From: Rob L. <ro...@la...> - 2006-02-15 22:14:07
|
So I finally got around to building 2.6.16-pre3 on my newly upgraded Ubuntu "Flatulent Badger" system, like so: tar xvjf linux-2.6.13-rc3.tar.bz2 cat > allno.config << EOF CONFIG_MODE_SKAS=y CONFIG_BINFMT_ELF=y CONFIG_HOSTFS=y CONFIG_SYSCTL=y CONFIG_STDERR_CONSOLE=y CONFIG_UNIX98_PTYS=y CONFIG_BLK_DEV_LOOP=y CONFIG_LBD=y CONFIG_EXT2_FS=y CONFIG_PROC_FS=y EOF make ARCH=um allnoconfig make ARCH=um Run it: ./linux rootfstype=hostfs rw init=/bin/sh And it's developed yet more verbose error messages: > Checking for /dev/anon on the host...Not available (open failed with errno > 2) Using 2.6 host AIO > io scheduler noop registered (default) > loop: loaded (max 8 devices) > Initialized stdio console driver > parse_chan_pair failed for device 1 > parse_chan_pair failed for device 2 > parse_chan_pair failed for device 3 > parse_chan_pair failed for device 4 > parse_chan_pair failed for device 5 > parse_chan_pair failed for device 6 > parse_chan_pair failed for device 7 > parse_chan_pair failed for device 8 > parse_chan_pair failed for device 9 > parse_chan_pair failed for device 10 > parse_chan_pair failed for device 11 > parse_chan_pair failed for device 12 > parse_chan_pair failed for device 13 > parse_chan_pair failed for device 14 > parse_chan_pair failed for device 15 > Console initialized on /dev/tty0 > VFS: Mounted root (hostfs filesystem). > idr_remove called for id=1 which is not allocated. > 0817f984: [<080f2c7f>] sub_remove+0xef/0x100 > 0817f9a0: [<0806305a>] set_signals+0x1a/0x30 > 0817f9a8: [<080905ef>] free_hot_cold_page+0xdf/0x120 > 0817f9ac: [<080a3633>] kfree+0x23/0x90 > 0817f9c4: [<080f2cbf>] idr_remove+0x2f/0xc0 > 0817f9dc: [<080fbd22>] release_dev+0x5e2/0x700 > 0817fa1c: [<0806b296>] __wake_up+0x46/0x60 > 0817fa44: [<0806305a>] set_signals+0x1a/0x30 > 0817fa4c: [<080fb1d5>] init_dev+0x1b5/0x5a0 > 0817fa98: [<080fbf6f>] tty_open+0x12f/0x380 > 0817fab4: [<080f340a>] kobject_get+0x1a/0x30 > 0817fac0: [<080afcff>] cdev_get+0xf/0x20 > 0817fac8: [<080afeef>] exact_lock+0xf/0x20 > 0817faec: [<080fbe40>] tty_open+0x0/0x380 > 0817faf4: [<080afdce>] chrdev_open+0x9e/0x140 > 0817fb14: [<080afd30>] chrdev_open+0x0/0x140 > 0817fb18: [<080a57e0>] __dentry_open+0x90/0x190 > 0817fb34: [<080a5a58>] nameidata_to_filp+0x38/0x50 > 0817fb50: [<080a593c>] do_filp_open+0x5c/0x60 > 0817fb78: [<080ec008>] get_request+0x238/0x2f0 > 0817fb98: [<080a5b4c>] get_unused_fd+0x6c/0xc0 > 0817fbc0: [<080a5cb1>] do_sys_open+0x51/0xa0 > 0817fbe0: [<080a5d27>] sys_open+0x27/0x30 > 0817fbe8: [<080ec008>] get_request+0x238/0x2f0 > 0817fbf4: [<0805c5fc>] handle_syscall+0xec/0x100 > 0817fbf8: [<080ec008>] get_request+0x238/0x2f0 > 0817fc14: [<080ec008>] get_request+0x238/0x2f0 > 0817fc34: [<080ec008>] get_request+0x238/0x2f0 > 0817fc64: [<080658d8>] handle_trap+0x28/0x130 > 0817fc8c: [<08065f44>] userspace+0x1c4/0x220 > 0817fcd8: [<08054f10>] init+0x0/0x170 > 0817fce4: [<0805c0c9>] new_thread_handler+0xe9/0x120 > > sh-3.00# Came up and ran, but looks really ugly. Entirely reproducible so far... Anybody got any ideas? Rob -- Never bet against the cheap plastic solution. |
From: Jeff D. <jd...@ad...> - 2006-02-17 20:28:32
|
On Wed, Feb 15, 2006 at 05:13:52PM -0500, Rob Landley wrote: > > parse_chan_pair failed for device 1 > > parse_chan_pair failed for device 2 > > parse_chan_pair failed for device 3 > > parse_chan_pair failed for device 4 > > parse_chan_pair failed for device 5 > > parse_chan_pair failed for device 6 > > parse_chan_pair failed for device 7 > > parse_chan_pair failed for device 8 > > parse_chan_pair failed for device 9 > > parse_chan_pair failed for device 10 > > parse_chan_pair failed for device 11 > > parse_chan_pair failed for device 12 > > parse_chan_pair failed for device 13 > > parse_chan_pair failed for device 14 > > parse_chan_pair failed for device 15 I don't know about the rest of this, but something like this just happened to me. It was attaching the consoles to a host portal which another UML was already using. I got semi-reasonable "that portal is already in use" errors in between the parse_chan_pair things. So, were you trying to reuse some host devices, and which ones? An improvement in the errors might be in order here. Jeff |
From: Rob L. <ro...@la...> - 2006-02-18 00:48:33
|
On Friday 17 February 2006 3:29 pm, Jeff Dike wrote: > On Wed, Feb 15, 2006 at 05:13:52PM -0500, Rob Landley wrote: > > > parse_chan_pair failed for device 1 > > > parse_chan_pair failed for device 2 > > > parse_chan_pair failed for device 3 > > > parse_chan_pair failed for device 4 > > > parse_chan_pair failed for device 5 > > > parse_chan_pair failed for device 6 > > > parse_chan_pair failed for device 7 > > > parse_chan_pair failed for device 8 > > > parse_chan_pair failed for device 9 > > > parse_chan_pair failed for device 10 > > > parse_chan_pair failed for device 11 > > > parse_chan_pair failed for device 12 > > > parse_chan_pair failed for device 13 > > > parse_chan_pair failed for device 14 > > > parse_chan_pair failed for device 15 > > I don't know about the rest of this, but something like this just > happened to me. It was attaching the consoles to a host portal which > another UML was already using. I got semi-reasonable "that portal is > already in use" errors in between the parse_chan_pair things. > > So, were you trying to reuse some host devices, and which ones? An > improvement in the errors might be in order here. I don't know. The did my standard "rootfstype=hostfs rw init=/bin/sh" thing, which should only have on process (a shell) on /dev/console. All that happened before that even managed to run /bin/sh. I don't think I had any other type of console support even built into the thing. (Console on stderr support, everything else disabled.) No idea what devices it was talking about... > Jeff Rob -- Never bet against the cheap plastic solution. |