![]() | |||||||||||||
Messages in this thread |
Hi! > > > Agreed. I wasn't seriously suggesting changing everywhere to be > > > GFP_NOWARN. Perhaps I should be more explicit in what I'm saying here. > > > The problem isn't just suspend trying to allocate memory. It's > > > _ANYTHING_ that might be running trying to allocate memory while we're > > > eating memory. (Remember that we don't just call shrink_all_memory, but > > > also allocate that memory so other processes don't grab it and stop us > > > making forward progress). As a result, they're going to scream when they > > > can't allocate a page. > > > > Hmm, that does not look too healthy. That means that userland programs > > will see all kinds of weird error conditions that normally > > "almost-can't-happen" during normal usage. > > Failure to allocate memory should be something any caller to get_*_page > deals with, so if they don't, are we to be blamed? Well, you'll have things like select() returning -ENOMEM. Applications will not be too happpy. We can probably live with that, but it is not nice. Pavel -- People were complaining that M$ turns users into beta-testers... ...jr ghea gurz vagb qrirybcref, naq gurl frrz gb yvxr vg gung jnl! - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2005-03-22 14:08 [from the cache] ©2003-2008 | |||||||||||||