Messages in this thread | | | Date | Fri, 17 Mar 2006 18:20:12 +0000 (GMT) | From | Hugh Dickins <> | Subject | Re: Remapping pages mapped to userspace (was: [PATCH 10 of 20] ipath - support for userspace apps using core driver) |
| |
On Fri, 17 Mar 2006, Linus Torvalds wrote: > > Anyway, zap_page_range() would do what you want, but it's not exported, > and I'm not convinced it's even something we want to export. You can only > zap a page range from within the context of the zappee, not from an > external module/driver. > > [ Maybe it works if somebody else calls it, maybe it doesn't. I wouldn't > bet on it, and more importantly, I can pretty much _guarantee_ that a > driver will get the "struct mm_struct" reference counting wrong. ]
But vmtruncate or unmap_mapping_range is quite used to operating on whatever mm's have mapped the file, so should be a safe route. Except here it's a device mapped VM_PFNMAP, so there might prove to be some gotchas. I'd be happy to make a nopage fault on VM_PFNMAP give SIGBUS, for sensible behaviour after the "truncate" - but recent history warns we're liable then to discover some app expecting otherwise ;)
Hugh - 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/
| |