3:33 a.m.
================================================================
Dear Bug Team!
I do a lot of ad hoc filtering of HTML files using shell pipelines. I test
those pipelines in XEmacs using shell-command-on-region. With 21.5 I notice
this process has become abysmally slow. I've whittled the problem down as
much as I can. If I visit /usr/share/file/magic (~ 360Kbytes on my Mac) and
execute
C-x h M-| cat RET
the command executes nearly instantaneously, as you'd expect. If, on the
hand, I execute this little Python script instead of cat:
import sys
for line in sys.stdin:
sys.stdout.write(line)
or this even simpler one:
import sys
sys.stdout.write(sys.stdin.read())
the command takes what seems like forever to execute. I just measured it at
about 19 seconds on my Mac PowerBook. I get the same result whether I use
Python 2.3, 2.4 or Python built from the latest CVS source.
I no longer have XEmacs 21.4 installed. If others can't reproduce this, let
me know and I'll download and try it. I don't recall this happening with
that version though. I can also give it a try without any fink (/sw/...)
libraries linked in.
--
Skip Montanaro - skip(a)pobox.com - http://www.mojam.com/
"The values to which people cling most stubbornly under inappropriate
conditions are those values that were previously the source of their
greatest triumphs over adversity." -- Jared Diamond in "Collapse"
================================================================
System Info to help track down your bug:
---------------------------------------
uname -a: Darwin montanaro.dyndns.org 8.5.0 Darwin Kernel Version 8.5.0: Sun Jan 22
10:38:46 PST 2006; root:xnu-792.6.61.obj~1/RELEASE_PPC Power Macintosh powerpc
./configure '--prefix=/Users/skip/local' '--with-optimization'
'--without-error-checking' 'LDFLAGS=-L/sw/lib'
'CPPFLAGS=-I/sw/include'
XEmacs 21.5-b25 "eggplant" (+CVS-20060318) configured for
`powerpc-apple-darwin8.5.0'.
Compilation Environment and Installation Defaults:
Source code location: /Users/skip/src/xemacs-21.5
Installation prefix: /Users/skip/local
Operating system description file: `s/darwin.h'
Machine description file: `m/powerpc.h'
Compiler version: powerpc-apple-darwin8-gcc-4.0.0 (GCC) 4.0.0 (Apple
Computer, Inc. build 5026)
- GCC specs file: specs.
- Compiler command: gcc -Wall -Wno-switch -Wundef -Wsign-compare
-Wno-char-subscripts -Wpacked -Wpointer-arith -Wunused-parameter -g -O3
libc version:
Relocating allocator for buffers: no
GNU version of malloc: no
- The GNU allocators don't work with this system configuration.
Window System:
Compiling in support for the X window system:
- X Windows headers location: /usr/X11R6/include
- X Windows libraries location: /usr/X11R6/lib
- Handling WM_COMMAND properly.
Compiling in support for the Athena widget set:
- Athena headers location: X11/Xaw
- Athena library to link: Xaw
Using Lucid menubars.
Using Lucid scrollbars.
Using Athena dialog boxes.
Using Athena native widgets.
TTY:
Compiling in support for ncurses.
Images:
Compiling in support for GIF images (builtin).
Compiling in support for XPM images.
Compiling in support for PNG images.
Compiling in support for JPEG images.
Compiling in support for TIFF images.
Sound:
Databases:
Compiling in support for Berkeley database.
Compiling in support for GNU DBM.
Compiling in support for LDAP.
Internationalization:
Mail:
Compiling in support for "file" mail spool file locking method.
Other Features:
Inhibiting IPv6 canonicalization at startup.
Compiling in support for dynamic shared object modules.
Using the new GC mark algorithms (KKCC).
WARNING: ---------------------------------------------------------
WARNING: The new algorithms are experimental. They are enabled by
WARNING: default for this release. Use `--disable-kkcc' to
WARNING: turn it off.
WARNING: ---------------------------------------------------------
Using the new portable dumper.
Dumping into executable.
Compiling in support for extra debugging code.
Load-Path Lisp Shadows:
----------------------
(/Users/skip/emacs/macedit
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/calc/macedit
/Users/skip/emacs/pdb
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/debug/pdb
/Users/skip/emacs/tempo
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/edit-utils/tempo
/Users/skip/emacs/python-mode
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/python-modes/python-mode
/Users/skip/emacs/slider
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/slider/slider
/Users/skip/emacs/po-mode
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/text-modes/po-mode
/Users/skip/emacs/vm-sort
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/vm/vm-sort
/Users/skip/emacs/imenu
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/xemacs-base/imenu
/Users/skip/emacs/ffap
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/xemacs-base/ffap
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/xemacs-base/regexp-opt
/Users/skip/local/lib/xemacs-21.5-b25/lisp/regexp-opt
/Users/skip/emacs/isearch-mode
/Users/skip/local/lib/xemacs-21.5-b25/lisp/isearch-mode
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/xemacs-base/easy-mmode
/Users/skip/local/lib/xemacs-21.5-b25/lisp/easy-mmode
/Users/skip/local/lib/xemacs/xemacs-packages/lisp/build/build-report
/Users/skip/local/lib/xemacs-21.5-b25/lisp/build-report)
Installed XEmacs Packages:
-------------------------
(xetla ver: 1.01 upstream: steve(a)eicq.org--2005/xetla--main--1.1--version-0)
(xemacs-base ver: 2.01 upstream: No-Upstream-Ver)
(w3 ver: 1.32 upstream: 4.0pre47)
(python-modes ver: 1.07 upstream: No-Upstream-Ver)
(gnus ver: 1.89 upstream: 5.10.7)
(fortran-modes ver: 1.05 upstream: No-Upstream-Ver)
(zenirc ver: 1.16 upstream: 2.112)
(xwem ver: 1.21 upstream: lg(a)xwem.org--2005/xwem--main--2.1--versionfix-1)
(xslt-process ver: 1.12 upstream: 1.2.1)
(xslide ver: 1.09 upstream: 0.2.2)
(xlib ver: 1.14 upstream: lg(a)xwem.org--2005/xlib--main--2.1--version-0)
(xemacs-devel ver: 1.72 upstream: No-Upstream-Ver)
(x-symbol ver: 1.1 upstream: 4.5.1)
(vm ver: 7.22 upstream: 7.17)
(viper ver: 1.47 upstream: 3.09)
(view-process ver: 1.13 upstream: 2.4)
(vhdl ver: 1.2 upstream: 3.33.2)
(vc-cc ver: 1.22 upstream: No-Upstream-Ver)
(vc ver: 1.41 upstream: No-Upstream-Ver)
(tramp ver: 1.28 upstream: 2.0.51)
(tpu ver: 1.14 upstream: 4.2X)
(tooltalk ver: 1.15 upstream: No-Upstream-Ver)
(tm ver: 1.38 upstream: No-Upstream-Ver)
(time ver: 1.14 upstream: 1.17)
(textools ver: 1.15 upstream: No-Upstream-Ver)
(text-modes ver: 1.9 upstream: No-Upstream-Ver)
(texinfo ver: 1.3 upstream: No-Upstream-Ver)
(supercite ver: 1.21 upstream: 3.55x3)
(strokes ver: 1.1 upstream: No-Upstream-Ver)
(speedbar ver: 1.28 upstream: 0.14beta4)
(sounds-wav ver: 1.12 upstream: No-Upstream-Ver)
(sounds-au ver: 1.12 upstream: No-Upstream-Ver)
(sml-mode ver: 0.12 upstream: 3.9.5)
(slider ver: 1.15 upstream: 0.3x1)
(sieve ver: 1.18 upstream: No-Upstream-Ver)
(sh-script ver: 1.21 upstream: 2.0f)
(sgml ver: 1.11 upstream: No-Upstream-Ver)
(semantic ver: 1.19 upstream: 1.4.2)
(scheme ver: 1.14 upstream: No-Upstream-Ver)
(sasl ver: 1.16 upstream: 1.14.4)
(ruby-modes ver: 1.02 upstream: 1.6.8)
(rmail ver: 1.14 upstream: No-Upstream-Ver)
(riece ver: 1.21 upstream: 2.0.2)
(reftex ver: 1.34 upstream: 4.21)
(re-builder ver: 1.05 upstream: 1.20)
(psgml-dtds ver: 1.03 upstream: No-Upstream-Ver)
(psgml ver: 1.44 upstream: 1.3.1)
(ps-print ver: 1.11 upstream: 6.5.6)
(prog-modes ver: 2.06 upstream: No-Upstream-Ver)
(pgg ver: 1.06 upstream: 0.1)
(perl-modes ver: 1.08 upstream: No-Upstream-Ver)
(pcomplete ver: 1.04 upstream: 1.1.6)
(pcl-cvs ver: 1.66 upstream: R-2_9_9)
(pc ver: 1.28 upstream: No-Upstream-Ver)
(os-utils ver: 1.36 upstream: No-Upstream-Ver)
(oo-browser ver: 1.04 upstream: 4.08)
(ocaml ver: 0.05 upstream: 3.06)
(net-utils ver: 1.48 upstream: N/A)
(mmm-mode ver: 1.02 upstream: 0.4.7)
(misc-games ver: 1.18 upstream: No-Upstream-Ver)
(mine ver: 1.16 upstream: 1.9)
(mh-e ver: 1.29 upstream: 7.4.2)
(mew ver: 1.19 upstream: 1.94.2)
(mailcrypt ver: 2.14 upstream: 3.5.8)
(mail-lib ver: 1.75 upstream: No-Upstream-Ver)
(liece ver: 1.15 upstream: 1.4.9)
(jde ver: 1.48 upstream: 2.3.2)
(ispell ver: 1.32 upstream: 3.6)
(ilisp ver: 1.33 upstream: 5.12.0)
(igrep ver: 1.14 upstream: 2.111)
(idlwave ver: 1.32 upstream: 5.1)
(ibuffer ver: 1.09 upstream: No-Upstream-Ver)
(hyperbole ver: 1.16 upstream: 4.18)
(hm--html-menus ver: 1.23 upstream: 5.9)
(haskell-mode ver: 1.07 upstream: 1.45)
(gnats ver: 1.17 upstream: 3.101)
(general-docs ver: 1.04 upstream: No-Upstream-Ver)
(games ver: 1.17 upstream: 1.04)
(fsf-compat ver: 1.15 upstream: No-Upstream-Ver)
(frame-icon ver: 1.11 upstream: No-Upstream-Ver)
(forms ver: 1.15 upstream: 2.37)
(footnote ver: 1.16 upstream: 0.18x)
(eudc ver: 1.39 upstream: 1.32)
(eterm ver: 1.17 upstream: No-Upstream-Ver)
(ess ver: 1.16 upstream: 5.2.7)
(eshell ver: 1.1 upstream: 2.4.1)
(escreen ver: 1.01 upstream: 1.16)
(erc ver: 0.19 upstream: Version 5.0 Revision: 1.776)
(emerge ver: 1.11 upstream: No-Upstream-Ver)
(elib ver: 1.11 upstream: 1.0)
(eieio ver: 1.05 upstream: 0.17)
(efs ver: 1.33 upstream: 1.23)
(edt ver: 1.13 upstream: No-Upstream-Ver)
(edit-utils ver: 2.32 upstream: No-Upstream-Ver)
(ediff ver: 1.6 upstream: 2.75)
(edebug ver: 1.21 upstream: No-Upstream-Ver)
(ecrypto ver: 0.2 upstream: 2.0)
(ecb ver: 1.22 upstream: 2.31)
(docbookide ver: 0.07000000000000001 upstream: 0.1)
(dired ver: 1.17 upstream: 7.13)
(dictionary ver: 1.16 upstream: 1.8)
(debug ver: 1.17 upstream: No-Upstream-Ver)
(crisp ver: 1.15 upstream: 1.34)
(cookie ver: 1.15 upstream: No-Upstream-Ver)
(clearcase ver: 1.1 upstream: /main/laptop/165)
(cc-mode ver: 1.45 upstream: 5.30.10)
(calendar ver: 1.23 upstream: No-Upstream-Ver)
(calc ver: 1.26 upstream: 2.02fX3)
(c-support ver: 1.22 upstream: No-Upstream-Ver)
(build ver: 1.14 upstream: 2.02)
(bbdb ver: 1.29 upstream: 2.34)
(auctex ver: 1.46 upstream: 11.55)
(apel ver: 1.32 upstream: 10.6)
(ada ver: 1.14 upstream: 2.27)
(Sun ver: 1.16 upstream: No-Upstream-Ver)
Installed Modules:
-----------------
Features:
--------
(xemacsbug shadow vm-summary vm-undo mail-abbrevs vm-menu sendmail vm-reply
vm-misc vm-mouse vm-window vm-delete vm-autoload vm-vars highlight-headers
vm-version vm vm-startup python-mode info-look overlay info ansi-color
html-mode html-helper-mode auc-menu tempo mail-utils annotations passwd
func-menu package-ui package-get package-admin config lazy-lock paren
blink-paren gnuserv font-lock cus-face complete mailcrypt rfc822 timezone
supercite mail-extr regi filladapt edmacro timer tramp-util compile
tramp-smb tramp-vc tramp regexp-opt executable advice advice-preload shell
password format-spec timer-funcs trampver efs-x19.15 efs-fnh efs-ovwrt
efs-cu efs-netrc efs-defun comint efs vc ring vc-hooks zenirc-autoloads
xwem-autoloads xslt-process-autoloads xslide-autoloads xlib-autoloads
xetla-autoloads xemacs-devel-autoloads xemacs-base-autoloads
x-symbol-autoloads w3-autoloads vm-autoloads viper-autoloads
view-process-autoloads vhdl-autoloads vc-cc-autoloads vc-autoloads
tramp-autoloads tpu-autoloads tooltalk-autoloads tm-autoloads time-autoloads
textools-autoloads text-modes-autoloads texinfo-autoloads
supercite-autoloads strokes-autoloads speedbar-autoloads
sounds-wav-autoloads sounds-au-autoloads sml-mode-autoloads slider-autoloads
sieve-autoloads sh-script-autoloads sgml-autoloads semantic-autoloads
scheme-autoloads sasl-autoloads ruby-modes-autoloads rmail-autoloads
riece-autoloads reftex-autoloads re-builder-autoloads python-modes-autoloads
psgml-dtds-autoloads psgml-autoloads ps-print-autoloads prog-modes-autoloads
pgg-autoloads perl-modes-autoloads pcomplete-autoloads pcl-cvs-autoloads
pc-autoloads os-utils-autoloads oo-browser-autoloads ocaml-autoloads
net-utils-autoloads mmm-mode-autoloads misc-games-autoloads mine-autoloads
mh-e-autoloads mew-autoloads mailcrypt-autoloads mail-lib-autoloads
liece-autoloads jde-autoloads ispell-autoloads ilisp-autoloads
igrep-autoloads idlwave-autoloads ibuffer-autoloads hyperbole-autoloads
hm--html-menus-autoloads haskell-mode-autoloads gnus-autoloads
gnats-autoloads general-docs-autoloads games-autoloads fsf-compat-autoloads
frame-icon-autoloads fortran-modes-autoloads forms-autoloads
footnote-autoloads eudc-autoloads eterm-autoloads ess-autoloads
eshell-autoloads escreen-autoloads erc-autoloads emerge-autoloads
elib-autoloads eieio-autoloads efs-autoloads edt-autoloads
edit-utils-autoloads ediff-autoloads edebug-autoloads ecrypto-autoloads
ecb-autoloads docbookide-autoloads dired-autoloads dictionary-autoloads
debug-autoloads crisp-autoloads cookie-autoloads clearcase-autoloads
cc-mode-autoloads calendar-autoloads calc-autoloads c-support-autoloads
build-autoloads bbdb-autoloads auctex-autoloads apel-autoloads ada-autoloads
Sun-autoloads modules-autoloads auto-autoloads loadhist rsz-minibuf
auto-show fontl-hooks code-cmds gutter-items menubar-items x-menubar
mode-motion mouse behavior itimer auto-save lisp-mode easymenu easy-mmode
iso8859-1 page buff-menu lib-complete cus-file derived newcomment env
text-props fontconfig frame obsolete cus-start custom widget cl-extra cl
cl-19 packages backquote unicode lucid-scrollbars cut-buffer lucid-menubars
athena-dialogs x c-balloon-help tty-frames tty toolbar scrollbar
unix-processes multicast network-streams subprocesses modules
menu-accelerator-support menubar berkeley-db dbm md5 xemacs gutter tiff png
gif jpeg xpm xbm lisp-float-type file-coding darwin dialog devices
window-system base64)
Recent keystrokes:
-----------------
k RET M-x v - v - f C-g C-x C-f t m p / t o u r s .
h t m l RET C-x h ESC | p y t h o n SPC p y c a t .
p y RET C-x o C-x k RET C-x C-f t m p / p y c a t .
p y RET C-n C-n C-n C-n C-x 4 m x e m a c s - b e t
a SPC C-x k RET M-x r e p o r t - x e TAB RET
Recent messages (most recent first):
-----------------------------------
Auto save file for draft message exists; consider M-x mail-recover
Parsing /Users/skip/.mailrc... done
Parsing /Users/skip/.mailrc...
Warning:
Defaulting to the GNU Emacs-derived `sendmail.el' mail client. This facility,
while part of base XEmacs, is heinously underfeatured, and not going to get
better in the medium term. We include it so that bug reports work without
packages; we suggest that you choose and/or install one of the other mail
clients from packages if you're doing something other than M-x
report-xemacs-bug , or even if you are reporting bugs regularly.
To choose a package from those installed, click on "Options" ->
"Internet" -> "Compose Mail With ..." and decide on one from the
list. Gnus and VM are full-featured and have active user communities.
To disable this warning and stick with the old behavior, you can explicitly
initialize `mail-user-agent' to 'sendmail-user-agent .
Loading xemacsbug...done
Loading xemacsbug...
Loading vm-summary...done
Loading vm-summary...
Loading vm-undo...done
Loading vm-undo...
Parsing /Users/skip/.mailrc... done
Parsing /Users/skip/.mailrc...
Loading mail-abbrevs...done
Loading mail-abbrevs...
Loading vm-menu...done
Loading vm-menu...
Loading sendmail...done
Loading sendmail...
Loading vm-reply...done
Loading vm-reply...
3:46 p.m.
"skip＠pobox.comFCC":~/Mail/archive/2006-03＠gwyn.tux.org writes:
I do a lot of ad hoc filtering of HTML files using shell pipelines.
I test
those pipelines in XEmacs using shell-command-on-region. With 21.5 I notice
this process has become abysmally slow. I've whittled the problem down as
much as I can. If I visit /usr/share/file/magic (~ 360Kbytes on my Mac) and
execute
C-x h M-| cat RET
the command executes nearly instantaneously, as you'd expect. If, on the
hand, I execute this little Python script instead of cat:
import sys
for line in sys.stdin:
sys.stdout.write(line)
...
the command takes what seems like forever to execute. I just measured it at
about 19 seconds on my Mac PowerBook. I get the same result whether I use
Python 2.3, 2.4 or Python built from the latest CVS source.
Maybe the following patch will help you. It makes running
subprocesses (eg: man) much faster for me. This patch was in
XEmacs 21.5 somewhere around 21.5.19 I think, but was reverted a
few versions back for some reason.
*** src/process-unix.c~ Sat Mar 1 02:25:36 2003
--- src/process-unix.c Mon Jul 12 22:00:37 2004
***************
*** 1530,1536 ****
/* Buffer is full. Wait, accepting input;
that may allow the program
to finish doing output and read more. */
! Faccept_process_output (Qnil, make_int (1), Qnil);
/* It could have *really* finished, deleting the process */
if (NILP(p->pipe_outstream))
return;
--- 1530,1537 ----
/* Buffer is full. Wait, accepting input;
that may allow the program
to finish doing output and read more. */
! // Faccept_process_output (Qnil, make_int (1), Qnil);
! Faccept_process_output (Qnil, Qnil, make_int (10));
/* It could have *really* finished, deleting the process */
if (NILP(p->pipe_outstream))
return;
--
--tony
3:56 a.m.
> I do a lot of ad hoc filtering of HTML files using shell
pipelines.
> I test those pipelines in XEmacs using shell-command-on-region. With
> 21.5 I notice this process has become abysmally slow.
tbennett> Maybe the following patch will help you.
Thanks, this worked perfectly. Maybe this should be reapplied to the
source.
Skip
4:16 a.m.
New subject: Reduce timeout in Faccept_process_output [was: XEmacs 21.5 doesn't like Python]
>>>> "tbennett" == tbennett
<tbennett(a)nvidia.com> writes:
tbennett> Maybe the following patch will help you. It makes
tbennett> running subprocesses (eg: man) much faster for me.
Since there is a timeout here already, it's probably safe to make it a
lot shorter. I'll forward it to xemacs-patches, and if there's no
objection apply in a couple days or so.
tbennett> This patch was in XEmacs 21.5 somewhere around 21.5.19 I
tbennett> think, but was reverted a few versions back for some reason.
This patch was never applied AFAICT, although you sent it to
xemacs-beta twice. Nobody ever responded to your posts, though. (I
think this was probably due to me assuming Jerry would handle it, and
Jerry---due to no fault of his own---was unable to handle anything for
a while.) I would guess you've just had it in your local tree for
years, and never noticed that it was never in a release!
Send any patch that you've done as much with as you intend to, and
believe is ready to be applied, to xemacs-patches. It's much more
likely to be treated as an honored guest there! :-)
In the future, that is; I'll handle this one from here.
--
School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
4:45 a.m.
New subject: Reduce timeout in Faccept_process_output
"Stephen J. Turnbull" <stephen(a)xemacs.org> writes:
This patch was never applied AFAICT, although you sent it to
xemacs-beta twice. Nobody ever responded to your posts, though. (I
think this was probably due to me assuming Jerry would handle it, and
Jerry---due to no fault of his own---was unable to handle anything for
a while.) I would guess you've just had it in your local tree for
years, and never noticed that it was never in a release!
You're right; I must have confused it with some other
process buffering changes.
Thanks for shepherding it thru.
--
--tony
7:55 a.m.
tbennett(a)nvidia.com writes:
*** src/process-unix.c~ Sat Mar 1 02:25:36 2003
--- src/process-unix.c Mon Jul 12 22:00:37 2004
***************
*** 1530,1536 ****
/* Buffer is full. Wait, accepting input;
that may allow the program
to finish doing output and read more. */
! Faccept_process_output (Qnil, make_int (1), Qnil);
/* It could have *really* finished, deleting the process */
if (NILP(p->pipe_outstream))
return;
--- 1530,1537 ----
/* Buffer is full. Wait, accepting input;
that may allow the program
to finish doing output and read more. */
! // Faccept_process_output (Qnil, make_int (1), Qnil);
Please change the above to a c-style comment before commit.
C++ comments are not allowed in the XEmacs code base, IIRC.
What about the windows platform?
Should a similar change be applied there too?
Has anyone already looked into this?
Best regards,
Adrian
! Faccept_process_output (Qnil, Qnil, make_int (10));
/* It could have *really* finished, deleting the process */
if (NILP(p->pipe_outstream))
return;
--
Adrian Aichner
mailto:adrian＠xemacs.org
http://www.xemacs.org/
2:38 p.m.
>>>> "APA" == Adrian Aichner
<adrian(a)xemacs.org> writes:
APA> What about the windows platform?
OK, I couldn't resist the challenge, and took a look.
APA> Should a similar change be applied there too?
It's already in process-nt.c, except that there's a duration doubling
loop (starting at 25ms with a maximum period of 1s). Does anybody
know whether this is a good idea? I suppose the idea is to avoid
polling every 25ms if the process is actually producing output only at
1s intervals, but it's not obvious to me that this really is a huge
resource drain.
Here's a patch for process-unix.c that incorporates this idea, for the
sake of discussion. (The differences with process-nt.c are that this
starts at 10ms, and the NT code doesn't need to fiddle with the
SIGPIPE handler, so it's actually much simpler.)
diff --git a/src/process-unix.c b/src/process-unix.c
index e0e64ed..121a7b8 100644
--- a/src/process-unix.c
+++ b/src/process-unix.c
＠＠ -1543,20 +1543,25 ＠＠ unix_send_process (Lisp_Object proc, str
Lstream_flush (XLSTREAM (DATA_OUTSTREAM(p)));
EMACS_SIGNAL (SIGPIPE, old_sigpipe);
}
- while (Lstream_was_blocked_p (XLSTREAM (p->pipe_outstream)))
- {
- /* Buffer is full. Wait, accepting input;
- that may allow the program
- to finish doing output and read more. */
- Faccept_process_output (Qnil, make_int (1), Qnil);
- /* It could have *really* finished, deleting the process */
- if (NILP(p->pipe_outstream))
- return;
- old_sigpipe =
- (SIGTYPE (*) (int)) EMACS_SIGNAL (SIGPIPE, send_process_trap);
- Lstream_flush (XLSTREAM (p->pipe_outstream));
- EMACS_SIGNAL (SIGPIPE, old_sigpipe);
- }
+ {
+ int wait_ms = 10;
+ while (Lstream_was_blocked_p (XLSTREAM (p->pipe_outstream)))
+ {
+ /* Buffer is full. Wait a bit, accepting input; that may
+ allow the program to finish doing output and read more.
+ Used to be 1s, but that's excruciatingly slow.
+ N.B. timeout_secs = Qnil is faster than Qzero. */
+ Faccept_process_output (Qnil, Qnil, make_int (wait_ms));
+ /* It could have *really* finished, deleting the process */
+ if (NILP(p->pipe_outstream))
+ return;
+ old_sigpipe =
+ (SIGTYPE (*) (int)) EMACS_SIGNAL (SIGPIPE, send_process_trap);
+ Lstream_flush (XLSTREAM (p->pipe_outstream));
+ EMACS_SIGNAL (SIGPIPE, old_sigpipe);
+ wait_ms = min (1000, 2 * wait_ms);
+ }
+ }
/* Perhaps should ABORT() if < 0? This should never happen. */
}
while (chunklen > 0);
diff --git a/src/ChangeLog b/src/ChangeLog
index fce3389..0ba6a97 100644
--- a/src/ChangeLog
+++ b/src/ChangeLog
＠＠ -1,3 +1,13 ＠＠
+2006-03-23 Stephen J. Turnbull <stephen(a)xemacs.org>
+
+ * process-unix.c (unix_send_process): Follow process-nt here, and
+ use an increasing timeout (up to 1s).
+
+2006-03-23 Tony Bennett <tbennett(a)nvidia.com>
+
+ * process-unix.c (unix_send_process): Reduce timeout in
+ Faccept_process_output to 10ms.
+
2006-03-20 Marcus Crestani <crestani(a)xemacs.org>
* dynarr.c (Dynarr_realloc): Malloc correct size.
--
School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Ask not how you can "do" free software business;
ask what your business can "do for" free software.
12:20 p.m.
New subject: Patch: fix wrong use of memset
It looks like memset () is not used correctly in
xemacs-21.5.25/src/vdb-posix.c.
Patch attached.
--
Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian
睡眠不足はいい仕事の敵だ。
4:21 p.m.
New subject: [AC21.5] fix wrong use of memset
APPROVE COMMIT 21.5
>>>>"MF" == Mike FABIAN <mfabian(a)suse.de>
writes:
MF> It looks like memset () is not used correctly in
MF> xemacs-21.5.25/src/vdb-posix.c.
Good catch, thanks.
Committed this:
src/ChangeLog addition:
2006-03-27 Mike Fabian <mfabian(a)suse.de>
* vdb-posix.c (vdb_install_signal_handler): Correct memset.
xemacs-21.5 source patch:
Diff command: cvs -q diff -u
Files affected: src/vdb-posix.c
Index: src/vdb-posix.c
===================================================================
RCS file: /pack/xemacscvs/XEmacs/xemacs/src/vdb-posix.c,v
retrieving revision 1.1
diff -u -r1.1 vdb-posix.c
--- src/vdb-posix.c 25 Nov 2005 01:42:07 -0000 1.1
+++ src/vdb-posix.c 27 Mar 2006 15:18:25 -0000
＠＠ -118,7 +118,7 ＠＠
}
#ifdef USE_SIGACTION
- memset(&act, sizeof(struct sigaction), 0);
+ memset(&act, 0, sizeof(struct sigaction));
act.sa_sigaction = vdb_fault_handler;
sigemptyset (&act.sa_mask);
act.sa_flags = SA_SIGINFO;
--
Marcus
6602
days inactive
6610
days old
8 comments
7 participants
participants (7)
-
"skip＠pobox.comFCC:~/Mail/archive/2006-03"＠gwyn.tux.org -
Adrian Aichner -
Marcus Crestani -
Mike FABIAN -
Skip Montanaro -
Stephen J. Turnbull -
Tony Bennett