LoongArch: BPF: Implement bpf_addr_space_cast instruction

LLVM generates bpf_addr_space_cast instruction while translating pointers
between native (zero) address space and __attribute__((address_space(N))).
The addr_space=0 is reserved as bpf_arena address space.

rY = addr_space_cast(rX, 0, 1) is processed by the verifier and converted
to normal 32-bit move: wX = wY

rY = addr_space_cast(rX, 1, 0) has to be converted by JIT.

With this, the following test cases passed:

    $ ./test_progs -a arena_htab,arena_list,arena_strsearch,verifier_arena,verifier_arena_large
    #4/1     arena_htab/arena_htab_llvm:OK
    #4/2     arena_htab/arena_htab_asm:OK
    #4       arena_htab:OK
    #5/1     arena_list/arena_list_1:OK
    #5/2     arena_list/arena_list_1000:OK
    #5       arena_list:OK
    #7/1     arena_strsearch/arena_strsearch:OK
    #7       arena_strsearch:OK
    #507/1   verifier_arena/basic_alloc1:OK
    #507/2   verifier_arena/basic_alloc2:OK
    #507/3   verifier_arena/basic_alloc3:OK
    #507/4   verifier_arena/basic_reserve1:OK
    #507/5   verifier_arena/basic_reserve2:OK
    #507/6   verifier_arena/reserve_twice:OK
    #507/7   verifier_arena/reserve_invalid_region:OK
    #507/8   verifier_arena/iter_maps1:OK
    #507/9   verifier_arena/iter_maps2:OK
    #507/10  verifier_arena/iter_maps3:OK
    #507     verifier_arena:OK
    #508/1   verifier_arena_large/big_alloc1:OK
    #508/2   verifier_arena_large/access_reserved:OK
    #508/3   verifier_arena_large/request_partially_reserved:OK
    #508/4   verifier_arena_large/free_reserved:OK
    #508/5   verifier_arena_large/big_alloc2:OK
    #508     verifier_arena_large:OK
    Summary: 5/20 PASSED, 0 SKIPPED, 0 FAILED

Acked-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Tested-by: Vincent Li <vincent.mc.li@gmail.com>
Signed-off-by: Hengqi Chen <hengqi.chen@gmail.com>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>

Author: chenhengqi

arch/loongarch/net/bpf_jit.c

@@ -559,6 +559,15 @@ static int build_insn(const struct bpf_insn *insn, struct jit_ctx *ctx, bool ext
 	/* dst = src */
 	case BPF_ALU | BPF_MOV | BPF_X:
 	case BPF_ALU64 | BPF_MOV | BPF_X:
+		if (insn_is_cast_user(insn)) {
+			move_reg(ctx, t1, src);
+			emit_zext_32(ctx, t1, true);
+			move_imm(ctx, dst, (ctx->user_vm_start >> 32) << 32, false);
+			emit_insn(ctx, beq, t1, LOONGARCH_GPR_ZERO, 1);
+			emit_insn(ctx, or, t1, dst, t1);
+			move_reg(ctx, dst, t1);
+			break;
+		}
 		switch (off) {
 		case 0:
 			move_reg(ctx, dst, src);
@@ -1955,6 +1964,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 	memset(&ctx, 0, sizeof(ctx));
 	ctx.prog = prog;
 	ctx.arena_vm_start = bpf_arena_get_kern_vm_start(prog->aux->arena);
+	ctx.user_vm_start = bpf_arena_get_user_vm_start(prog->aux->arena);
 
 	ctx.offset = kvcalloc(prog->len + 1, sizeof(u32), GFP_KERNEL);
 	if (ctx.offset == NULL) {
@@ -2110,6 +2120,11 @@ bool bpf_jit_bypass_spec_v4(void)
 	return true;
 }
 
+bool bpf_jit_supports_arena(void)
+{
+	return true;
+}
+
 /* Indicate the JIT backend supports mixing bpf2bpf and tailcalls. */
 bool bpf_jit_supports_subprog_tailcalls(void)
 {

arch/loongarch/net/bpf_jit.h

@@ -21,6 +21,7 @@ struct jit_ctx {
 	union loongarch_instruction *ro_image;
 	u32 stack_size;
 	u64 arena_vm_start;
+	u64 user_vm_start;
 };
 
 struct jit_data {