riscv, bpf: Fix unpredictable kernel crash about RV64 struct_ops

We encountered a kernel crash triggered by the bpf_tcp_ca testcase as
show below:

Unable to handle kernel paging request at virtual address ff60000088554500
Oops [#1]
...
CPU: 3 PID: 458 Comm: test_progs Tainted: G           OE      6.8.0-rc1-kselftest_plain #1
Hardware name: riscv-virtio,qemu (DT)
epc : 0xff60000088554500
 ra : tcp_ack+0x288/0x1232
epc : ff60000088554500 ra : ffffffff80cc7166 sp : ff2000000117ba50
 gp : ffffffff82587b60 tp : ff60000087be0040 t0 : ff60000088554500
 t1 : ffffffff801ed24e t2 : 0000000000000000 s0 : ff2000000117bbc0
 s1 : 0000000000000500 a0 : ff20000000691000 a1 : 0000000000000018
 a2 : 0000000000000001 a3 : ff60000087be03a0 a4 : 0000000000000000
 a5 : 0000000000000000 a6 : 0000000000000021 a7 : ffffffff8263f880
 s2 : 000000004ac3c13b s3 : 000000004ac3c13a s4 : 0000000000008200
 s5 : 0000000000000001 s6 : 0000000000000104 s7 : ff2000000117bb00
 s8 : ff600000885544c0 s9 : 0000000000000000 s10: ff60000086ff0b80
 s11: 000055557983a9c0 t3 : 0000000000000000 t4 : 000000000000ffc4
 t5 : ffffffff8154f170 t6 : 0000000000000030
status: 0000000200000120 badaddr: ff60000088554500 cause: 000000000000000c
Code: c796 67d7 0000 0000 0052 0002 c13b 4ac3 0000 0000 (0001) 0000
---[ end trace 0000000000000000 ]---

The reason is that commit 2cd3e3772e41 ("x86/cfi,bpf: Fix bpf_struct_ops
CFI") changes the func_addr of arch_prepare_bpf_trampoline in struct_ops
from NULL to non-NULL, while we use func_addr on RV64 to differentiate
between struct_ops and regular trampoline. When the struct_ops testcase
is triggered, it emits wrong prologue and epilogue, and lead to
unpredictable issues. After commit 2cd3e3772e41, we can use
BPF_TRAMP_F_INDIRECT to distinguish them as it always be set in
struct_ops.

Fixes: 2cd3e3772e41 ("x86/cfi,bpf: Fix bpf_struct_ops CFI")
Signed-off-by: Pu Lehui <pulehui@huawei.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Tested-by: Björn Töpel <bjorn@rivosinc.com>
Acked-by: Björn Töpel <bjorn@kernel.org>
Link: https://lore.kernel.org/bpf/20240123023207.1917284-1-pulehui@huaweicloud.com

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index 58dc64d..719a97e 100644 (file)
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -795,6 +795,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
        struct bpf_tramp_links *fentry = &tlinks[BPF_TRAMP_FENTRY];
        struct bpf_tramp_links *fexit = &tlinks[BPF_TRAMP_FEXIT];
        struct bpf_tramp_links *fmod_ret = &tlinks[BPF_TRAMP_MODIFY_RETURN];
+       bool is_struct_ops = flags & BPF_TRAMP_F_INDIRECT;
        void *orig_call = func_addr;
        bool save_ret;
        u32 insn;
@@ -878,7 +879,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 
        stack_size = round_up(stack_size, 16);
 
-       if (func_addr) {
+       if (!is_struct_ops) {
                /* For the trampoline called from function entry,
                 * the frame of traced function and the frame of
                 * trampoline need to be considered.
@@ -998,7 +999,7 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 
        emit_ld(RV_REG_S1, -sreg_off, RV_REG_FP, ctx);
 
-       if (func_addr) {
+       if (!is_struct_ops) {
                /* trampoline called from function entry */
                emit_ld(RV_REG_T0, stack_size - 8, RV_REG_SP, ctx);
                emit_ld(RV_REG_FP, stack_size - 16, RV_REG_SP, ctx);