On Sat, Oct 27, 2007 at 12:44:51PM +0200, devzero@xxxxxx wrote: > it seems that losetup doesn`t yet honour recent kernel change (see > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=73285082745045bcd64333c1fbaa88f8490f2626 > ) to allow more than 256 loop-devices. Ah, I remember this thing from lkml. > loop-device creation is dynamically now, is there already work in > progress for support in losetup ? If I good remember there is not any other who how create a loop device than by "max_loop" or "mknod /dev/loopN". It means we needn't to add to losetup(8) any extra support for dynamically allocated loop devices. So, we need increment hardcoded limit of loop devices only. Right? > i would have send a fix, but i`m unsure if just changing limitations in lomount.c > > show_used_loop_devices (void) { > .... > for (j = 0; j < SIZE(loop_formats); j++) { > for(i = 0; i < 256; i++) { > sprintf(dev, loop_formats[j], i); > > and > > find_unused_loop_device (void) { > .... > for (j = 0; j < SIZE(loop_formats); j++) { > for(i = 0; i < 256; i++) { > sprintf(dev, loop_formats[j], i); > > > is the way to go since MINOR limit of device nodes seems to be at > 1048575 and that`s probably a little bit to large value to loop > trough. Read the code. It doesn't check for all devices in range 0..256 -- it breaks the for() loop when a /dev/loopN doesn't exist. For example on my system it doesn't check for more than nine device. I don't see a problem with a huge limit. Karel -- Karel Zak <kzak@xxxxxxxxxx> - To unsubscribe from this list: send the line "unsubscribe util-linux-ng" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
