On Thu, 2006-02-16 at 15:35, Dave Jones wrote: > > > One of the horrors of binary kernel modules is they don't keep up > > > with the steady release of upstream kernels, so what might work > > > fine on one release might break horribly in next weeks updates, > > > sometimes in particularly drastic ways like memory corruption > > > when then finds its way written out to disk. > > > > That's one way of thinking about it. The other is that one of > > the horrors of running Linux is that every kernel release may > > break previously used interfaces and force you to replace all > > your tested modules. > > Even if there were an established ABI that the upstream kernel maintained, > there would still be incompatibilities between versions. So if you make a mistake a few times you should always keep repeating that? > Look at any OS that maintains such an illusion for proof of this. > Every time a vendor has done this, there's been some internal struct > or function that wasn't exported, that 3rd parties wanted to poke > around with, which inevitably does completely the wrong thing when > things are changed. Can you give examples of that 'any OS'? It certainly isn't an issue with Windows or OSX. I personally maintain a few hundred machines that have been through every vendor update for years without a single problem I could blame on binary-only drivers or modules. > The lack of ABI argument is totally bogus. It's no panacea to > 3rd-party code magically getting better. The real problem is by > keeping it binary, no-one but them can fix it, and when they > aren't inclined to do so, who loses? Nothing is perfect, but I see this mentioned much more often as a problem with Linux than an advantage. Can you give some examples of drivers or modules that are better in the second-guessed source versions than the vendor's own versions that they do for the OS's that accept binaries? I recall in particular a lot of trouble with various aic7xxx linux modules that didn't have equivalent problems in the binaries for Windows. And I'm still not all that sure about the reliability of firewire support even though it's sort-of back after months of not working at all in FC4 kernels. You are right that perhaps somebody 'could' fix it but it's not a win unless they actually do. -- Les Mikesell lesmikesell@xxxxxxxxx
