From mboxrd@z Thu Jan 1 00:00:00 1970 From: Karel Zak Date: Fri, 03 Apr 2020 09:11:24 +0000 Subject: Re: Upcoming: Notifications, FS notifications and fsinfo() Message-Id: <20200403091124.zxo7cckcvygzwvgl@ws.net.home> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit List-Id: References: <20200330211700.g7evnuvvjenq3fzm@wittgenstein> <1445647.1585576702@warthog.procyon.org.uk> <2418286.1585691572@warthog.procyon.org.uk> <20200401090445.6t73dt7gz36bv4rh@ws.net.home> <2488530.1585749351@warthog.procyon.org.uk> <2488734.1585749502@warthog.procyon.org.uk> In-Reply-To: To: Miklos Szeredi Cc: David Howells , Christian Brauner , Linus Torvalds , Al Viro , dray@redhat.com, Miklos Szeredi , Steven Whitehouse , Jeff Layton , Ian Kent , andres@anarazel.de, keyrings@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Lennart Poettering , Aleksa Sarai On Wed, Apr 01, 2020 at 05:25:54PM +0200, Miklos Szeredi wrote: > fsinfo(2) will never be substantially cheaper than reading and parsing > /mnt/MNT_ID/info. In fact reading a large part of the mount table > using fsinfo(2) will be substantially slower than parsing > /proc/self/mountinfo (this doesn't actually do the parsing but that > would add a very small amount of overhead): I think nobody wants to use fsinfo() or mountfs as replacement to whole /proc/self/mountinfo. It does not make sense. We need per-mountpoint API, for whole mount table use-cases (like findmnt or lsblk) the current mountinfo is good enough. Karel -- Karel Zak http://karelzak.blogspot.com