|
From: | Ognyan Kulev |
Subject: | Re: address@hidden: ls --translators (coreutils)] |
Date: | Fri, 19 Sep 2003 12:37:38 +0300 |
User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030908 Debian/1.4-4 |
Alfred M. Szmidt wrote:
> + if (trans_stat.st_mode & S_IPTRANS) > + { > + char buf[1024], *trans = buf; Is this 1024-byte buffer guaranteed to be large enough? Good one. I poked around in the Hurd source code, and couldn't really figure out if it would be guaranteed (some places use a 2048 byte buffer). I think that this should be less than the block size of the current file-system (according to [hurd]/ext2fs/inode.c and [hurd]/ufs/inode.c). Could some Hurd god comment on this and what should/could be done?
It it has to be no more than block size of ext2fs, then it better be 4096. This is the maximum block size of ext2fs. (I think that Linux block layer can't map blocks larger than page size, which is 4K in i386. I'm not 100% sure though.)
Regards -- Ognyan Kulev <address@hidden,fsa-bg.org,jabber.net}> 7D9F 66E6 68B7 A62B 0FCF EB04 80BF 3A8C A252 9782
[Prev in Thread] | Current Thread | [Next in Thread] |