Merge branch 'allocate-bpf-trampoline-on-bpf_prog_pack'

Song Liu says:

====================
Allocate bpf trampoline on bpf_prog_pack

This set enables allocating bpf trampoline from bpf_prog_pack on x86. The
majority of this work, however, is the refactoring of trampoline code.
This is needed because we need to handle 4 archs and 2 users (trampoline
and struct_ops).

1/7 through 6/7 refactors trampoline code. A few helpers are added.
7/7 finally let bpf trampoline on x86 use bpf_prog_pack.

Changes in v7:
1. Use kvmalloc for rw_image in x86/arch_prepare_bpf_trampoline. (Alexei)
2. Add comment to explain why we cannot use kvmalloc in
   x86/arch_bpf_trampoline_size. (Alexei)

Changes in v6:
1. Rebase.
2. Add Acked-by and Tested-by from Jiri Olsa and Björn Töpel.

Changes in v5:
1. Adjust size of trampoline ksym. (Jiri)
2. Use "unsigned int size" arg in image management helpers.(Daniel)

Changes in v4:
1. Dropped 1/8 in v3, which is already merged in bpf-next.
2. Add Reviewed-by from Björn Töpel.

Changes in v3:
1. Fix bug in s390. (Thanks to Ilya Leoshkevich).
2. Fix build error in riscv. (kernel test robot).

Changes in v2:
1. Add missing changes in net/bpf/bpf_dummy_struct_ops.c.
2. Reduce one dry run in arch_prepare_bpf_trampoline. (Xu Kuohai)
3. Other small fixes.
====================

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Alexei Starovoitov <[email protected]>

Author: Alexei Starovoitov

arch/arm64/net/bpf_jit_comp.c

@@ -1828,7 +1828,7 @@ static void restore_args(struct jit_ctx *ctx, int args_off, int nregs)
  *
  */
 static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
-			      struct bpf_tramp_links *tlinks, void *orig_call,
+			      struct bpf_tramp_links *tlinks, void *func_addr,
 			      int nregs, u32 flags)
 {
 	int i;
@@ -1926,7 +1926,7 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 
 	if (flags & BPF_TRAMP_F_IP_ARG) {
 		/* save ip address of the traced function */
-		emit_addr_mov_i64(A64_R(10), (const u64)orig_call, ctx);
+		emit_addr_mov_i64(A64_R(10), (const u64)func_addr, ctx);
 		emit(A64_STR64I(A64_R(10), A64_SP, ip_off), ctx);
 	}
 
@@ -2026,18 +2026,10 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	return ctx->idx;
 }
 
-int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
-				void *image_end, const struct btf_func_model *m,
-				u32 flags, struct bpf_tramp_links *tlinks,
-				void *orig_call)
+static int btf_func_model_nregs(const struct btf_func_model *m)
 {
-	int i, ret;
 	int nregs = m->nr_args;
-	int max_insns = ((long)image_end - (long)image) / AARCH64_INSN_SIZE;
-	struct jit_ctx ctx = {
-		.image = NULL,
-		.idx = 0,
-	};
+	int i;
 
 	/* extra registers needed for struct argument */
 	for (i = 0; i < MAX_BPF_FUNC_ARGS; i++) {
@@ -2046,22 +2038,49 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
 			nregs += (m->arg_size[i] + 7) / 8 - 1;
 	}
 
+	return nregs;
+}
+
+int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
+			     struct bpf_tramp_links *tlinks, void *func_addr)
+{
+	struct jit_ctx ctx = {
+		.image = NULL,
+		.idx = 0,
+	};
+	struct bpf_tramp_image im;
+	int nregs, ret;
+
+	nregs = btf_func_model_nregs(m);
 	/* the first 8 registers are used for arguments */
 	if (nregs > 8)
 		return -ENOTSUPP;
 
-	ret = prepare_trampoline(&ctx, im, tlinks, orig_call, nregs, flags);
+	ret = prepare_trampoline(&ctx, &im, tlinks, func_addr, nregs, flags);
 	if (ret < 0)
 		return ret;
 
-	if (ret > max_insns)
-		return -EFBIG;
+	return ret < 0 ? ret : ret * AARCH64_INSN_SIZE;
+}
 
-	ctx.image = image;
-	ctx.idx = 0;
+int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
+				void *image_end, const struct btf_func_model *m,
+				u32 flags, struct bpf_tramp_links *tlinks,
+				void *func_addr)
+{
+	int ret, nregs;
+	struct jit_ctx ctx = {
+		.image = image,
+		.idx = 0,
+	};
+
+	nregs = btf_func_model_nregs(m);
+	/* the first 8 registers are used for arguments */
+	if (nregs > 8)
+		return -ENOTSUPP;
 
 	jit_fill_hole(image, (unsigned int)(image_end - image));
-	ret = prepare_trampoline(&ctx, im, tlinks, orig_call, nregs, flags);
+	ret = prepare_trampoline(&ctx, im, tlinks, func_addr, nregs, flags);
 
 	if (ret > 0 && validate_code(&ctx) < 0)
 		ret = -EINVAL;

arch/riscv/net/bpf_jit_comp64.c

@@ -1029,23 +1029,28 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	return ret;
 }
 
-int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
-				void *image_end, const struct btf_func_model *m,
-				u32 flags, struct bpf_tramp_links *tlinks,
-				void *func_addr)
+int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
+			     struct bpf_tramp_links *tlinks, void *func_addr)
 {
-	int ret;
+	struct bpf_tramp_image im;
 	struct rv_jit_context ctx;
+	int ret;
 
 	ctx.ninsns = 0;
 	ctx.insns = NULL;
 	ctx.ro_insns = NULL;
-	ret = __arch_prepare_bpf_trampoline(im, m, tlinks, func_addr, flags, &ctx);
-	if (ret < 0)
-		return ret;
+	ret = __arch_prepare_bpf_trampoline(&im, m, tlinks, func_addr, flags, &ctx);
 
-	if (ninsns_rvoff(ret) > (long)image_end - (long)image)
-		return -EFBIG;
+	return ret < 0 ? ret : ninsns_rvoff(ctx.ninsns);
+}
+
+int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
+				void *image_end, const struct btf_func_model *m,
+				u32 flags, struct bpf_tramp_links *tlinks,
+				void *func_addr)
+{
+	int ret;
+	struct rv_jit_context ctx;
 
 	ctx.ninsns = 0;
 	/*

arch/s390/net/bpf_jit_comp.c

@@ -2637,37 +2637,49 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
 	return 0;
 }
 
+int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
+			     struct bpf_tramp_links *tlinks, void *orig_call)
+{
+	struct bpf_tramp_image im;
+	struct bpf_tramp_jit tjit;
+	int ret;
+
+	memset(&tjit, 0, sizeof(tjit));
+
+	ret = __arch_prepare_bpf_trampoline(&im, &tjit, m, flags,
+					    tlinks, orig_call);
+
+	return ret < 0 ? ret : tjit.common.prg;
+}
+
 int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *image,
 				void *image_end, const struct btf_func_model *m,
 				u32 flags, struct bpf_tramp_links *tlinks,
 				void *func_addr)
 {
 	struct bpf_tramp_jit tjit;
 	int ret;
-	int i;
 
-	for (i = 0; i < 2; i++) {
-		if (i == 0) {
-			/* Compute offsets, check whether the code fits. */
-			memset(&tjit, 0, sizeof(tjit));
-		} else {
-			/* Generate the code. */
-			tjit.common.prg = 0;
-			tjit.common.prg_buf = image;
-		}
-		ret = __arch_prepare_bpf_trampoline(im, &tjit, m, flags,
-						    tlinks, func_addr);
-		if (ret < 0)
-			return ret;
-		if (tjit.common.prg > (char *)image_end - (char *)image)
-			/*
-			 * Use the same error code as for exceeding
-			 * BPF_MAX_TRAMP_LINKS.
-			 */
-			return -E2BIG;
-	}
+	/* Compute offsets, check whether the code fits. */
+	memset(&tjit, 0, sizeof(tjit));
+	ret = __arch_prepare_bpf_trampoline(im, &tjit, m, flags,
+					    tlinks, func_addr);
+
+	if (ret < 0)
+		return ret;
+	if (tjit.common.prg > (char *)image_end - (char *)image)
+		/*
+		 * Use the same error code as for exceeding
+		 * BPF_MAX_TRAMP_LINKS.
+		 */
+		return -E2BIG;
+
+	tjit.common.prg = 0;
+	tjit.common.prg_buf = image;
+	ret = __arch_prepare_bpf_trampoline(im, &tjit, m, flags,
+					    tlinks, func_addr);
 
-	return tjit.common.prg;
+	return ret < 0 ? ret : tjit.common.prg;
 }
 
 bool bpf_jit_supports_subprog_tailcalls(void)