Am 07.12.2009 10:47, schrieb Avi Kivity: > On 12/06/2009 08:41 PM, malc wrote: >> Sure. My point is that sometimes failure to allocate is due to bugs, >> invalid input etc, and conversion to OOM aborts en masse, might have >> not been the best possible route to take, but most likely it was better >> than doing nothing. >> > > I agree. Early oom handling does limit opportunities for recovery in > the cases where it is possible/easy. We can have an alternative API > that doesn't do oom handling for those cases where it is desirable.
You could simply use normal malloc as an alternative API there. After all, the only thing that qemu_malloc adds is the OOM check (and currently also the zero check). Kevin
