50133: use read-ahead and lseek-rewind for efficient line-buffered input

3 files changed, 86 insertions, 1 deletions

diff --git a/ChangeLog b/ChangeLog
index f5301882a..0faf8fda7 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,9 @@
 2022-04-28  Bart Schaefer  <schaefer@zsh.org>
 
+	* 50133 (Bart, PWS, Jun-ichi): Src/input.c, configure.ac: when
+	lseek(2) is available, use it to check for and rewind read-ahead
+	for more efficient line-buffered input.
+
 	* 50126: Etc/BUGS, Src/exec.c: Fix multios in current-shell "exec"
 
 	* 50101: Src/Modules/system.c: sysread -o with param matches doc
diff --git a/Src/input.c b/Src/input.c
index c59232681..9898a7177 100644
--- a/Src/input.c
+++ b/Src/input.c
@@ -217,12 +217,34 @@ shinbufrestore(void)
 static int
 shingetchar(void)
 {
-    int nread;
+    int nread, rsize = isset(SHINSTDIN) ? 1 : SHINBUFSIZE;
 
     if (shinbufptr < shinbufendptr)
 	return STOUC(*shinbufptr++);
 
     shinbufreset();
+#ifdef USE_LSEEK
+    if (rsize == 1 && lseek(SHIN, 0, SEEK_CUR) != (off_t)-1)
+	rsize = SHINBUFSIZE;
+    if (rsize > 1) {
+	do {
+	    errno = 0;
+	    nread = read(SHIN, shinbuffer, rsize);
+	} while (nread < 0 && errno == EINTR);
+	if (nread <= 0)
+	    return -1;
+	if (isset(SHINSTDIN) &&
+	    (shinbufendptr = memchr(shinbuffer, '\n', nread))) {
+	    shinbufendptr++;
+	    rsize = (shinbufendptr - shinbuffer);
+	    if (nread > rsize &&
+		lseek(SHIN, -(nread - rsize), SEEK_CUR) < 0)
+		zerr("lseek(%d, %d): %e", SHIN, -(nread - rsize), errno);
+	} else
+	    shinbufendptr = shinbuffer + nread;
+	return STOUC(*shinbufptr++);
+    }
+#endif
     for (;;) {
        errno = 0;
        nread = read(SHIN, shinbufendptr, 1);
diff --git a/configure.ac b/configure.ac
index 8bba78c56..c72148d06 100644
--- a/configure.ac
+++ b/configure.ac
@@ -2210,6 +2210,65 @@ if test x$zsh_cv_sys_fifo = xyes; then
 fi
 
 dnl -----------
+dnl check that lseek() correctly reports seekability.
+dnl -----------
+AC_CACHE_CHECK(if lseek() correctly reports seekability,
+zsh_cv_sys_lseek,
+[AC_RUN_IFELSE([AC_LANG_SOURCE([[
+#include <stdio.h>
+#include <sys/types.h>
+#include <fcntl.h>
+#include <unistd.h>
+#include <sys/socket.h>
+#include <sys/stat.h>
+int main() {
+    int pipefd[2], fd;
+    off_t ret;
+    char* tmpfile = "seekfiletest.tmp";
+    if ((fd = open(tmpfile, O_CREAT, S_IRUSR)) < 0) {
+	fprintf(stderr, "creating file failed\n");
+	return 1;
+    }
+    ret = lseek(fd, 0, SEEK_CUR);
+    close(fd);
+    unlink(tmpfile);
+    if (ret == (off_t)-1) {
+	fprintf(stderr, "lseek on regular file failed\n");
+	return 1;
+    }
+    if (pipe(pipefd) < 0) {
+	fprintf(stderr, "creating pipe failed\n");
+	return 1;
+    }
+    write(pipefd[1], "abcdefgh", 8);
+    ret = lseek(pipefd[0], 0, SEEK_CUR);
+    close(pipefd[0]);
+    close(pipefd[1]);
+    if (ret != (off_t)-1) {
+	fprintf(stderr, "lseek on pipe succeeded\n");
+	return 1;
+    }
+    if ((fd = socket(AF_UNIX, SOCK_STREAM, 0)) < 0) {
+	fprintf(stderr, "creating UNIX domain socket failed\n");
+	return 1;
+    }
+    ret = lseek(fd, 0, SEEK_CUR);
+    close(fd);
+    if (ret != (off_t)-1) {
+	fprintf(stderr, "lseek on UNIX domain socket succeeded\n");
+	return 1;
+    }
+    return 0;
+}
+]])],[zsh_cv_sys_lseek=yes],[zsh_cv_sys_lseek=no],[zsh_cv_sys_lseek=yes])
+])
+AH_TEMPLATE([USE_LSEEK],
+[Define to 1 if lseek() can be used for SHIN.])
+if test x$zsh_cv_sys_lseek = xyes; then
+  AC_DEFINE(USE_LSEEK)
+fi
+
+dnl -----------
 dnl test for whether link() works
 dnl for instance, BeOS R4.51 doesn't support hard links yet
 dnl -----------