linux-mips
[Top] [All Lists]

Re: reenabling interrupts on return from function

To: Justin Carlson <justin@cs.cmu.edu>
Subject: Re: reenabling interrupts on return from function
From: Justin Carlson <justin@cs.cmu.edu>
Date: 15 Jun 2002 15:22:28 -0700
Cc: linux-mips@oss.sgi.com
In-reply-to: <1024177741.1549.4.camel@localhost.localdomain>
References: <1024177741.1549.4.camel@localhost.localdomain>
Sender: owner-linux-mips@oss.sgi.com
On Sat, 2002-06-15 at 14:49, Justin Carlson wrote:
> I'm obviously missing something basic here.
> 
> Looking at stackframe.h, I see this code as a part of RESTORE_SOME
> 
> 
>               mfc0    t0, CP0_STATUS;                  \
>               .set    pop;                             \
>               ori     t0, 0x1f;                        \
>               xori    t0, 0x1f;                        \
>               mtc0    t0, CP0_STATUS;                  
> 

OK, this was a stupid question; the answer was staring me in the face
(the restoration of the status register from the stack), and I didn't
see it.

However, I still don't see the point of the above code.  Why do we
explicitly clear bits 4-0 of the status register just before reloading
it from the system stack?  

-Justin

<Prev in Thread] Current Thread [Next in Thread>