unofficial mirror of guile-devel@gnu.org 
 help / color / mirror / Atom feed
From: "Neil Jerram" <neiljerram@googlemail.com>
To: "Ludovic Courtès" <ludo@gnu.org>, "Greg Troxel" <gdt@ir.bbn.com>
Cc: guile-devel@gnu.org
Subject: Re: Stack calibration
Date: Mon, 13 Oct 2008 22:37:20 +0100	[thread overview]
Message-ID: <49dd78620810131437i2cd4c746n7793c5d9f7cf7f0b@mail.gmail.com> (raw)
In-Reply-To: <49dd78620810121416g43be72b0pc98a1f682a726a24@mail.gmail.com>

2008/10/12 Neil Jerram <neiljerram@googlemail.com>:
>
> Here's the new patch.  Please (as ever!) let me know what you think.

One update to this below.  It isn't actually necessary or helpful, for
this case, to pull in the GOOPS interface to evaluator traps.

    Neil

diff --git a/libguile/measure-hwm.scm b/libguile/measure-hwm.scm
index 364740b..53a30d5 100644
--- a/libguile/measure-hwm.scm
+++ b/libguile/measure-hwm.scm
@@ -35,16 +35,16 @@
 ;; Define a variable to hold the measured stack high water mark (HWM).
 (define top-repl-hwm-measured 0)

-;; Install an evaluator trap to measure the stack size at every
+;; Use an evaluator trap to measure the stack size at every
 ;; evaluation step, and increase top-repl-hwm-measured if it is less
 ;; than the measured stack size.
-(use-modules (ice-9 debugging traps)
-            (oop goops))
-(install-trap (make <entry-trap>
-               #:behaviour (lambda (trap-context)
-                             (let ((stack-size (%get-stack-size)))
-                               (if (< top-repl-hwm-measured stack-size)
-                                   (set! top-repl-hwm-measured stack-size))))))
+(trap-set! enter-frame-handler
+          (lambda _
+            (let ((stack-size (%get-stack-size)))
+              (if (< top-repl-hwm-measured stack-size)
+                  (set! top-repl-hwm-measured stack-size)))))
+(trap-enable 'enter-frame)
+(trap-enable 'traps)

 ;; Call (turn-on-debugging) and (top-repl) in order to simulate as
 ;; closely as possible what happens - and in particular, how much
@@ -74,7 +74,7 @@

 ;; This is the value of top-repl-hwm-measured that we get on a
 ;; `canonical' build platform.  (See text below for what that means.)
-(define top-repl-hwm-i686-pc-linux-gnu 9685)
+(define top-repl-hwm-i686-pc-linux-gnu 9461)

 ;; Using the above results, output code that tests can run in order to
 ;; configure the stack limit correctly for the current build platform.




  reply	other threads:[~2008-10-13 21:37 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <47B2A8DF.9070004@tammer.net>
     [not found] ` <87tzkd8bvz.fsf@gnu.org>
     [not found]   ` <87ejbh8ben.fsf@gnu.org>
     [not found]     ` <47B2D88F.1040505@tammer.net>
     [not found]       ` <87ir0tvx6e.fsf@inria.fr>
2008-02-13 20:40         ` stack overflow Neil Jerram
2008-02-14  8:48           ` Ludovic Courtès
2008-02-14 10:26             ` Mikael Djurfeldt
2008-02-14 11:25               ` Ludovic Courtès
2008-02-14 11:39                 ` Mikael Djurfeldt
2008-02-25 21:52                   ` Neil Jerram
2008-07-16 12:34                     ` Ludovic Courtès
2008-09-12 20:47                     ` Stack calibration Ludovic Courtès
2008-09-27 18:20                       ` Neil Jerram
2008-09-28 20:05                         ` Ludovic Courtès
2008-09-30 22:10                           ` Neil Jerram
2008-10-02  8:25                             ` Andy Wingo
2008-10-02  8:38                               ` Neil Jerram
2008-10-02 22:30                             ` Neil Jerram
2008-10-06 22:32                               ` Ludovic Courtès
2008-10-06 23:11                                 ` Neil Jerram
2008-10-09 22:53                                   ` Neil Jerram
2008-10-10 13:22                                     ` Greg Troxel
2008-10-10 18:04                                       ` Neil Jerram
2008-10-10 18:28                                         ` Greg Troxel
2008-10-10 18:41                                           ` Neil Jerram
2008-10-11 17:22                                     ` Ludovic Courtès
2008-10-12 15:59                                       ` Neil Jerram
2008-10-12 21:16                                         ` Neil Jerram
2008-10-13 21:37                                           ` Neil Jerram [this message]
2008-10-14  7:25                                           ` Ludovic Courtès
2008-10-17 20:49                                             ` Neil Jerram
2008-10-14  7:19                                         ` Ludovic Courtès
2008-09-28 20:07                         ` Ludovic Courtès
2008-09-30 22:11                           ` Neil Jerram
2008-02-17  1:38           ` stack overflow Han-Wen Nienhuys
2008-02-17  9:20             ` Mikael Djurfeldt
2009-03-27 21:19 stack calibration Andy Wingo
2009-03-27 22:04 ` Mike Gran
2009-03-27 22:29   ` Julian Graham
2009-03-30 20:43 ` Neil Jerram
2009-03-31  3:39   ` Andy Wingo
2009-03-31 22:47     ` Neil Jerram
2009-04-03 17:44       ` Andy Wingo
2009-03-31 17:45   ` Greg Troxel
2009-04-17  9:35   ` Andy Wingo
2009-03-31 16:20 ` Ludovic Courtès

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

  List information: https://www.gnu.org/software/guile/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=49dd78620810131437i2cd4c746n7793c5d9f7cf7f0b@mail.gmail.com \
    --to=neiljerram@googlemail.com \
    --cc=gdt@ir.bbn.com \
    --cc=guile-devel@gnu.org \
    --cc=ludo@gnu.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 read-only IMAP folder(s) and NNTP newsgroup(s).