unix: print a message when a fatal signal happens

Print a message for SIGBUS, SIGSEGV, and SIGILL when they happen.
These signals are always fatal, but it's very useful to know which of
them happened.

Also, it prints the location in the binary which can then be parsed by
`tinygo run` (see https://github.com/tinygo-org/tinygo/pull/4383).

While this does add some extra binary size, it's for Linux and MacOS
(systems that typically have plenty of RAM/storage) and could be very
useful when debugging some low-level crash such as a runtime bug.

compileopts/target.go

@@ -388,6 +388,8 @@ func defaultTarget(options *Options) (*TargetSpec, error) {
 			"-arch", llvmarch,
 			"-platform_version", "macos", platformVersion, platformVersion,
 		)
+		spec.ExtraFiles = append(spec.ExtraFiles,
+			"src/runtime/runtime_unix.c")
 	case "linux":
 		spec.Linker = "ld.lld"
 		spec.RTLib = "compiler-rt"
@@ -407,6 +409,8 @@ func defaultTarget(options *Options) (*TargetSpec, error) {
 			// proper threading.
 			spec.CFlags = append(spec.CFlags, "-mno-outline-atomics")
 		}
+		spec.ExtraFiles = append(spec.ExtraFiles,
+			"src/runtime/runtime_unix.c")
 	case "windows":
 		spec.Linker = "ld.lld"
 		spec.Libc = "mingw-w64"

src/runtime/arch_386.go

@@ -9,7 +9,12 @@ const deferExtraRegs = 0
 
 const callInstSize = 5 // "call someFunction" is 5 bytes
 
-const linux_MAP_ANONYMOUS = 0x20
+const (
+	linux_MAP_ANONYMOUS = 0x20
+	linux_SIGBUS        = 7
+	linux_SIGILL        = 4
+	linux_SIGSEGV       = 11
+)
 
 // Align on word boundary.
 func align(ptr uintptr) uintptr {

src/runtime/arch_amd64.go

@@ -9,7 +9,12 @@ const deferExtraRegs = 0
 
 const callInstSize = 5 // "call someFunction" is 5 bytes
 
-const linux_MAP_ANONYMOUS = 0x20
+const (
+	linux_MAP_ANONYMOUS = 0x20
+	linux_SIGBUS        = 7
+	linux_SIGILL        = 4
+	linux_SIGSEGV       = 11
+)
 
 // Align a pointer.
 // Note that some amd64 instructions (like movaps) expect 16-byte aligned

src/runtime/arch_arm.go

@@ -11,7 +11,12 @@ const deferExtraRegs = 0
 
 const callInstSize = 4 // "bl someFunction" is 4 bytes
 
-const linux_MAP_ANONYMOUS = 0x20
+const (
+	linux_MAP_ANONYMOUS = 0x20
+	linux_SIGBUS        = 7
+	linux_SIGILL        = 4
+	linux_SIGSEGV       = 11
+)
 
 // Align on the maximum alignment for this platform (double).
 func align(ptr uintptr) uintptr {

src/runtime/arch_arm64.go

@@ -9,7 +9,12 @@ const deferExtraRegs = 0
 
 const callInstSize = 4 // "bl someFunction" is 4 bytes
 
-const linux_MAP_ANONYMOUS = 0x20
+const (
+	linux_MAP_ANONYMOUS = 0x20
+	linux_SIGBUS        = 7
+	linux_SIGILL        = 4
+	linux_SIGSEGV       = 11
+)
 
 // Align on word boundary.
 func align(ptr uintptr) uintptr {

src/runtime/arch_mips.go

@@ -9,7 +9,12 @@ const deferExtraRegs = 0
 
 const callInstSize = 8 // "jal someFunc" is 4 bytes, plus a MIPS delay slot
 
-const linux_MAP_ANONYMOUS = 0x800
+const (
+	linux_MAP_ANONYMOUS = 0x800
+	linux_SIGBUS        = 10
+	linux_SIGILL        = 4
+	linux_SIGSEGV       = 11
+)
 
 // It appears that MIPS has a maximum alignment of 8 bytes.
 func align(ptr uintptr) uintptr {

src/runtime/arch_mipsle.go

@@ -9,7 +9,12 @@ const deferExtraRegs = 0
 
 const callInstSize = 8 // "jal someFunc" is 4 bytes, plus a MIPS delay slot
 
-const linux_MAP_ANONYMOUS = 0x800
+const (
+	linux_MAP_ANONYMOUS = 0x800
+	linux_SIGBUS        = 10
+	linux_SIGILL        = 4
+	linux_SIGSEGV       = 11
+)
 
 // It appears that MIPS has a maximum alignment of 8 bytes.
 func align(ptr uintptr) uintptr {

src/runtime/os_darwin.go

@@ -22,6 +22,14 @@ const (
 	clock_MONOTONIC_RAW = 4
 )
 
+// Source:
+// https://opensource.apple.com/source/xnu/xnu-7195.141.2/bsd/sys/signal.h.auto.html
+const (
+	sig_SIGBUS  = 10
+	sig_SIGILL  = 4
+	sig_SIGSEGV = 11
+)
+
 // https://opensource.apple.com/source/xnu/xnu-7195.141.2/EXTERNAL_HEADERS/mach-o/loader.h.auto.html
 type machHeader struct {
 	magic      uint32

src/runtime/os_linux.go

@@ -23,6 +23,12 @@ const (
 	clock_MONOTONIC_RAW = 4
 )
 
+const (
+	sig_SIGBUS  = linux_SIGBUS
+	sig_SIGILL  = linux_SIGILL
+	sig_SIGSEGV = linux_SIGSEGV
+)
+
 // For the definition of the various header structs, see:
 // https://refspecs.linuxfoundation.org/elf/elf.pdf
 // Also useful:

src/runtime/runtime_unix.c

@@ -0,0 +1,56 @@
+//go:build none
+
+// This file is included on Darwin and Linux (despite the //go:build line above).
+
+#define _GNU_SOURCE
+#define _XOPEN_SOURCE
+#include <signal.h>
+#include <unistd.h>
+#include <stdint.h>
+#include <ucontext.h>
+#include <string.h>
+
+void tinygo_handle_fatal_signal(int sig, uintptr_t addr);
+
+static void signal_handler(int sig, siginfo_t *info, void *context) {
+	ucontext_t* uctx = context;
+	uintptr_t addr = 0;
+	#if __APPLE__
+		#if __arm64__
+			addr = uctx->uc_mcontext->__ss.__pc;
+		#elif __x86_64__
+			addr = uctx->uc_mcontext->__ss.__rip;
+		#else
+			#error unknown architecture
+		#endif
+	#elif __linux__
+		// Note: this can probably be simplified using the MC_PC macro in musl,
+		// but this works for now.
+		#if __arm__
+			addr = uctx->uc_mcontext.arm_pc;
+		#elif __i386__
+			addr = uctx->uc_mcontext.gregs[REG_EIP];
+		#elif __x86_64__
+			addr = uctx->uc_mcontext.gregs[REG_RIP];
+		#else // aarch64, mips, maybe others
+			addr = uctx->uc_mcontext.pc;
+		#endif
+	#else
+		#error unknown platform
+	#endif
+	tinygo_handle_fatal_signal(sig, addr);
+}
+
+void tinygo_register_fatal_signals(void) {
+	struct sigaction act = { 0 };
+	// SA_SIGINFO:   we want the 2 extra parameters
+	// SA_RESETHAND: only catch the signal once (the handler will re-raise the signal)
+	act.sa_flags = SA_SIGINFO | SA_RESETHAND;
+	act.sa_sigaction = &signal_handler;
+
+	// Register the signal handler for common issues. There are more signals,
+	// which can be added if needed.
+	sigaction(SIGBUS, &act, NULL);
+	sigaction(SIGILL, &act, NULL);
+	sigaction(SIGSEGV, &act, NULL);
+}

src/runtime/runtime_unix.go

@@ -26,6 +26,9 @@ func abort()
 //export exit
 func exit(code int)
 
+//export raise
+func raise(sig int32)
+
 //export clock_gettime
 func libc_clock_gettime(clk_id int32, ts *timespec)
 
@@ -74,6 +77,10 @@ func main(argc int32, argv *unsafe.Pointer) int {
 	main_argc = argc
 	main_argv = argv
 
+	// Register some fatal signals, so that we can print slightly better error
+	// messages.
+	tinygo_register_fatal_signals()
+
 	// Obtain the initial stack pointer right before calling the run() function.
 	// The run function has been moved to a separate (non-inlined) function so
 	// that the correct stack pointer is read.
@@ -119,6 +126,50 @@ func runMain() {
 	run()
 }
 
+//export tinygo_register_fatal_signals
+func tinygo_register_fatal_signals()
+
+// Print fatal errors when they happen, including the instruction location.
+// With the particular formatting below, `tinygo run` can extract the location
+// where the signal happened and try to show the source location based on DWARF
+// information.
+//
+//export tinygo_handle_fatal_signal
+func tinygo_handle_fatal_signal(sig int32, addr uintptr) {
+	if panicStrategy() == panicStrategyTrap {
+		trap()
+	}
+
+	// Print signal including the faulting instruction.
+	if addr != 0 {
+		printstring("panic: runtime error at ")
+		printptr(addr)
+	} else {
+		printstring("panic: runtime error")
+	}
+	printstring(": caught signal ")
+	switch sig {
+	case sig_SIGBUS:
+		println("SIGBUS")
+	case sig_SIGILL:
+		println("SIGILL")
+	case sig_SIGSEGV:
+		println("SIGSEGV")
+	default:
+		println(sig)
+	}
+
+	// TODO: it might be interesting to also print the invalid address for
+	// SIGSEGV and SIGBUS.
+
+	// Do *not* abort here, instead raise the same signal again. The signal is
+	// registered with SA_RESETHAND which means it executes only once. So when
+	// we raise the signal again below, the signal isn't handled specially but
+	// is handled in the default way (probably exiting the process, maybe with a
+	// core dump).
+	raise(sig)
+}
+
 //go:extern environ
 var environ *unsafe.Pointer