Merging upstream version 0.0~git20240411.9e8c542.

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

14 files changed, 216 insertions, 31 deletions

diff --git a/libc-top-half/musl/arch/wasm32/bits/setjmp.h b/libc-top-half/musl/arch/wasm32/bits/setjmp.h
new file mode 100644
index 0000000..63973a8
--- /dev/null
+++ b/libc-top-half/musl/arch/wasm32/bits/setjmp.h
@@ -0,0 +1 @@
+typedef unsigned long __jmp_buf[8];
diff --git a/libc-top-half/musl/include/dlfcn.h b/libc-top-half/musl/include/dlfcn.h
index 13ab71d..e802de8 100644
--- a/libc-top-half/musl/include/dlfcn.h
+++ b/libc-top-half/musl/include/dlfcn.h
@@ -12,19 +12,29 @@ extern "C" {
 #define RTLD_NOLOAD 4
 #define RTLD_NODELETE 4096
 #define RTLD_GLOBAL 256
+#ifdef __wasilibc_unmodified_upstream
 #define RTLD_LOCAL  0
+#else
+/* For WASI, we give `RTLD_LOCAL` a non-zero value, avoiding ambiguity and
+ * allowing us to defer the decision of whether `RTLD_LOCAL` or `RTLD_GLOBAL`
+ * should be the default when neither is specified.
+ */
+#define RTLD_LOCAL  8
+#endif
 
 #define RTLD_NEXT    ((void *)-1)
 #define RTLD_DEFAULT ((void *)0)
 
+#ifdef __wasilibc_unmodified_upstream
 #define RTLD_DI_LINKMAP 2
+#endif
 
 int    dlclose(void *);
 char  *dlerror(void);
 void  *dlopen(const char *, int);
 void  *dlsym(void *__restrict, const char *__restrict);
 
-#if defined(_GNU_SOURCE) || defined(_BSD_SOURCE)
+#if defined(__wasilibc_unmodified_upstream) && (defined(_GNU_SOURCE) || defined(_BSD_SOURCE))
 typedef struct {
 	const char *dli_fname;
 	void *dli_fbase;
diff --git a/libc-top-half/musl/include/netdb.h b/libc-top-half/musl/include/netdb.h
index d096c78..f9797bc 100644
--- a/libc-top-half/musl/include/netdb.h
+++ b/libc-top-half/musl/include/netdb.h
@@ -118,8 +118,13 @@ struct hostent *gethostbyaddr (const void *, socklen_t, int);
 #ifdef __GNUC__
 __attribute__((const))
 #endif
+#ifdef __wasilibc_unmodified_upstream
 int *__h_errno_location(void);
 #define h_errno (*__h_errno_location())
+#elif (defined __wasilibc_use_wasip2)
+extern _Thread_local int h_errno;
+#define h_errno h_errno
+#endif
 #define HOST_NOT_FOUND 1
 #define TRY_AGAIN      2
 #define NO_RECOVERY    3
diff --git a/libc-top-half/musl/include/pthread.h b/libc-top-half/musl/include/pthread.h
index 05101e8..2c35d0b 100644
--- a/libc-top-half/musl/include/pthread.h
+++ b/libc-top-half/musl/include/pthread.h
@@ -97,7 +97,9 @@ int pthread_equal(pthread_t, pthread_t);
 int pthread_setcancelstate(int, int *);
 int pthread_setcanceltype(int, int *);
 void pthread_testcancel(void);
+#ifdef __wasilibc_unmodified_upstream /* WASI has no cancellation support. */
 int pthread_cancel(pthread_t);
+#endif
 
 #ifdef __wasilibc_unmodified_upstream /* WASI has no CPU scheduling support. */
 int pthread_getschedparam(pthread_t, int *__restrict, struct sched_param *__restrict);
diff --git a/libc-top-half/musl/include/setjmp.h b/libc-top-half/musl/include/setjmp.h
index f505f8e..b57999e 100644
--- a/libc-top-half/musl/include/setjmp.h
+++ b/libc-top-half/musl/include/setjmp.h
@@ -7,7 +7,12 @@ extern "C" {
 
 #include <features.h>
 
-#ifdef __wasilibc_unmodified_upstream /* WASI has no setjmp */
+#ifndef __wasilibc_unmodified_upstream
+/* WASI has no setjmp */
+#if !defined(__wasm_exception_handling__)
+#error Setjmp/longjmp support requires Exception handling support, which is [not yet standardized](https://github.com/WebAssembly/proposals?tab=readme-ov-file#phase-3---implementation-phase-cg--wg). To enable it, compile with `-mllvm -wasm-enable-sjlj` and use an engine that implements the Exception handling proposal.
+#endif
+#endif
 #include <bits/setjmp.h>
 
 typedef struct __jmp_buf_tag {
@@ -40,9 +45,6 @@ int setjmp (jmp_buf) __setjmp_attr;
 _Noreturn void longjmp (jmp_buf, int);
 
 #define setjmp setjmp
-#else
-#warning setjmp is not yet implemented for WASI
-#endif
 
 #undef __setjmp_attr
 
diff --git a/libc-top-half/musl/include/stdlib.h b/libc-top-half/musl/include/stdlib.h
index 1bcb9ab..553e9ae 100644
--- a/libc-top-half/musl/include/stdlib.h
+++ b/libc-top-half/musl/include/stdlib.h
@@ -128,9 +128,7 @@ int rand_r (unsigned *);
 
 #if defined(_XOPEN_SOURCE) || defined(_GNU_SOURCE) \
  || defined(_BSD_SOURCE)
-#ifdef __wasilibc_unmodified_upstream /* WASI has no absolute paths */
 char *realpath (const char *__restrict, char *__restrict);
-#endif
 long int random (void);
 void srandom (unsigned int);
 char *initstate (unsigned int, char *, size_t);
diff --git a/libc-top-half/musl/include/sys/socket.h b/libc-top-half/musl/include/sys/socket.h
index 4d574c6..29189e3 100644
--- a/libc-top-half/musl/include/sys/socket.h
+++ b/libc-top-half/musl/include/sys/socket.h
@@ -1,5 +1,8 @@
 #ifndef	_SYS_SOCKET_H
 #define	_SYS_SOCKET_H
+
+#include <__wasi_snapshot.h>
+
 #ifdef __wasilibc_unmodified_upstream /* Use alternate WASI libc headers */
 #else
 #include <__header_sys_socket.h>
@@ -395,30 +398,33 @@ struct sockaddr_storage {
 #include <__struct_sockaddr_storage.h>
 #endif
 
-#ifdef __wasilibc_unmodified_upstream /* WASI has no socket/socketpair */
+#if (defined __wasilibc_unmodified_upstream) || (defined __wasilibc_use_wasip2)
 int socket (int, int, int);
+#endif
+
+#ifdef __wasilibc_unmodified_upstream /* WASI has no socketpair */
 int socketpair (int, int, int, int [2]);
 #endif
 
 int shutdown (int, int);
 
-#ifdef __wasilibc_unmodified_upstream /* WASI has no bind/connect/listen/accept */
-int bind (int, const struct sockaddr *, socklen_t);
+#if (defined __wasilibc_unmodified_upstream) || (defined __wasilibc_use_wasip2)
 int connect (int, const struct sockaddr *, socklen_t);
+int bind (int, const struct sockaddr *, socklen_t);
 int listen (int, int);
 #endif
 
 int accept (int, struct sockaddr *__restrict, socklen_t *__restrict);
 int accept4(int, struct sockaddr *__restrict, socklen_t *__restrict, int);
 
-#ifdef __wasilibc_unmodified_upstream /* WASI has no getsockname/getpeername */
+#if (defined __wasilibc_unmodified_upstream) || (defined __wasilibc_use_wasip2)
 int getsockname (int, struct sockaddr *__restrict, socklen_t *__restrict);
 int getpeername (int, struct sockaddr *__restrict, socklen_t *__restrict);
 #endif
 
 ssize_t send (int, const void *, size_t, int);
 ssize_t recv (int, void *, size_t, int);
-#ifdef __wasilibc_unmodified_upstream /* WASI has no sendto/recvfrom */
+#if (defined __wasilibc_unmodified_upstream) || (defined __wasilibc_use_wasip2)
 ssize_t sendto (int, const void *, size_t, int, const struct sockaddr *, socklen_t);
 ssize_t recvfrom (int, void *__restrict, size_t, int, struct sockaddr *__restrict, socklen_t *__restrict);
 #endif
@@ -428,7 +434,7 @@ ssize_t recvmsg (int, struct msghdr *, int);
 #endif
 
 int getsockopt (int, int, int, void *__restrict, socklen_t *__restrict);
-#ifdef __wasilibc_unmodified_upstream /* WASI has no setsockopt */
+#if (defined __wasilibc_unmodified_upstream) || (defined __wasilibc_use_wasip2)
 int setsockopt (int, int, int, const void *, socklen_t);
 #endif
 
diff --git a/libc-top-half/musl/include/sys/stat.h b/libc-top-half/musl/include/sys/stat.h
index 72e1626..c0090f7 100644
--- a/libc-top-half/musl/include/sys/stat.h
+++ b/libc-top-half/musl/include/sys/stat.h
@@ -78,11 +78,9 @@ int stat(const char *__restrict, struct stat *__restrict);
 int fstat(int, struct stat *);
 int lstat(const char *__restrict, struct stat *__restrict);
 int fstatat(int, const char *__restrict, struct stat *__restrict, int);
-#ifdef __wasilibc_unmodified_upstream /* WASI has no chmod */
 int chmod(const char *, mode_t);
 int fchmod(int, mode_t);
 int fchmodat(int, const char *, mode_t, int);
-#endif
 #ifdef __wasilibc_unmodified_upstream /* WASI has no umask */
 mode_t umask(mode_t);
 #endif
diff --git a/libc-top-half/musl/src/env/__init_tls.c b/libc-top-half/musl/src/env/__init_tls.c
index c3e407c..4f4c221 100644
--- a/libc-top-half/musl/src/env/__init_tls.c
+++ b/libc-top-half/musl/src/env/__init_tls.c
@@ -31,25 +31,35 @@ extern unsigned char __global_base;
 extern weak unsigned char __stack_high;
 extern weak unsigned char __stack_low;
 
-static inline void setup_default_stack_size()
+struct stack_bounds {
+	void *base;
+	size_t size;
+};
+
+static inline struct stack_bounds get_stack_bounds()
 {
-	ptrdiff_t stack_size;
+	struct stack_bounds bounds;
 
-	if (&__stack_high)
-		stack_size = &__stack_high - &__stack_low;
-	else {
+	if (&__stack_high) {
+		bounds.base = &__stack_high;
+		bounds.size = &__stack_high - &__stack_low;
+	} else {
 		unsigned char *sp;
 		__asm__(
 			".globaltype __stack_pointer, i32\n"
 			"global.get __stack_pointer\n"
 			"local.set %0\n"
 			: "=r"(sp));
-		stack_size = sp > &__global_base ? &__heap_base - &__data_end : (ptrdiff_t)&__global_base;
+		if (sp > &__global_base) {
+			bounds.base = &__heap_base;
+			bounds.size = &__heap_base - &__data_end;
+		} else {
+			bounds.base = &__global_base;
+			bounds.size = (size_t)&__global_base;
+		}
 	}
 
-	__default_stacksize =
-		stack_size < DEFAULT_STACK_MAX ?
-		stack_size : DEFAULT_STACK_MAX;
+	return bounds;
 }
 
 void __wasi_init_tp() {
@@ -68,8 +78,14 @@ int __init_tp(void *p)
 	td->detach_state = DT_JOINABLE;
 	td->tid = __syscall(SYS_set_tid_address, &__thread_list_lock);
 #else
-	setup_default_stack_size();
+	struct stack_bounds bounds = get_stack_bounds();
+	__default_stacksize =
+		bounds.size < DEFAULT_STACK_MAX ?
+		bounds.size : DEFAULT_STACK_MAX;
 	td->detach_state = DT_JOINABLE;
+	td->stack = bounds.base;
+	td->stack_size = bounds.size;
+	td->guard_size = 0;
 	/*
 	 * Initialize the TID to a value which doesn't conflict with
 	 * host-allocated TIDs, so that TID-based locks can work.
diff --git a/libc-top-half/musl/src/internal/locale_impl.h b/libc-top-half/musl/src/internal/locale_impl.h
index 7f79b7f..4649a43 100644
--- a/libc-top-half/musl/src/internal/locale_impl.h
+++ b/libc-top-half/musl/src/internal/locale_impl.h
@@ -28,7 +28,15 @@ extern hidden const struct __locale_struct __c_dot_utf8_locale;
 hidden const struct __locale_map *__get_locale(int, const char *);
 hidden const char *__mo_lookup(const void *, size_t, const char *);
 hidden const char *__lctrans(const char *, const struct __locale_map *);
+#ifdef __wasilibc_unmodified_upstream
 hidden const char *__lctrans_cur(const char *);
+#else
+// We make this visible in the wasi-libc build because
+// libwasi-emulated-signal.so needs to import it from libc.so.  If we ever
+// decide to merge libwasi-emulated-signal.so into libc.so, this will no longer
+// be necessary.
+const char *__lctrans_cur(const char *);
+#endif
 hidden const char *__lctrans_impl(const char *, const struct __locale_map *);
 hidden int __loc_is_allocated(locale_t);
 hidden char *__gettextdomain(void);
diff --git a/libc-top-half/musl/src/misc/dl.c b/libc-top-half/musl/src/misc/dl.c
new file mode 100644
index 0000000..266594f
--- /dev/null
+++ b/libc-top-half/musl/src/misc/dl.c
@@ -0,0 +1,45 @@
+/* This file is used to build libdl.so with stub versions of `dlopen`, `dlsym`,
+ * etc.  The intention is that this stubbed libdl.so can be used to build
+ * libraries and applications which use `dlopen` without committing to a
+ * specific runtime implementation.  Later, it can be replaced with a real,
+ * working libdl.so (e.g. at runtime or component composition time).
+ * 
+ * For example, the `wasm-tools component link` subcommand can be used to create
+ * a component that bundles any `dlopen`-able libraries in such a way that their
+ * function exports can be resolved symbolically at runtime using an
+ * implementation of libdl.so designed for that purpose.  In other cases, a
+ * runtime might provide Emscripten-style dynamic linking via URLs or else a
+ * more traditional, filesystem-based implementation.  Finally, even this
+ * stubbed version of libdl.so can be used at runtime in cases where dynamic
+ * library resolution cannot or should not be supported (and the application can
+ * handle this situation gracefully). */
+
+#include <stddef.h>
+#include <dlfcn.h>
+
+static const char *error = NULL;
+
+weak int dlclose(void *library)
+{
+	error = "dlclose not implemented";
+	return -1;
+}
+
+weak char *dlerror(void)
+{
+	const char *var = error;
+	error = NULL;
+	return (char*) var;
+}
+
+weak void *dlopen(const char *name, int flags)
+{
+	error = "dlopen not implemented";
+	return NULL;
+}
+
+weak void *dlsym(void *library, const char *name)
+{
+	error = "dlsym not implemented";
+	return NULL;
+}
diff --git a/libc-top-half/musl/src/setjmp/wasm32/rt.c b/libc-top-half/musl/src/setjmp/wasm32/rt.c
new file mode 100644
index 0000000..24e4e33
--- /dev/null
+++ b/libc-top-half/musl/src/setjmp/wasm32/rt.c
@@ -0,0 +1,83 @@
+/*
+ * a runtime implementation for
+ * https://github.com/llvm/llvm-project/pull/84137
+ * https://docs.google.com/document/d/1ZvTPT36K5jjiedF8MCXbEmYjULJjI723aOAks1IdLLg/edit
+ */
+
+#include <stddef.h>
+#include <stdint.h>
+
+/*
+ * function prototypes
+ */
+void __wasm_setjmp(void *env, uint32_t label, void *func_invocation_id);
+uint32_t __wasm_setjmp_test(void *env, void *func_invocation_id);
+void __wasm_longjmp(void *env, int val);
+
+/*
+ * jmp_buf should have large enough size and alignment to contain
+ * this structure.
+ */
+struct jmp_buf_impl {
+        void *func_invocation_id;
+        uint32_t label;
+
+        /*
+         * this is a temorary storage used by the communication between
+         * __wasm_sjlj_longjmp and WebAssemblyLowerEmscriptenEHSjL-generated
+         * logic.
+         * ideally, this can be replaced with multivalue.
+         */
+        struct arg {
+                void *env;
+                int val;
+        } arg;
+};
+
+void
+__wasm_setjmp(void *env, uint32_t label, void *func_invocation_id)
+{
+        struct jmp_buf_impl *jb = env;
+        if (label == 0) { /* ABI contract */
+                __builtin_trap();
+        }
+        if (func_invocation_id == NULL) { /* sanity check */
+                __builtin_trap();
+        }
+        jb->func_invocation_id = func_invocation_id;
+        jb->label = label;
+}
+
+uint32_t
+__wasm_setjmp_test(void *env, void *func_invocation_id)
+{
+        struct jmp_buf_impl *jb = env;
+        if (jb->label == 0) { /* ABI contract */
+                __builtin_trap();
+        }
+        if (func_invocation_id == NULL) { /* sanity check */
+                __builtin_trap();
+        }
+        if (jb->func_invocation_id == func_invocation_id) {
+                return jb->label;
+        }
+        return 0;
+}
+
+void
+__wasm_longjmp(void *env, int val)
+{
+        struct jmp_buf_impl *jb = env;
+        struct arg *arg = &jb->arg;
+        /*
+         * C standard says:
+         * The longjmp function cannot cause the setjmp macro to return
+         * the value 0; if val is 0, the setjmp macro returns the value 1.
+         */
+        if (val == 0) {
+                val = 1;
+        }
+        arg->env = env;
+        arg->val = val;
+        __builtin_wasm_throw(1, arg); /* 1 == C_LONGJMP */
+}
diff --git a/libc-top-half/musl/src/thread/pthread_create.c b/libc-top-half/musl/src/thread/pthread_create.c
index 5de9f5a..450fe15 100644
--- a/libc-top-half/musl/src/thread/pthread_create.c
+++ b/libc-top-half/musl/src/thread/pthread_create.c
@@ -13,6 +13,7 @@
 #endif
 
 #include <stdalign.h>
+#include <assert.h>
 
 static void dummy_0()
 {
@@ -558,13 +559,17 @@ int __pthread_create(pthread_t *restrict res, const pthread_attr_t *restrict att
 			__wait(&args->control, 0, 3, 0);
 	}
 #else
+#define WASI_THREADS_MAX_TID 0x1FFFFFFF
 	/* `wasi_thread_spawn` will either return a host-provided thread ID (TID)
-	 * (`>= 0`) or an error code (`< 0`). As in the unmodified version, all
-	 * spawn failures translate to EAGAIN; unlike the modified version, there is
-	 * no need to "start up" the child thread--the host does this. If the spawn
-	 * did succeed, then we store the TID atomically, since this parent thread
-	 * is racing with the child thread to set this field; this way, whichever
-	 * thread reaches this point first can continue without waiting. */
+	 * (`<1, 0x1FFFFFFF>`) or an error code (`< 0`). Please note that `0` is
+	 * reserved for compatibility reasons and must not be returned by the runtime.
+	 * As in the unmodified version, all spawn failures translate to EAGAIN;
+	 * unlike the modified version, there is no need to "start up" the child
+	 * thread--the host does this. If the spawn did succeed, then we store the
+	 * TID atomically, since this parent thread is racing with the child thread
+	 * to set this field; this way, whichever thread reaches this point first
+	 * can continue without waiting. */
+	assert(ret != 0 && ret <= WASI_THREADS_MAX_TID);
 	if (ret < 0) {
 		ret = -EAGAIN;
 	} else {
diff --git a/libc-top-half/musl/src/thread/pthread_getattr_np.c b/libc-top-half/musl/src/thread/pthread_getattr_np.c
index 2881831..c23e5d7 100644
--- a/libc-top-half/musl/src/thread/pthread_getattr_np.c
+++ b/libc-top-half/musl/src/thread/pthread_getattr_np.c
@@ -1,7 +1,9 @@
 #define _GNU_SOURCE
 #include "pthread_impl.h"
 #include "libc.h"
+#ifdef __wasilibc_unmodified_upstream
 #include <sys/mman.h>
+#endif
 
 int pthread_getattr_np(pthread_t t, pthread_attr_t *a)
 {
@@ -12,6 +14,7 @@ int pthread_getattr_np(pthread_t t, pthread_attr_t *a)
 		a->_a_stackaddr = (uintptr_t)t->stack;
 		a->_a_stacksize = t->stack_size;
 	} else {
+#ifdef __wasilibc_unmodified_upstream
 		char *p = (void *)libc.auxv;
 		size_t l = PAGE_SIZE;
 		p += -(uintptr_t)p & PAGE_SIZE-1;
@@ -19,6 +22,9 @@ int pthread_getattr_np(pthread_t t, pthread_attr_t *a)
 		while (mremap(p-l-PAGE_SIZE, PAGE_SIZE, 2*PAGE_SIZE, 0)==MAP_FAILED && errno==ENOMEM)
 			l += PAGE_SIZE;
 		a->_a_stacksize = l;
+#else
+		return ENOSYS;
+#endif
 	}
 	return 0;
 }