Graham Stoney wrote:
I've just noticed some potential memory corruption bugs associated with
the CPU6 workaround in the PPC HardHat-1.2 kernel.
In what LSP of the 1.2 CDK? Mark has been predominantly discussing
the 8240/755 Sandpoint and providing pointers to software that doesn't
represent integration of the 8xx software on the CDK.
Is this a problem in one of the MPC8xx LSPs of the CDK?
...meaning that they will clobber an unrelated value in memory 8 bytes
after cpu6_bug.
I'm travelling and don't have an 8xx LSP handy, but there was supposed
to be sufficient space allocated here to prevent this condition. Thanks
for the note and I will check the software to ensure this is correct in
the MPC8xx CDKs.
-- Dan
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/