forge/linux-pinenote
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
linux-pinenote/arch/powerpc/net/bpf_jit_64.S

223 lines
5.9 KiB
ArmAsm
Raw Normal View History

net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
/* bpf_jit.S: Packet/header access helper functions
* for PPC64 BPF compiler.
*
* Copyright 2011 Matt Evans <matt@ozlabs.org>, IBM Corporation
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; version 2
* of the License.
*/
#include <asm/ppc_asm.h>
#include "bpf_jit.h"
/*
* All of these routines are called directly from generated code,
* whose register usage is:
*
* r3 skb
* r4,r5 A,X
* r6 *** address parameter to helper ***
* r7-r10 scratch
* r14 skb->data
* r15 skb headlen
* r16-31 M[]
*/
/*
* To consider: These helpers are so small it could be better to just
* generate them inline. Inline code can do the simple headlen check
* then branch directly to slow_path_XXX if required. (In fact, could
* load a spare GPR with the address of slow_path_generic and pass size
* as an argument, making the call site a mtlr, li and bllr.)
*/
.globl sk_load_word
sk_load_word:
cmpdi r_addr, 0
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
blt bpf_slow_path_word_neg
.globl sk_load_word_positive_offset
sk_load_word_positive_offset:
net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
/* Are we accessing past headlen? */
subi r_scratch1, r_HL, 4
cmpd r_scratch1, r_addr
blt bpf_slow_path_word
/* Nope, just hitting the header. cr0 here is eq or gt! */
lwzx r_A, r_D, r_addr
/* When big endian we don't need to byteswap. */
blr /* Return success, cr0 != LT */
.globl sk_load_half
sk_load_half:
cmpdi r_addr, 0
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
blt bpf_slow_path_half_neg
.globl sk_load_half_positive_offset
sk_load_half_positive_offset:
net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
subi r_scratch1, r_HL, 2
cmpd r_scratch1, r_addr
blt bpf_slow_path_half
lhzx r_A, r_D, r_addr
blr
.globl sk_load_byte
sk_load_byte:
cmpdi r_addr, 0
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
blt bpf_slow_path_byte_neg
.globl sk_load_byte_positive_offset
sk_load_byte_positive_offset:
net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
cmpd r_HL, r_addr
ble bpf_slow_path_byte
lbzx r_A, r_D, r_addr
blr
/*
* BPF_S_LDX_B_MSH: ldxb 4*([offset]&0xf)
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
* r_addr is the offset value
net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
*/
.globl sk_load_byte_msh
sk_load_byte_msh:
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
cmpdi r_addr, 0
blt bpf_slow_path_byte_msh_neg
.globl sk_load_byte_msh_positive_offset
sk_load_byte_msh_positive_offset:
net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
cmpd r_HL, r_addr
ble bpf_slow_path_byte_msh
lbzx r_X, r_D, r_addr
rlwinm r_X, r_X, 2, 32-4-2, 31-2
blr
/* Call out to skb_copy_bits:
* We'll need to back up our volatile regs first; we have
* local variable space at r1+(BPF_PPC_STACK_BASIC).
* Allocate a new stack frame here to remain ABI-compliant in
* stashing LR.
*/
#define bpf_slow_path_common(SIZE) \
mflr r0; \
std r0, 16(r1); \
/* R3 goes in parameter space of caller's frame */ \
std r_skb, (BPF_PPC_STACKFRAME+48)(r1); \
std r_A, (BPF_PPC_STACK_BASIC+(0*8))(r1); \
std r_X, (BPF_PPC_STACK_BASIC+(1*8))(r1); \
addi r5, r1, BPF_PPC_STACK_BASIC+(2*8); \
stdu r1, -BPF_PPC_SLOWPATH_FRAME(r1); \
/* R3 = r_skb, as passed */ \
mr r4, r_addr; \
li r6, SIZE; \
bl skb_copy_bits; \
powerpc: Fix BPF_JIT code to link with multiple TOCs If the kernel is big enough (eg. allyesconfig), the linker may need to switch TOCs when calling from the BPF JIT code out to the external helpers (skb_copy_bits() & bpf_internal_load_pointer_neg_helper()). In order to do that we need to leave space after the bl for the linker to insert a reload of our TOC pointer. Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Acked-by: Matt Evans <matt@ozlabs.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2012-06-21 17:50:27 +00:00
nop; \
net: filter: BPF 'JIT' compiler for PPC64 An implementation of a code generator for BPF programs to speed up packet filtering on PPC64, inspired by Eric Dumazet's x86-64 version. Filter code is generated as an ABI-compliant function in module_alloc()'d mem with stackframe & prologue/epilogue generated if required (simple filters don't need anything more than an li/blr). The filter's local variables, M[], live in registers. Supports all BPF opcodes, although "complicated" loads from negative packet offsets (e.g. SKF_LL_OFF) are not yet supported. There are a couple of further optimisations left for future work; many-pass assembly with branch-reach reduction and a register allocator to push M[] variables into volatile registers would improve the code quality further. This currently supports big-endian 64-bit PowerPC only (but is fairly simple to port to PPC32 or LE!). Enabled in the same way as x86-64: echo 1 > /proc/sys/net/core/bpf_jit_enable Or, enabled with extra debug output: echo 2 > /proc/sys/net/core/bpf_jit_enable Signed-off-by: Matt Evans <matt@ozlabs.org> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-07-20 15:51:00 +00:00
/* R3 = 0 on success */ \
addi r1, r1, BPF_PPC_SLOWPATH_FRAME; \
ld r0, 16(r1); \
ld r_A, (BPF_PPC_STACK_BASIC+(0*8))(r1); \
ld r_X, (BPF_PPC_STACK_BASIC+(1*8))(r1); \
mtlr r0; \
cmpdi r3, 0; \
blt bpf_error; /* cr0 = LT */ \
ld r_skb, (BPF_PPC_STACKFRAME+48)(r1); \
/* Great success! */
bpf_slow_path_word:
bpf_slow_path_common(4)
/* Data value is on stack, and cr0 != LT */
lwz r_A, BPF_PPC_STACK_BASIC+(2*8)(r1)
blr
bpf_slow_path_half:
bpf_slow_path_common(2)
lhz r_A, BPF_PPC_STACK_BASIC+(2*8)(r1)
blr
bpf_slow_path_byte:
bpf_slow_path_common(1)
lbz r_A, BPF_PPC_STACK_BASIC+(2*8)(r1)
blr
bpf_slow_path_byte_msh:
bpf_slow_path_common(1)
lbz r_X, BPF_PPC_STACK_BASIC+(2*8)(r1)
rlwinm r_X, r_X, 2, 32-4-2, 31-2
blr
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
/* Call out to bpf_internal_load_pointer_neg_helper:
* We'll need to back up our volatile regs first; we have
* local variable space at r1+(BPF_PPC_STACK_BASIC).
* Allocate a new stack frame here to remain ABI-compliant in
* stashing LR.
*/
#define sk_negative_common(SIZE) \
mflr r0; \
std r0, 16(r1); \
/* R3 goes in parameter space of caller's frame */ \
std r_skb, (BPF_PPC_STACKFRAME+48)(r1); \
std r_A, (BPF_PPC_STACK_BASIC+(0*8))(r1); \
std r_X, (BPF_PPC_STACK_BASIC+(1*8))(r1); \
stdu r1, -BPF_PPC_SLOWPATH_FRAME(r1); \
/* R3 = r_skb, as passed */ \
mr r4, r_addr; \
li r5, SIZE; \
bl bpf_internal_load_pointer_neg_helper; \
powerpc: Fix BPF_JIT code to link with multiple TOCs If the kernel is big enough (eg. allyesconfig), the linker may need to switch TOCs when calling from the BPF JIT code out to the external helpers (skb_copy_bits() & bpf_internal_load_pointer_neg_helper()). In order to do that we need to leave space after the bl for the linker to insert a reload of our TOC pointer. Signed-off-by: Michael Ellerman <michael@ellerman.id.au> Acked-by: Matt Evans <matt@ozlabs.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2012-06-21 17:50:27 +00:00
nop; \
bpf jit: Let the powerpc jit handle negative offsets Now the helper function from filter.c for negative offsets is exported, it can be used it in the jit to handle negative offsets. First modify the asm load helper functions to handle: - know positive offsets - know negative offsets - any offset then the compiler can be modified to explicitly use these helper when appropriate. This fixes the case of a negative X register and allows to lift the restriction that bpf programs with negative offsets can't be jited. Tested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Jan Seiffert <kaffeemonster@googlemail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-04-29 19:02:19 +00:00
/* R3 != 0 on success */ \
addi r1, r1, BPF_PPC_SLOWPATH_FRAME; \
ld r0, 16(r1); \
ld r_A, (BPF_PPC_STACK_BASIC+(0*8))(r1); \
ld r_X, (BPF_PPC_STACK_BASIC+(1*8))(r1); \
mtlr r0; \
cmpldi r3, 0; \
beq bpf_error_slow; /* cr0 = EQ */ \
mr r_addr, r3; \
ld r_skb, (BPF_PPC_STACKFRAME+48)(r1); \
/* Great success! */
bpf_slow_path_word_neg:
lis r_scratch1,-32 /* SKF_LL_OFF */
cmpd r_addr, r_scratch1 /* addr < SKF_* */
blt bpf_error /* cr0 = LT */
.globl sk_load_word_negative_offset
sk_load_word_negative_offset:
sk_negative_common(4)
lwz r_A, 0(r_addr)
blr
bpf_slow_path_half_neg:
lis r_scratch1,-32 /* SKF_LL_OFF */
cmpd r_addr, r_scratch1 /* addr < SKF_* */
blt bpf_error /* cr0 = LT */
.globl sk_load_half_negative_offset
sk_load_half_negative_offset:
sk_negative_common(2)
lhz r_A, 0(r_addr)
blr
bpf_slow_path_byte_neg:
lis r_scratch1,-32 /* SKF_LL_OFF */
cmpd r_addr, r_scratch1 /* addr < SKF_* */
blt bpf_error /* cr0 = LT */
.globl sk_load_byte_negative_offset
sk_load_byte_negative_offset:
sk_negative_common(1)
lbz r_A, 0(r_addr)
blr
bpf_slow_path_byte_msh_neg:
lis r_scratch1,-32 /* SKF_LL_OFF */
cmpd r_addr, r_scratch1 /* addr < SKF_* */
blt bpf_error /* cr0 = LT */
.globl sk_load_byte_msh_negative_offset
sk_load_byte_msh_negative_offset:
sk_negative_common(1)
lbz r_X, 0(r_addr)
rlwinm r_X, r_X, 2, 32-4-2, 31-2
blr
bpf_error_slow:
/* fabricate a cr0 = lt */
li r_scratch1, -1
cmpdi r_scratch1, 0
bpf_error:
/* Entered with cr0 = lt */
li r3, 0
/* Generated code will 'blt epilogue', returning 0. */
blr
Reference in a new issue Copy permalink