netdev@vger.kernel.org
Patchwork: http://patchwork.linux-mips.org/patch/10530/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
  • Loading branch information
Markos Chandras authored and Ralf Baechle committed Jun 21, 2015
1 parent beaf70b commit 266a88e
Show file tree
Hide file tree
Showing 4 changed files with 302 additions and 145 deletions.
2 changes: 1 addition & 1 deletion arch/mips/net/Makefile
View file
Original file line number Diff line number Diff line change
@@ -1,3 +1,3 @@
# MIPS networking code

obj-$(CONFIG_BPF_JIT) += bpf_jit.o
obj-$(CONFIG_BPF_JIT) += bpf_jit.o bpf_jit_asm.o
174 changes: 45 additions & 129 deletions arch/mips/net/bpf_jit.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -29,11 +29,14 @@
#include "bpf_jit.h"

/* ABI
*
* r_skb_hl SKB header length
* r_data SKB data pointer
* r_off Offset
* r_A BPF register A
* r_X BPF register X
* r_skb *skb
* r_M *scratch memory
* r_skb_len SKB length
*
* On entry (*bpf_func)(*skb, *filter)
* a0 = MIPS_R_A0 = skb;
Expand Down Expand Up @@ -75,6 +78,8 @@
#define SEEN_X SEEN_SREG(4)
#define SEEN_SKB SEEN_SREG(5)
#define SEEN_MEM SEEN_SREG(6)
/* SEEN_SK_DATA also implies skb_hl an skb_len */
#define SEEN_SKB_DATA (SEEN_SREG(7) | SEEN_SREG(1) | SEEN_SREG(0))

/* Arguments used by JIT */
#define ARGS_USED_BY_JIT 2 /* only applicable to 64-bit */
Expand Down Expand Up @@ -537,20 +542,6 @@ static void save_bpf_jit_regs(struct jit_ctx *ctx, unsigned offset)
/* Adjust the stack pointer */
emit_stack_offset(-align_sp(offset), ctx);

if (ctx->flags & SEEN_CALL) {
/* Argument save area */
if (config_enabled(CONFIG_64BIT))
/* Bottom of current frame */
real_off = align_sp(offset) - SZREG;
else
/* Top of previous frame */
real_off = align_sp(offset) + SZREG;
emit_store_stack_reg(MIPS_R_A0, r_sp, real_off, ctx);
emit_store_stack_reg(MIPS_R_A1, r_sp, real_off + SZREG, ctx);

real_off = 0;
}

tmp_flags = sflags = ctx->flags >> SEEN_SREG_SFT;
/* sflags is essentially a bitmap */
while (tmp_flags) {
Expand Down Expand Up @@ -583,19 +574,6 @@ static void restore_bpf_jit_regs(struct jit_ctx *ctx,
int i, real_off = 0;
u32 sflags, tmp_flags;

if (ctx->flags & SEEN_CALL) {
if (config_enabled(CONFIG_64BIT))
/* Bottom of current frame */
real_off = align_sp(offset) - SZREG;
else
/* Top of previous frame */
real_off = align_sp(offset) + SZREG;
emit_load_stack_reg(MIPS_R_A0, r_sp, real_off, ctx);
emit_load_stack_reg(MIPS_R_A1, r_sp, real_off + SZREG, ctx);

real_off = 0;
}

tmp_flags = sflags = ctx->flags >> SEEN_SREG_SFT;
/* sflags is a bitmap */
i = 0;
Expand Down Expand Up @@ -629,17 +607,7 @@ static unsigned int get_stack_depth(struct jit_ctx *ctx)
sp_off += 4 * BPF_MEMWORDS; /* BPF_MEMWORDS are 32-bit */

if (ctx->flags & SEEN_CALL)
/*
* The JIT code make calls to external functions using 2
* arguments. Therefore, for o32 we don't need to allocate
* space because we don't care if the argumetns are lost
* across calls. We do need however to preserve incoming
* arguments but the space is already allocated for us by
* the caller. On the other hand, for n64, we need to allocate
* this space ourselves. We need to preserve $ra as well.
*/
sp_off += config_enabled(CONFIG_64BIT) ?
(ARGS_USED_BY_JIT + 1) * SZREG : SZREG;
sp_off += SZREG; /* Space for our ra register */

return sp_off;
}
Expand All @@ -656,6 +624,19 @@ static void build_prologue(struct jit_ctx *ctx)
if (ctx->flags & SEEN_SKB)
emit_reg_move(r_skb, MIPS_R_A0, ctx);

if (ctx->flags & SEEN_SKB_DATA) {
/* Load packet length */
emit_load(r_skb_len, r_skb, offsetof(struct sk_buff, len),
ctx);
emit_load(r_tmp, r_skb, offsetof(struct sk_buff, data_len),
ctx);
/* Load the data pointer */
emit_load_ptr(r_skb_data, r_skb,
offsetof(struct sk_buff, data), ctx);
/* Load the header length */
emit_subu(r_skb_hl, r_skb_len, r_tmp, ctx);
}

if (ctx->flags & SEEN_X)
emit_jit_reg_move(r_X, r_zero, ctx);

Expand All @@ -678,43 +659,17 @@ static void build_epilogue(struct jit_ctx *ctx)
emit_nop(ctx);
}

static u64 jit_get_skb_b(struct sk_buff *skb, unsigned offset)
{
u8 ret;
int err;

err = skb_copy_bits(skb, offset, &ret, 1);

return (u64)err << 32 | ret;
}

static u64 jit_get_skb_h(struct sk_buff *skb, unsigned offset)
{
u16 ret;
int err;

err = skb_copy_bits(skb, offset, &ret, 2);

return (u64)err << 32 | ntohs(ret);
}

static u64 jit_get_skb_w(struct sk_buff *skb, unsigned offset)
{
u32 ret;
int err;

err = skb_copy_bits(skb, offset, &ret, 4);

return (u64)err << 32 | ntohl(ret);
}
#define CHOOSE_LOAD_FUNC(K, func) \
((int)K < 0 ? ((int)K >= SKF_LL_OFF ? func##_negative : func) : \
func##_positive)

static int build_body(struct jit_ctx *ctx)
{
void *load_func[] = {jit_get_skb_b, jit_get_skb_h, jit_get_skb_w};
const struct bpf_prog *prog = ctx->skf;
const struct sock_filter *inst;
unsigned int i, off, load_order, condt;
unsigned int i, off, condt;
u32 k, b_off __maybe_unused;
u8 (*sk_load_func)(unsigned long *skb, int offset);

for (i = 0; i < prog->len; i++) {
u16 code;
Expand Down Expand Up @@ -748,71 +703,46 @@ static int build_body(struct jit_ctx *ctx)
break;
case BPF_LD | BPF_W | BPF_ABS:
/* A <- P[k:4] */
load_order = 2;
sk_load_func = CHOOSE_LOAD_FUNC(k, sk_load_word);
goto load;
case BPF_LD | BPF_H | BPF_ABS:
/* A <- P[k:2] */
load_order = 1;
sk_load_func = CHOOSE_LOAD_FUNC(k, sk_load_half);
goto load;
case BPF_LD | BPF_B | BPF_ABS:
/* A <- P[k:1] */
load_order = 0;
sk_load_func = CHOOSE_LOAD_FUNC(k, sk_load_byte);
load:
/* the interpreter will deal with the negative K */
if ((int)k < 0)
return -ENOTSUPP;

emit_load_imm(r_off, k, ctx);
load_common:
/*
* We may got here from the indirect loads so
* return if offset is negative.
*/
emit_slt(r_s0, r_off, r_zero, ctx);
emit_bcond(MIPS_COND_NE, r_s0, r_zero,
b_imm(prog->len, ctx), ctx);
emit_reg_move(r_ret, r_zero, ctx);

ctx->flags |= SEEN_CALL | SEEN_OFF |
SEEN_SKB | SEEN_A;
SEEN_SKB | SEEN_A | SEEN_SKB_DATA;

emit_load_func(r_s0, (ptr)load_func[load_order],
ctx);
emit_load_func(r_s0, (ptr)sk_load_func, ctx);
emit_reg_move(MIPS_R_A0, r_skb, ctx);
emit_jalr(MIPS_R_RA, r_s0, ctx);
/* Load second argument to delay slot */
emit_reg_move(MIPS_R_A1, r_off, ctx);
/* Check the error value */
if (config_enabled(CONFIG_64BIT)) {
/* Get error code from the top 32-bits */
emit_dsrl32(r_s0, r_val, 0, ctx);
/* Branch to 3 instructions ahead */
emit_bcond(MIPS_COND_NE, r_s0, r_zero, 3 << 2,
ctx);
} else {
/* Branch to 3 instructions ahead */
emit_bcond(MIPS_COND_NE, r_err, r_zero, 3 << 2,
ctx);
}
emit_nop(ctx);
/* We are good */
emit_b(b_imm(i + 1, ctx), ctx);
emit_jit_reg_move(r_A, r_val, ctx);
emit_bcond(MIPS_COND_EQ, r_ret, 0, b_imm(i + 1, ctx),
ctx);
/* Load return register on DS for failures */
emit_reg_move(r_ret, r_zero, ctx);
/* Return with error */
emit_b(b_imm(prog->len, ctx), ctx);
emit_reg_move(r_ret, r_zero, ctx);
emit_nop(ctx);
break;
case BPF_LD | BPF_W | BPF_IND:
/* A <- P[X + k:4] */
load_order = 2;
sk_load_func = sk_load_word;
goto load_ind;
case BPF_LD | BPF_H | BPF_IND:
/* A <- P[X + k:2] */
load_order = 1;
sk_load_func = sk_load_half;
goto load_ind;
case BPF_LD | BPF_B | BPF_IND:
/* A <- P[X + k:1] */
load_order = 0;
sk_load_func = sk_load_byte;
load_ind:
ctx->flags |= SEEN_OFF | SEEN_X;
emit_addiu(r_off, r_X, k, ctx);
Expand All @@ -834,14 +764,10 @@ static int build_body(struct jit_ctx *ctx)
emit_load(r_X, r_skb, off, ctx);
break;
case BPF_LDX | BPF_B | BPF_MSH:
/* the interpreter will deal with the negative K */
if ((int)k < 0)
return -ENOTSUPP;

/* X <- 4 * (P[k:1] & 0xf) */
ctx->flags |= SEEN_X | SEEN_CALL | SEEN_SKB;
/* Load offset to a1 */
emit_load_func(r_s0, (ptr)jit_get_skb_b, ctx);
emit_load_func(r_s0, (ptr)sk_load_byte, ctx);
/*
* This may emit two instructions so it may not fit
* in the delay slot. So use a0 in the delay slot.
Expand All @@ -850,25 +776,15 @@ static int build_body(struct jit_ctx *ctx)
emit_jalr(MIPS_R_RA, r_s0, ctx);
emit_reg_move(MIPS_R_A0, r_skb, ctx); /* delay slot */
/* Check the error value */
if (config_enabled(CONFIG_64BIT)) {
/* Top 32-bits of $v0 on 64-bit */
emit_dsrl32(r_s0, r_val, 0, ctx);
emit_bcond(MIPS_COND_NE, r_s0, r_zero,
3 << 2, ctx);
} else {
emit_bcond(MIPS_COND_NE, r_err, r_zero,
3 << 2, ctx);
}
/* No need for delay slot */
emit_bcond(MIPS_COND_NE, r_ret, 0,
b_imm(prog->len, ctx), ctx);
emit_reg_move(r_ret, r_zero, ctx);
/* We are good */
/* X <- P[1:K] & 0xf */
emit_andi(r_X, r_val, 0xf, ctx);
emit_andi(r_X, r_A, 0xf, ctx);
/* X << 2 */
emit_b(b_imm(i + 1, ctx), ctx);
emit_sll(r_X, r_X, 2, ctx); /* delay slot */
/* Return with error */
emit_b(b_imm(prog->len, ctx), ctx);
emit_load_imm(r_ret, 0, ctx); /* delay slot */
break;
case BPF_ST:
/* M[k] <- A */
Expand Down Expand Up @@ -942,7 +858,7 @@ static int build_body(struct jit_ctx *ctx)
/* Check if r_X is zero */
emit_bcond(MIPS_COND_EQ, r_X, r_zero,
b_imm(prog->len, ctx), ctx);
emit_load_imm(r_val, 0, ctx); /* delay slot */
emit_load_imm(r_ret, 0, ctx); /* delay slot */
emit_div(r_A, r_X, ctx);
break;
case BPF_ALU | BPF_MOD | BPF_X:
Expand All @@ -951,7 +867,7 @@ static int build_body(struct jit_ctx *ctx)
/* Check if r_X is zero */
emit_bcond(MIPS_COND_EQ, r_X, r_zero,
b_imm(prog->len, ctx), ctx);
emit_load_imm(r_val, 0, ctx); /* delay slot */
emit_load_imm(r_ret, 0, ctx); /* delay slot */
emit_mod(r_A, r_X, ctx);
break;
case BPF_ALU | BPF_OR | BPF_K:
Expand Down
33 changes: 18 additions & 15 deletions arch/mips/net/bpf_jit.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,6 @@
/* Registers used by JIT */
#define MIPS_R_ZERO 0
#define MIPS_R_V0 2
#define MIPS_R_V1 3
#define MIPS_R_A0 4
#define MIPS_R_A1 5
#define MIPS_R_T4 12
Expand Down Expand Up @@ -43,20 +42,6 @@
#define MIPS_COND_X (0x1 << 5)
#define MIPS_COND_K (0x1 << 6)

/* ABI specific return values */
#ifdef CONFIG_32BIT /* O32 */
#ifdef CONFIG_CPU_LITTLE_ENDIAN
#define r_err MIPS_R_V1
#define r_val MIPS_R_V0
#else /* CONFIG_CPU_LITTLE_ENDIAN */
#define r_err MIPS_R_V0
#define r_val MIPS_R_V1
#endif
#else /* N64 */
#define r_err MIPS_R_V0
#define r_val MIPS_R_V0
#endif

#define r_ret MIPS_R_V0

/*
Expand All @@ -65,11 +50,14 @@
* any of the $s0-$s6 registers will only be preserved if
* they are going to actually be used.
*/
#define r_skb_hl MIPS_R_S0 /* skb header length */
#define r_skb_data MIPS_R_S1 /* skb actual data */
#define r_off MIPS_R_S2
#define r_A MIPS_R_S3
#define r_X MIPS_R_S4
#define r_skb MIPS_R_S5
#define r_M MIPS_R_S6
#define r_skb_len MIPS_R_S7
#define r_s0 MIPS_R_T4 /* scratch reg 1 */
#define r_s1 MIPS_R_T5 /* scratch reg 2 */
#define r_tmp_imm MIPS_R_T6 /* No need to preserve this */
Expand All @@ -78,4 +66,19 @@
#define r_sp MIPS_R_SP
#define r_ra MIPS_R_RA

#ifndef __ASSEMBLY__

/* Declare ASM helpers */

#define DECLARE_LOAD_FUNC(func) \
extern u8 func(unsigned long *skb, int offset); \
extern u8 func##_negative(unsigned long *skb, int offset); \
extern u8 func##_positive(unsigned long *skb, int offset)

DECLARE_LOAD_FUNC(sk_load_word);
DECLARE_LOAD_FUNC(sk_load_half);
DECLARE_LOAD_FUNC(sk_load_byte);

#endif

#endif /* BPF_JIT_MIPS_OP_H */
Loading

0 comments on commit 266a88e

Please sign in to comment.