On Wed, Nov 3, 2010 at 1:57 PM, Michael Brown <mbrown at fensystems.co.uk> wrote: > On Wednesday 03 Nov 2010 17:26:35 Joshua Oreman wrote: >> This may be a better way of fixing the regression that was previously >> fixed by reverting the DHCP changes in question. > > Could you explain to me what the benefit is of the "gather-multiple offers" > approach? ?Part of the reason why I chose to revert those changes is that I > didn't really understand the motivation for them. ?(I wouldn't have reverted > them for that reason alone, but combined with the fact that two breakages > could be git-bisected to that patch series, it seemed like the best option at > the time.) > > I'm happy to accept that your way could be better, but I need to understand it > first.
The benefit is in compatibility with Intel PXE, which acts similarly. I'm not incredibly attached to the approach, I just wanted to point out that this patch might fix the problems with it if we consider supporting the same broken setups Intel PXE supports to be a desirable thing. Josh > > Michael >
