Merge branch 'bpf-arm64-support-for-timed-may_goto'

Puranjay Mohan says:

====================
bpf, arm64: support for timed may_goto

Changes in v2->v3:
v2: https://lore.kernel.org/all/[e-mail]/
- Rebased on bpf-next/master
- Added Acked-by: tags from Xu and Kumar

Changes in v1->v2:
v1: https://lore.kernel.org/bpf/[e-mail]/
- Added comment in arch_bpf_timed_may_goto() about BPF_REG_FP setup (Xu
Kuohai)

This set adds support for the timed may_goto instruction for the arm64.
The timed may_goto instruction is implemented by the verifier by
reserving 2 8byte slots in the program stack and then calling
arch_bpf_timed_may_goto() in a loop with the stack offset of these two
slots in BPF_REG_AX. It expects the function to put a timestamp in the
first slot and the returned count in BPF_REG_AX is put into the second
slot by a store instruction emitted by the verifier.

arch_bpf_timed_may_goto() is special as it receives the parameter in
BPF_REG_AX and is expected to return the result in BPF_REG_AX as well.
It can't clobber any caller saved registers because verifier doesn't
save anything before emitting the call.

So, arch_bpf_timed_may_goto() is implemented in assembly so the exact
registers that are stored/restored can be controlled (BPF caller saved
registers here) and it also needs to take care of moving arguments and
return values to and from BPF_REG_AX <-> arm64 R0.

So, arch_bpf_timed_may_goto() acts as a trampoline to call
bpf_check_timed_may_goto() which does the main logic of placing the
timestamp and returning the count.

All tests that use may_goto instruction pass after the changing some of
them in patch 2

#404 stream_errors:OK
[...]
#406/2 stream_success/stream_cond_break:OK
[...]
#494/23 verifier_bpf_fastcall/may_goto_interaction_x86_64:SKIP
#494/24 verifier_bpf_fastcall/may_goto_interaction_arm64:OK
[...]
#539/1 verifier_may_goto_1/may_goto 0:OK
#539/2 verifier_may_goto_1/batch 2 of may_goto 0:OK
#539/3 verifier_may_goto_1/may_goto batch with offsets 2/1/0:OK
#539/4 verifier_may_goto_1/may_goto batch with offsets 2/0:OK
#539 verifier_may_goto_1:OK
#540/1 verifier_may_goto_2/C code with may_goto 0:OK
#540 verifier_may_goto_2:OK
Summary: 7/16 PASSED, 25 SKIPPED, 0 FAILED
====================

Link: https://patch.msgid.link/[e-mail]
Signed-off-by: Alexei Starovoitov <[e-mail]>

arch/arm64/net/Makefile

···
 #
 # ARM64 networking code
 #
-obj-$(CONFIG_BPF_JIT) += bpf_jit_comp.o
+obj-$(CONFIG_BPF_JIT) += bpf_jit_comp.o bpf_timed_may_goto.o

arch/arm64/net/bpf_jit_comp.c

···
 		if (ret < 0)
 			return ret;
 		emit_call(func_addr, ctx);
-		emit(A64_MOV(1, r0, A64_R(0)), ctx);
+		/*
+		 * Call to arch_bpf_timed_may_goto() is emitted by the
+		 * verifier and called with custom calling convention with
+		 * first argument and return value in BPF_REG_AX (x9).
+		 */
+		if (func_addr != (u64)arch_bpf_timed_may_goto)
+			emit(A64_MOV(1, r0, A64_R(0)), ctx);
 		break;
 	}
 	/* tail call */
···
 	 * no need to provide any additional instructions. Therefore, skip
 	 * inserting nospec insns against Spectre v4.
 	 */
+	return true;
+}
+
+bool bpf_jit_supports_timed_may_goto(void)
+{
 	return true;
 }
 

arch/arm64/net/bpf_timed_may_goto.S

···
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2025 Puranjay Mohan <[e-mail]> */
+
+#include <linux/linkage.h>
+
+SYM_FUNC_START(arch_bpf_timed_may_goto)
+	/* Allocate stack space and emit frame record */
+	stp     x29, x30, [sp, #-64]!
+	mov     x29, sp
+
+	/* Save BPF registers R0 - R5 (x7, x0-x4)*/
+	stp	x7, x0, [sp, #16]
+	stp	x1, x2, [sp, #32]
+	stp	x3, x4, [sp, #48]
+
+	/*
+	 * Stack depth was passed in BPF_REG_AX (x9), add it to the BPF_FP
+	 * (x25) to get the pointer to count and timestamp and pass it as the
+	 * first argument in x0.
+	 *
+	 * Before generating the call to arch_bpf_timed_may_goto, the verifier
+	 * generates a load instruction using FP, i.e. REG_AX = *(u64 *)(FP -
+	 * stack_off_cnt), so BPF_REG_FP (x25) is always set up by the arm64
+	 * jit in this case.
+	 */
+	add	x0, x9, x25
+	bl	bpf_check_timed_may_goto
+	/* BPF_REG_AX(x9) will be stored into count, so move return value to it. */
+	mov	x9, x0
+
+	/* Restore BPF registers R0 - R5 (x7, x0-x4) */
+	ldp	x7, x0, [sp, #16]
+	ldp	x1, x2, [sp, #32]
+	ldp	x3, x4, [sp, #48]
+
+	/* Restore FP and LR */
+	ldp     x29, x30, [sp], #64
+
+	ret
+SYM_FUNC_END(arch_bpf_timed_may_goto)

tools/testing/selftests/bpf/prog_tests/stream.c

···
 		ASSERT_OK(ret, "ret");
 		ASSERT_OK(opts.retval, "retval");
 
-#if !defined(__x86_64__) && !defined(__s390x__)
+#if !defined(__x86_64__) && !defined(__s390x__) && !defined(__aarch64__)
 		ASSERT_TRUE(1, "Timed may_goto unsupported, skip.");
 		if (i == 0) {
 			ret = bpf_prog_stream_read(prog_fd, 2, buf, sizeof(buf), &ropts);

tools/testing/selftests/bpf/progs/verifier_bpf_fastcall.c

···
 
 SEC("raw_tp")
 __arch_arm64
-__log_level(4) __msg("stack depth 16")
-/* may_goto counter at -16 */
-__xlated("0: *(u64 *)(r10 -16) =")
-__xlated("1: r1 = 1")
-__xlated("2: call bpf_get_smp_processor_id")
+__log_level(4) __msg("stack depth 24")
+/* may_goto counter at -24 */
+__xlated("0: *(u64 *)(r10 -24) =")
+/* may_goto timestamp at -16 */
+__xlated("1: *(u64 *)(r10 -16) =")
+__xlated("2: r1 = 1")
+__xlated("3: call bpf_get_smp_processor_id")
 /* may_goto expansion starts */
-__xlated("3: r11 = *(u64 *)(r10 -16)")
-__xlated("4: if r11 == 0x0 goto pc+3")
-__xlated("5: r11 -= 1")
-__xlated("6: *(u64 *)(r10 -16) = r11")
+__xlated("4: r11 = *(u64 *)(r10 -24)")
+__xlated("5: if r11 == 0x0 goto pc+6")
+__xlated("6: r11 -= 1")
+__xlated("7: if r11 != 0x0 goto pc+2")
+__xlated("8: r11 = -24")
+__xlated("9: call unknown")
+__xlated("10: *(u64 *)(r10 -24) = r11")
 /* may_goto expansion ends */
-__xlated("7: *(u64 *)(r10 -8) = r1")
-__xlated("8: exit")
+__xlated("11: *(u64 *)(r10 -8) = r1")
+__xlated("12: exit")
 __success
 __naked void may_goto_interaction_arm64(void)
 {

tools/testing/selftests/bpf/progs/verifier_may_goto_1.c

···
 __description("may_goto 0")
 __arch_x86_64
 __arch_s390x
+__arch_arm64
 __xlated("0: r0 = 1")
 __xlated("1: exit")
 __success
···
 __description("batch 2 of may_goto 0")
 __arch_x86_64
 __arch_s390x
+__arch_arm64
 __xlated("0: r0 = 1")
 __xlated("1: exit")
 __success
···
 __description("may_goto batch with offsets 2/1/0")
 __arch_x86_64
 __arch_s390x
+__arch_arm64
 __xlated("0: r0 = 1")
 __xlated("1: exit")
 __success
···
 }
 
 SEC("raw_tp")
-__description("may_goto batch with offsets 2/0 - x86_64 and s390x")
+__description("may_goto batch with offsets 2/0")
 __arch_x86_64
 __arch_s390x
+__arch_arm64
 __xlated("0: *(u64 *)(r10 -16) = 65535")
 __xlated("1: *(u64 *)(r10 -8) = 0")
 __xlated("2: r11 = *(u64 *)(r10 -16)")
···
 __xlated("10: r0 = 2")
 __xlated("11: exit")
 __success
-__naked void may_goto_batch_2_x86_64_s390x(void)
-{
-	asm volatile (
-	".8byte %[may_goto1];"
-	".8byte %[may_goto3];"
-	"r0 = 1;"
-	"r0 = 2;"
-	"exit;"
-	:
-	: __imm_insn(may_goto1, BPF_RAW_INSN(BPF_JMP | BPF_JCOND, 0, 0, 2 /* offset */, 0)),
-	  __imm_insn(may_goto3, BPF_RAW_INSN(BPF_JMP | BPF_JCOND, 0, 0, 0 /* offset */, 0))
-	: __clobber_all);
-}
-
-SEC("raw_tp")
-__description("may_goto batch with offsets 2/0 - arm64")
-__arch_arm64
-__xlated("0: *(u64 *)(r10 -8) = 8388608")
-__xlated("1: r11 = *(u64 *)(r10 -8)")
-__xlated("2: if r11 == 0x0 goto pc+3")
-__xlated("3: r11 -= 1")
-__xlated("4: *(u64 *)(r10 -8) = r11")
-__xlated("5: r0 = 1")
-__xlated("6: r0 = 2")
-__xlated("7: exit")
-__success
-__naked void may_goto_batch_2_arm64(void)
+__naked void may_goto_batch_2(void)
 {
 	asm volatile (
 	".8byte %[may_goto1];"