From: UnixOS2 Archive To: "UnixOS2 Archive" Date: Sun, 17 Mar 2002 04:19:47 EST-10EDT,10,-1,0,7200,3,-1,0,7200,3600 Subject: [UnixOS2_Archive] No. 166 ************************************************** Saturday 16 March 2002 Number 166 ************************************************** Subjects for today 1 Re: Emacs docs in English : Masaru Nomiya 2 Re: Porting Joe's Own Editor (JOE) : Stepan Kazakov 3 Re: Re: Emacs docs in English : Masaru Nomiya 4 Re: Re: Emacs docs in English : Masaru Nomiya 5 Re: Autoconf 2.52h : Henry Sobotka 6 Re: Porting Joe's Own Editor (JOE) : Marty" 7 Re: Porting Joe's Own Editor (JOE) : Brian Smith" 8 Re: Porting Joe's Own Editor (JOE) : Marty" 9 Re: Porting Joe's Own Editor (JOE) : Brian Smith" 10 Re: Porting Joe's Own Editor (JOE) : John Poltorak 11 Re: Porting Joe's Own Editor (JOE) : Thomas Dickey 12 Re: Re: Emacs docs in English : John Poltorak 13 Re: Porting Joe's Own Editor (JOE) : Thomas Dickey 14 Emacs .el & .elc files : John Poltorak 15 Re: OS/2 v. Linux performance : Dave and Natalie" 16 Re: Emacs .el & .elc files : Stefan Neis 17 Re: Porting Joe's Own Editor (JOE) : John Poltorak 18 Building XFree86 v3.3.6 : John Poltorak 19 Re: Porting Joe's Own Editor (JOE) : John Poltorak 20 Re: OS/2 v. Linux performance : Adrian Gschwend" 21 Re: OS/2 v. Linux performance : Adrian Gschwend" 22 Re: OS/2 v. Linux performance : Adrian Gschwend" **= Email 1 ==========================** Date: Sun, 17 Mar 2002 01:01:47 +0900 From: Masaru Nomiya Subject: Re: Emacs docs in English Hello, In the Message; Subject : Re: Emacs docs in English Message-ID : <20020316150659.X54364 at eyup.org> Date & Time: Sat, 16 Mar 2002 15:06:59 +0000 [John] == John Poltorak has written: John> I would if somebody could translate these directions for me:- John> ^[$B!&$=$l$H$b#M#L$XEPO?4uK>$G$7$g$&$+!)^[(B John> ^[$BEPO?$N4uK>$J$i%7%s%?%C%/%9$,0c$$$^$9!#^[(B John> subscribe ^[$B$"$J$?$NL>A0^[(B John> ^[$BNc^[(B: subscribe Hayakawa Aoi John> ^[$B$N$h$&$J%7%s%?%C%/%9$GAw$C$F2<$5$$!#^[(B John> Hi, I am the fml ML manager for the ML . Written in Japanese only! I am so sorry. This message means; If you want to join this ML, send your mail with the body subscribe John Poltorak Then you will receive a mail to confirm. Just send back with the top line, maybe Confirm ........ PS. I must translate .... --- Masaru Nomiya mail-to: nomiya at ttmy.ne.jp "No WIndows, no gains!" ..... "Why, I am wrong?" -- Bill -- **= Email 2 ==========================** Date: Sun, 17 Mar 2002 01:17:49 -0500 From: Stepan Kazakov Subject: Re: Porting Joe's Own Editor (JOE) Marty wrote: > > BTW which TOP do you mean? > The one I use is: "TOP 0.3 - Ian.Hargreaves at ibm.net" > I'm not sure if this was a port or a workalike though, now that I look at it. latest good top/2 -> http://home.clara.net/orac/os2.htm -- madded. [Red Hot Chili Hackers] **= Email 3 ==========================** Date: Sun, 17 Mar 2002 01:55:41 +0900 From: Masaru Nomiya Subject: Re: Re: Emacs docs in English Hello, In the Message; Subject : Re: Re: Emacs docs in English Message-ID : <20020316160809.D54364 at eyup.org> Date & Time: Sat, 16 Mar 2002 16:08:09 +0000 [John] == John Poltorak has written: John> Is it OK to post msgs in English? Yes, please. Mr. Nakagawa, present maintenor of Emacs 20.7, would welcome you. Regards, --- Masaru Nomiya mail-to: nomiya at ttmy.ne.jp "No WIndows, no gains!" ..... "Why, I am wrong?" -- Bill -- **= Email 4 ==========================** Date: Sun, 17 Mar 2002 11:25:22 +0900 From: Masaru Nomiya Subject: Re: Re: Emacs docs in English Hello, In the Message; Subject : Re: Re: Emacs docs in English Message-ID : <20020316185520.E54364 at eyup.org> Date & Time: Sat, 16 Mar 2002 18:55:20 +0000 [John] == John Poltorak has written: John> WellI tried to subscribe, but got a confirmation reply which I don't know John> how to respond to. It had something like this:- John> confirm 84682771 Anne Shirley John> ^[$B$r4^$`!X$3$N%a!<%j%s%0%j%9%H$KEPO?$r$7$F$b$h$$$+!)!Y$H$$$&3NG'$r5a$a$ [...] John> confirm ^[$B%Q%9%o!<%I^[(B(^[$B?t;z^[(B) ^[$B$"$J$?$NL>A0^[(B Isn't there a sentence confirm 23469876098...878 John Poltorak in the returned mail from os2-emacs-ja-admnin at m17n.org? Send a mail with the body confirm 23469876098...878 John Poltorak to os2-emacs-ja-ctl at m17n.org, not to os2-emacs-ja-admnin at m17n.org. Regards, --- Masaru Nomiya mail-to: nomiya at ttmy.ne.jp "No Windows, no gains!" ... "Why, I am wrong?" -- Bill -- **= Email 5 ==========================** Date: Sun, 17 Mar 2002 11:50:52 -0500 From: Henry Sobotka Subject: Re: Autoconf 2.52h Thomas Hoffmann wrote: > > Just on a side note: IIRC the C9x draft at least has a paragraph called > "7.12.3.4: The isnan macro". > > So I think EMX is on the side of the standard, but as it is (was?) a > draft I am not completely sure. > > As I experienced failing autoconf/configure code for the R software package > at this very same topic, maybe one should complain about this glib or > whatever practice of treating isnan as a function. Interesting. I came across the problem in GSL while looking into why AC_CHECK_FUNCS(isnan) was failing. http://www.gnu.org/manual/glibc-2.2.3/html_node/libc_394.html even recommends using the ISO C99 macros in new code. h~ **= Email 6 ==========================** Date: Sun, 17 Mar 2002 12:56:35 -0500 (EST) From: "Marty" Subject: Re: Porting Joe's Own Editor (JOE) On Sun, 17 Mar 2002 14:37:12 +0000, John Poltorak wrote: > Is F1 supposed to work? It doesn't here. I don't know what to expect with > it. I'm not sure why they changed it to say F1, but the versions I've worked with before said CTRL-K H was help, and that works here. Of course, the screen gets all messed up when it tries to draw the help window because the terminal emulation is messed up. Hit CTRL-R to refresh the screen after it is drawn. **= Email 7 ==========================** Date: Sun, 17 Mar 2002 13:55:03 -0600 (CST) From: "Brian Smith" Subject: Re: Porting Joe's Own Editor (JOE) On Sun, 17 Mar 2002 14:42:01 -0500, Thomas Dickey wrote: >On Sun, Mar 17, 2002 at 02:19:58PM -0500, Marty wrote: >> On Sun, 17 Mar 2002 19:00:07 +0000, John Poltorak wrote: >> >> > Is there any way for JOE to use the whole screen? >> > >> > I normally run in mode co80,70, but JOE will only use 25 lines. >> >> Detecting that should be a function of the termcap library if it's working correctly > >no - the termcap library does not detect the screen size: it only returns >the value in the termcap file. The curses library (if JOE used that) would >try to determine the screensize. > >> (which it isn't). On Solaris, HP-UX, AIX, and Linux, JOE automatically resizes >> itself if you resize the xterm it's running in. > >But that mechanism isn't available from an OS/2 console window. In OS/2 ports what I usually do is something like this: #ifdef __EMX__ vmode.cb = sizeof(VIOMODEINFO); VioGetMode(&vmode, 0); default_pair[0] = ' '; default_pair[1] = 7; memset(current_term, 0, sizeof(struct term_struct)); current_term->TI_lines = vmode.row; current_term->TI_cols = vmode.col; li = current_term->TI_lines; co = current_term->TI_cols; #endif During program startup. This won't allow the size to change during running of the program but it will aquire the correct startup values. Brian Smith **= Email 8 ==========================** Date: Sun, 17 Mar 2002 14:19:58 -0500 (EST) From: "Marty" Subject: Re: Porting Joe's Own Editor (JOE) On Sun, 17 Mar 2002 19:00:07 +0000, John Poltorak wrote: > Is there any way for JOE to use the whole screen? > > I normally run in mode co80,70, but JOE will only use 25 lines. Detecting that should be a function of the termcap library if it's working correctly (which it isn't). On Solaris, HP-UX, AIX, and Linux, JOE automatically resizes itself if you resize the xterm it's running in. I think you can force it to a certain width or height in the joerc file though. Just to clarify, you are getting the drawing glitches that I've been seeing, right? On a side note, I just booted up with eCS FP2. I'd post this to the eCS server, but I can't get to it for some reason. Whatever you do, DON'T use a RAMIFS drive as a temp drive when doing the fixpack installation. After FSERVICE runs, any access to the RAMIFS drive will lock the system. I was able to change TEMP in my Config.Sys, backout, and reapply the fixpack successfully, thank goodness. **= Email 9 ==========================** Date: Sun, 17 Mar 2002 14:27:13 -0600 (CST) From: "Brian Smith" Subject: Re: Porting Joe's Own Editor (JOE) On Sun, 17 Mar 2002 19:58:54 +0000, John Poltorak wrote: >> In OS/2 ports what I usually do is something like this: >> >> #ifdef __EMX__ >> vmode.cb = sizeof(VIOMODEINFO); >> VioGetMode(&vmode, 0); >> default_pair[0] = ' '; >> default_pair[1] = 7; >> >> memset(current_term, 0, sizeof(struct term_struct)); >> >> current_term->TI_lines = vmode.row; >> current_term->TI_cols = vmode.col; >> li = current_term->TI_lines; >> co = current_term->TI_cols; >> #endif >> >> During program startup. > >Could the code above be incorporated into a termcap library? # if defined (TIOCGWINSZ) { struct winsize window; if (ioctl(tty_des, TIOCGWINSZ, &window) < 0) { current_term->TI_lines = li; current_term->TI_cols = co; } else { if ((current_term->TI_lines = window.ws_row) == 0) current_term->TI_lines = li; if ((current_term->TI_cols = (window.ws_col)) == 0) current_term->TI_cols = co; } } # else I suppose it could be worked into the termcap library somehow but I am not familiar with that code so I can't say for sure. However this is how it is normally done in unix if the TIOCGWINSZ is available. I am not sure if this works in EMX... I'm going to guess no. Brian **= Email 10 ==========================** Date: Sun, 17 Mar 2002 14:37:12 +0000 From: John Poltorak Subject: Re: Porting Joe's Own Editor (JOE) On Sat, Mar 16, 2002 at 02:55:16PM -0500, Marty wrote: > On Sat, 16 Mar 2002 19:24:45 +0000, John Poltorak wrote: > > > On Sat, Mar 16, 2002 at 02:01:07PM -0500, Marty wrote: > > > I've updated it here: > > > http://emuos2.netlabs.org/generic/joe2.9.7-patches-os2.zip > > > > Great. I'll take a look at it soon. > > > > Is it OK to include it in a UnixOS/2 package? > > When it's done, sure. Right now, the terminal emulation is too messed up to be > useful though. :-( Is F1 supposed to work? It doesn't here. I don't know what to expect with it. > > I was trying to build an updated libterm a few weeks ago myself, but > > couldn't find a termcap.c newer than the one in \emx\bsd\libterm. I > > expected to find one somewhere in FreeBSD but never came across it, so I'm > > a bit puzzled how libterm is built on FreeBSD. > > > > BTW which TOP do you mean? > > The one I use is: "TOP 0.3 - Ian.Hargreaves at ibm.net" > > I'm not sure if this was a port or a workalike though, now that I look at it. There is an updated version by Paul Ratcliffe as well as another port called POT which is more Unix-like included in:- ftp://hobbes.nmsu.edu/pub/os2/util/process/cpuload-19990420.zip but no source... -- John **= Email 11 ==========================** Date: Sun, 17 Mar 2002 14:42:01 -0500 From: Thomas Dickey Subject: Re: Porting Joe's Own Editor (JOE) On Sun, Mar 17, 2002 at 02:19:58PM -0500, Marty wrote: > On Sun, 17 Mar 2002 19:00:07 +0000, John Poltorak wrote: > > > Is there any way for JOE to use the whole screen? > > > > I normally run in mode co80,70, but JOE will only use 25 lines. > > Detecting that should be a function of the termcap library if it's working correctly no - the termcap library does not detect the screen size: it only returns the value in the termcap file. The curses library (if JOE used that) would try to determine the screensize. > (which it isn't). On Solaris, HP-UX, AIX, and Linux, JOE automatically resizes > itself if you resize the xterm it's running in. But that mechanism isn't available from an OS/2 console window. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 12 ==========================** Date: Sun, 17 Mar 2002 14:59:07 +0000 From: John Poltorak Subject: Re: Re: Emacs docs in English On Sun, Mar 17, 2002 at 11:25:22AM +0900, Masaru Nomiya wrote: > Hello, > > In the Message; > > Subject : Re: Re: Emacs docs in English > Message-ID : <20020316185520.E54364 at eyup.org> > Date & Time: Sat, 16 Mar 2002 18:55:20 +0000 > > [John] == John Poltorak has written: > > John> WellI tried to subscribe, but got a confirmation reply which I don't know > John> how to respond to. It had something like this:- > > John> confirm 84682771 Anne Shirley > > John> ^[$B$r4^$`!X$3$N%a!<%j%s%0%j%9%H$KEPO?$r$7$F$b$h$$$+!)!Y$H$$$&3NG'$r5a$a$ > [...] > John> confirm ^[$B%Q%9%o!<%I^[(B(^[$B?t;z^[(B) ^[$B$"$J$?$NL>A0^[(B > > Isn't there a sentence > > confirm 23469876098...878 John Poltorak No. If there had been I would replied with it. > in the returned mail from os2-emacs-ja-admnin at m17n.org? > Send a mail with the body > > confirm 23469876098...878 John Poltorak I've tried every combination I can think of but still haven't managed to subscribe. > to os2-emacs-ja-ctl at m17n.org, not to os2-emacs-ja-admnin at m17n.org. > > Regards, > > --- > Masaru Nomiya mail-to: nomiya at ttmy.ne.jp > > "No Windows, no gains!" ... "Why, I am wrong?" > > -- Bill -- -- John **= Email 13 ==========================** Date: Sun, 17 Mar 2002 15:26:54 -0500 From: Thomas Dickey Subject: Re: Porting Joe's Own Editor (JOE) On Sun, Mar 17, 2002 at 07:58:54PM +0000, John Poltorak wrote: > On Sun, Mar 17, 2002 at 01:55:03PM -0600, Brian Smith wrote: > > On Sun, 17 Mar 2002 14:42:01 -0500, Thomas Dickey wrote: > > >no - the termcap library does not detect the screen size: it only returns > > >the value in the termcap file. The curses library (if JOE used that) would > > >try to determine the screensize. (curses does this, termcap does not). > > During program startup. > > Could the code above be incorporated into a termcap library? not really. my point: termcap does not supply runtime information of this type. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 14 ==========================** Date: Sun, 17 Mar 2002 15:46:37 +0000 From: John Poltorak Subject: Emacs .el & .elc files Are emacs .elc files easily compilable from .el files? If so how are they built? Is there any reason to include .elc files in an archive? Slackware is inconsistant, there are no .elc files in the elisp package but the emacleim is full of them. Anyone know why? See:- ftp://ftp.mirror.ac.uk/sites/ftp.slackware.com/pub/slackware/slackware/slakware/e1/elisp.tgz ftp://ftp.mirror.ac.uk/sites/ftp.slackware.com/pub/slackware/slackware/slakware/e1/emacleim.tgz -- John **= Email 15 ==========================** Date: Sun, 17 Mar 2002 16:19:43 -0800 From: "Dave and Natalie" Subject: Re: OS/2 v. Linux performance One thing that OS/2 may do much better then Linux is scale up. Once the CPU count gets larger then 8 I'd think OS/2 may very well outperform everything else on Intel. Imagine WSEB on a 64 cpu machine . Sure wish more benchmarks included OS/2 in their comparisions. I also think that OS/2 performs better under heavy load then most others. Linux gets pretty unuseable here if there is too much going on, OS/2 just gets slow. Dave **= Email 16 ==========================** Date: Sun, 17 Mar 2002 18:45:37 +0100 (CET) From: Stefan Neis Subject: Re: Emacs .el & .elc files On Sun, 17 Mar 2002, John Poltorak wrote: > Are emacs .elc files easily compilable from .el files? If so how are they > built? Inside emacs, do M-x byte-recompile-directory I suppose it's possible to to it via "emacs -batch" as well, but don't remember the syntax... > Is there any reason to include .elc files in an archive? Not really - at least not more than including .lib in addition to .a libraries in an EMX distribution. ;-) Regards, Stefan **= Email 17 ==========================** Date: Sun, 17 Mar 2002 19:00:07 +0000 From: John Poltorak Subject: Re: Porting Joe's Own Editor (JOE) On Sun, Mar 17, 2002 at 12:56:35PM -0500, Marty wrote: > On Sun, 17 Mar 2002 14:37:12 +0000, John Poltorak wrote: > > > Is F1 supposed to work? It doesn't here. I don't know what to expect with > > it. > > I'm not sure why they changed it to say F1, but the versions I've worked with before > said CTRL-K H was help, and that works here. Ctl-K H works OK > Of course, the screen gets all messed > up when it tries to draw the help window because the terminal emulation is messed > up. Hit CTRL-R to refresh the screen after it is drawn. Is there any way for JOE to use the whole screen? I normally run in mode co80,70, but JOE will only use 25 lines. -- John **= Email 18 ==========================** Date: Sun, 17 Mar 2002 19:18:11 +0000 From: John Poltorak Subject: Building XFree86 v3.3.6 I'm thinking of rebuilding XFree86 v3.3.6 and would be interested in anyone's experience of attempting this... Reading through OS2.notes it seems as though there are plenty of opportunities to go wrong, but I'll give it a go. I'd just like to know how easy it is to find out where any errors have occurred. Is there a build log anywhere which I can check against? -- John **= Email 19 ==========================** Date: Sun, 17 Mar 2002 19:58:54 +0000 From: John Poltorak Subject: Re: Porting Joe's Own Editor (JOE) On Sun, Mar 17, 2002 at 01:55:03PM -0600, Brian Smith wrote: > On Sun, 17 Mar 2002 14:42:01 -0500, Thomas Dickey wrote: > > >On Sun, Mar 17, 2002 at 02:19:58PM -0500, Marty wrote: > >> On Sun, 17 Mar 2002 19:00:07 +0000, John Poltorak wrote: > >> > >> > Is there any way for JOE to use the whole screen? > >> > > >> > I normally run in mode co80,70, but JOE will only use 25 lines. > >> > >> Detecting that should be a function of the termcap library if it's working > correctly > > > >no - the termcap library does not detect the screen size: it only returns > >the value in the termcap file. The curses library (if JOE used that) would > >try to determine the screensize. > > > >> (which it isn't). On Solaris, HP-UX, AIX, and Linux, JOE automatically resizes > >> itself if you resize the xterm it's running in. > > > >But that mechanism isn't available from an OS/2 console window. > > In OS/2 ports what I usually do is something like this: > > #ifdef __EMX__ > vmode.cb = sizeof(VIOMODEINFO); > VioGetMode(&vmode, 0); > default_pair[0] = ' '; > default_pair[1] = 7; > > memset(current_term, 0, sizeof(struct term_struct)); > > current_term->TI_lines = vmode.row; > current_term->TI_cols = vmode.col; > li = current_term->TI_lines; > co = current_term->TI_cols; > #endif > > During program startup. Could the code above be incorporated into a termcap library? > Brian Smith > -- John **= Email 20 ==========================** Date: Sun, 17 Mar 2002 22:00:20 +0100 (CET) From: "Adrian Gschwend" Subject: Re: OS/2 v. Linux performance On Thu, 14 Mar 2002 10:34:21 +0100 (CET), Peter Jespersen wrote: >Any comparisons should be done with Apache 2.0, otherwise I'm afaid that OS/2 would have too big an disadvantage! >Perhaps that is why no large scale comparisons has been made! IBM's port of Apache does support threading, like Apache 2.0 will. That means just one process is started and it does the rest in threads. Works notably faster on the netlabs.org server. Unfortunately they didn't release the source of it, the apachesrc-directory contains a zip file with src but just the EMX port of Brian... Stupid IBM cu Adrian -- Adrian Gschwend at OS/2 Netlabs ICQ: 22419590 ktk at netlabs.org ------- The OS/2 OpenSource Project: http://www.netlabs.org **= Email 21 ==========================** Date: Sun, 17 Mar 2002 22:03:41 +0100 (CET) From: "Adrian Gschwend" Subject: Re: OS/2 v. Linux performance On Thu, 14 Mar 2002 12:24:45 +0000, John Poltorak wrote: >> first of all : OS/2 tcp/ip stack must be slower then Linux stack, because of >> OS/2 modularity > >Are there any benchtests to prove this? I have heard that the gcc version >of Mozilla is faster than the VAC version. not sure about that, I remember some IP-Stack tests (related to network adapter drivers) in the german c't magazine and OS/2 performed very good in this test. It's already some years ago that they did this test but it was already the 32-bit stack IIRC. cu Adrian -- Adrian Gschwend at OS/2 Netlabs ICQ: 22419590 ktk at netlabs.org ------- The OS/2 OpenSource Project: http://www.netlabs.org **= Email 22 ==========================** Date: Sun, 17 Mar 2002 22:07:05 +0100 (CET) From: "Adrian Gschwend" Subject: Re: OS/2 v. Linux performance On Thu, 14 Mar 2002 14:58:55 +0100, Holger Veit wrote: >The paths to the system should be shorter, but the real improvement, as already >mentioned will be achieveable through native threading rather than emulated fork. does that mean that instead of a "real" fork it creates a new thread instead? Is that suposed to work? (don't know much about multiprocessing/threading yet so sorry if this is a stupid question :-) cu Adrian -- Adrian Gschwend at OS/2 Netlabs ICQ: 22419590 ktk at netlabs.org ------- The OS/2 OpenSource Project: http://www.netlabs.org