On Sun, Jan 29, 2012 at 09:34:37PM +0000, Dave Mitchell wrote: > On Sun, Jan 29, 2012 at 01:03:52PM -0800, Father Chrysostomos via RT wrote: > > Is there any chance your incomplete work could be merged to blead? > > > > After all, it's already better that what we now have in blead. > > I'd prefer not to. We're only 3 weeks away from user-visible changes > freeze, and 2 months away from 5.16-RC0, so it's highly likely that the > rest of my planned work won't make into 5.16. Which would mean we have a Much that I like your fixes as soon as possible, I think I'd be uncomfortable with them going into blead if they were *all* ready now (which of course they aren't), and extremely uncomfortable with them going in either incomplete, or complete but closer to 5.16. I hope they'll be one of the first things merged into 5.17 And that there aren't too many 80%s left. > Also, I am unclear as to whether the 'use re eval' security mechanism > works 100% correctly in this in-between state; I know it will be secure at > the end when I rip out the PL_reginterp_cnt mechanism, but I haven't > bothered to ensure safety while its in a hybrid state. Get that wrong and we'll be fighting CVEs. That's not a risk that seems worth it. Nicholas ClarkThread Previous | Thread Next