Marcio, Please stop thinking CORBA is heavy machinery, only useful for complex situations. CORBA is useful for the simplest things too. And the whole point of...
Dominic WILLIAMS
d.williams@...
Jan 2, 2001 7:49 am
2838
... I've not looked at the separate implementation. It's for SE, but does it support both POSIX and Win32? Perhaps that implementation can use this library?...
Berend de Boer
berend@...
Jan 2, 2001 6:50 pm
2839
I think Miguel's SE implementation requires POSIX Threads, which has a lame story under Windows. Zthreads built a layer that works on both Win32 native and ...
Marcio Marchini
mqm@...
Jan 2, 2001 7:53 pm
2840
... But surely the pthreads library could be implemented using Win32 native threads? Why invent a new threads API? -- Patrick Doyle doylep@......
Patrick Doyle
doylep@...
Jan 2, 2001 8:17 pm
2841
... It could. It should. But it does not seem to be there yet. If you can get it to work well, let me know. Make sure it is small too. I'd like to be able to...
Marcio Marchini
mqm@...
Jan 2, 2001 8:26 pm
2842
... Yes, looks like a good idea. Any takers for a port? :-) Groetjes, Berend. (-:...
Berend de Boer
berend@...
Jan 2, 2001 8:28 pm
2843
... Keep in mind that SmallEiffel supports more than just Unix clones and Windows variants. Threads will also have to work on AmigaOS, ancient MS-DOS, Next,...
Greg C
gmc333@...
Jan 3, 2001 12:16 am
2844
Hello, I'd make few tests with SE 0.76 arrays. Compilers: MSVC 6.0, Cygnus 20.1 Athlon 700, Windows NT. class TEST creation make feature a: ARRAY[INTEGER] --a:...
Alexei Nikitin
anikitin@...
Jan 3, 2001 12:04 pm
2845
We're in the final stages of making sure yaesockets runs with ISE Eiffel as well. So, it now uses GOBO for portability. The Emu app still only compiles under...
Marcio Marchini
mqm@...
Jan 3, 2001 3:30 pm
2846
I think a good approach would be to implement the simplest thing first. I think that something along Thread State Library (see previous msgs) would require...
Marcio Marchini
mqm@...
Jan 3, 2001 4:15 pm
2847
... Out of curiosity, I repeated Alexei's test on my Linux machine. With gcc -O3, all three methods are equally fast, so gcc is apparently able to optimize...
Tom Wainwright
thomas.wainwright@...
Jan 3, 2001 7:36 pm
2848
Would this be a bug in SE ? marcio compile -clean -no_style_warning hashes make crc32.obj C:\eiffel\eposix\libeposix.lib Error hashes1.c: 8956 illegal...
Marcio Marchini
mqm@...
Jan 4, 2001 1:02 am
2849
... ... = ? Often, -no_split and -fomit-frame-pointer give good results. -funroll-loops gives here a really good improvement. 3.50 for ARRAY, 2.84 for...
Philippe RIBET
philippe.ribet@...
Jan 4, 2001 2:30 pm
2850
Here's a simple test case. The problem seems to be with expanded types. If INT_PAIR is no longer expanded, and MY_CLIENT is changed to create an instance that...
Marcio Marchini
mqm@...
Jan 4, 2001 4:22 pm
2851
... Thanks, I wasn't aware of -funroll-loops. ... Not on my system, but you inspired me to try some other combinations of options. Interestingly, including...
Tom Wainwright
thomas.wainwright@...
Jan 4, 2001 6:12 pm
2852
The following test program produces an error in the generated C code. I originally found the problem in the -0.75 Beta, but the error also occurs with release...
Greg C
gmc333@...
Jan 4, 2001 7:22 pm
2853
One of the main problems getting the yaesockets examples working under ISE is because of something very simple. In class STRING, resize(new_count: INTEGER) is ...
Marcio Marchini
mqm@...
Jan 5, 2001 8:41 am
2854
... For me, this problem is not due to cache miss but missalign. Default gcc behavior is to use 4 bytes alignment. Pentium is much more efficient with 8 bytes...
Philippe RIBET
philippe.ribet@...
Jan 5, 2001 9:24 am
2855
Hi, I've had (and still have) similar problems with expanded classes, too. I reported it to the SE team some time ago (when -0.78 or -0.77 had been released)...
Pascal Poncin
Pascal.Poncin@...
Jan 5, 2001 9:18 pm
2856
I would like to submit the following insert_string feature for the string class: insert_string(str: STRING; index: INTEGER) is -- Insert `str' after position...
J. Scott Edwards
sedwards@...
Jan 6, 2001 3:23 pm
2857
... I originally found the problem in the -0.75 Beta, but the error also occurs with release -0.76. I'm using gcc as the backend compiler, running on Windows...
Kolari S. Bhat
kolari.s.bhat@...
Jan 6, 2001 4:22 pm
2858
... I couldn't identify the exact error source any further. The following is how far I have reached at the error source: Original max function's ensure...
Kolari S. Bhat
kolari.s.bhat@...
Jan 7, 2001 2:52 am
2859
I started making a version of SE on sourceforge and I wanted it to be able to co-exist on the same system with SE without disturbing SE. I also wanted to try...
J. Scott Edwards
sedwards@...
Jan 8, 2001 3:04 pm
2860
We are a group of students developing a small scale programming project in an European university and we have encountered with some trouble when working with ...
Marc Andreu
marcandreu@...
Jan 10, 2001 5:35 pm
2861
lcc (SE's default back-end if you used the SE install shield) comes with a Boehm GC DLL. As far as I know WEdit itself uses Boehm. But I was never able to get...
Marcio Marchini
mqm@...
Jan 10, 2001 5:41 pm
2862
... Yes, we do use the Boehm GC. So I suggest to download the elj-win32 distribution and than install it. If the GC stuff is not enabled in the delivered...
Friedrich Dominicus
frido@...
Jan 11, 2001 7:33 am
2863
I have written some Eiffel programs for a Benchmark. The benchmarks can be found here: http://www.bagley.org/~doug/shootout/ Now I tries the string-concat...
Friedrich Dominicus
frido@...
Jan 11, 2001 8:19 am
2864
... There is a bug. Actually, memory chunks are not freed. As your program needs chunks of memory which have increasing size, there are a lot of free (unused)...
Dominique Colnet
Dominique.Colnet@...
Jan 11, 2001 8:53 am
2865
Dominique Colnet <Dominique.Colnet@...> writes: Thanks Dominique. ... Sorry, but jokes used to be funny. I have some problems with laughing it is lost...
Friedrich Dominicus
frido@...
Jan 11, 2001 11:07 am
2866
Well, here was my original question: http://www.egroups.com/message/smalleiffel/2281 Then I got this answer: http://www.egroups.com/message/smalleiffel/2283 ...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
