reuse open(2) from rb_file_load_ok on POSIX-like system

When loading Ruby source files, we can save the result of
successful opens as open(2)/openat(2) are a fairly expensive
syscalls.  This also avoids a time-of-check-to-time-of-use
(TOCTTOU) problem.

This reduces open(2) syscalls during `require'; but should be
most apparent when users have a small $LOAD_PATH.  Users with
large $LOAD_PATH will benefit less since there'll be more
open(2) failures due to ENOENT.

With `strace -c -e openat ruby -e exit' under Linux, this
results in a ~14% reduction of openat(2) syscalls
(glibc uses openat(2) to implement open(2)).

 % time     seconds  usecs/call     calls    errors syscall
 ------ ----------- ----------- --------- --------- ----------------
   0.00    0.000000           0       296       110 openat
   0.00    0.000000           0       254       110 openat

Additionally, the introduction of `struct ruby_file_load_state'
may make future optimizations more apparent.

This change cannot benefit binary (.so) loading since the
dlopen(3) API requires a filename and I'm not aware of an
alternative that takes a pre-existing FD.  In typical
situations, Ruby source files outnumber the mount of .so
files.

file.c

@@ -6359,7 +6359,7 @@ ruby_is_fd_loadable(int fd)
 
 #ifndef _WIN32
 int
-rb_file_load_ok(const char *path)
+rb_file_load_ok(const char *path, struct ruby_file_load_state *fls)
 {
     int ret = 1;
     /*
@@ -6381,10 +6381,21 @@ rb_file_load_ok(const char *path)
     }
     rb_update_max_fd(fd);
     ret = ruby_is_fd_loadable(fd);
-    (void)close(fd);
+
+#if defined(DOSISH) || defined(__CYGWIN__)
+    fls = NULL; /* need to set xflag via open_load_file */
+#endif
+    if (ret && fls) {
+        /* TODO: avoid path object alloc in rb_io_fdopen */
+        fls->filev = rb_io_fdopen(fd, mode, path);
+        fls->is_fifo = ret < 0 ? 1 : 0;
+        fls->is_nonblock = mode == O_RDONLY ? 0 : 1;
+    } else {
+        (void)close(fd);
+    }
     return ret;
 }
-#endif
+#endif /* !_WIN32 */
 
 static int
 is_explicit_relative(const char *path)
@@ -6409,6 +6420,13 @@ copy_path_class(VALUE path, VALUE orig)
 
 int
 rb_find_file_ext(VALUE *filep, const char *const *ext)
+{
+    return ruby_find_file_ext(filep, ext, NULL);
+}
+
+int
+ruby_find_file_ext(VALUE *filep, const char *const *ext,
+                struct ruby_file_load_state *fls)
 {
     const char *f = StringValueCStr(*filep);
     VALUE fname = *filep, load_path, tmp;
@@ -6429,7 +6447,7 @@ rb_find_file_ext(VALUE *filep, const char *const *ext)
         fnlen = RSTRING_LEN(fname);
         for (i=0; ext[i]; i++) {
             rb_str_cat2(fname, ext[i]);
-            if (rb_file_load_ok(RSTRING_PTR(fname))) {
+            if (rb_file_load_ok(RSTRING_PTR(fname), fls)) {
                 *filep = copy_path_class(fname, *filep);
                 return (int)(i+1);
             }
@@ -6454,7 +6472,7 @@ rb_find_file_ext(VALUE *filep, const char *const *ext)
             RB_GC_GUARD(str) = rb_get_path(str);
             if (RSTRING_LEN(str) == 0) continue;
             rb_file_expand_path_internal(fname, str, 0, 0, tmp);
-            if (rb_file_load_ok(RSTRING_PTR(tmp))) {
+            if (rb_file_load_ok(RSTRING_PTR(tmp), fls)) {
                 *filep = copy_path_class(tmp, *filep);
                 return (int)(j+1);
             }
@@ -6468,6 +6486,12 @@ rb_find_file_ext(VALUE *filep, const char *const *ext)
 
 VALUE
 rb_find_file(VALUE path)
+{
+    return ruby_find_file(path, NULL);
+}
+
+VALUE
+ruby_find_file(VALUE path, struct ruby_file_load_state *fls)
 {
     VALUE tmp, load_path;
     const char *f = StringValueCStr(path);
@@ -6481,7 +6505,7 @@ rb_find_file(VALUE path)
     }
 
     if (expanded || rb_is_absolute_path(f) || is_explicit_relative(f)) {
-        if (!rb_file_load_ok(f)) return 0;
+        if (!rb_file_load_ok(f, fls)) return 0;
         if (!expanded)
             path = copy_path_class(file_expand_path_1(path), path);
         return path;
@@ -6499,7 +6523,7 @@ rb_find_file(VALUE path)
             if (RSTRING_LEN(str) > 0) {
                 rb_file_expand_path_internal(path, str, 0, 0, tmp);
                 f = RSTRING_PTR(tmp);
-                if (rb_file_load_ok(f)) goto found;
+                if (rb_file_load_ok(f, fls)) goto found;
             }
         }
         rb_str_resize(tmp, 0);

internal/file.h

@@ -11,6 +11,14 @@
 #include "ruby/ruby.h"          /* for VALUE */
 #include "ruby/encoding.h"      /* for rb_encodinng */
 
+struct ruby_file_load_state {
+    /* TODO: consider stuffing `VALUE fname' here */
+    VALUE filev;
+    unsigned int is_fifo:1;
+    unsigned int is_nonblock:1;
+    /* TODO: DOSISH / __CYGWIN__ maintainer may add xflag here */
+};
+
 /* file.c */
 extern const char ruby_null_device[];
 VALUE rb_home_dir_of(VALUE user, VALUE result);
@@ -18,12 +26,16 @@ VALUE rb_default_home_dir(VALUE result);
 VALUE rb_realpath_internal(VALUE basedir, VALUE path, int strict);
 VALUE rb_check_realpath(VALUE basedir, VALUE path, rb_encoding *origenc);
 void rb_file_const(const char*, VALUE);
-int rb_file_load_ok(const char *);
+int rb_file_load_ok(const char *, struct ruby_file_load_state *);
 VALUE rb_file_expand_path_fast(VALUE, VALUE);
 VALUE rb_file_expand_path_internal(VALUE, VALUE, int, int, VALUE);
 VALUE rb_get_path_check_to_string(VALUE);
 VALUE rb_get_path_check_convert(VALUE);
 int ruby_is_fd_loadable(int fd);
+int ruby_disable_nonblock(int fd);
+int ruby_find_file_ext(VALUE *filep, const char *const *ext,
+                        struct ruby_file_load_state *);
+VALUE ruby_find_file(VALUE path, struct ruby_file_load_state *);
 
 RUBY_SYMBOL_EXPORT_BEGIN
 /* file.c (export) */

internal/parse.h

@@ -10,10 +10,13 @@
  */
 #include "ruby/ruby.h"          /* for VALUE */
 struct rb_iseq_struct;          /* in vm_core.h */
+struct ruby_file_load_state;    /* internal/file.h */
 
 /* parse.y */
 VALUE rb_parser_set_yydebug(VALUE, VALUE);
 void *rb_parser_load_file(VALUE parser, VALUE name);
+void *rb_parser_load_state(VALUE parser, VALUE name,
+                        struct ruby_file_load_state *);
 void rb_parser_keep_script_lines(VALUE vparser);
 void rb_parser_error_tolerant(VALUE vparser);
 void rb_parser_keep_tokens(VALUE vparser);

load.c

@@ -676,7 +676,8 @@ rb_provide(const char *feature)
 NORETURN(static void load_failed(VALUE));
 
 static inline void
-load_iseq_eval(rb_execution_context_t *ec, VALUE fname)
+load_iseq_eval(rb_execution_context_t *ec, VALUE fname,
+                struct ruby_file_load_state *fls)
 {
     const rb_iseq_t *iseq = rb_iseq_load_iseq(fname);
 
@@ -686,7 +687,7 @@ load_iseq_eval(rb_execution_context_t *ec, VALUE fname)
         rb_ast_t *ast;
         VALUE parser = rb_parser_new();
         rb_parser_set_context(parser, NULL, FALSE);
-        ast = (rb_ast_t *)rb_parser_load_file(parser, fname);
+        ast = (rb_ast_t *)rb_parser_load_state(parser, fname, fls);
         iseq = rb_iseq_new_top(&ast->body, rb_fstring_lit("<top (required)>"),
                                fname, rb_realpath_internal(Qnil, fname, 1), NULL);
         rb_ast_dispose(ast);
@@ -698,7 +699,8 @@ load_iseq_eval(rb_execution_context_t *ec, VALUE fname)
 }
 
 static inline enum ruby_tag_type
-load_wrapping(rb_execution_context_t *ec, VALUE fname, VALUE load_wrapper)
+load_wrapping(rb_execution_context_t *ec, VALUE fname, VALUE load_wrapper,
+              struct ruby_file_load_state *fls)
 {
     enum ruby_tag_type state;
     rb_thread_t *th = rb_ec_thread_ptr(ec);
@@ -718,7 +720,7 @@ load_wrapping(rb_execution_context_t *ec, VALUE fname, VALUE load_wrapper)
     EC_PUSH_TAG(ec);
     state = EC_EXEC_TAG();
     if (state == TAG_NONE) {
-        load_iseq_eval(ec, fname);
+        load_iseq_eval(ec, fname, fls);
     }
     EC_POP_TAG();
 
@@ -744,7 +746,7 @@ raise_load_if_failed(rb_execution_context_t *ec, enum ruby_tag_type state)
 }
 
 static void
-rb_load_internal(VALUE fname, VALUE wrap)
+rb_load_internal(VALUE fname, VALUE wrap, struct ruby_file_load_state *fls)
 {
     rb_execution_context_t *ec = GET_EC();
     enum ruby_tag_type state = TAG_NONE;
@@ -752,10 +754,10 @@ rb_load_internal(VALUE fname, VALUE wrap)
         if (!RB_TYPE_P(wrap, T_MODULE)) {
             wrap = rb_module_new();
         }
-        state = load_wrapping(ec, fname, wrap);
+        state = load_wrapping(ec, fname, wrap, fls);
     }
     else {
-        load_iseq_eval(ec, fname);
+        load_iseq_eval(ec, fname, fls);
     }
     raise_load_if_failed(ec, state);
 }
@@ -763,9 +765,10 @@ rb_load_internal(VALUE fname, VALUE wrap)
 void
 rb_load(VALUE fname, int wrap)
 {
-    VALUE tmp = rb_find_file(FilePathValue(fname));
+    struct ruby_file_load_state fls = { .filev = Qfalse };
+    VALUE tmp = ruby_find_file(FilePathValue(fname), &fls);
     if (!tmp) load_failed(fname);
-    rb_load_internal(tmp, RBOOL(wrap));
+    rb_load_internal(tmp, RBOOL(wrap), &fls);
 }
 
 void
@@ -817,6 +820,7 @@ static VALUE
 rb_f_load(int argc, VALUE *argv, VALUE _)
 {
     VALUE fname, wrap, path, orig_fname;
+    struct ruby_file_load_state fls = { .filev = Qfalse };
 
     rb_scan_args(argc, argv, "11", &fname, &wrap);
 
@@ -824,13 +828,13 @@ rb_f_load(int argc, VALUE *argv, VALUE _)
     fname = rb_str_encode_ospath(orig_fname);
     RUBY_DTRACE_HOOK(LOAD_ENTRY, RSTRING_PTR(orig_fname));
 
-    path = rb_find_file(fname);
+    path = ruby_find_file(fname, &fls);
     if (!path) {
-        if (!rb_file_load_ok(RSTRING_PTR(fname)))
+        if (!rb_file_load_ok(RSTRING_PTR(fname), &fls))
             load_failed(orig_fname);
         path = fname;
     }
-    rb_load_internal(path, wrap);
+    rb_load_internal(path, wrap, &fls);
 
     RUBY_DTRACE_HOOK(LOAD_RETURN, RSTRING_PTR(orig_fname));
 
@@ -961,7 +965,8 @@ rb_f_require_relative(VALUE obj, VALUE fname)
 typedef int (*feature_func)(rb_vm_t *vm, const char *feature, const char *ext, int rb, int expanded, const char **fn);
 
 static int
-search_required(rb_vm_t *vm, VALUE fname, volatile VALUE *path, feature_func rb_feature_p)
+search_required(rb_vm_t *vm, VALUE fname, volatile VALUE *path,
+                feature_func rb_feature_p, struct ruby_file_load_state *fls)
 {
     VALUE tmp;
     char *ext, *ftptr;
@@ -976,7 +981,7 @@ search_required(rb_vm_t *vm, VALUE fname, volatile VALUE *path, feature_func rb_
                 if (loading) *path = rb_filesystem_str_new_cstr(loading);
                 return 'r';
             }
-            if ((tmp = rb_find_file(fname)) != 0) {
+            if ((tmp = ruby_find_file(fname, fls)) != 0) {
                 ext = strrchr(ftptr = RSTRING_PTR(tmp), '.');
                 if (!rb_feature_p(vm, ftptr, ext, TRUE, TRUE, &loading) || loading)
                     *path = tmp;
@@ -1017,7 +1022,7 @@ search_required(rb_vm_t *vm, VALUE fname, volatile VALUE *path, feature_func rb_
         return 'r';
     }
     tmp = fname;
-    type = rb_find_file_ext(&tmp, ft == 's' ? ruby_ext : loadable_ext);
+    type = ruby_find_file_ext(&tmp, ft == 's' ? ruby_ext : loadable_ext, fls);
 #if EXTSTATIC
     if (!ft && type != 1) { // not already a feature and not found as a dynamic library
         VALUE lookup_name = tmp;
@@ -1103,7 +1108,7 @@ rb_resolve_feature_path(VALUE klass, VALUE fname)
 
     fname = rb_get_path(fname);
     path = rb_str_encode_ospath(fname);
-    found = search_required(GET_VM(), path, &path, no_feature_p);
+    found = search_required(GET_VM(), path, &path, no_feature_p, NULL);
 
     switch (found) {
       case 'r':
@@ -1163,6 +1168,7 @@ require_internal(rb_execution_context_t *ec, VALUE fname, int exception, bool wa
     VALUE realpaths = get_loaded_features_realpaths(th->vm);
     volatile bool reset_ext_config = false;
     struct rb_ext_config prev_ext_config;
+    struct ruby_file_load_state fls = { .filev = Qfalse };
 
     fname = rb_get_path(fname);
     path = rb_str_encode_ospath(fname);
@@ -1177,7 +1183,7 @@ require_internal(rb_execution_context_t *ec, VALUE fname, int exception, bool wa
         int found;
 
         RUBY_DTRACE_HOOK(FIND_REQUIRE_ENTRY, RSTRING_PTR(fname));
-        found = search_required(th->vm, path, &saved_path, rb_feature_p);
+        found = search_required(th->vm, path, &saved_path, rb_feature_p, &fls);
         RUBY_DTRACE_HOOK(FIND_REQUIRE_RETURN, RSTRING_PTR(fname));
         path = saved_path;
 
@@ -1200,10 +1206,14 @@ require_internal(rb_execution_context_t *ec, VALUE fname, int exception, bool wa
             else {
                 switch (found) {
                   case 'r':
-                    load_iseq_eval(ec, path);
+                    load_iseq_eval(ec, path, &fls);
                     break;
 
                   case 's':
+                    if (fls.filev != Qfalse) {
+                        rb_io_close(fls.filev);
+                        fls.filev = Qfalse;
+                    }
                     reset_ext_config = true;
                     ext_config_push(th, &prev_ext_config);
                     handle = (long)rb_vm_call_cfunc(rb_vm_top_self(), load_ext,

ruby.c

@@ -2539,6 +2539,29 @@ rb_parser_load_file(VALUE parser, VALUE fname_v)
     return load_file(parser, fname_v, f, 0, &opt);
 }
 
+void *
+rb_parser_load_state(VALUE parser, VALUE fname_v,
+                     struct ruby_file_load_state *fls)
+{
+    if (fls->filev != Qfalse) {
+        ruby_cmdline_options_t opt;
+
+        cmdline_options_init(&opt);
+        /* TODO: xflag for DOSISH || __CYGWIN__ */
+        if (fls->is_nonblock) {
+            struct rb_io_t *fptr;
+
+            RB_IO_POINTER(fls->filev, fptr);
+            disable_nonblock(fptr->fd);
+        }
+        if (fls->is_fifo) {
+            rb_io_wait(fls->filev, RB_INT2NUM(RUBY_IO_READABLE), Qnil);
+        }
+        return load_file(parser, fname_v, fls->filev, 0, &opt);
+    }
+    return rb_parser_load_file(parser, fname_v);
+}
+
 /*
  *  call-seq:
  *     Process.argv0  -> frozen_string

win32/file.c

@@ -10,6 +10,7 @@
 #include <wchar.h>
 #include <shlwapi.h>
 #include "win32/file.h"
+#include "internal/file.h"
 
 #ifndef INVALID_FILE_ATTRIBUTES
 # define INVALID_FILE_ATTRIBUTES ((DWORD)-1)
@@ -592,8 +593,9 @@ rb_file_expand_path_internal(VALUE fname, VALUE dname, int abs_mode, int long_na
     return result;
 }
 
+/* TODO: using @fls is an exercise for a win32 hacker */
 int
-rb_file_load_ok(const char *path)
+rb_file_load_ok(const char *path, struct ruby_file_load_state *fls)
 {
     DWORD attr;
     int ret = 1;