login
Header Space

 
 

Re: sparc vs. gcc 4.3

Score:
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: <harvey.harrison@...>
Cc: <bunk@...>, <wli@...>, <sparclinux@...>, <linux-kernel@...>
Date: Monday, March 3, 2008 - 6:02 pm

From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Mon, 03 Mar 2008 13:41:11 -0800


Won't work, the asm needs to be inlined because it needs to:

1) Know where local variables are located

2) Has to run in the context of the stack frame the
   call site runs in so that the tail-call it's doing works

And actually all of this is why the sparc64 fix turned out to
be so invasive, and I'll need to do the same for sparc32
as a result.
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
sparc vs. gcc 4.3, Adrian Bunk, (Mon Mar 3, 3:26 pm)
Re: sparc vs. gcc 4.3, David Miller, (Mon Mar 3, 7:02 pm)
Re: sparc vs. gcc 4.3, David Miller, (Mon Mar 3, 5:03 pm)
Re: sparc vs. gcc 4.3, Harvey Harrison, (Mon Mar 3, 5:41 pm)
Re: sparc vs. gcc 4.3, David Miller, (Mon Mar 3, 6:02 pm)
speck-geostationary