similar to: Memory allocation, IBM-AIX and R-1.6.2 - addendum

Dear all, I compiled R-1.6.2 for IBM-AIX (using the native compilers) and I am facing problems to instanciate (rather) large matrices. I have: > m <- matrix(0, 640*640, 102) Error: cannot allocate vector of size 326400 Kb I am not truly familiar with AIX, but this does not seem to be caused by kernel/user limitations: ulimit -Ha gives: core file

Dear all, The 'make check' step fails for the pacakge mva on IBM AIX. The tail of the Rout log file looks like: > for(factors in 2:4) print(update(Harman23.FA, factors = factors)) Call: factanal(factors = factors, covmat = Harman23.cor) Uniquenesses: height arm.span forearm lower.leg weight 0.170 0.107 0.166

The short version: When I port our application from LLVM 7 to 8, some JIT-compiled code seems to break the implicit instanciation of C++ templates. That is, when I try to resolve a symbol defined by the code, I get undefined symbols corresponding to methods that are actually function templates. If the code adds explicit template instanciations, the problem goes away. Does anybody have an

Hi, I am trying to compile R for 64 bits on a mainframe running IBM AIX. Setting the environment variable OBJECT_MODE=64 leads to hanging while checking whether leap seconds are treated according to POSIX... Interestingly, it works fine when OBJECT_MODE=32... Any pointer would help me much... Thanks, L. -- -------------------------------------------------------------- currently at the

Well, I agree the front end must be responsible for (not) instantiating the function templates. However the problem shows up only when JIT compiling, for some reason, and only with LLVM 8. Is there perhaps something else that needs to be done before resolving symbols, in addition to running constructors, to ensure that instantiated function templates are taken into account? Or perhaps there is

Hi, does anybody know wether samba is running on an IBM mainframe under os/390? Are there any specific things to consider or do it just like for a "normal" unix? thanks Stephan Hendl -- LDS Brandenburg Dr. Stephan Hendl fon: +49-(0)331-39 471 EMail: stehan.hendl@lds.brandenburg.de

Mark Rousell wrote: > On 30/10/2018 17:14, Simon Matter wrote: <snip> > Yup. When I looked at IBM Power machines before (maybe about a year ago, > not sure), there was actually a pricing tool on the website. You could go > through various options for machines (GPUs, CPUs, storage, memory, etc.) > and get a price. Annoyingly I didn't record detailed pricing info but, as

Hi, I''m running into something along the lines of bug 5383. I''m hoping I''m doing something obviously wrong and that someone can point me in the right direction. I''ve got a custom paint event handler: def on_paint puts "mainframe on_paint" paint do |dc| dc.clear @cl.draw dc end end Then I''ve got a key press

I have found the cause of our problems with implicit template instantiation, and it looks to me like a bug in Orc JIT in LLVM 8. (Perhaps Lang could confirm?) We use createLegacyLookupResolver() to create our symbol resolver, passing it a lookup function. Amongst other things, our function calls LegacyRTDyldObjectLinkingLayer::findSymbol() to look up symbols in the JIT. When a module is

Hello, I'm attempting to port OpenSSH to IBM's S/390 mainframe. Things have gone well but I have come a little unstuck with the internal PRNG. Although the commands in ssh_prng_cmds are being executed the select() seems to be returning 0 and therefore the cose is assuming that the forked process has timed out. This could be a difference in the way that select is implemented on OS/390.

On 30/10/2018 16:40, mark wrote: > Linux was IBM's silver > bullet on a free platter. I mean, *how* many operatings systems do you > want to support...? Yup, it must cost them a pretty penny to maintain all those proprietary operating systems (especially when you include their mainframe ones). I suspect that Linux will eventually replace i and AIX -- eventually. But I bet there are

No, the problem is that Finalized is true. It is set to true at the start of the finalize() method. That's why I added a test causing that line to be executed if the symbol's address is zero... On Tue, 17 Sep 2019, 00:05 Lang Hames, <lhames at gmail.com> wrote: > Hi Geoff, > > Oof. I don't know what this is yet, but I bet it's going to be awful. > Quick apology

we're testing a new NFS product for our mainframe system (Conectivitiy Systems TCP/IP for VSE) and have successfully mapped a virtual drive to our unix system (AIX 4.2). We use Samba (v.1.9.18p2) to share files with the Windows 95 and Windows NT systems in the building. We have attempted to map the "mapped" drive from the mainframe via Samba. We're able to see files in the

hi, i did some tests about performance in jspeex ... uls@mainframe:~/eclipse/avrelay$ java TestClient Diff1: 0 10 Diff2: 141 started. uls@mainframe:~/eclipse/avrelay$ <p>{ SpeexEncoder e1=new SpeexEncoder(); e1.init(0,1,44100,1); byte[] by1=new byte[320]; long l1=System.currentTimeMillis(); long l2=System.currentTimeMillis(); System.out.println("Diff1:

Help please. I am in the process of configuring Samba 3.0.14a onto 4 AIX 5.3 systems. It seems to be working OK on 3 of the 4, but on the last one, I cannot start smbd. nmbd seems to start fine. The smb.conf file matches the other servers except that the server name is changed. Here's the error showing up on the log.smbd file. [2007/05/02 15:19:47, 0] smbd/server.c:main(835) standard

Hi We are running Samba on an IBM zSeries mainframe on Linux kernel 2.2.16 and are running into problems when using winbindd and security=domain. We are doing a proof of concept using opensource and I really hope we can pull it off! Before I raise a bug report, I would appreciate it if someone can tell me if this is a known problem. I have not found anything on it yet. We have winbindd set up

Would anybody have any experience with that ? (that's the first time I use AIX) The configure step is incredibly slow... any hint ? L.

When I empty and refill an FXTable multiple times with a large data set (10,000 rows x 1 column in the example below, around 1000 rows x 6 columns in a more complicated app), the application eventually fails with a segmentation fault. The number of repetitions required to produce the segfault is rather random, ranging from just a couple reps to 15 or 20 in some trials. The program disappears

hi! All: I wonder if someone has done this before... I am writing several functions that conduct statistical analysis using a GUI interface by Tcl/Tk, they share many identical GUI components. What I am trying to do now is to simplify the code by writing a GUI repository for all the components they use, thus save effort for code maintenance. Since they all use: mainFrame <-

Hi (and Merry Christmas), I am trying to compile R on an AIX IBM machine using the native C and FORTRAN compilers... and everything goes smoothly (congratulations for the configure and make files, this is very nice)... until the pacakge 'cluster' (see dump below).... Any hint ? Laurent 1501-510 Compilation successful for file pam.f. cc -I/data1/laurent/R-1.6.1/include