![]() | |||||||||||||
Messages in this thread |
On Mon, 17 Nov 2003, William Lee Irwin III wrote: > On Mon, Nov 17, 2003 at 08:21:34AM +0000, Tigran Aivazian wrote: > > Now, since there is no way to detect EOF, other than by reading an extra > > page and discovering that it belongs to the next iteration, we have to do > > the lseek(fd, 0, SEEK_SET) anyway. > > So, the "auto-rewinding" read would only help the cases where application > > doesn't need to differentiate between samples and is happy to just > > continuously read chunks packed into pages one by one as fast as > > possible. In this case it doesn't need to lseek to 0, so auto-rewinding on > > kernel side would prevent it from slowing down. > > If you're going to repeatedly read from 0 pread() sounds like a good > alternative to read() + lseek(), plus no kernel changes required to > get rid of the lseek(). The reason why I didn't use pread(2) is because I have to do multiple calls to read(2). There is no way that I know of to pack more than a single page into a single read(2) with seq_file API. Yes, I remember Al saying "it's not a page" but in practice it still appears to be limited to a page unless someone shows a sample seq_file module which can provide more than a page of data on a single read(2). The implementations I have looked at in the kernel (e.g. mm/slab.c) are limited to a single page per read(2). Kind regards Tigran - 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 12:58 [from the cache] ©2003-2008 | |||||||||||||