Do we really need these defines, I found that it
is not used anywhere and defined as deadbeef on
some architectures. Does it make sense to remove
these variables from the kernel source?
Balbir
_________________________________________________________________
Send and receive Hotmail on your mobile device: http://mobile.msn.com
From: "Balbir Singh" <[email protected]>
Date: Fri, 08 Feb 2002 12:36:59 -0800
Do we really need these defines, I found that it
is not used anywhere and defined as deadbeef on
some architectures. Does it make sense to remove
these variables from the kernel source?
Perhaps your copy of grep is buggy, check out
fs/proc/array.c which does make use of those macros.
My fault, I was using lxr.linux.no. I was also confused by
the way it is defined in include/asm-parisc (2.4.17)
#define KSTK_EIP(tsk) (0xdeadbeef)
#define KSTK_ESP(tsk) (0xdeadbeef)
Thanks for pointing this out,
Balbir
>From: "David S. Miller" <[email protected]>
>To: [email protected]
>CC: [email protected], [email protected]
>Subject: Re: KSTK_EIP and KSTK_ESP
>Date: Fri, 08 Feb 2002 12:55:22 -0800 (PST)
>
> From: "Balbir Singh" <[email protected]>
> Date: Fri, 08 Feb 2002 12:36:59 -0800
>
> Do we really need these defines, I found that it
> is not used anywhere and defined as deadbeef on
> some architectures. Does it make sense to remove
> these variables from the kernel source?
>
>Perhaps your copy of grep is buggy, check out
>fs/proc/array.c which does make use of those macros.
>
_________________________________________________________________
Chat with friends online, try MSN Messenger: http://messenger.msn.com
Balbir Singh writes:
> Do we really need these defines, I found that it
> is not used anywhere and defined as deadbeef on
> some architectures. Does it make sense to remove
> these variables from the kernel source?
You should implement these. The names may be x86-specific,
but the purpose is not.
EIP -- user instruction pointer or (eeew!) program counter
ESP -- user stack pointer, as defined by your ABI