From 580031ff9952b7dbf48dedba6b56a100ae002bef Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Mon, 27 Mar 2023 17:42:32 -0700 Subject: [PATCH 1/4] bpf: tcp: Use sock_gen_put instead of sock_put in bpf_iter_tcp While reviewing the udp-iter batching patches, noticed the bpf_iter_tcp calling sock_put() is incorrect. It should call sock_gen_put instead because bpf_iter_tcp is iterating the ehash table which has the req sk and tw sk. This patch replaces all sock_put with sock_gen_put in the bpf_iter_tcp codepath. Fixes: 04c7820b776f ("bpf: tcp: Bpf iter batching and lock_sock") Signed-off-by: Martin KaFai Lau Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230328004232.2134233-1-martin.lau@linux.dev --- net/ipv4/tcp_ipv4.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index ea370afa70ed..b9d55277cb85 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -2780,7 +2780,7 @@ static int tcp_prog_seq_show(struct bpf_prog *prog, struct bpf_iter_meta *meta, static void bpf_iter_tcp_put_batch(struct bpf_tcp_iter_state *iter) { while (iter->cur_sk < iter->end_sk) - sock_put(iter->batch[iter->cur_sk++]); + sock_gen_put(iter->batch[iter->cur_sk++]); } static int bpf_iter_tcp_realloc_batch(struct bpf_tcp_iter_state *iter, @@ -2941,7 +2941,7 @@ static void *bpf_iter_tcp_seq_next(struct seq_file *seq, void *v, loff_t *pos) * st->bucket. See tcp_seek_last_pos(). */ st->offset++; - sock_put(iter->batch[iter->cur_sk++]); + sock_gen_put(iter->batch[iter->cur_sk++]); } if (iter->cur_sk < iter->end_sk) From a6f6a95f25803500079513780d11a911ce551d76 Mon Sep 17 00:00:00 2001 From: George Guo Date: Tue, 28 Mar 2023 15:13:35 +0800 Subject: [PATCH 2/4] LoongArch, bpf: Fix jit to skip speculation barrier opcode Just skip the opcode(BPF_ST | BPF_NOSPEC) in the BPF JIT instead of failing to JIT the entire program, given LoongArch currently has no couterpart of a speculation barrier instruction. To verify the issue, use the ltp testcase as shown below. Also, Wang says: I can confirm there's currently no speculation barrier equivalent on LonogArch. (Loongson says there are builtin mitigations for Spectre-V1 and V2 on their chips, and AFAIK efforts to port the exploits to mips/LoongArch have all failed a few years ago.) Without this patch: $ ./bpf_prog02 [...] bpf_common.c:123: TBROK: Failed verification: ??? (524) [...] Summary: passed 0 failed 0 broken 1 skipped 0 warnings 0 With this patch: $ ./bpf_prog02 [...] Summary: passed 0 failed 0 broken 0 skipped 0 warnings 0 Fixes: 5dc615520c4d ("LoongArch: Add BPF JIT support") Signed-off-by: George Guo Signed-off-by: Daniel Borkmann Acked-by: WANG Xuerui Cc: Tiezhu Yang Link: https://lore.kernel.org/bpf/20230328071335.2664966-1-guodongtai@kylinos.cn --- arch/loongarch/net/bpf_jit.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/loongarch/net/bpf_jit.c b/arch/loongarch/net/bpf_jit.c index 288003a9f0ca..d586df48ecc6 100644 --- a/arch/loongarch/net/bpf_jit.c +++ b/arch/loongarch/net/bpf_jit.c @@ -1022,6 +1022,10 @@ static int build_insn(const struct bpf_insn *insn, struct jit_ctx *ctx, bool ext emit_atomic(insn, ctx); break; + /* Speculation barrier */ + case BPF_ST | BPF_NOSPEC: + break; + default: pr_err("bpf_jit: unknown opcode %02x\n", code); return -EINVAL; From 738a96c4a8c36950803fdd27e7c30aca92dccefd Mon Sep 17 00:00:00 2001 From: Xu Kuohai Date: Sat, 1 Apr 2023 19:41:44 -0400 Subject: [PATCH 3/4] bpf, arm64: Fixed a BTI error on returning to patched function When BPF_TRAMP_F_CALL_ORIG is set, BPF trampoline uses BLR to jump back to the instruction next to call site to call the patched function. For BTI-enabled kernel, the instruction next to call site is usually PACIASP, in this case, it's safe to jump back with BLR. But when the call site is not followed by a PACIASP or bti, a BTI exception is triggered. Here is a fault log: Unhandled 64-bit el1h sync exception on CPU0, ESR 0x0000000034000002 -- BTI CPU: 0 PID: 263 Comm: test_progs Tainted: GF Hardware name: linux,dummy-virt (DT) pstate: 40400805 (nZcv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=-c) pc : bpf_fentry_test1+0xc/0x30 lr : bpf_trampoline_6442573892_0+0x48/0x1000 sp : ffff80000c0c3a50 x29: ffff80000c0c3a90 x28: ffff0000c2e6c080 x27: 0000000000000000 x26: 0000000000000000 x25: 0000000000000000 x24: 0000000000000050 x23: 0000000000000000 x22: 0000ffffcfd2a7f0 x21: 000000000000000a x20: 0000ffffcfd2a7f0 x19: 0000000000000000 x18: 0000000000000000 x17: 0000000000000000 x16: 0000000000000000 x15: 0000ffffcfd2a7f0 x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000 x11: 0000000000000000 x10: ffff80000914f5e4 x9 : ffff8000082a1528 x8 : 0000000000000000 x7 : 0000000000000000 x6 : 0101010101010101 x5 : 0000000000000000 x4 : 00000000fffffff2 x3 : 0000000000000001 x2 : ffff8001f4b82000 x1 : 0000000000000000 x0 : 0000000000000001 Kernel panic - not syncing: Unhandled exception CPU: 0 PID: 263 Comm: test_progs Tainted: GF Hardware name: linux,dummy-virt (DT) Call trace: dump_backtrace+0xec/0x144 show_stack+0x24/0x7c dump_stack_lvl+0x8c/0xb8 dump_stack+0x18/0x34 panic+0x1cc/0x3ec __el0_error_handler_common+0x0/0x130 el1h_64_sync_handler+0x60/0xd0 el1h_64_sync+0x78/0x7c bpf_fentry_test1+0xc/0x30 bpf_fentry_test1+0xc/0x30 bpf_prog_test_run_tracing+0xdc/0x2a0 __sys_bpf+0x438/0x22a0 __arm64_sys_bpf+0x30/0x54 invoke_syscall+0x78/0x110 el0_svc_common.constprop.0+0x6c/0x1d0 do_el0_svc+0x38/0xe0 el0_svc+0x30/0xd0 el0t_64_sync_handler+0x1ac/0x1b0 el0t_64_sync+0x1a0/0x1a4 Kernel Offset: disabled CPU features: 0x0000,00034c24,f994fdab Memory Limit: none And the instruction next to call site of bpf_fentry_test1 is ADD, not PACIASP: : bti c nop nop add w0, w0, #0x1 paciasp For BPF prog, JIT always puts a PACIASP after call site for BTI-enabled kernel, so there is no problem. To fix it, replace BLR with RET to bypass the branch target check. Fixes: efc9909fdce0 ("bpf, arm64: Add bpf trampoline for arm64") Reported-by: Florent Revest Signed-off-by: Xu Kuohai Signed-off-by: Daniel Borkmann Tested-by: Florent Revest Acked-by: Florent Revest Link: https://lore.kernel.org/bpf/20230401234144.3719742-1-xukuohai@huaweicloud.com --- arch/arm64/net/bpf_jit.h | 4 ++++ arch/arm64/net/bpf_jit_comp.c | 3 ++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/arm64/net/bpf_jit.h b/arch/arm64/net/bpf_jit.h index a6acb94ea3d6..c2edadb8ec6a 100644 --- a/arch/arm64/net/bpf_jit.h +++ b/arch/arm64/net/bpf_jit.h @@ -281,4 +281,8 @@ /* DMB */ #define A64_DMB_ISH aarch64_insn_gen_dmb(AARCH64_INSN_MB_ISH) +/* ADR */ +#define A64_ADR(Rd, offset) \ + aarch64_insn_gen_adr(0, offset, Rd, AARCH64_INSN_ADR_TYPE_ADR) + #endif /* _BPF_JIT_H */ diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 62f805f427b7..b26da8efa616 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -1900,7 +1900,8 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im, restore_args(ctx, args_off, nargs); /* call original func */ emit(A64_LDR64I(A64_R(10), A64_SP, retaddr_off), ctx); - emit(A64_BLR(A64_R(10)), ctx); + emit(A64_ADR(A64_LR, AARCH64_INSN_SIZE * 2), ctx); + emit(A64_RET(A64_R(10)), ctx); /* store return value */ emit(A64_STR64I(A64_R(0), A64_SP, retval_off), ctx); /* reserve a nop for bpf_tramp_image_put */ From 919e659ed12568b5b8ba6c2ffdd82d8d31fc28af Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Thu, 6 Apr 2023 12:40:19 +0200 Subject: [PATCH 4/4] selftests/bpf: fix xdp_redirect xdp-features selftest for veth driver xdp-features supported by veth driver are no more static, but they depends on veth configuration (e.g. if GRO is enabled/disabled or TX/RX queue configuration). Take it into account in xdp_redirect xdp-features selftest for veth driver. Fixes: fccca038f300 ("veth: take into account device reconfiguration for xdp_features flag") Signed-off-by: Lorenzo Bianconi Link: https://lore.kernel.org/r/bc35455cfbb1d4f7f52536955ded81ad47d8dc54.1680777371.git.lorenzo@kernel.org Signed-off-by: Martin KaFai Lau --- .../bpf/prog_tests/xdp_do_redirect.c | 30 +++++++++++++++++-- 1 file changed, 27 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c index 7271a18ab3e2..8251a0fc6ee9 100644 --- a/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c +++ b/tools/testing/selftests/bpf/prog_tests/xdp_do_redirect.c @@ -167,8 +167,7 @@ void test_xdp_do_redirect(void) if (!ASSERT_EQ(query_opts.feature_flags, NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | - NETDEV_XDP_ACT_NDO_XMIT | NETDEV_XDP_ACT_RX_SG | - NETDEV_XDP_ACT_NDO_XMIT_SG, + NETDEV_XDP_ACT_RX_SG, "veth_src query_opts.feature_flags")) goto out; @@ -176,11 +175,36 @@ void test_xdp_do_redirect(void) if (!ASSERT_OK(err, "veth_dst bpf_xdp_query")) goto out; + if (!ASSERT_EQ(query_opts.feature_flags, + NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_RX_SG, + "veth_dst query_opts.feature_flags")) + goto out; + + /* Enable GRO */ + SYS("ethtool -K veth_src gro on"); + SYS("ethtool -K veth_dst gro on"); + + err = bpf_xdp_query(ifindex_src, XDP_FLAGS_DRV_MODE, &query_opts); + if (!ASSERT_OK(err, "veth_src bpf_xdp_query gro on")) + goto out; + if (!ASSERT_EQ(query_opts.feature_flags, NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | NETDEV_XDP_ACT_NDO_XMIT | NETDEV_XDP_ACT_RX_SG | NETDEV_XDP_ACT_NDO_XMIT_SG, - "veth_dst query_opts.feature_flags")) + "veth_src query_opts.feature_flags gro on")) + goto out; + + err = bpf_xdp_query(ifindex_dst, XDP_FLAGS_DRV_MODE, &query_opts); + if (!ASSERT_OK(err, "veth_dst bpf_xdp_query gro on")) + goto out; + + if (!ASSERT_EQ(query_opts.feature_flags, + NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT | + NETDEV_XDP_ACT_NDO_XMIT | NETDEV_XDP_ACT_RX_SG | + NETDEV_XDP_ACT_NDO_XMIT_SG, + "veth_dst query_opts.feature_flags gro on")) goto out; memcpy(skel->rodata->expect_dst, &pkt_udp.eth.h_dest, ETH_ALEN);