I think this is better functionality. If anything is scripted, this patch will make it easier instead of having to test if the file exists (or if the binary exits due to the file not already existing). I actually think most binaries work like this... Just my $0.02
Dave On 10/22/19, Joerg Vehlow <[email protected]> wrote: > No interest in util-linux compatibility? > > Am 23.09.2019 um 07:39 schrieb Joerg Vehlow: >> From: Joerg Vehlow <[email protected]> >> >> util-linux's losetup creates a loop device, if it is missing. >> This is missing in busybox's implementation. >> >> Testcase: >> losetup /dev/loop1000 <somepath> >> If /dev/loop1000 does not exist busybox give an error but >> util-linux's losetup just creates the file >> >> Signed-off-by: Joerg Vehlow <[email protected]> >> --- >> util-linux/losetup.c | 15 +++++++++++++++ >> 1 file changed, 15 insertions(+) >> >> diff --git a/util-linux/losetup.c b/util-linux/losetup.c >> index cc6c2b1d5..790c41cab 100644 >> --- a/util-linux/losetup.c >> +++ b/util-linux/losetup.c >> @@ -147,10 +147,25 @@ int losetup_main(int argc UNUSED_PARAM, char >> **argv) >> d = *argv++; >> >> if (argv[0]) { >> + struct stat statbuf; >> + int n; >> unsigned flags = (opt & OPT_r) ? >> BB_LO_FLAGS_READ_ONLY : 0; >> + >> if (opt & OPT_P) { >> flags |= BB_LO_FLAGS_PARTSCAN; >> } >> + >> + // If the loop device does not exist, create it >> + if (stat(d, &statbuf) != 0 && errno == ENOENT) { >> + if (sscanf(d, LOOP_FORMAT, &n) != 1) { >> + errno = 0; >> + bb_perror_msg_and_die("Loop device >> %s does not exist and does not start with " LOOP_NAME, d); >> + } >> + if (mknod(d, S_IFBLK|0644, makedev(7, n)) >> != 0) { >> + bb_simple_perror_msg_and_die(d); >> + } >> + } >> + >> if (set_loop(&d, argv[0], offset, flags) < 0) >> bb_simple_perror_msg_and_die(argv[0]); >> return EXIT_SUCCESS; > > _______________________________________________ > busybox mailing list > [email protected] > http://lists.busybox.net/mailman/listinfo/busybox > _______________________________________________ busybox mailing list [email protected] http://lists.busybox.net/mailman/listinfo/busybox
