[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Gm2] latest cvs changes
From: |
Gaius Mulley |
Subject: |
[Gm2] latest cvs changes |
Date: |
Sun, 23 Aug 2009 09:56:14 +0100 |
User-agent: |
Gnus/5.11 (Gnus v5.11) Emacs/22.2 (gnu/linux) |
Hi,
here are the latest changes. Summary fixes:
- argument ordering bug fixed.
- explicitly use bash when necessary.
- macport section added and later improved.
- bug fixes to variant record checking.
- SMathLib0.{def,mod} new files.
- documentation updated and web pages updated.
- ISO M2 language now implemented.
If you use a mac then you might be interested in the macports section.
At the end of this section there is a script which will automatically
pull down the latest gm2 from a cvs tarball, build the compiler, run
gm2.paranoid and all the regression tests.
regards,
Gaius
* Make-lang.in: added SMathLib0.def to list of library
modules.
* gm2.texi: added build log for macports description.
* gm2-compiler/M2CaseList.mod: only test case ranges if
the high field is used, otherwise test equality. Check
that case types are also known and their maximum and
minimum values.
* gm2-compiler/SymbolTable.mod: only place record fields
into the Record.LocalSymbol tree if the name is not NulName.
* gm2-libs/SMathLib0.mod: many fixes to SMathLib0 using the
correct libm function name expf, sqrtf etc.
* gm2-libs-iso/Processes.def: tidied up comment formatting.
* gm2/gm2.texi: improved the building under macports section.
* gm2/www/index.ms: updated to reflect recent changes.
Removed older build platforms.
* gm2/gm2.texi: added macport section. Also updated
state of gm2 within the documentation. The languages
are complete, libraries are incomplete.
* gm2/Make-lang.in: added dependency on
$(objdir)/gm2/gm2-libs for gm2/gm2-libs/gm2-libs-host.h.
Added dependencies on cc1gm2$(exeext) when creating
SYSTEM.def. Explicitly invoke bash to run bash scripts
rather than assume that the developers SHELL is bash.
* gm2/gm2spec.c: fixed an off by one bug when scanning
arguments. This manifested itself as a linking bug
(as reported by John O Goyo <address@hidden>)
when arguments were presented in a different order.
'-o' skipped one to many arguments.
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [Gm2] latest cvs changes,
Gaius Mulley <=