From: Thomas Gleixner <tglx@linutronix.de>
To: Roman Zippel <zippel@linux-m68k.org>
Cc: linux-kernel@vger.kernel.org, Andrew Morton <akpm@osdl.org>,
rostedt@goodmis.org, johnstul@us.ibm.com, mingo@elte.hu
Subject: Re: [patch 00/21] hrtimer - High-resolution timer subsystem
Date: Mon, 12 Dec 2005 19:37:53 +0100 [thread overview]
Message-ID: <1134412673.4205.230.camel@tglx.tec.linutronix.de> (raw)
In-Reply-To: <1134405768.4205.190.camel@tglx.tec.linutronix.de>
> On Mon, 2005-12-12 at 14:39 +0100, Roman Zippel wrote:
> > > Actually the change adds more code lines and removes one field of the
> > > hrtimer structure, but it has exactly the same functionality: Fast
> > > access to the first expiring timer without walking the rb_tree.
> >
> > Together with the state field this would save 12 bytes, which is IMO very
> > well worth considering.
> > You seem to have some plans for it, the best hint I've found for it is:
> >
> > + (This seperate list is also useful for high-resolution timers where we
> > + need seperate pending and expired queues while keeping the time-order
> > + intact.)"
> >
> > Could you please elaborate on this?
>
> Sure. I have already removed the list_head for the non high resolution
> case as it turned out that it does not hurt the high resolution
> implementation.
>
> For the high resolution implementation we have to move the expired
> timers to a seperate list, as we do not want to do complex callback
> functions from the event interrupt itself. But we have to reprogramm the
> next event interrupt, so we need simple access to the timer which
> expires first.
>
> The initial implementation did simply move the timer from the pending
> list to the expired list without doing the rb_tree removal inside of the
> event interrupt handler. That way the next event for reprogramming was
> the first event in the pending list.
>
> The new rebased version with the pending list removed does the rb_tree
> removal inside the event interrupt and enqueues the timer, for which the
> callback function has to be executed in the softirq, to the expired
> list. One exception are simple wakeup callback functions, as they are
> reasonably fast and we save two context switches. The next event for
> reprogramming the event interrupt is retrieved by the pointer in the
> base structure.
>
> This way the list head is only necessary for the high resolution case.
>
> The state field is not removed
Oops, I somehow managed to remove the rest of this paragraph :(
The state field is not removed because I'm not a big fan of those
overloaded fields and I prefer to pay the 4 byte penalty for the
seperation.
Of course if there is the absolute requirement to reduce the size, I'm
not insisting on keeping it.
tglx
next prev parent reply other threads:[~2005-12-12 18:31 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-06 0:01 [patch 00/21] hrtimer - High-resolution timer subsystem tglx
2005-12-06 0:01 ` [patch 01/21] Move div_long_long_rem out of jiffies.h tglx
2005-12-06 0:01 ` [patch 02/21] Remove duplicate div_long_long_rem implementation tglx
2005-12-06 0:01 ` [patch 03/21] Deinline mktime and set_normalized_timespec tglx
2005-12-06 0:01 ` [patch 04/21] Clean up mktime and make arguments const tglx
2005-12-06 0:01 ` [patch 05/21] Export deinlined mktime tglx
2005-12-06 0:01 ` [patch 06/21] Remove unused clock constants tglx
2005-12-06 0:01 ` [patch 07/21] Coding style clean up of " tglx
2005-12-06 0:01 ` [patch 08/21] Coding style and white space cleanup tglx
2005-12-06 0:01 ` [patch 09/21] Make clockid_t arguments const tglx
2005-12-06 0:01 ` [patch 10/21] Coding style and white space cleanup tglx
2005-12-06 0:01 ` [patch 11/21] Create and use timespec_valid macro tglx
2005-12-06 0:01 ` [patch 12/21] Validate timespec of do_sys_settimeofday tglx
2005-12-06 0:01 ` [patch 13/21] Introduce nsec_t type and conversion functions tglx
2005-12-06 0:01 ` [patch 14/21] Introduce ktime_t time format tglx
2005-12-06 0:01 ` [patch 15/21] hrtimer core code tglx
2005-12-15 3:43 ` Matt Helsley
2005-12-06 0:01 ` [patch 16/21] hrtimer documentation tglx
2005-12-06 0:01 ` [patch 17/21] Switch itimers to hrtimer tglx
2005-12-06 0:01 ` [patch 18/21] Create hrtimer nanosleep API tglx
2005-12-06 0:01 ` [patch 19/21] Switch sys_nanosleep to hrtimer tglx
2005-12-06 0:01 ` [patch 20/21] Switch clock_nanosleep to hrtimer nanosleep API tglx
2005-12-06 0:01 ` [patch 21/21] Convert posix timers completely tglx
2005-12-06 17:32 ` [patch 00/21] hrtimer - High-resolution timer subsystem Roman Zippel
2005-12-06 19:07 ` Ingo Molnar
2005-12-07 3:05 ` Roman Zippel
2005-12-08 5:18 ` Paul Jackson
2005-12-08 8:12 ` Ingo Molnar
2005-12-08 9:26 ` Ingo Molnar
2005-12-08 13:08 ` Roman Zippel
2005-12-08 15:36 ` Steven Rostedt
2005-12-06 22:10 ` Thomas Gleixner
2005-12-07 3:11 ` Roman Zippel
2005-12-06 22:28 ` Thomas Gleixner
2005-12-07 9:31 ` Andrew Morton
2005-12-07 10:11 ` Ingo Molnar
2005-12-07 10:20 ` Ingo Molnar
2005-12-07 10:23 ` Nick Piggin
2005-12-07 10:49 ` Ingo Molnar
2005-12-07 11:09 ` Nick Piggin
2005-12-07 11:33 ` Ingo Molnar
2005-12-07 11:40 ` Nick Piggin
2005-12-07 13:06 ` Roman Zippel
2005-12-07 12:40 ` Roman Zippel
2005-12-07 23:12 ` Nick Piggin
2005-12-07 12:18 ` Roman Zippel
2005-12-07 16:55 ` Ingo Molnar
2005-12-07 17:17 ` Roman Zippel
2005-12-07 17:57 ` Ingo Molnar
2005-12-07 18:18 ` Roman Zippel
2005-12-07 18:02 ` Paul Baxter
2005-12-09 17:23 ` Thomas Gleixner
2005-12-12 13:39 ` Roman Zippel
2005-12-12 16:42 ` Thomas Gleixner
2005-12-12 18:37 ` Thomas Gleixner [this message]
2005-12-13 1:25 ` George Anzinger
2005-12-13 9:18 ` Thomas Gleixner
2005-12-15 1:35 ` Roman Zippel
2005-12-15 2:29 ` George Anzinger
2005-12-19 14:56 ` Roman Zippel
2005-12-19 20:54 ` George Anzinger
2005-12-21 23:03 ` Roman Zippel
2005-12-22 4:30 ` George Anzinger
2005-12-14 20:48 ` Roman Zippel
2005-12-14 22:30 ` Thomas Gleixner
2005-12-15 0:55 ` George Anzinger
2005-12-15 14:18 ` Steven Rostedt
2005-12-19 14:50 ` Roman Zippel
2005-12-19 22:05 ` Thomas Gleixner
2005-12-13 12:45 Nicolas Mailhot
2005-12-13 23:38 ` George Anzinger
2005-12-14 8:58 ` Kyle Moffett
2005-12-14 10:03 ` Nicolas Mailhot
2005-12-15 1:11 ` George Anzinger
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=1134412673.4205.230.camel@tglx.tec.linutronix.de \
--to=tglx@linutronix.de \
--cc=akpm@osdl.org \
--cc=johnstul@us.ibm.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rostedt@goodmis.org \
--cc=zippel@linux-m68k.org \
/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).