Re: Maxima works with gcl 2.7.1
David Scherfgen via Maxima-discuss <[email protected]> Fri, 25 Apr 2025 01:01:14 +0200
| Newsgroups | gmane.comp.mathematics.maxima.general,gmane.lisp.gcl.devel |
|---|---|
| Message-ID | <CAMTHLKjkx7KwqiHm7mB71LH1+tcJYS1WbGSB_XvRWwpSfyxxwQ@mail.gmail.com> |
I only use this for Maxima development. Right now, my main priority is to get GCL >=2.7.1 working - somehow. If I could choose, I'd prefer a GCL that simply doesn't use the hugepages, a performance penalty would be OK. Am Fr., 25. Apr. 2025 um 00:55 Uhr schrieb Camm Maguire < [email protected]>: > Greetings! > > > Greetings! Great! This will be an interesting test case never before > > encountered. > > > The issue is that allocating the minimum 1 page to each of the cell > > types brings us to a minimum heap of 10Gb. I need to meditate on how to > > reserve memory beneath 2Gb in such a setting. > > OK I've figured out how to reserve low memory for code with such a large > pagesize if this is what is desired. But before I do it I just want to > make sure it is what you want. I do think it reasonable that the > minimum heap be around 10 pages. This would mean that all your gcl > executables will be at least this big (e.g. 10G). Perhaps that is not a > concern, or perhaps it is. Please let me know how you would like to see > this work. > > Take care, > -- > Camm Maguire [email protected] > ========================================================================== > "The earth is but one country, and mankind its citizens." -- Baha'u'llah > _______________________________________________ Maxima-discuss mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/maxima-discuss