bpf, x86: Propagate tailcall info only for subprogs

Asphaltt · Alexei Starovoitov · commit a1087da9d11e · 2024-11-12T17:24:03.000-08:00
In x64 JIT, propagate tailcall info only for subprogs, not for helpers or kfuncs. Acked-by: Yonghong Song <yonghong.song@linux.dev> Signed-off-by: Leon Hwang <leon.hwang@linux.dev> Link: https://lore.kernel.org/r/20241107134529.8602-2-leon.hwang@linux.dev Signed-off-by: Alexei Starovoitov <ast@kernel.org>
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
@@ -2186,7 +2186,7 @@ st:			if (is_imm8(insn->off))
 			u8 *ip = image + addrs[i - 1];
 
 			func = (u8 *) __bpf_call_base + imm32;
-			if (tail_call_reachable) {
+			if (src_reg == BPF_PSEUDO_CALL && tail_call_reachable) {
 				LOAD_TAIL_CALL_CNT_PTR(stack_depth);
 				ip += 7;
 			}

Original file line number	Diff line number	Diff line change
`@@ -2186,7 +2186,7 @@ st: if (is_imm8(insn->off))`
`2186`	`2186`	`u8 *ip = image + addrs[i - 1];`
`2187`	`2187`
`2188`	`2188`	`func = (u8 *) __bpf_call_base + imm32;`
`2189`		`- if (tail_call_reachable) {`
	`2189`	`+ if (src_reg == BPF_PSEUDO_CALL && tail_call_reachable) {`
`2190`	`2190`	`LOAD_TAIL_CALL_CNT_PTR(stack_depth);`
`2191`	`2191`	`ip += 7;`
`2192`	`2192`	`}`