diff options
| author | grehan <grehan@FreeBSD.org> | 2005-11-07 06:55:48 +0000 |
|---|---|---|
| committer | grehan <grehan@FreeBSD.org> | 2005-11-07 06:55:48 +0000 |
| commit | 0a9e3aee3535cf3561f560600680c21b8b9ce8d8 (patch) | |
| tree | 1cdb8a3ccd2c4fbd27983332d76ebd15712b9122 /sys/powerpc | |
| parent | 5ae83a3b6a9bfa94da179b38c43a4c8c64267994 (diff) | |
| download | FreeBSD-src-0a9e3aee3535cf3561f560600680c21b8b9ce8d8.zip FreeBSD-src-0a9e3aee3535cf3561f560600680c21b8b9ce8d8.tar.gz | |
Finally (!?) get to the bottom of the mysterious G3 boot-time panics.
After a number of tests using nop's to change the alignment, it was
confirmed that the mtibat instructions should be cache-aligned.
FreeScale app note AN2540 indicates that the isync before and after
the mtdbat is the right thing to do, but sync/isync isn't required
before the mtibat so it has been removed.
Fix by using a ".balign 32" to pull the code in question to the correct
alignment.
MFC after: 3 days
Diffstat (limited to 'sys/powerpc')
| -rw-r--r-- | sys/powerpc/powerpc/pmap.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/sys/powerpc/powerpc/pmap.c b/sys/powerpc/powerpc/pmap.c index 7f4455e..45319f0 100644 --- a/sys/powerpc/powerpc/pmap.c +++ b/sys/powerpc/powerpc/pmap.c @@ -569,7 +569,7 @@ pmap_bootstrap(vm_offset_t kernelstart, vm_offset_t kernelend) */ batu = BATU(0x00000000, BAT_BL_256M, BAT_Vs); batl = BATL(0x00000000, BAT_M, BAT_PP_RW); - __asm ("sync; isync; \n" + __asm (".balign 32; \n" "mtibatu 0,%0; mtibatl 0,%1; isync; \n" "mtdbatu 0,%0; mtdbatl 0,%1; isync" :: "r"(batu), "r"(batl)); |
