I'd like to understand the implications of this bug.
Obviously applications using the futex system can be affected, but
does anybody know whether GNU software packages suffer from this problem.
I mean glibc (nptl) uses futexes, so does gdb and gcc. will this bug hurt
them ?
Paul Mackerras wrot
Paul Mackerras wrote:
Can anyone see any reason why the FUTEX_OP_SET case can't just do a
__put_user?
Because you need to return the old value. Despite the name,
it's an exchange operation not a set.
r~
___
Linuxppc-dev mailing list
Linuxppc-dev@oz
Richard Henderson pointed out that the powerpc __futex_atomic_op has a
bug: it will write the wrong value if the stwcx. fails and it has to
retry the lwarx/stwcx. loop, since 'oparg' will have been overwritten
by the result from the first time around the loop. This happens
because it uses the same
