| Message ID | 20210917153047.177141-9-hbathini@linux.ibm.com (mailing list archive) |
|---|---|
| State | Changes Requested |
| Headers | show |
| Series | bpf powerpc: Add BPF_PROBE_MEM support in powerpc JIT compiler | expand |
| Context | Check | Description |
|---|---|---|
| snowpatch_ozlabs/github-powerpc_ppctests | success | Successfully ran 8 jobs. |
| snowpatch_ozlabs/github-powerpc_selftests | success | Successfully ran 8 jobs. |
| snowpatch_ozlabs/github-powerpc_clang | success | Successfully ran 7 jobs. |
| snowpatch_ozlabs/github-powerpc_sparse | warning | Found 4 issues from 2 of 4 jobs. |
| snowpatch_ozlabs/github-powerpc_kernel_qemu | success | Successfully ran 24 jobs. |
Le 17/09/2021 à 17:30, Hari Bathini a écrit : > With KUAP enabled, any kernel code which wants to access userspace > needs to be surrounded by disable-enable KUAP. But that is not > happening for BPF_PROBE_MEM load instruction. Though PPC32 does not > support read protection, considering the fact that PTR_TO_BTF_ID > (which uses BPF_PROBE_MEM mode) could either be a valid kernel pointer > or NULL but should never be a pointer to userspace address, execute > BPF_PROBE_MEM load only if addr > TASK_SIZE_MAX, otherwise set > dst_reg=0 and move on. Same comment as patch 6. > > This will catch NULL, valid or invalid userspace pointers. Only bad > kernel pointer will be handled by BPF exception table. > > [Alexei suggested for x86] > Suggested-by: Alexei Starovoitov <ast@kernel.org> > Signed-off-by: Hari Bathini <hbathini@linux.ibm.com> > --- > > Changes in v2: > * New patch to handle bad userspace pointers on PPC32. > > > arch/powerpc/net/bpf_jit_comp32.c | 39 +++++++++++++++++++++++++++++++ > 1 file changed, 39 insertions(+) > > diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c > index c6262289dcc4..faa8047fcf4a 100644 > --- a/arch/powerpc/net/bpf_jit_comp32.c > +++ b/arch/powerpc/net/bpf_jit_comp32.c > @@ -821,6 +821,45 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * > /* dst = *(u64 *)(ul) (src + off) */ > case BPF_LDX | BPF_MEM | BPF_DW: > 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 > TASK_SIZE_MAX, otherwise set dst_reg=0 and move on. > + */ > + if (BPF_MODE(code) == BPF_PROBE_MEM) { > + bool extra_insn_needed = false; > + unsigned int adjusted_idx; > + > + /* > + * For BPF_DW case, "li reg_h,0" would be needed when > + * !fp->aux->verifier_zext. Adjust conditional branch'ing > + * address accordingly. > + */ > + if ((size == BPF_DW) && !fp->aux->verifier_zext) > + extra_insn_needed = true; Don't make it too complicated. That's a fallback that should never happen, no need to optimise. You can put that instruction all the time (or put a NOP) and keep the jumps always the same. > + > + /* > + * Need to jump two instructions instead of one for BPF_DW case > + * as there are two load instructions for dst_reg_h & dst_reg > + * respectively. > + */ > + adjusted_idx = (size == BPF_DW) ? 1 : 0; Same comment as patch 6, drop adjusted_idx and do an if/else directly for the PPC_JMP. > + > + EMIT(PPC_RAW_ADDI(b2p[TMP_REG], src_reg, off)); > + PPC_LI32(_R0, TASK_SIZE_MAX); > + EMIT(PPC_RAW_CMPLW(b2p[TMP_REG], _R0)); > + PPC_BCC(COND_GT, (ctx->idx + 4 + (extra_insn_needed ? 1 : 0)) * 4); > + EMIT(PPC_RAW_LI(dst_reg, 0)); > + /* > + * Note that "li reg_h,0" is emitted for BPF_B/H/W case, > + * if necessary. So, jump there insted of emitting an > + * additional "li reg_h,0" instruction. > + */ > + if (extra_insn_needed) > + EMIT(PPC_RAW_LI(dst_reg_h, 0)); > + PPC_JMP((ctx->idx + 2 + adjusted_idx) * 4); > + } > + > switch (size) { > case BPF_B: > EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off)); >
diff --git a/arch/powerpc/net/bpf_jit_comp32.c b/arch/powerpc/net/bpf_jit_comp32.c index c6262289dcc4..faa8047fcf4a 100644 --- a/arch/powerpc/net/bpf_jit_comp32.c +++ b/arch/powerpc/net/bpf_jit_comp32.c @@ -821,6 +821,45 @@ int bpf_jit_build_body(struct bpf_prog *fp, u32 *image, struct codegen_context * /* dst = *(u64 *)(ul) (src + off) */ case BPF_LDX | BPF_MEM | BPF_DW: 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 > TASK_SIZE_MAX, otherwise set dst_reg=0 and move on. + */ + if (BPF_MODE(code) == BPF_PROBE_MEM) { + bool extra_insn_needed = false; + unsigned int adjusted_idx; + + /* + * For BPF_DW case, "li reg_h,0" would be needed when + * !fp->aux->verifier_zext. Adjust conditional branch'ing + * address accordingly. + */ + if ((size == BPF_DW) && !fp->aux->verifier_zext) + extra_insn_needed = true; + + /* + * Need to jump two instructions instead of one for BPF_DW case + * as there are two load instructions for dst_reg_h & dst_reg + * respectively. + */ + adjusted_idx = (size == BPF_DW) ? 1 : 0; + + EMIT(PPC_RAW_ADDI(b2p[TMP_REG], src_reg, off)); + PPC_LI32(_R0, TASK_SIZE_MAX); + EMIT(PPC_RAW_CMPLW(b2p[TMP_REG], _R0)); + PPC_BCC(COND_GT, (ctx->idx + 4 + (extra_insn_needed ? 1 : 0)) * 4); + EMIT(PPC_RAW_LI(dst_reg, 0)); + /* + * Note that "li reg_h,0" is emitted for BPF_B/H/W case, + * if necessary. So, jump there insted of emitting an + * additional "li reg_h,0" instruction. + */ + if (extra_insn_needed) + EMIT(PPC_RAW_LI(dst_reg_h, 0)); + PPC_JMP((ctx->idx + 2 + adjusted_idx) * 4); + } + switch (size) { case BPF_B: EMIT(PPC_RAW_LBZ(dst_reg, src_reg, off));
With KUAP enabled, any kernel code which wants to access userspace needs to be surrounded by disable-enable KUAP. But that is not happening for BPF_PROBE_MEM load instruction. Though PPC32 does not support read protection, considering the fact that PTR_TO_BTF_ID (which uses BPF_PROBE_MEM mode) could either be a valid kernel pointer or NULL but should never be a pointer to userspace address, execute BPF_PROBE_MEM load only if addr > TASK_SIZE_MAX, otherwise set dst_reg=0 and move on. This will catch NULL, valid or invalid userspace pointers. Only bad kernel pointer will be handled by BPF exception table. [Alexei suggested for x86] Suggested-by: Alexei Starovoitov <ast@kernel.org> Signed-off-by: Hari Bathini <hbathini@linux.ibm.com> --- Changes in v2: * New patch to handle bad userspace pointers on PPC32. arch/powerpc/net/bpf_jit_comp32.c | 39 +++++++++++++++++++++++++++++++ 1 file changed, 39 insertions(+)