Thanks for Mike Westerhof <mwester@dls.net>

----------
 This patchset updates the libbb stuff to use a vfork() version of
 gz_open, called gzvopen.  This is done because a standard fork will
 duplicate the entire address space.  This will invoke the OOM
 (out of memory) killer on small-memory machines, because most often
 by the time we unzip any package, we've read the entire package
 database into memory already.  By using vfork() and immediatly
 execing the external gunzip utility, we avoid the need to clone the
 entire address space.

 Yes, this is actually **LESS** efficient than the original way!
 But there is no way to (currently) dodge the OOM killer on a
 per-process basis, so the alternatives are to either change the
 OOM killer behavior system-wide, or to use this workaround.

 Mike Westerhof, Dec 2008


git-svn-id: http://opkg.googlecode.com/svn/trunk@190 e8e0d7a0-c8d9-11dd-a880-a1081c7ac358

3 files changed, 83 insertions, 6 deletions

diff --git a/libbb/gz_open.c b/libbb/gz_open.c
index dbaf3bb..ba4d6a5 100644
--- a/libbb/gz_open.c
+++ b/libbb/gz_open.c
@@ -56,3 +56,78 @@ extern FILE *gz_open(FILE *compressed_file, int *pid)
 	}
 	return(fdopen(unzip_pipe[0], "r"));
 }
+
+/* gz_open implementation using gunzip and a vfork/exec -- dodges OOM killer */
+extern FILE *gzvopen(FILE *compressed_file, int *pid)
+{
+	int unzip_pipe[2];
+	off_t floc;
+	int cfile;
+
+	/* create a new file descriptor for the input stream
+	 * (it *must* be associated with a file)
+	 * and seek to the same position in that fd as the stream.
+	 */
+	cfile = dup(fileno(compressed_file));
+	floc = ftello(compressed_file);
+	lseek(cfile, floc, SEEK_SET);
+
+	/* create the pipe */
+	if (pipe(unzip_pipe)!=0) {
+		error_msg("gzvopen(): pipe error");
+		return(NULL);
+	}
+
+	*pid = vfork();
+
+	if (*pid < 0) {
+		error_msg("gzvopen(): fork failed");
+		return(NULL);
+	}
+
+	if (*pid==0) {
+		/* child process - reads STDIN, writes to pipe */
+
+		/* close unused read end of pipe */
+		close(unzip_pipe[0]);
+
+		/* connect child's stdout to the pipe write end */
+		dup2(unzip_pipe[1], 1);
+
+		/* connect child's stdin to the fd passed in to us */
+		dup2(cfile, 0);
+
+		/* execute the gunzip utility */
+		execlp("gunzip","gunzip",NULL);
+
+		/* if we get here, we had a failure - since we are
+		 * using vfork(), we cannot call exit(), must call _exit().
+		 */
+		_exit(-1);
+	}
+
+	/* Parent process is executing here */
+
+	/* we have no more need of the duplicate fd */
+	close(cfile);
+
+	/* close the write end of the pipe */
+	close(unzip_pipe[1]);
+
+	/* return the read end of the pipe as a FILE */
+	return(fdopen(unzip_pipe[0], "r"));
+}
+
+extern void gzvclose(int gunzip_pid)
+{
+        if (kill(gunzip_pid, SIGTERM) == -1) {
+		perror("gzvclose()");
+                fprintf(stderr,"%s: unable to kill gunzip pid.\n",
+			__FUNCTION__);
+        }
+
+        if (waitpid(gunzip_pid, NULL, 0) == -1) {
+                fprintf(stderr,"%s: unable to wait on gunzip pid.\n",
+			__FUNCTION__);
+        }
+}
diff --git a/libbb/libbb.h b/libbb/libbb.h
index 860dacd..238e5a7 100644
--- a/libbb/libbb.h
+++ b/libbb/libbb.h
@@ -270,6 +270,8 @@ char *fgets_str(FILE *file, const char *terminating_string);
 extern int unzip(FILE *l_in_file, FILE *l_out_file);
 extern void gz_close(int gunzip_pid);
 extern FILE *gz_open(FILE *compressed_file, int *pid);
+extern void gzvclose(int gunzip_pid);
+extern FILE *gzvopen(FILE *compressed_file, int *pid);
 
 extern struct hostent *xgethostbyname(const char *name);
 extern int create_icmp_socket(void);
diff --git a/libbb/unarchive.c b/libbb/unarchive.c
index bb70ccb..3251c71 100644
--- a/libbb/unarchive.c
+++ b/libbb/unarchive.c
@@ -749,7 +749,7 @@ char *deb_extract(const char *package_filename, FILE *out_stream,
 		while ((ar_header = get_header_ar(deb_stream)) != NULL) {
 			if (strcmp(ared_file, ar_header->name) == 0) {
 				/* open a stream of decompressed data */
-				uncompressed_stream = gz_open(deb_stream, &gunzip_pid);
+				uncompressed_stream = gzvopen(deb_stream, &gunzip_pid);
 				archive_offset = 0;
 				output_buffer = unarchive(uncompressed_stream, out_stream, get_header_tar, free_header_tar, extract_function, prefix, file_list);
 			}
@@ -757,7 +757,7 @@ char *deb_extract(const char *package_filename, FILE *out_stream,
 			free (ar_header->name);
 			free (ar_header);
 		}
-		gz_close(gunzip_pid);
+		gzvclose(gunzip_pid);
 		fclose(deb_stream);
 		fclose(uncompressed_stream);
 		free(ared_file);
@@ -769,7 +769,7 @@ char *deb_extract(const char *package_filename, FILE *out_stream,
 		file_header_t *tar_header;
 		archive_offset = 0;
 		fseek(deb_stream, 0, SEEK_SET);
-		unzipped_opkg_stream = gz_open(deb_stream, &unzipped_opkg_pid);
+		unzipped_opkg_stream = gzvopen(deb_stream, &unzipped_opkg_pid);
 		
                 /*fprintf(stderr, __FUNCTION__ ": processing opkg %s -- ared_file=%s\n", package_filename, ared_file);*/
 		/* walk through outer tar file to find ared_file */
@@ -779,7 +779,7 @@ char *deb_extract(const char *package_filename, FILE *out_stream,
                                 name_offset = 2;
 			if (strcmp(ared_file, tar_header->name+name_offset) == 0) {
 				/* open a stream of decompressed data */
-				uncompressed_stream = gz_open(unzipped_opkg_stream, &gunzip_pid);
+				uncompressed_stream = gzvopen(unzipped_opkg_stream, &gunzip_pid);
 				archive_offset = 0;
                                 /*fprintf(stderr, __FUNCTION__ ":%d: here -- found file\n", __LINE__);*/
 				output_buffer = unarchive(uncompressed_stream, 
@@ -791,14 +791,14 @@ char *deb_extract(const char *package_filename, FILE *out_stream,
 							  file_list);
                                 /*fprintf(stderr, __FUNCTION__ ":%d: unarchive complete\n", __LINE__);*/
 				free_header_tar(tar_header);
-				gz_close(gunzip_pid);
+				gzvclose(gunzip_pid);
 				fclose(uncompressed_stream);
 				break;
 			}
 			seek_sub_file(unzipped_opkg_stream, tar_header->size);
 			free_header_tar(tar_header);
 		}
-		gz_close(unzipped_opkg_pid);
+		gzvclose(unzipped_opkg_pid);
 		fclose(unzipped_opkg_stream);
 		fclose(deb_stream);
 		free(ared_file);