[PATCH] invalid instructions in kernel mode
kumar.gala at freescale.com
Tue Apr 12 01:39:20 EST 2005
After some debug on Friday and this weekend I understand what's going
on. The reason we hit the code path in the kernel is due to the fact
that we are actually executing an integer load/store that is
misaligned, however when we read the instruction from memory we get a
float point instruction. This is due to the fact that test app is not
properly handling its self generated code.
Note, I was testing with math-emu turned on.
It seems to me that when we introduce CONFIG_PPC_FPU that we ifdef
around the code and return 0 in the 44x/e500 cases w/o FPUs
On Apr 8, 2005, at 12:36 PM, Fillod Stephane wrote:
> Kumar Gala wrote:
> > Are you running this via a ramdisk or nfs? If ramdisk can you post
> > somewhere that I can get to?
> I'm running via NFS.
> > Also, can you email the list with the kernel oops that shows up.
> I'll do that next week.
> Anyway, reading the source suffices to realize there's a problem when
> MATH_EMULATION is disabled on a FPU-less system with user programs
> using load/store fp instructions.
More information about the Linuxppc-dev