diff mbox series

[v3,2/3] input: Use lseek on stdin when possible

Message ID 9a8cc9e122ed6673281e21fc2581dd72eb6034c6.1717291846.git.herbert@gondor.apana.org.au (mailing list archive)
State Accepted
Delegated to: Herbert Xu
Headers show
Series Improve performance when reading stdin | expand

Commit Message

Herbert Xu June 2, 2024, 1:31 a.m. UTC
For files that can be sought, use lseek instead of reading one
byte at a time.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
---
 src/eval.c    |   3 ++
 src/init.h    |   1 +
 src/input.c   | 108 +++++++++++++++++++++++++++++++++++---------------
 src/input.h   |   4 ++
 src/jobs.c    |   3 ++
 src/mkinit.c  |   6 +++
 src/options.c |   2 +-
 src/redir.c   |  26 +++++++-----
 src/trap.c    |   1 +
 9 files changed, 111 insertions(+), 43 deletions(-)
diff mbox series

Patch

diff --git a/src/eval.c b/src/eval.c
index 140a734..ad2aa03 100644
--- a/src/eval.c
+++ b/src/eval.c
@@ -907,6 +907,8 @@  bail:
 		goto bail;
 
 	default:
+		flush_input();
+
 		/* Fork off a child process if necessary. */
 		if (!(flags & EV_EXIT) || have_traps()) {
 			INTOFF;
@@ -1142,6 +1144,7 @@  execcmd(int argc, char **argv)
 		iflag = 0;		/* exit on error */
 		mflag = 0;
 		optschanged();
+		flush_input();
 		shellexec(argv + 1, pathval(), 0);
 	}
 	return 0;
diff --git a/src/init.h b/src/init.h
index 4f98b5d..e117895 100644
--- a/src/init.h
+++ b/src/init.h
@@ -39,4 +39,5 @@  union node;
 void init(void);
 void exitreset(void);
 void forkreset(union node *);
+void postexitreset(void);
 void reset(void);
diff --git a/src/input.c b/src/input.c
index 193235d..b84ecec 100644
--- a/src/input.c
+++ b/src/input.c
@@ -32,11 +32,13 @@ 
  * SUCH DAMAGE.
  */
 
-#include <stdio.h>	/* defines BUFSIZ */
 #include <fcntl.h>
-#include <unistd.h>
+#include <stdbool.h>
+#include <stdio.h>	/* defines BUFSIZ */
 #include <stdlib.h>
 #include <string.h>
+#include <termios.h>
+#include <unistd.h>
 
 /*
  * This file implements the input routines used by the parser.
@@ -59,12 +61,21 @@ 
 
 #define IBUFSIZ (BUFSIZ + PUNGETC_MAX + 1)
 
+struct stdin_state {
+	tcflag_t canon;
+	off_t seekable;
+	struct termios tios;
+};
 
 MKINIT struct parsefile basepf;	/* top level input file */
 MKINIT char basebuf[IBUFSIZ];	/* buffer for top level input file */
 MKINIT struct parsefile *toppf = &basepf;
+MKINIT struct stdin_state stdin_state;
 struct parsefile *parsefile = &basepf;	/* current input file */
 int whichprompt;		/* 1 == PS1, 2 == PS2 */
+int stdin_istty;
+
+MKINIT void input_init(void);
 
 STATIC void pushfile(void);
 static void popstring(void);
@@ -74,6 +85,7 @@  static int preadbuffer(void);
 
 #ifdef mkinit
 INCLUDE <stdio.h>
+INCLUDE <termios.h>
 INCLUDE <unistd.h>
 INCLUDE "input.h"
 INCLUDE "error.h"
@@ -82,6 +94,8 @@  INCLUDE "syntax.h"
 INIT {
 	basepf.nextc = basepf.buf = basebuf;
 	basepf.linno = 1;
+
+	input_init();
 }
 
 RESET {
@@ -104,8 +118,32 @@  FORKRESET {
 		parsefile->fd = 0;
 	}
 }
+
+POSTEXITRESET {
+	flush_input();
+}
 #endif
 
+void input_init(void)
+{
+	struct stdin_state *st = &stdin_state;
+	int istty;
+
+	istty = tcgetattr(0, &st->tios) + 1;
+	st->seekable = istty ? 0 : lseek(0, 0, SEEK_CUR) + 1;
+	st->canon = istty ? st->tios.c_lflag & ICANON : 0;
+	stdin_istty = istty;
+}
+
+static bool stdin_bufferable(void)
+{
+	struct stdin_state *st = &stdin_state;
+
+	if (stdin_istty < 0)
+		input_init();
+
+	return st->canon || st->seekable;
+}
 
 static void freestrings(struct strpush *sp)
 {
@@ -191,6 +229,7 @@  static int
 preadfd(void)
 {
 	char *buf = parsefile->buf;
+	int fd = parsefile->fd;
 	int unget;
 	int pnr;
 	int nr;
@@ -214,7 +253,7 @@  preadfd(void)
 retry:
 	nr = pnr;
 #ifndef SMALL
-	if (parsefile->fd == 0 && el) {
+	if (fd == 0 && el) {
 		static const char *rl_cp;
 		static int el_len;
 
@@ -237,38 +276,23 @@  retry:
 				rl_cp = 0;
 		}
 
-	} else
-#endif
-	if (parsefile->fd)
-		nr = read(parsefile->fd, buf, nr);
-	else {
-		nr = 0;
-
-		do {
-			int err;
-
-			err = read(0, buf, 1);
-			if (err <= 0) {
-				if (nr)
-					break;
-
-				nr = err;
-				if (errno != EWOULDBLOCK)
-					break;
-				if (stdin_clear_nonblock() < 0)
-					break;
-
-				out2str("sh: turning off NDELAY mode\n");
-				goto retry;
-			}
-
-			nr++;
-		} while (0);
+		return nr;
 	}
+#endif
+
+	if (!fd && !stdin_bufferable())
+		nr = 1;
+
+	nr = read(fd, buf, nr);
 
 	if (nr < 0) {
 		if (errno == EINTR && !(basepf.prev && pending_sig))
 			goto retry;
+		if (fd == 0 && errno == EWOULDBLOCK &&
+		    stdin_clear_nonblock() >= 0) {
+			out2str("sh: turning off NDELAY mode\n");
+			goto retry;
+		}
 	}
 	return nr;
 }
@@ -302,6 +326,8 @@  static int preadbuffer(void)
 	something = !first;
 
 	more = input_get_lleft(parsefile);
+
+	INTOFF;
 	if (more <= 0) {
 		int nr;
 
@@ -313,6 +339,7 @@  again:
 			input_set_lleft(parsefile, parsefile->nleft = 0);
 			if (!IS_DEFINED_SMALL && nr > 0)
 				goto save;
+			INTON;
 			return PEOF;
 		}
 	}
@@ -365,11 +392,10 @@  save:
 
 	if (parsefile->fd == 0 && hist && something) {
 		HistEvent he;
-		INTOFF;
 		history(hist, &he, first ? H_ENTER : H_APPEND,
 			parsefile->nextc);
-		INTON;
 	}
+	INTON;
 
 	if (vflag) {
 		out2str(parsefile->nextc);
@@ -590,3 +616,21 @@  popallfiles(void)
 {
 	unwindfiles(toppf);
 }
+
+void __attribute__((noinline)) flush_input(void)
+{
+	int left = basepf.nleft + input_get_lleft(&basepf);
+
+	if (stdin_state.seekable && left) {
+		INTOFF;
+		lseek(0, -left, SEEK_CUR);
+		input_set_lleft(&basepf, basepf.nleft = 0);
+		INTON;
+	}
+}
+
+void reset_input(void)
+{
+	flush_input();
+	stdin_istty = -1;
+}
diff --git a/src/input.h b/src/input.h
index c59d784..af1c1be 100644
--- a/src/input.h
+++ b/src/input.h
@@ -35,6 +35,7 @@ 
  */
 
 #include <limits.h>
+#include <stdbool.h>
 
 #ifdef SMALL
 #define IS_DEFINED_SMALL 1
@@ -94,6 +95,7 @@  struct parsefile {
 };
 
 extern struct parsefile *parsefile;
+extern int stdin_istty;
 
 /*
  * The input line number.  Input.c just defines this variable, and saves
@@ -113,6 +115,8 @@  void pushstdin(void);
 void popfile(void);
 void unwindfiles(struct parsefile *);
 void popallfiles(void);
+void flush_input(void);
+void reset_input(void);
 
 static inline int input_get_lleft(struct parsefile *pf)
 {
diff --git a/src/jobs.c b/src/jobs.c
index 02ec6f4..247fab5 100644
--- a/src/jobs.c
+++ b/src/jobs.c
@@ -969,6 +969,9 @@  forkshell(struct job *jp, union node *n, int mode)
 	int pid;
 
 	TRACE(("forkshell(%%%d, %p, %d) called\n", jobno(jp), n, mode));
+
+	flush_input();
+
 	pid = fork();
 	if (pid == 0)
 		forkchild(jp, n, mode);
diff --git a/src/mkinit.c b/src/mkinit.c
index 870b64d..2514ebf 100644
--- a/src/mkinit.c
+++ b/src/mkinit.c
@@ -119,6 +119,11 @@  char forkreset[] = "\
  * This routine is called when we enter a subshell.\n\
  */\n";
 
+char postexitreset[] = "\
+/*\n\
+ * This routine is called in exitshell.\n\
+ */\n";
+
 char reset[] = "\
 /*\n\
  * This routine is called when an error or an interrupt occurs in an\n\
@@ -130,6 +135,7 @@  struct event event[] = {
 	{"INIT", "init", init},
 	{"EXITRESET", "exitreset", exitreset},
 	{"FORKRESET", "forkreset", forkreset, "union node *n"},
+	{"POSTEXITRESET", "postexitreset", postexitreset},
 	{"RESET", "reset", reset},
 	{NULL, NULL}
 };
diff --git a/src/options.c b/src/options.c
index 34e500a..c74e4fe 100644
--- a/src/options.c
+++ b/src/options.c
@@ -142,7 +142,7 @@  procargs(int argc, char **argv)
 			sh_error("-c requires an argument");
 		sflag = 1;
 	}
-	if (iflag == 2 && sflag == 1 && isatty(0) && isatty(1))
+	if (iflag == 2 && sflag == 1 && stdin_istty && isatty(1))
 		iflag = 1;
 	if (mflag == 2)
 		mflag = iflag;
diff --git a/src/redir.c b/src/redir.c
index c57d745..8d1c8f6 100644
--- a/src/redir.c
+++ b/src/redir.c
@@ -46,16 +46,17 @@ 
  * Code for dealing with input/output redirection.
  */
 
-#include "main.h"
-#include "shell.h"
-#include "nodes.h"
-#include "jobs.h"
-#include "options.h"
-#include "expand.h"
-#include "redir.h"
-#include "output.h"
-#include "memalloc.h"
 #include "error.h"
+#include "expand.h"
+#include "input.h"
+#include "jobs.h"
+#include "main.h"
+#include "memalloc.h"
+#include "nodes.h"
+#include "options.h"
+#include "output.h"
+#include "redir.h"
+#include "shell.h"
 #include "system.h"
 #include "trap.h"
 
@@ -142,6 +143,8 @@  redirect(union node *redir, int flags)
 			continue;
 
 		fd = n->nfile.fd;
+		if (fd == 0)
+			reset_input();
 
 		if (sv) {
 			int closed;
@@ -415,8 +418,11 @@  popredir(int drop)
 				close(i);
 			break;
 		default:
-			if (!drop)
+			if (!drop) {
+				if (i == 0)
+					reset_input();
 				dup2(rp->renamed[i], i);
+			}
 			close(rp->renamed[i]);
 			break;
 		}
diff --git a/src/trap.c b/src/trap.c
index 4eb9849..aebffa0 100644
--- a/src/trap.c
+++ b/src/trap.c
@@ -426,6 +426,7 @@  exitshell(void)
 	}
 out:
 	exitreset();
+	postexitreset();
 	/*
 	 * Disable job control so that whoever had the foreground before we
 	 * started can get it back.