bug-gmp
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Configure bug under Win9x-Cygwin systems.


From: Jim . Fougeron
Subject: Configure bug under Win9x-Cygwin systems.
Date: Tue, 23 Apr 2002 13:29:52 -0500

Hello,

First thanks for a terrific tool.

Second, I have had dismal success (until today) building under a Cygwin
under Win9x environment.  Building under a Win2k PC works fine, but the 9x
configure would take a LONG time (2 hours or more). During this configure
(and all Cygwin usage from this point on) the PC would semi "lockup" (mouse
unresponsive for 10-15 seconds, then would work for a second or two, and
then unresponsive, also other apps were unresponsive). The full Cygwin
environment worked this way until the PC was rebooted, even after exiting
and restarting the bash shell.

I think I have found the problem.  It is in the check for command line
parameter length.  I have changed .\configure from:

  case $host_os in
  msdosdjgpp*)
    # On DJGPP, this test can blow up pretty badly due to problems in libc
    # (any single argument exceeding 2000 bytes causes a buffer overrun
    # during glob expansion).  Even if it were fixed, the result of this
    # check would be larger than it should be.
    lt_cv_sys_max_cmd_len=12288;    # 12K is about right
    ;;

TO

  case $host_os in
  msdosdjgpp* | cygwin*)
    # On DJGPP, this test can blow up pretty badly due to problems in libc
    # (any single argument exceeding 2000 bytes causes a buffer overrun
    # during glob expansion).  Even if it were fixed, the result of this
    # check would be larger than it should be.
    lt_cv_sys_max_cmd_len=12288;    # 12K is about right
    ;;

Now, I KNOW that this change is not correct, but it worked around the
problem on my system, and now I have built 6 times (different CPU types),
and things are working like they should.  I do not know shell scripting (or
configure work) very well, but I can hack out a few things.  This bug does
not seem to affect my Win2k boxes.

Originally, it was saying I had like 365k for command line (under Win9x),
and 12288 for command line under Win2k.  I do not know how to code into the
configure a check for Win9x (which is the only thing I am suggesting should
skip this check).  My hack is certainly not something for the real GMP real
release, but I am sure there must be some way of doing this skip of the
check IF a Win9x Cygwin (or probably Mingw) build is taking place.

Versions:

GMP  4.0.1 pre-packaged
gcc  version 2.95.3-5 (cygwin special)

NOTE I tried different config "switches" and was not able to work around
the listed "bug".  I have tried --enable-static --disable-shared -C (or
--cache_file=jim_cache) --disable-dependency-tracking (also tried
--enable-dependency-tracking) --disable-cxx --disable-libtool-lock
--enable-fast-install    None of these made any difference. As soon as
configure hit the command line length check, the PC semi-died.

Now I simply use --build-cpu-cygwin --enable-alloca -C --disable-cxx
--disable-shared --enable-static.    I will also try building DLL's tonight
with MinGW32 under Cygwin and see how that works.

Thank you, Jim.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]