executor: arm64: sys/linux: introduce syzos API

Allow guest payload to call syzos API functions. The available calls
are enumerated by SYZOS_API_* constants, and have a form of:

  struct api_call {
         uint64 call;
         uint64 struct_size;
         /* arbitrary call-related data here */
  };

Complex instruction sequences are too easy to break, so most of the time
fuzzer won't be able to efficiently mutate them.
We replace kvm_text_arm64 with a sequence of `struct api_call`, making it
possible to intermix assembly instructions (SYZOS_API_CODE) with
higher-level constructs.

Right now the supported calls are:
 - SYZOS_API_UEXIT - abort from KVM_RUN (1 argument: exit code, uint64)
 - SYZOS_API_CODE - execute an ARM64 assembly blob
                    (1 argument: inline array of int32's)

3 files changed, 76 insertions, 12 deletions

diff --git a/executor/common_kvm_arm64.h b/executor/common_kvm_arm64.h
index ba02e244b..e6bb2b665 100644
--- a/executor/common_kvm_arm64.h
+++ b/executor/common_kvm_arm64.h
@@ -9,6 +9,7 @@
 #include "kvm.h"
 
 // Register encodings from https://docs.kernel.org/virt/kvm/api.html.
+#define KVM_ARM64_REGS_X0 0x6030000000100000UL
 #define KVM_ARM64_REGS_PC 0x6030000000100040UL
 #define KVM_ARM64_REGS_SP_EL1 0x6030000000100044UL
 
@@ -60,13 +61,10 @@ static struct addr_size alloc_guest_mem(struct addr_size* free, size_t size)
 	return ret;
 }
 
-static void fill_with_ret(void* addr, int size)
-{
-	uint32* insn = (uint32*)addr;
-
-	for (int i = 0; i < size / 4; i++)
-		insn[i] = 0xd65f03c0; // RET
-}
+struct api_fn {
+	int index;
+	void* fn;
+};
 
 // syz_kvm_setup_cpu(fd fd_kvmvm, cpufd fd_kvmcpu, usermem vma[24], text ptr[in, array[kvm_text, 1]], ntext len[text], flags flags[kvm_setup_flags], opts ptr[in, array[kvm_setup_opt, 0:2]], nopt len[opts])
 static volatile long syz_kvm_setup_cpu(volatile long a0, volatile long a1, volatile long a2, volatile long a3, volatile long a4, volatile long a5, volatile long a6, volatile long a7)
@@ -108,6 +106,7 @@ static volatile long syz_kvm_setup_cpu(volatile long a0, volatile long a1, volat
 
 	// Guest physical memory layout:
 	// 0x00000000 - unused pages
+	// 0xdddd0000 - unmapped region to trigger a page faults for uexits etc. (1 page)
 	// 0xeeee0000 - user code (1 page)
 	// 0xeeee8000 - executor guest code (4 pages)
 	// 0xffff1000 - EL1 stack (1 page)
@@ -119,12 +118,11 @@ static volatile long syz_kvm_setup_cpu(volatile long a0, volatile long a1, volat
 	vm_set_user_memory_region(vmfd, slot++, KVM_MEM_READONLY, ARM64_ADDR_EXECUTOR_CODE, host_text.size, (uintptr_t)host_text.addr);
 
 	struct addr_size next = alloc_guest_mem(&allocator, page_size);
-	// Fill the guest code page with RET instructions to be on the safe side.
-	fill_with_ret(next.addr, next.size);
 	if (text_size > next.size)
 		text_size = next.size;
 	memcpy(next.addr, text, text_size);
 	vm_set_user_memory_region(vmfd, slot++, KVM_MEM_READONLY, ARM64_ADDR_USER_CODE, next.size, (uintptr_t)next.addr);
+
 	next = alloc_guest_mem(&allocator, page_size);
 	vm_set_user_memory_region(vmfd, slot++, 0, ARM64_ADDR_EL1_STACK_BOTTOM, next.size, (uintptr_t)next.addr);
 
@@ -143,6 +141,8 @@ static volatile long syz_kvm_setup_cpu(volatile long a0, volatile long a1, volat
 	// PC points to the relative offset of guest_main() within the guest code.
 	vcpu_set_reg(cpufd, KVM_ARM64_REGS_PC, ARM64_ADDR_EXECUTOR_CODE + ((uint64)guest_main - (uint64)&__start_guest));
 	vcpu_set_reg(cpufd, KVM_ARM64_REGS_SP_EL1, ARM64_ADDR_EL1_STACK_BOTTOM + page_size - 128);
+	// Pass parameters to guest_main().
+	vcpu_set_reg(cpufd, KVM_ARM64_REGS_X0, text_size);
 
 	return 0;
 }
diff --git a/executor/common_kvm_arm64_syzos.h b/executor/common_kvm_arm64_syzos.h
index b9edf0069..022378b34 100644
--- a/executor/common_kvm_arm64_syzos.h
+++ b/executor/common_kvm_arm64_syzos.h
@@ -11,10 +11,70 @@
 // Start/end of the guest section.
 extern char *__start_guest, *__stop_guest;
 
+typedef enum {
+	SYZOS_API_UEXIT,
+	SYZOS_API_CODE,
+	SYZOS_API_STOP, // Must be the last one
+} syzos_api_id;
+
+struct api_call_header {
+	uint64 call;
+	uint64 size;
+};
+
+struct api_call_uexit {
+	struct api_call_header header;
+	uint64 exit_code;
+};
+
+struct api_call_code {
+	struct api_call_header header;
+	uint32 insns[];
+};
+
+void guest_uexit(uint64 exit_code);
+void guest_execute_code(uint32* insns, uint64 size);
+
 // Main guest function that performs necessary setup and passes the control to the user-provided
 // payload.
-GUEST_CODE static void guest_main()
+GUEST_CODE static void guest_main(uint64 size)
+{
+	uint64 addr = ARM64_ADDR_USER_CODE;
+
+	while (size >= sizeof(struct api_call_header)) {
+		struct api_call_header* cmd = (struct api_call_header*)addr;
+		if (cmd->call >= SYZOS_API_STOP)
+			return;
+		if (cmd->size > size)
+			return;
+		switch (cmd->call) {
+		case SYZOS_API_UEXIT: {
+			struct api_call_uexit* ucmd = (struct api_call_uexit*)cmd;
+			guest_uexit(ucmd->exit_code);
+			break;
+		}
+		case SYZOS_API_CODE: {
+			struct api_call_code* ccmd = (struct api_call_code*)cmd;
+			guest_execute_code(ccmd->insns, cmd->size - sizeof(struct api_call_header));
+			break;
+		}
+		}
+		addr += cmd->size;
+		size -= cmd->size;
+	};
+}
+
+GUEST_CODE void guest_execute_code(uint32* insns, uint64 size)
+{
+	volatile void (*fn)() = (volatile void (*)())insns;
+	fn();
+}
+
+// Perform a userspace exit that can be handled by the host.
+// The host returns from ioctl(KVM_RUN) with kvm_run.exit_reason=KVM_EXIT_MMIO,
+// and can handle the call depending on the data passed as exit code.
+GUEST_CODE void guest_uexit(uint64 exit_code)
 {
-	void (*guest_payload)() = (void (*)())ARM64_ADDR_USER_CODE;
-	guest_payload();
+	volatile uint64* ptr = (volatile uint64*)ARM64_ADDR_UEXIT;
+	*ptr = exit_code;
 }
diff --git a/executor/kvm.h b/executor/kvm.h
index 49a493818..1afbcd40e 100644
--- a/executor/kvm.h
+++ b/executor/kvm.h
@@ -76,6 +76,10 @@
 #define NEXT_INSN $0xbadc0de
 #define PREFIX_SIZE 0xba1d
 
+// Write to this page to trigger a page fault and stop KVM_RUN.
+#define ARM64_ADDR_EXIT 0xdddd0000
+// Dedicated address within the exit page for the uexit command.
+#define ARM64_ADDR_UEXIT (ARM64_ADDR_EXIT + 256)
 #define ARM64_ADDR_USER_CODE 0xeeee0000
 #define ARM64_ADDR_EXECUTOR_CODE 0xeeee8000
 #define ARM64_ADDR_EL1_STACK_BOTTOM 0xffff1000