On Fri, 2011-09-23 at 18:10 +0200, Miklos Szeredi wrote: > David Howells <dhowells@xxxxxxxxxx> writes: > > >> Get a grip, people. Stop over-analyzing things. Stop bothering to > >> mention what Solaris does, when the *MUCH* bigger issue is what > >> *Linux* has done for years and years. > > > > So the word is Linux may not change it's behaviour in this regard from what > > existed prior to the d_automount patch, period? > > You can change behavior if it won't cause regressions. So fix userspace > first, introduce change afterwards. Easy to say, not so easy to do. > > >> Stop saying "we'll revert Miklos patch" in the same sentence where you > >> then seem to not even understand that the *original* behavior was the > >> one that Miklos patch re-introduced. > > > > Miklos's patch did *NOT* re-introduce the original behaviour. It introduced a > > third state. It merely moved the regression and brought about a more serious > > one. > > Yes, and there have been lots of proposals on how to fix it. There just > doesn't seem to be a consensus. Indeed, that's probably because of (as Linus says) the empirically derived division of system calls that do and do not trigger an automount, for cases where there is a question of whether to automount or not. The old ill defined moving target. > > I suggest that we first restore the original behavior for all > filesystems. And it doesn't mean all the d_automount infrastructure has > to be thrown out. A really simple fix is to pass the lookup flags (or > some derived automount flags) to d_automount and fix up the very few > instances to reflect the old semantics. Untested patch attached. Perhaps, but that allows modules to circumvent VFS policy which is what allowed this situation to come up in the first place. Ian -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
