diff options
Diffstat (limited to 'arch/powerpc/net/bpf_jit_comp64.c')
-rw-r--r-- | arch/powerpc/net/bpf_jit_comp64.c | 591 |
1 files changed, 360 insertions, 231 deletions
diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c index 8571aafcc9e1..29ee306d6302 100644 --- a/arch/powerpc/net/bpf_jit_comp64.c +++ b/arch/powerpc/net/bpf_jit_comp64.c @@ -17,7 +17,60 @@ #include <linux/bpf.h> #include <asm/security_features.h> -#include "bpf_jit64.h" +#include "bpf_jit.h" + +/* + * Stack layout: + * Ensure the top half (upto local_tmp_var) stays consistent + * with our redzone usage. + * + * [ prev sp ] <------------- + * [ nv gpr save area ] 5*8 | + * [ tail_call_cnt ] 8 | + * [ local_tmp_var ] 16 | + * fp (r31) --> [ ebpf stack space ] upto 512 | + * [ frame header ] 32/112 | + * sp (r1) ---> [ stack pointer ] -------------- + */ + +/* for gpr non volatile registers BPG_REG_6 to 10 */ +#define BPF_PPC_STACK_SAVE (5*8) +/* for bpf JIT code internal usage */ +#define BPF_PPC_STACK_LOCALS 24 +/* stack frame excluding BPF stack, ensure this is quadword aligned */ +#define BPF_PPC_STACKFRAME (STACK_FRAME_MIN_SIZE + \ + BPF_PPC_STACK_LOCALS + BPF_PPC_STACK_SAVE) + +/* BPF register usage */ +#define TMP_REG_1 (MAX_BPF_JIT_REG + 0) +#define TMP_REG_2 (MAX_BPF_JIT_REG + 1) + +/* BPF to ppc register mappings */ +void bpf_jit_init_reg_mapping(struct codegen_context *ctx) +{ + /* function return value */ + ctx->b2p[BPF_REG_0] = _R8; + /* function arguments */ + ctx->b2p[BPF_REG_1] = _R3; + ctx->b2p[BPF_REG_2] = _R4; + ctx->b2p[BPF_REG_3] = _R5; + ctx->b2p[BPF_REG_4] = _R6; + ctx->b2p[BPF_REG_5] = _R7; + /* non volatile registers */ + ctx->b2p[BPF_REG_6] = _R27; + ctx->b2p[BPF_REG_7] = _R28; + ctx->b2p[BPF_REG_8] = _R29; + ctx->b2p[BPF_REG_9] = _R30; + /* frame pointer aka BPF_REG_10 */ + ctx->b2p[BPF_REG_FP] = _R31; + /* eBPF jit internal registers */ + ctx->b2p[BPF_REG_AX] = _R12; + ctx->b2p[TMP_REG_1] = _R9; + ctx->b2p[TMP_REG_2] = _R10; +} + +/* PPC NVR range -- update this if we ever use NVRs below r27 */ +#define BPF_PPC_NVR_MIN _R27 static inline bool bpf_has_stack_frame(struct codegen_context *ctx) { @@ -27,7 +80,7 @@ static inline bool bpf_has_stack_frame(struct codegen_context *ctx) * - the bpf program uses its stack area * The latter condition is deduced from the usage of BPF_REG_FP */ - return ctx->seen & SEEN_FUNC || bpf_is_seen_register(ctx, b2p[BPF_REG_FP]); + return ctx->seen & SEEN_FUNC || bpf_is_seen_register(ctx, bpf_to_ppc(BPF_REG_FP)); } /* @@ -73,22 +126,23 @@ void bpf_jit_build_prologue(u32 *image, struct codegen_context *ctx) { int i; + if (IS_ENABLED(CONFIG_PPC64_ELF_ABI_V2)) + EMIT(PPC_RAW_LD(_R2, _R13, offsetof(struct paca_struct, kernel_toc))); + /* * Initialize tail_call_cnt if we do tail calls. * Otherwise, put in NOPs so that it can be skipped when we are * invoked through a tail call. */ if (ctx->seen & SEEN_TAILCALL) { - EMIT(PPC_RAW_LI(b2p[TMP_REG_1], 0)); + EMIT(PPC_RAW_LI(bpf_to_ppc(TMP_REG_1), 0)); /* this goes in the redzone */ - PPC_BPF_STL(b2p[TMP_REG_1], 1, -(BPF_PPC_STACK_SAVE + 8)); + EMIT(PPC_RAW_STD(bpf_to_ppc(TMP_REG_1), _R1, -(BPF_PPC_STACK_SAVE + 8))); } else { EMIT(PPC_RAW_NOP()); EMIT(PPC_RAW_NOP()); } -#define BPF_TAILCALL_PROLOGUE_SIZE 8 - if (bpf_has_stack_frame(ctx)) { /* * We need a stack frame, but we don't necessarily need to @@ -96,10 +150,10 @@ void bpf_jit_build_prologue(u32 *image, struct codegen_context *ctx) */ if (ctx->seen & SEEN_FUNC) { EMIT(PPC_RAW_MFLR(_R0)); - PPC_BPF_STL(0, 1, PPC_LR_STKOFF); + EMIT(PPC_RAW_STD(_R0, _R1, PPC_LR_STKOFF)); } - PPC_BPF_STLU(1, 1, -(BPF_PPC_STACKFRAME + ctx->stack_size)); + EMIT(PPC_RAW_STDU(_R1, _R1, -(BPF_PPC_STACKFRAME + ctx->stack_size))); } /* @@ -108,12 +162,12 @@ void bpf_jit_build_prologue(u32 *image, struct codegen_context *ctx) * in the protected zone below the previous stack frame */ for (i = BPF_REG_6; i <= BPF_REG_10; i++) - if (bpf_is_seen_register(ctx, b2p[i])) - PPC_BPF_STL(b2p[i], 1, bpf_jit_stack_offsetof(ctx, b2p[i])); + if (bpf_is_seen_register(ctx, bpf_to_ppc(i))) + EMIT(PPC_RAW_STD(bpf_to_ppc(i), _R1, bpf_jit_stack_offsetof(ctx, bpf_to_ppc(i)))); /* Setup frame pointer to point to the bpf stack area */ - if (bpf_is_seen_register(ctx, b2p[BPF_REG_FP])) - EMIT(PPC_RAW_ADDI(b2p[BPF_REG_FP], 1, + if (bpf_is_seen_register(ctx, bpf_to_ppc(BPF_REG_FP))) + EMIT(PPC_RAW_ADDI(bpf_to_ppc(BPF_REG_FP), _R1, STACK_FRAME_MIN_SIZE + ctx->stack_size)); } @@ -123,15 +177,15 @@ static void bpf_jit_emit_common_epilogue(u32 *image, struct codegen_context *ctx /* Restore NVRs */ for (i = BPF_REG_6; i <= BPF_REG_10; i++) - if (bpf_is_seen_register(ctx, b2p[i])) - PPC_BPF_LL(b2p[i], 1, bpf_jit_stack_offsetof(ctx, b2p[i])); + if (bpf_is_seen_register(ctx, bpf_to_ppc(i))) + EMIT(PPC_RAW_LD(bpf_to_ppc(i), _R1, bpf_jit_stack_offsetof(ctx, bpf_to_ppc(i)))); /* Tear down our stack frame */ if (bpf_has_stack_frame(ctx)) { - EMIT(PPC_RAW_ADDI(1, 1, BPF_PPC_STACKFRAME + ctx->stack_size)); + EMIT(PPC_RAW_ADDI(_R1, _R1, BPF_PPC_STACKFRAME + ctx->stack_size)); if (ctx->seen & SEEN_FUNC) { - PPC_BPF_LL(0, 1, PPC_LR_STKOFF); - EMIT(PPC_RAW_MTLR(0)); + EMIT(PPC_RAW_LD(_R0, _R1, PPC_LR_STKOFF)); + EMIT(PPC_RAW_MTLR(_R0)); } } } @@ -141,42 +195,45 @@ void bpf_jit_build_epilogue(u32 *image, struct codegen_context *ctx) bpf_jit_emit_common_epilogue(image, ctx); /* Move result to r3 */ - EMIT(PPC_RAW_MR(3, b2p[BPF_REG_0])); + EMIT(PPC_RAW_MR(_R3, bpf_to_ppc(BPF_REG_0))); EMIT(PPC_RAW_BLR()); } -static void bpf_jit_emit_func_call_hlp(u32 *image, struct codegen_context *ctx, - u64 func) +static int bpf_jit_emit_func_call_hlp(u32 *image, struct codegen_context *ctx, u64 func) { -#ifdef PPC64_ELF_ABI_v1 - /* func points to the function descriptor */ - PPC_LI64(b2p[TMP_REG_2], func); - /* Load actual entry point from function descriptor */ - PPC_BPF_LL(b2p[TMP_REG_1], b2p[TMP_REG_2], 0); - /* ... and move it to CTR */ - EMIT(PPC_RAW_MTCTR(b2p[TMP_REG_1])); - /* - * Load TOC from function descriptor at offset 8. - * We can clobber r2 since we get called through a - * function pointer (so caller will save/restore r2) - * and since we don't use a TOC ourself. - */ - PPC_BPF_LL(2, b2p[TMP_REG_2], 8); -#else - /* We can clobber r12 */ - PPC_FUNC_ADDR(12, func); - EMIT(PPC_RAW_MTCTR(12)); -#endif + unsigned long func_addr = func ? ppc_function_entry((void *)func) : 0; + long reladdr; + + if (WARN_ON_ONCE(!core_kernel_text(func_addr))) + return -EINVAL; + + reladdr = func_addr - kernel_toc_addr(); + if (reladdr > 0x7FFFFFFF || reladdr < -(0x80000000L)) { + pr_err("eBPF: address of %ps out of range of kernel_toc.\n", (void *)func); + return -ERANGE; + } + + EMIT(PPC_RAW_ADDIS(_R12, _R2, PPC_HA(reladdr))); + EMIT(PPC_RAW_ADDI(_R12, _R12, PPC_LO(reladdr))); + EMIT(PPC_RAW_MTCTR(_R12)); EMIT(PPC_RAW_BCTRL()); + + return 0; } -void bpf_jit_emit_func_call_rel(u32 *image, struct codegen_context *ctx, u64 func) +int bpf_jit_emit_func_call_rel(u32 *image, struct codegen_context *ctx, u64 func) { unsigned int i, ctx_idx = ctx->idx; + if (WARN_ON_ONCE(func && is_module_text_address(func))) + return -EINVAL; + + /* skip past descriptor if elf v1 */ + func += FUNCTION_DESCR_SIZE; + /* Load function address into r12 */ - PPC_LI64(12, func); + PPC_LI64(_R12, func); /* For bpf-to-bpf function calls, the callee's address is unknown * until the last extra pass. As seen above, we use PPC_LI64() to @@ -191,20 +248,10 @@ void bpf_jit_emit_func_call_rel(u32 *image, struct codegen_context *ctx, u64 fun for (i = ctx->idx - ctx_idx; i < 5; i++) EMIT(PPC_RAW_NOP()); -#ifdef PPC64_ELF_ABI_v1 - /* - * Load TOC from function descriptor at offset 8. - * We can clobber r2 since we get called through a - * function pointer (so caller will save/restore r2) - * and since we don't use a TOC ourself. - */ - PPC_BPF_LL(2, 12, 8); - /* Load actual entry point from function descriptor */ - PPC_BPF_LL(12, 12, 0); -#endif - - EMIT(PPC_RAW_MTCTR(12)); + EMIT(PPC_RAW_MTCTR(_R12)); EMIT(PPC_RAW_BCTRL()); + + return 0; } static int bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32 out) @@ -215,54 +262,53 @@ static int bpf_jit_emit_tail_call(u32 *image, struct codegen_context *ctx, u32 o * r4/BPF_REG_2 - pointer to bpf_array * r5/BPF_REG_3 - index in bpf_array */ - int b2p_bpf_array = b2p[BPF_REG_2]; - int b2p_index = b2p[BPF_REG_3]; + int b2p_bpf_array = bpf_to_ppc(BPF_REG_2); + int b2p_index = bpf_to_ppc(BPF_REG_3); + int bpf_tailcall_prologue_size = 8; + + if (IS_ENABLED(CONFIG_PPC64_ELF_ABI_V2)) + bpf_tailcall_prologue_size += 4; /* skip past the toc load */ /* * if (index >= array->map.max_entries) * goto out; */ - EMIT(PPC_RAW_LWZ(b2p[TMP_REG_1], b2p_bpf_array, offsetof(struct bpf_array, map.max_entries))); + EMIT(PPC_RAW_LWZ(bpf_to_ppc(TMP_REG_1), b2p_bpf_array, offsetof(struct bpf_array, map.max_entries))); EMIT(PPC_RAW_RLWINM(b2p_index, b2p_index, 0, 0, 31)); - EMIT(PPC_RAW_CMPLW(b2p_index, b2p[TMP_REG_1])); - PPC_BCC(COND_GE, out); + EMIT(PPC_RAW_CMPLW(b2p_index, bpf_to_ppc(TMP_REG_1))); + PPC_BCC_SHORT(COND_GE, out); /* * if (tail_call_cnt >= MAX_TAIL_CALL_CNT) * goto out; */ - PPC_BPF_LL(b2p[TMP_REG_1], 1, bpf_jit_stack_tailcallcnt(ctx)); - EMIT(PPC_RAW_CMPLWI(b2p[TMP_REG_1], MAX_TAIL_CALL_CNT)); - PPC_BCC(COND_GE, out); + EMIT(PPC_RAW_LD(bpf_to_ppc(TMP_REG_1), _R1, bpf_jit_stack_tailcallcnt(ctx))); + EMIT(PPC_RAW_CMPLWI(bpf_to_ppc(TMP_REG_1), MAX_TAIL_CALL_CNT)); + PPC_BCC_SHORT(COND_GE, out); /* * tail_call_cnt++; */ - EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], b2p[TMP_REG_1], 1)); - PPC_BPF_STL(b2p[TMP_REG_1], 1, bpf_jit_stack_tailcallcnt(ctx)); + EMIT(PPC_RAW_ADDI(bpf_to_ppc(TMP_REG_1), bpf_to_ppc(TMP_REG_1), 1)); + EMIT(PPC_RAW_STD(bpf_to_ppc(TMP_REG_1), _R1, bpf_jit_stack_tailcallcnt(ctx))); /* prog = array->ptrs[index]; */ - EMIT(PPC_RAW_MULI(b2p[TMP_REG_1], b2p_index, 8)); - EMIT(PPC_RAW_ADD(b2p[TMP_REG_1], b2p[TMP_REG_1], b2p_bpf_array)); - PPC_BPF_LL(b2p[TMP_REG_1], b2p[TMP_REG_1], offsetof(struct bpf_array, ptrs)); + EMIT(PPC_RAW_MULI(bpf_to_ppc(TMP_REG_1), b2p_index, 8)); + EMIT(PPC_RAW_ADD(bpf_to_ppc(TMP_REG_1), bpf_to_ppc(TMP_REG_1), b2p_bpf_array)); + EMIT(PPC_RAW_LD(bpf_to_ppc(TMP_REG_1), bpf_to_ppc(TMP_REG_1), offsetof(struct bpf_array, ptrs))); /* * if (prog == NULL) * goto out; */ - EMIT(PPC_RAW_CMPLDI(b2p[TMP_REG_1], 0)); - PPC_BCC(COND_EQ, out); + EMIT(PPC_RAW_CMPLDI(bpf_to_ppc(TMP_REG_1), 0)); + PPC_BCC_SHORT(COND_EQ, out); /* goto *(prog->bpf_func + prologue_size); */ - PPC_BPF_LL(b2p[TMP_REG_1], b2p[TMP_REG_1], offsetof(struct bpf_prog, bpf_func)); -#ifdef PPC64_ELF_ABI_v1 - /* skip past the function descriptor */ - EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], b2p[TMP_REG_1], - FUNCTION_DESCR_SIZE + BPF_TAILCALL_PROLOGUE_SIZE)); -#else - EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], b2p[TMP_REG_1], BPF_TAILCALL_PROLOGUE_SIZE)); -#endif - EMIT(PPC_RAW_MTCTR(b2p[TMP_REG_1])); + EMIT(PPC_RAW_LD(bpf_to_ppc(TMP_REG_1), bpf_to_ppc(TMP_REG_1), offsetof(struct bpf_prog, bpf_func))); + EMIT(PPC_RAW_ADDI(bpf_to_ppc(TMP_REG_1), bpf_to_ppc(TMP_REG_1), + FUNCTION_DESCR_SIZE + bpf_tailcall_prologue_size)); + EMIT(PPC_RAW_MTCTR(bpf_to_ppc(TMP_REG_1))); /* tear down stack, restore NVRs, ... */ bpf_jit_emit_common_epilogue(image, ctx); @@ -297,7 +343,7 @@ asm ( /* Assemble the body code between the prologue & epilogue */ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context *ctx, - u32 *addrs, bool extra_pass) + u32 *addrs, int pass) { enum stf_barrier_type stf_barrier = stf_barrier_type_get(); const struct bpf_insn *insn = fp->insnsi; @@ -309,8 +355,12 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * for (i = 0; i < flen; i++) { u32 code = insn[i].code; - u32 dst_reg = b2p[insn[i].dst_reg]; - u32 src_reg = b2p[insn[i].src_reg]; + u32 dst_reg = bpf_to_ppc(insn[i].dst_reg); + u32 src_reg = bpf_to_ppc(insn[i].src_reg); + u32 size = BPF_SIZE(code); + u32 tmp1_reg = bpf_to_ppc(TMP_REG_1); + u32 tmp2_reg = bpf_to_ppc(TMP_REG_2); + u32 save_reg, ret_reg; s16 off = insn[i].off; s32 imm = insn[i].imm; bool func_addr_fixed; @@ -318,6 +368,7 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * u64 imm64; u32 true_cond; u32 tmp_idx; + int j; /* * addrs[] maps a BPF bytecode address into a real offset from @@ -360,8 +411,8 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * } else if (imm >= -32768 && imm < 32768) { EMIT(PPC_RAW_ADDI(dst_reg, dst_reg, IMM_L(imm))); } else { - PPC_LI32(b2p[TMP_REG_1], imm); - EMIT(PPC_RAW_ADD(dst_reg, dst_reg, b2p[TMP_REG_1])); + PPC_LI32(tmp1_reg, imm); + EMIT(PPC_RAW_ADD(dst_reg, dst_reg, tmp1_reg)); } goto bpf_alu32_trunc; case BPF_ALU | BPF_SUB | BPF_K: /* (u32) dst -= (u32) imm */ @@ -371,8 +422,8 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * } else if (imm > -32768 && imm <= 32768) { EMIT(PPC_RAW_ADDI(dst_reg, dst_reg, IMM_L(-imm))); } else { - PPC_LI32(b2p[TMP_REG_1], imm); - EMIT(PPC_RAW_SUB(dst_reg, dst_reg, b2p[TMP_REG_1])); + PPC_LI32(tmp1_reg, imm); + EMIT(PPC_RAW_SUB(dst_reg, dst_reg, tmp1_reg)); } goto bpf_alu32_trunc; case BPF_ALU | BPF_MUL | BPF_X: /* (u32) dst *= (u32) src */ @@ -387,32 +438,28 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * if (imm >= -32768 && imm < 32768) EMIT(PPC_RAW_MULI(dst_reg, dst_reg, IMM_L(imm))); else { - PPC_LI32(b2p[TMP_REG_1], imm); + PPC_LI32(tmp1_reg, imm); if (BPF_CLASS(code) == BPF_ALU) - EMIT(PPC_RAW_MULW(dst_reg, dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_MULW(dst_reg, dst_reg, tmp1_reg)); else - EMIT(PPC_RAW_MULD(dst_reg, dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_MULD(dst_reg, dst_reg, tmp1_reg)); } goto bpf_alu32_trunc; case BPF_ALU | BPF_DIV | BPF_X: /* (u32) dst /= (u32) src */ case BPF_ALU | BPF_MOD | BPF_X: /* (u32) dst %= (u32) src */ if (BPF_OP(code) == BPF_MOD) { - EMIT(PPC_RAW_DIVWU(b2p[TMP_REG_1], dst_reg, src_reg)); - EMIT(PPC_RAW_MULW(b2p[TMP_REG_1], src_reg, - b2p[TMP_REG_1])); - EMIT(PPC_RAW_SUB(dst_reg, dst_reg, b2p[TMP_REG_1])); + EMIT(PPC_RAW_DIVWU(tmp1_reg, dst_reg, src_reg)); + EMIT(PPC_RAW_MULW(tmp1_reg, src_reg, tmp1_reg)); + EMIT(PPC_RAW_SUB(dst_reg, dst_reg, tmp1_reg)); } else EMIT(PPC_RAW_DIVWU(dst_reg, dst_reg, src_reg)); goto bpf_alu32_trunc; case BPF_ALU64 | BPF_DIV | BPF_X: /* dst /= src */ case BPF_ALU64 | BPF_MOD | BPF_X: /* dst %= src */ if (BPF_OP(code) == BPF_MOD) { - EMIT(PPC_RAW_DIVDU(b2p[TMP_REG_1], dst_reg, src_reg)); - EMIT(PPC_RAW_MULD(b2p[TMP_REG_1], src_reg, - b2p[TMP_REG_1])); - EMIT(PPC_RAW_SUB(dst_reg, dst_reg, b2p[TMP_REG_1])); + EMIT(PPC_RAW_DIVDU(tmp1_reg, dst_reg, src_reg)); + EMIT(PPC_RAW_MULD(tmp1_reg, src_reg, tmp1_reg)); + EMIT(PPC_RAW_SUB(dst_reg, dst_reg, tmp1_reg)); } else EMIT(PPC_RAW_DIVDU(dst_reg, dst_reg, src_reg)); break; @@ -431,35 +478,23 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * } } - PPC_LI32(b2p[TMP_REG_1], imm); + PPC_LI32(tmp1_reg, imm); switch (BPF_CLASS(code)) { case BPF_ALU: if (BPF_OP(code) == BPF_MOD) { - EMIT(PPC_RAW_DIVWU(b2p[TMP_REG_2], - dst_reg, - b2p[TMP_REG_1])); - EMIT(PPC_RAW_MULW(b2p[TMP_REG_1], - b2p[TMP_REG_1], - b2p[TMP_REG_2])); - EMIT(PPC_RAW_SUB(dst_reg, dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_DIVWU(tmp2_reg, dst_reg, tmp1_reg)); + EMIT(PPC_RAW_MULW(tmp1_reg, tmp1_reg, tmp2_reg)); + EMIT(PPC_RAW_SUB(dst_reg, dst_reg, tmp1_reg)); } else - EMIT(PPC_RAW_DIVWU(dst_reg, dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_DIVWU(dst_reg, dst_reg, tmp1_reg)); break; case BPF_ALU64: if (BPF_OP(code) == BPF_MOD) { - EMIT(PPC_RAW_DIVDU(b2p[TMP_REG_2], - dst_reg, - b2p[TMP_REG_1])); - EMIT(PPC_RAW_MULD(b2p[TMP_REG_1], - b2p[TMP_REG_1], - b2p[TMP_REG_2])); - EMIT(PPC_RAW_SUB(dst_reg, dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_DIVDU(tmp2_reg, dst_reg, tmp1_reg)); + EMIT(PPC_RAW_MULD(tmp1_reg, tmp1_reg, tmp2_reg)); + EMIT(PPC_RAW_SUB(dst_reg, dst_reg, tmp1_reg)); } else - EMIT(PPC_RAW_DIVDU(dst_reg, dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_DIVDU(dst_reg, dst_reg, tmp1_reg)); break; } goto bpf_alu32_trunc; @@ -481,8 +516,8 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * EMIT(PPC_RAW_ANDI(dst_reg, dst_reg, IMM_L(imm))); else { /* Sign-extended */ - PPC_LI32(b2p[TMP_REG_1], imm); - EMIT(PPC_RAW_AND(dst_reg, dst_reg, b2p[TMP_REG_1])); + PPC_LI32(tmp1_reg, imm); + EMIT(PPC_RAW_AND(dst_reg, dst_reg, tmp1_reg)); } goto bpf_alu32_trunc; case BPF_ALU | BPF_OR | BPF_X: /* dst = (u32) dst | (u32) src */ @@ -493,8 +528,8 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * case BPF_ALU64 | BPF_OR | BPF_K:/* dst = dst | imm */ if (imm < 0 && BPF_CLASS(code) == BPF_ALU64) { /* Sign-extended */ - PPC_LI32(b2p[TMP_REG_1], imm); - EMIT(PPC_RAW_OR(dst_reg, dst_reg, b2p[TMP_REG_1])); + PPC_LI32(tmp1_reg, imm); + EMIT(PPC_RAW_OR(dst_reg, dst_reg, tmp1_reg)); } else { if (IMM_L(imm)) EMIT(PPC_RAW_ORI(dst_reg, dst_reg, IMM_L(imm))); @@ -510,8 +545,8 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * case BPF_ALU64 | BPF_XOR | BPF_K: /* dst ^= imm */ if (imm < 0 && BPF_CLASS(code) == BPF_ALU64) { /* Sign-extended */ - PPC_LI32(b2p[TMP_REG_1], imm); - EMIT(PPC_RAW_XOR(dst_reg, dst_reg, b2p[TMP_REG_1])); + PPC_LI32(tmp1_reg, imm); + EMIT(PPC_RAW_XOR(dst_reg, dst_reg, tmp1_reg)); } else { if (IMM_L(imm)) EMIT(PPC_RAW_XORI(dst_reg, dst_reg, IMM_L(imm))); @@ -612,11 +647,11 @@ bpf_alu32_trunc: switch (imm) { case 16: /* Rotate 8 bits left & mask with 0x0000ff00 */ - EMIT(PPC_RAW_RLWINM(b2p[TMP_REG_1], dst_reg, 8, 16, 23)); + EMIT(PPC_RAW_RLWINM(tmp1_reg, dst_reg, 8, 16, 23)); /* Rotate 8 bits right & insert LSB to reg */ - EMIT(PPC_RAW_RLWIMI(b2p[TMP_REG_1], dst_reg, 24, 24, 31)); + EMIT(PPC_RAW_RLWIMI(tmp1_reg, dst_reg, 24, 24, 31)); /* Move result back to dst_reg */ - EMIT(PPC_RAW_MR(dst_reg, b2p[TMP_REG_1])); + EMIT(PPC_RAW_MR(dst_reg, tmp1_reg)); break; case 32: /* @@ -624,25 +659,29 @@ bpf_alu32_trunc: * 2 bytes are already in their final position * -- byte 2 and 4 (of bytes 1, 2, 3 and 4) */ - EMIT(PPC_RAW_RLWINM(b2p[TMP_REG_1], dst_reg, 8, 0, 31)); + EMIT(PPC_RAW_RLWINM(tmp1_reg, dst_reg, 8, 0, 31)); /* Rotate 24 bits and insert byte 1 */ - EMIT(PPC_RAW_RLWIMI(b2p[TMP_REG_1], dst_reg, 24, 0, 7)); + EMIT(PPC_RAW_RLWIMI(tmp1_reg, dst_reg, 24, 0, 7)); /* Rotate 24 bits and insert byte 3 */ - EMIT(PPC_RAW_RLWIMI(b2p[TMP_REG_1], dst_reg, 24, 16, 23)); - EMIT(PPC_RAW_MR(dst_reg, b2p[TMP_REG_1])); + EMIT(PPC_RAW_RLWIMI(tmp1_reg, dst_reg, 24, 16, 23)); + EMIT(PPC_RAW_MR(dst_reg, tmp1_reg)); break; case 64: - /* - * Way easier and faster(?) to store the value - * into stack and then use ldbrx - * - * ctx->seen will be reliable in pass2, but - * the instructions generated will remain the - * same across all passes - */ - PPC_BPF_STL(dst_reg, 1, bpf_jit_stack_local(ctx)); - EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], 1, bpf_jit_stack_local(ctx))); - EMIT(PPC_RAW_LDBRX(dst_reg, 0, b2p[TMP_REG_1])); + /* Store the value to stack and then use byte-reverse loads */ + EMIT(PPC_RAW_STD(dst_reg, _R1, bpf_jit_stack_local(ctx))); + EMIT(PPC_RAW_ADDI(tmp1_reg, _R1, bpf_jit_stack_local(ctx))); + if (cpu_has_feature(CPU_FTR_ARCH_206)) { + EMIT(PPC_RAW_LDBRX(dst_reg, 0, tmp1_reg)); + } else { + EMIT(PPC_RAW_LWBRX(dst_reg, 0, tmp1_reg)); + if (IS_ENABLED(CONFIG_CPU_LITTLE_ENDIAN)) + EMIT(PPC_RAW_SLDI(dst_reg, dst_reg, 32)); + EMIT(PPC_RAW_LI(tmp2_reg, 4)); + EMIT(PPC_RAW_LWBRX(tmp2_reg, tmp2_reg, tmp1_reg)); + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN)) + EMIT(PPC_RAW_SLDI(tmp2_reg, tmp2_reg, 32)); + EMIT(PPC_RAW_OR(dst_reg, dst_reg, tmp2_reg)); + } break; } break; @@ -680,15 +719,14 @@ emit_clear: break; case STF_BARRIER_SYNC_ORI: EMIT(PPC_RAW_SYNC()); - EMIT(PPC_RAW_LD(b2p[TMP_REG_1], _R13, 0)); + EMIT(PPC_RAW_LD(tmp1_reg, _R13, 0)); EMIT(PPC_RAW_ORI(_R31, _R31, 0)); break; case STF_BARRIER_FALLBACK: - EMIT(PPC_RAW_MFLR(b2p[TMP_REG_1])); - PPC_LI64(12, dereference_kernel_function_descriptor(bpf_stf_barrier)); - EMIT(PPC_RAW_MTCTR(12)); + ctx->seen |= SEEN_FUNC; + PPC_LI64(_R12, dereference_kernel_function_descriptor(bpf_stf_barrier)); + EMIT(PPC_RAW_MTCTR(_R12)); EMIT(PPC_RAW_BCTRL()); - EMIT(PPC_RAW_MTLR(b2p[TMP_REG_1])); break; case STF_BARRIER_NONE: break; @@ -701,76 +739,122 @@ emit_clear: case BPF_STX | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = src */ case BPF_ST | BPF_MEM | BPF_B: /* *(u8 *)(dst + off) = imm */ if (BPF_CLASS(code) == BPF_ST) { - EMIT(PPC_RAW_LI(b2p[TMP_REG_1], imm)); - src_reg = b2p[TMP_REG_1]; + EMIT(PPC_RAW_LI(tmp1_reg, imm)); + src_reg = tmp1_reg; } EMIT(PPC_RAW_STB(src_reg, dst_reg, off)); break; case BPF_STX | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = src */ case BPF_ST | BPF_MEM | BPF_H: /* (u16 *)(dst + off) = imm */ if (BPF_CLASS(code) == BPF_ST) { - EMIT(PPC_RAW_LI(b2p[TMP_REG_1], imm)); - src_reg = b2p[TMP_REG_1]; + EMIT(PPC_RAW_LI(tmp1_reg, imm)); + src_reg = tmp1_reg; } EMIT(PPC_RAW_STH(src_reg, dst_reg, off)); break; case BPF_STX | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = src */ case BPF_ST | BPF_MEM | BPF_W: /* *(u32 *)(dst + off) = imm */ if (BPF_CLASS(code) == BPF_ST) { - PPC_LI32(b2p[TMP_REG_1], imm); - src_reg = b2p[TMP_REG_1]; + PPC_LI32(tmp1_reg, imm); + src_reg = tmp1_reg; } EMIT(PPC_RAW_STW(src_reg, dst_reg, off)); break; case BPF_STX | BPF_MEM | BPF_DW: /* (u64 *)(dst + off) = src */ case BPF_ST | BPF_MEM | BPF_DW: /* *(u64 *)(dst + off) = imm */ if (BPF_CLASS(code) == BPF_ST) { - PPC_LI32(b2p[TMP_REG_1], imm); - src_reg = b2p[TMP_REG_1]; + PPC_LI32(tmp1_reg, imm); + src_reg = tmp1_reg; + } + if (off % 4) { + EMIT(PPC_RAW_LI(tmp2_reg, off)); + EMIT(PPC_RAW_STDX(src_reg, dst_reg, tmp2_reg)); + } else { + EMIT(PPC_RAW_STD(src_reg, dst_reg, off)); } - PPC_BPF_STL(src_reg, dst_reg, off); break; /* * BPF_STX ATOMIC (atomic ops) */ case BPF_STX | BPF_ATOMIC | BPF_W: - if (imm != BPF_ADD) { - pr_err_ratelimited( - "eBPF filter atomic op code %02x (@%d) unsupported\n", - code, i); - return -ENOTSUPP; - } - - /* *(u32 *)(dst + off) += src */ + case BPF_STX | BPF_ATOMIC | BPF_DW: + save_reg = tmp2_reg; + ret_reg = src_reg; - /* Get EA into TMP_REG_1 */ - EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], dst_reg, off)); + /* Get offset into TMP_REG_1 */ + EMIT(PPC_RAW_LI(tmp1_reg, off)); tmp_idx = ctx->idx * 4; /* load value from memory into TMP_REG_2 */ - EMIT(PPC_RAW_LWARX(b2p[TMP_REG_2], 0, b2p[TMP_REG_1], 0)); - /* add value from src_reg into this */ - EMIT(PPC_RAW_ADD(b2p[TMP_REG_2], b2p[TMP_REG_2], src_reg)); - /* store result back */ - EMIT(PPC_RAW_STWCX(b2p[TMP_REG_2], 0, b2p[TMP_REG_1])); - /* we're done if this succeeded */ - PPC_BCC_SHORT(COND_NE, tmp_idx); - break; - case BPF_STX | BPF_ATOMIC | BPF_DW: - if (imm != BPF_ADD) { + if (size == BPF_DW) + EMIT(PPC_RAW_LDARX(tmp2_reg, tmp1_reg, dst_reg, 0)); + else + EMIT(PPC_RAW_LWARX(tmp2_reg, tmp1_reg, dst_reg, 0)); + + /* Save old value in _R0 */ + if (imm & BPF_FETCH) + EMIT(PPC_RAW_MR(_R0, tmp2_reg)); + + switch (imm) { + case BPF_ADD: + case BPF_ADD | BPF_FETCH: + EMIT(PPC_RAW_ADD(tmp2_reg, tmp2_reg, src_reg)); + break; + case BPF_AND: + case BPF_AND | BPF_FETCH: + EMIT(PPC_RAW_AND(tmp2_reg, tmp2_reg, src_reg)); + break; + case BPF_OR: + case BPF_OR | BPF_FETCH: + EMIT(PPC_RAW_OR(tmp2_reg, tmp2_reg, src_reg)); + break; + case BPF_XOR: + case BPF_XOR | BPF_FETCH: + EMIT(PPC_RAW_XOR(tmp2_reg, tmp2_reg, src_reg)); + break; + case BPF_CMPXCHG: + /* + * Return old value in BPF_REG_0 for BPF_CMPXCHG & + * in src_reg for other cases. + */ + ret_reg = bpf_to_ppc(BPF_REG_0); + + /* Compare with old value in BPF_R0 */ + if (size == BPF_DW) + EMIT(PPC_RAW_CMPD(bpf_to_ppc(BPF_REG_0), tmp2_reg)); + else + EMIT(PPC_RAW_CMPW(bpf_to_ppc(BPF_REG_0), tmp2_reg)); + /* Don't set if different from old value */ + PPC_BCC_SHORT(COND_NE, (ctx->idx + 3) * 4); + fallthrough; + case BPF_XCHG: + save_reg = src_reg; + break; + default: pr_err_ratelimited( "eBPF filter atomic op code %02x (@%d) unsupported\n", code, i); - return -ENOTSUPP; + return -EOPNOTSUPP; } - /* *(u64 *)(dst + off) += src */ - EMIT(PPC_RAW_ADDI(b2p[TMP_REG_1], dst_reg, off)); - tmp_idx = ctx->idx * 4; - EMIT(PPC_RAW_LDARX(b2p[TMP_REG_2], 0, b2p[TMP_REG_1], 0)); - EMIT(PPC_RAW_ADD(b2p[TMP_REG_2], b2p[TMP_REG_2], src_reg)); - EMIT(PPC_RAW_STDCX(b2p[TMP_REG_2], 0, b2p[TMP_REG_1])); + /* store new value */ + if (size == BPF_DW) + EMIT(PPC_RAW_STDCX(save_reg, tmp1_reg, dst_reg)); + else + EMIT(PPC_RAW_STWCX(save_reg, tmp1_reg, dst_reg)); + /* we're done if this succeeded */ PPC_BCC_SHORT(COND_NE, tmp_idx); + + if (imm & BPF_FETCH) { + EMIT(PPC_RAW_MR(ret_reg, _R0)); + /* + * Skip unnecessary zero-extension for 32-bit cmpxchg. + * For context, see commit 39491867ace5. + */ + if (size != BPF_DW && imm == BPF_CMPXCHG && + insn_is_zext(&insn[i + 1])) + addrs[++i] = ctx->idx * 4; + } break; /* @@ -778,25 +862,70 @@ emit_clear: */ /* dst = *(u8 *)(ul) (src + off) */ case BPF_LDX | BPF_MEM | BPF_B: - EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off)); - if (insn_is_zext(&insn[i + 1])) - addrs[++i] = ctx->idx * 4; - break; + case BPF_LDX | BPF_PROBE_MEM | BPF_B: /* dst = *(u16 *)(ul) (src + off) */ case BPF_LDX | BPF_MEM | BPF_H: - EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off)); - if (insn_is_zext(&insn[i + 1])) - addrs[++i] = ctx->idx * 4; - break; + case BPF_LDX | BPF_PROBE_MEM | BPF_H: /* dst = *(u32 *)(ul) (src + off) */ case BPF_LDX | BPF_MEM | BPF_W: - EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off)); - if (insn_is_zext(&insn[i + 1])) - addrs[++i] = ctx->idx * 4; - break; + case BPF_LDX | BPF_PROBE_MEM | BPF_W: /* dst = *(u64 *)(ul) (src + off) */ case BPF_LDX | BPF_MEM | BPF_DW: - PPC_BPF_LL(dst_reg, src_reg, off); + case BPF_LDX | BPF_PROBE_MEM | BPF_DW: + /* + * As PTR_TO_BTF_ID that uses BPF_PROBE_MEM mode could either be a valid + * kernel pointer or NULL but not a userspace address, execute BPF_PROBE_MEM + * load only if addr is kernel address (see is_kernel_addr()), otherwise + * set dst_reg=0 and move on. + */ + if (BPF_MODE(code) == BPF_PROBE_MEM) { + EMIT(PPC_RAW_ADDI(tmp1_reg, src_reg, off)); + if (IS_ENABLED(CONFIG_PPC_BOOK3E_64)) + PPC_LI64(tmp2_reg, 0x8000000000000000ul); + else /* BOOK3S_64 */ + PPC_LI64(tmp2_reg, PAGE_OFFSET); + EMIT(PPC_RAW_CMPLD(tmp1_reg, tmp2_reg)); + PPC_BCC_SHORT(COND_GT, (ctx->idx + 3) * 4); + EMIT(PPC_RAW_LI(dst_reg, 0)); + /* + * Check if 'off' is word aligned for BPF_DW, because + * we might generate two instructions. + */ + if (BPF_SIZE(code) == BPF_DW && (off & 3)) + PPC_JMP((ctx->idx + 3) * 4); + else + PPC_JMP((ctx->idx + 2) * 4); + } + + switch (size) { + case BPF_B: + EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off)); + break; + case BPF_H: + EMIT(PPC_RAW_LHZ(dst_reg, src_reg, off)); + break; + case BPF_W: + EMIT(PPC_RAW_LWZ(dst_reg, src_reg, off)); + break; + case BPF_DW: + if (off % 4) { + EMIT(PPC_RAW_LI(tmp1_reg, off)); + EMIT(PPC_RAW_LDX(dst_reg, src_reg, tmp1_reg)); + } else { + EMIT(PPC_RAW_LD(dst_reg, src_reg, off)); + } + break; + } + + if (size != BPF_DW && insn_is_zext(&insn[i + 1])) + addrs[++i] = ctx->idx * 4; + + if (BPF_MODE(code) == BPF_PROBE_MEM) { + ret = bpf_add_extable_entry(fp, image, pass, ctx, ctx->idx - 1, + 4, dst_reg); + if (ret) + return ret; + } break; /* @@ -806,9 +935,13 @@ emit_clear: case BPF_LD | BPF_IMM | BPF_DW: /* dst = (u64) imm */ imm64 = ((u64)(u32) insn[i].imm) | (((u64)(u32) insn[i+1].imm) << 32); + tmp_idx = ctx->idx; + PPC_LI64(dst_reg, imm64); + /* padding to allow full 5 instructions for later patching */ + for (j = ctx->idx - tmp_idx; j < 5; j++) + EMIT(PPC_RAW_NOP()); /* Adjust for two bpf instructions */ addrs[++i] = ctx->idx * 4; - PPC_LI64(dst_reg, imm64); break; /* @@ -820,8 +953,11 @@ emit_clear: * the epilogue. If we _are_ the last instruction, * we'll just fall through to the epilogue. */ - if (i != flen - 1) - PPC_JMP(exit_addr); + if (i != flen - 1) { + ret = bpf_jit_emit_exit_insn(image, ctx, tmp1_reg, exit_addr); + if (ret) + return ret; + } /* else fall through to the epilogue */ break; @@ -831,17 +967,21 @@ emit_clear: case BPF_JMP | BPF_CALL: ctx->seen |= SEEN_FUNC; - ret = bpf_jit_get_func_addr(fp, &insn[i], extra_pass, + ret = bpf_jit_get_func_addr(fp, &insn[i], false, &func_addr, &func_addr_fixed); if (ret < 0) return ret; if (func_addr_fixed) - bpf_jit_emit_func_call_hlp(image, ctx, func_addr); + ret = bpf_jit_emit_func_call_hlp(image, ctx, func_addr); else - bpf_jit_emit_func_call_rel(image, ctx, func_addr); + ret = bpf_jit_emit_func_call_rel(image, ctx, func_addr); + + if (ret) + return ret; + /* move return value from r3 to BPF_REG_0 */ - EMIT(PPC_RAW_MR(b2p[BPF_REG_0], 3)); + EMIT(PPC_RAW_MR(bpf_to_ppc(BPF_REG_0), _R3)); break; /* @@ -947,14 +1087,10 @@ cond_branch: case BPF_JMP | BPF_JSET | BPF_X: case BPF_JMP32 | BPF_JSET | BPF_X: if (BPF_CLASS(code) == BPF_JMP) { - EMIT(PPC_RAW_AND_DOT(b2p[TMP_REG_1], dst_reg, - src_reg)); + EMIT(PPC_RAW_AND_DOT(tmp1_reg, dst_reg, src_reg)); } else { - int tmp_reg = b2p[TMP_REG_1]; - - EMIT(PPC_RAW_AND(tmp_reg, dst_reg, src_reg)); - EMIT(PPC_RAW_RLWINM_DOT(tmp_reg, tmp_reg, 0, 0, - 31)); + EMIT(PPC_RAW_AND(tmp1_reg, dst_reg, src_reg)); + EMIT(PPC_RAW_RLWINM_DOT(tmp1_reg, tmp1_reg, 0, 0, 31)); } break; case BPF_JMP | BPF_JNE | BPF_K: @@ -983,14 +1119,12 @@ cond_branch: EMIT(PPC_RAW_CMPLDI(dst_reg, imm)); } else { /* sign-extending load */ - PPC_LI32(b2p[TMP_REG_1], imm); + PPC_LI32(tmp1_reg, imm); /* ... but unsigned comparison */ if (is_jmp32) - EMIT(PPC_RAW_CMPLW(dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_CMPLW(dst_reg, tmp1_reg)); else - EMIT(PPC_RAW_CMPLD(dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_CMPLD(dst_reg, tmp1_reg)); } break; } @@ -1015,13 +1149,11 @@ cond_branch: else EMIT(PPC_RAW_CMPDI(dst_reg, imm)); } else { - PPC_LI32(b2p[TMP_REG_1], imm); + PPC_LI32(tmp1_reg, imm); if (is_jmp32) - EMIT(PPC_RAW_CMPW(dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_CMPW(dst_reg, tmp1_reg)); else - EMIT(PPC_RAW_CMPD(dst_reg, - b2p[TMP_REG_1])); + EMIT(PPC_RAW_CMPD(dst_reg, tmp1_reg)); } break; } @@ -1030,19 +1162,16 @@ cond_branch: /* andi does not sign-extend the immediate */ if (imm >= 0 && imm < 32768) /* PPC_ANDI is _only/always_ dot-form */ - EMIT(PPC_RAW_ANDI(b2p[TMP_REG_1], dst_reg, imm)); + EMIT(PPC_RAW_ANDI(tmp1_reg, dst_reg, imm)); else { - int tmp_reg = b2p[TMP_REG_1]; - - PPC_LI32(tmp_reg, imm); + PPC_LI32(tmp1_reg, imm); if (BPF_CLASS(code) == BPF_JMP) { - EMIT(PPC_RAW_AND_DOT(tmp_reg, dst_reg, - tmp_reg)); + EMIT(PPC_RAW_AND_DOT(tmp1_reg, dst_reg, + tmp1_reg)); } else { - EMIT(PPC_RAW_AND(tmp_reg, dst_reg, - tmp_reg)); - EMIT(PPC_RAW_RLWINM_DOT(tmp_reg, tmp_reg, - 0, 0, 31)); + EMIT(PPC_RAW_AND(tmp1_reg, dst_reg, tmp1_reg)); + EMIT(PPC_RAW_RLWINM_DOT(tmp1_reg, tmp1_reg, + 0, 0, 31)); } } break; |