![]() | |||||||||||||
Messages in this thread |
Andrew Morton wrote on Monday, December 04, 2006 11:36 AM > On Mon, 4 Dec 2006 08:26:36 -0800 > "Chen, Kenneth W" <kenneth.w.chen@intel.com> wrote: > > > So it's not possible to call down to generic_file_aio_read/write with invalid > > iov segment. Patch proposed to delete these redundant code. > > erp, please don't touch that code. > > The writev deadlock fixes which went into 2.6.17 trashed writev() > performance and Nick and I are slowly trying to get it back, while fixing > the has-been-there-forever pagefault-in-write() deadlock. > > This is all proving very hard to do, and we don't need sweeping code > cleanups happening under our feet ;) > > I'll bring those patches back in next -mm, but not very confidently. OK, I will wait until that bug settles down and resubmit. I really don't see the value of walking the iov multiple times doing the same thing. I will also dig up the archive to see if I can help in any way. - Ken - 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: 2006-12-04 19:55 [from the cache] ©2003-2008 | |||||||||||||