anaconda: possible recursive locking detected

[Jason Dravet]

During my last attempt at installing rawhide I got the following in the 
anaconda.log:

08:10:17 INFO    : moving (1) to step installpackages
08:10:17 INFO    : Preparing to install packages
08:10:17 DEBUG   : Member: vixie-cron.i386 4-4.1-58.fc6 - u
08:10:19 DEBUG   : Adding Package vixie-cron - 4:4.1-58.fc6.i386 in mode u
08:10:19 DEBUG   : Member: rhpl.i386 0-0.188-1 - u
08:10:21 DEBUG   : Adding Package rhpl - 0.188-1.i386 in mode u
08:10:21 DEBUG   : Member: ncurses.i386 0-5.5-23.20060715 - u
08:10:24 DEBUG   : Adding Package ncurses - 5.5-23.20060715.i386 in mode u
08:10:24 DEBUG   : Member: elfutils-libs.i386 0-0.123-1.fc6 - u

It takes about an hour to finish filling the anaconda.log with package info 
like above.  Then the download begins.  It might have to do with:

<4>
<4>=============================================
<4>[ INFO: possible recursive locking detected ]
<4>2.6.17-1.2571.fc6 #1
<4>---------------------------------------------
<4>anaconda/432 is trying to acquire lock:
<4> (&bdev->bd_mutex){--..}, at: [<c04658bb>] __blkdev_put+0x1f/0x11f
<4>
<4>but task is already holding lock:
<4> (&bdev->bd_mutex){--..}, at: [<c0465b4e>] do_open+0x6b/0x3b2
<4>
<4>other info that might help us debug this:
<4>1 lock held by anaconda/432:
<4> #0:  (&bdev->bd_mutex){--..}, at: [<c0465b4e>] do_open+0x6b/0x3b2
<4>
<4>stack backtrace:
<4> [<c04037db>] show_trace_log_lvl+0x58/0x159
<4> [<c0403d9e>] show_trace+0xd/0x10
<4> [<c0403e3b>] dump_stack+0x19/0x1b
<4> [<c042bddb>] __lock_acquire+0x765/0x97c
<4> [<c042c563>] lock_acquire+0x4b/0x6c
<4> [<c05f37ee>] mutex_lock_nested+0xcb/0x214
<4> [<c04658bb>] __blkdev_put+0x1f/0x11f
<4> [<c04659d4>] blkdev_put+0xa/0xc
<4> [<c0465e26>] do_open+0x343/0x3b2
<4> [<c0466030>] blkdev_open+0x1f/0x48
<4> [<c045d83c>] __dentry_open+0xb8/0x186
<4> [<c045d978>] nameidata_to_filp+0x1c/0x2e
<4> [<c045d9b9>] do_filp_open+0x2f/0x36
<4> [<c045da00>] do_sys_open+0x40/0xbb
<4> [<c045daa7>] sys_open+0x16/0x18
<4> [<c0402e57>] syscall_call+0x7/0xb
<4>DWARF2 unwinder stuck at syscall_call+0x7/0xb
<4>Leftover inexact backtrace:
<4> [<c0403d9e>] show_trace+0xd/0x10
<4> [<c0403e3b>] dump_stack+0x19/0x1b
<4> [<c042bddb>] __lock_acquire+0x765/0x97c
<4> [<c042c563>] lock_acquire+0x4b/0x6c
<4> [<c05f37ee>] mutex_lock_nested+0xcb/0x214
<4> [<c04658bb>] __blkdev_put+0x1f/0x11f
<4> [<c04659d4>] blkdev_put+0xa/0xc
<4> [<c0465e26>] do_open+0x343/0x3b2
<4> [<c0466030>] blkdev_open+0x1f/0x48
<4> [<c045d83c>] __dentry_open+0xb8/0x186
<4> [<c045d978>] nameidata_to_filp+0x1c/0x2e
<4> [<c045d9b9>] do_filp_open+0x2f/0x36
<4> [<c045da00>] do_sys_open+0x40/0xbb
<4> [<c045daa7>] sys_open+0x16/0x18
<4> [<c0402e57>] syscall_call+0x7/0xb
<6>kjournald starting.  Commit interval 5 seconds

https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=203137

Thanks,
Jason