From: cinap_lenrek@felloff.net
To: 9fans@9fans.net
Subject: Re: [9fans] PDP11 (Was: Re: what heavy negativity!)
Date: Wed, 10 Oct 2018 19:34:39 +0200 [thread overview]
Message-ID: <F022808EFD5F0B8A438B332081972DD2@felloff.net> (raw)
In-Reply-To: 1806122E-7160-403A-A2F0-027E5E77E31A@bitblocks.com
> But the reason I want this is to reduce latency to the first
> access, especially for very large files. With read() I have
> to wait until the read completes. With mmap() processing can
> start much earlier and can be interleaved with background
> data fetch or prefetch. With read() a lot more resources
> are tied down. If I need random access and don't need to
> read all of the data, the application has to do pread(),
> pwrite() a lot thus complicating it. With mmap() I can just
> map in the whole file and excess reading (beyond what the
> app needs) will not be a large fraction.
you think doing single 4K page sized reads in the pagefault
handler is better than doing precise >4K reads from your
application? possibly in a background thread so you can
overlap processing with data fetching?
the advantage of mmap is not prefetch. its about not to do
any I/O when data is already in the *SHARED* buffer cache!
which plan9 does not have (except the mntcache, but that is
optional and only works for the disk fileservers that maintain
ther file qid ver info consistently). its *IS* really a linux
thing where all block device i/o goes thru the buffer cache.
--
cinap
next reply other threads:[~2018-10-10 17:34 UTC|newest]
Thread overview: 104+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-10 17:34 cinap_lenrek [this message]
2018-10-10 21:54 ` Steven Stallion
2018-10-10 22:26 ` [9fans] zero copy & 9p (was " Bakul Shah
2018-10-10 22:52 ` Steven Stallion
2018-10-11 20:43 ` Lyndon Nerenberg
2018-10-11 22:28 ` hiro
2018-10-12 6:04 ` Ori Bernstein
2018-10-13 18:01 ` Charles Forsyth
2018-10-13 21:11 ` hiro
2018-10-14 5:25 ` FJ Ballesteros
2018-10-14 7:34 ` hiro
2018-10-14 7:38 ` Francisco J Ballesteros
2018-10-14 8:00 ` hiro
2018-10-15 16:48 ` Charles Forsyth
2018-10-15 17:01 ` hiro
2018-10-15 17:29 ` hiro
2018-10-15 23:06 ` Charles Forsyth
2018-10-16 0:09 ` erik quanstrom
2018-10-17 18:14 ` Charles Forsyth
2018-10-10 22:29 ` [9fans] " Kurt H Maier
2018-10-10 22:55 ` Steven Stallion
2018-10-11 11:19 ` Aram Hăvărneanu
2018-10-11 0:26 ` Skip Tavakkolian
2018-10-11 1:03 ` Steven Stallion
2018-10-14 9:46 ` Ole-Hjalmar Kristensen
2018-10-14 10:37 ` hiro
2018-10-14 17:34 ` Ole-Hjalmar Kristensen
2018-10-14 19:17 ` hiro
2018-10-15 9:29 ` Giacomo Tesio
-- strict thread matches above, loose matches on Subject: below --
2018-10-10 22:19 cinap_lenrek
2018-10-10 16:14 cinap_lenrek
2018-10-10 0:15 cinap_lenrek
2018-10-10 0:22 ` Lyndon Nerenberg
2018-10-09 19:49 cinap_lenrek
2018-10-09 19:56 ` hiro
2018-10-09 19:47 cinap_lenrek
2018-10-09 22:01 ` erik quanstrom
2018-10-09 23:43 ` Lyndon Nerenberg
2018-10-10 5:52 ` hiro
2018-10-10 8:13 ` Digby R.S. Tarvin
2018-10-10 9:14 ` hiro
2018-10-10 13:59 ` Steve Simon
2018-10-10 21:32 ` Digby R.S. Tarvin
2018-10-11 17:43 ` Lyndon Nerenberg
2018-10-11 19:11 ` hiro
2018-10-11 19:27 ` Lyndon Nerenberg
2018-10-11 19:56 ` hiro
2018-10-10 5:57 ` hiro
2018-10-08 3:38 Lucio De Re
2018-10-08 4:29 ` Digby R.S. Tarvin
2018-10-08 7:20 ` hiro
2018-10-08 12:03 ` Charles Forsyth
2018-10-08 17:20 ` hiro
2018-10-08 21:55 ` Digby R.S. Tarvin
2018-10-08 23:03 ` Dan Cross
2018-10-09 0:14 ` Bakul Shah
2018-10-09 1:34 ` Christopher Nielsen
2018-10-09 3:28 ` Lucio De Re
2018-10-09 8:23 ` hiro
2018-10-09 9:45 ` Ethan Gardener
2018-10-09 17:50 ` Bakul Shah
2018-10-09 18:57 ` Ori Bernstein
2018-10-10 7:32 ` Giacomo Tesio
2018-10-09 17:45 ` Lyndon Nerenberg
2018-10-09 18:49 ` hiro
2018-10-09 19:14 ` Lyndon Nerenberg
2018-10-09 22:05 ` erik quanstrom
2018-10-11 17:54 ` Lyndon Nerenberg
2018-10-11 18:04 ` Kurt H Maier
2018-10-11 19:23 ` hiro
2018-10-11 19:24 ` hiro
2018-10-11 19:25 ` hiro
2018-10-11 19:26 ` Skip Tavakkolian
2018-10-11 19:39 ` Lyndon Nerenberg
2018-10-11 19:44 ` Skip Tavakkolian
2018-10-11 19:47 ` Lyndon Nerenberg
2018-10-11 19:57 ` hiro
2018-10-11 20:23 ` Lyndon Nerenberg
2018-10-10 10:42 ` Ethan Gardener
2018-10-09 19:23 ` Lyndon Nerenberg
2018-10-09 19:34 ` hiro
2018-10-09 19:36 ` hiro
2018-10-09 19:40 ` Lyndon Nerenberg
2018-10-10 0:18 ` Dan Cross
2018-10-10 5:45 ` hiro
2018-10-09 22:06 ` erik quanstrom
2018-10-10 6:24 ` Bakul Shah
2018-10-10 13:58 ` erik quanstrom
2018-10-09 22:42 ` Dan Cross
2018-10-09 19:09 ` Bakul Shah
2018-10-09 19:30 ` Lyndon Nerenberg
2018-10-09 3:08 ` Digby R.S. Tarvin
2018-10-09 11:58 ` Ethan Gardener
2018-10-09 13:59 ` erik quanstrom
2018-10-09 22:22 ` Digby R.S. Tarvin
2018-10-10 10:38 ` Ethan Gardener
2018-10-10 23:15 ` Digby R.S. Tarvin
2018-10-11 18:10 ` Lyndon Nerenberg
2018-10-11 20:55 ` Digby R.S. Tarvin
2018-10-11 21:03 ` Lyndon Nerenberg
2018-10-09 14:02 ` erik quanstrom
2018-10-08 8:12 ` Nils M Holm
2018-10-08 9:12 ` Digby R.S. Tarvin
2018-10-08 8:09 ` Nils M Holm
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=F022808EFD5F0B8A438B332081972DD2@felloff.net \
--to=cinap_lenrek@felloff.net \
--cc=9fans@9fans.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).