Re: [Crash-utility] Fix bug of list -h
Dne Pá 1. června 2012 05:52:17 qiaonuohan napsal(a):
Hello Dave,
Hello qiaonuohan,
When investigating list command, I found the "-h" is
involved but not
discussed in help page. And then I tried to use, but I find some
problems with it.
In a previous post, I pointed out that "list -h" is undocumented and hardly
useful. In fact, there is a comment in tools.c that explains "list -h":
* If the structures are linked using list_head structures, the -h or -H
* options must be used. In that case, the "start" address is:
* a pointer to the embedded list_head structure (-h), or a pointer to a
* LIST_HEAD() structure (-H).
This matches exactly what crash actually does, although I can't find a use
case for that.
The first, I made patch to show it. The "ld->start"
should be the
pointer to the structure list_head.
And the second one, I am not sure about the reason why you don't
display the node related to the address user input. I will take the
tasks of task_struct as an example.
crash> task_struct.tasks ffff8800371a0ac0
tasks = {
next = 0xffffffff81a8d468,
prev = 0xffff88004a9e0f88
}
crash> list task_struct.tasks -s task_struct.tasks -h ffff8800371a0ac0
ffffffff81a8d020
tasks = {
next = 0xffff88004eaf1908,
prev = 0xffff8800371a0f08
}
This doesn't match my experience. If I try to use the address of the
containing task_struct with "list -h", I'll get invalid output. I'd have
to
use the address of the embedded list_head, but then it also includes the
structure in which that list_head was embedded. E.g. on my 32-bit laptop I
get:
crash> task_struct.tasks 0xeede04f0
tasks = {
next = 0xd79d6f20,
prev = 0xd55e6220
}
crash> struct -o task_struct.tasks
struct task_struct {
[432] struct list_head tasks;
}
crash> list task_struct.tasks -s task_struct.tasks -h 0xeede04f0+432
eede04f0
tasks = {
next = 0xd79d6f20,
prev = 0xd55e6220
}
d79d6d70
tasks = {
next = 0xc32c5320,
prev = 0xeede06a0
}
...
Note that the first element in the list (the one which contains the starting
list_head) is included.
Based on the above, I'm afraid don't quite get the intended goal of your
patch.
Petr Tesarik