public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work
@ 2004-03-16 9:54 stl at caltech dot edu
2004-03-20 9:40 ` [Bug driver/14601] " stl at caltech dot edu
` (8 more replies)
0 siblings, 9 replies; 10+ messages in thread
From: stl at caltech dot edu @ 2004-03-16 9:54 UTC (permalink / raw)
To: gcc-bugs
I currently use MinGW gcc 3.3 crossed over from i686-pc-linux-gnu. It lives in
C:\MinGW. If I compile a Hello World program with g++ -v, it says (in part):
ignoring nonexistent directory "C:/MinGW/mingw32/include"
ignoring nonexistent directory "/home/stl/temp/dest/include/c++/3.3"
ignoring nonexistent directory "/home/stl/temp/dest/include/c++/3.3/mingw32"
ignoring nonexistent directory "/home/stl/temp/dest/include/c++/3.3/backward"
ignoring nonexistent directory "/home/stl/temp/dest/include"
ignoring nonexistent directory "/home/stl/temp/dest/include"
ignoring nonexistent directory "/home/stl/temp/dest/lib/gcc-
lib/mingw32/3.3/include"
ignoring nonexistent directory "/home/stl/temp/dest/mingw32/include"
ignoring nonexistent directory "/usr/local/mingw32/include"
#include "..." search starts here:
#include <...> search starts here:
C:/MinGW/include/c++/3.3
C:/MinGW/include/c++/3.3/mingw32
C:/MinGW/include/c++/3.3/backward
C:/MinGW/include
C:/MinGW/lib/gcc-lib/mingw32/3.3/include
End of search list.
It produces a hello.exe that works.
I bootstrapped 3.3.3 in MinGW/MSYS. I put it in C:\Temp\gcc\gcc-3.3.3-meow. It
says:
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3.3"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3.3/i686-pc-
mingw32"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3.3/backward"
ignoring nonexistent directory "c:/temp/gcc/dest/include"
ignoring nonexistent directory "/c/temp/gcc/dest/include"
ignoring nonexistent directory "c:/temp/gcc/dest/lib/gcc-lib/i686-pc-
mingw32/3.3.3/include"
ignoring nonexistent directory "c:/temp/gcc/dest/i686-pc-mingw32/include"
#include "..." search starts here:
#include <...> search starts here:
C:/Temp/gcc/gcc-3.3.3-meow/lib/gcc-lib/i686-pc-mingw32/3.3.3/include
/mingw/include
End of search list.
It then fails to find <iostream>.
All of the MinGW/MSYS bootstrapped gcc's I've been accumulating - 3.3, 3.3.1,
3.3.3, and 3.4-20040310 with the PR12974 and PR14291 patches - have this same
problem.
3.3 living in C:\Temp\gcc\gcc-3.3-hiss says:
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3/i686-pc-
mingw32"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3/backward"
ignoring nonexistent directory "c:/temp/gcc/dest/include"
ignoring nonexistent directory "/c/temp/gcc/dest/include"
ignoring nonexistent directory "c:/temp/gcc/dest/lib/gcc-lib/i686-pc-
mingw32/3.3/include"
ignoring nonexistent directory "c:/temp/gcc/dest/i686-pc-mingw32/include"
ignoring nonexistent directory "/usr/local/mingw32/include"
#include "..." search starts here:
#include <...> search starts here:
C:/Temp/gcc/gcc-3.3-hiss/lib/gcc-lib/i686-pc-mingw32/3.3/include
End of search list.
3.3.1 living in C:\Temp\gcc\gcc-3.3.1-purr says:
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3.1"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3.1/i686-pc-
mingw32"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.3.1/backward"
ignoring nonexistent directory "c:/temp/gcc/dest/include"
ignoring nonexistent directory "/c/temp/gcc/dest/include"
ignoring nonexistent directory "c:/temp/gcc/dest/lib/gcc-lib/i686-pc-
mingw32/3.3.1/include"
ignoring nonexistent directory "c:/temp/gcc/dest/i686-pc-mingw32/include"
#include "..." search starts here:
#include <...> search starts here:
C:/Temp/gcc/gcc-3.3.1-purr/lib/gcc-lib/i686-pc-mingw32/3.3.1/include
/mingw/include
End of search list.
3.4.0 living in C:\Temp\gcc\gcc-3.4.0-oink says:
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.4.0"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.4.0/i686-pc-
mingw32"
ignoring nonexistent directory "c:/temp/gcc/dest/include/c++/3.4.0/backward"
ignoring nonexistent directory "c:/temp/gcc/dest/include"
ignoring nonexistent directory "/c/temp/gcc/dest/include"
ignoring nonexistent directory "c:/temp/gcc/dest/lib/gcc/i686-pc-
mingw32/3.4.0/include"
ignoring nonexistent directory "c:/temp/gcc/dest/i686-pc-mingw32/include"
#include "..." search starts here:
#include <...> search starts here:
C:\Temp\gcc\gcc-3.4-oink\bin\../lib/gcc/i686-pc-mingw32/3.4.0/include
/mingw/include
End of search list.
Interestingly, compiling hello.c using stdio.h with gcc 3.3.3 worked. I don't
know why.
Is this a problem with gcc? Am I not configuring my bootstraps properly? Since
I can cross a compiler over and then move it to any old directory (i.e.
C:\MinGW) I like, I figure that I should be able to bootstrap with a
destination of C:\Temp\gcc\dest and then move it to any old directory.
This is how I configured 3.3 during the cross:
Reading specs from C:/MinGW/bin/../lib/gcc-lib/mingw32/3.3/specs
Configured with: /home/stl/temp/src/configure --prefix=/home/stl/temp/dest --
build=i686-pc-linux-gnu --host=mingw32 --ta
rget=mingw32 --enable-languages=c,c++ --with-gcc --with-gnu-as --with-gnu-ld --
with-as=/home/stl/xmingw32/bin/mingw32-as
--with-ld=/home/stl/xmingw32/bin/mingw32-ld --enable-threads=win32 --disable-
nls --disable-win32-registry --disable-sha
red --enable-sjlj-exceptions --disable-debug --without-newlib
Thread model: win32
gcc version 3.3 (mingw special 20030514-1)
This is how I configured the 3.3.3 bootstrap:
Using built-in specs.
Configured with: ../src/configure --prefix=/c/temp/gcc/dest --enable-
languages=c,c++ --enable-sjlj-exceptions --enable-t
hreads=win32 --disable-win32-registry --disable-shared --without-newlib --
disable-nls --disable-debug
Thread model: win32
gcc version 3.3.3 (mingw special)
--
Summary: MinGW/MSYS bootstrapped compiler doesn't work
Product: gcc
Version: 3.3.3
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: driver
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: stl at caltech dot edu
CC: gcc-bugs at gcc dot gnu dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] MinGW/MSYS bootstrapped compiler doesn't work
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
@ 2004-03-20 9:40 ` stl at caltech dot edu
2004-03-21 3:16 ` [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't stl at caltech dot edu
` (7 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: stl at caltech dot edu @ 2004-03-20 9:40 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From stl at caltech dot edu 2004-03-20 09:40 -------
After much frustrating work, I have crossed 3.3.3 and I can verify that it is
directory-independent: it doesn't care what directory it is put in, it can
always find its include files. The bootstrapped 3.3.3 is tied to the directory
it was configured to be installed it.
This is undesirable behavior: why should a native-native compiler be LESS
useful than a crossed-native?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
2004-03-20 9:40 ` [Bug driver/14601] " stl at caltech dot edu
@ 2004-03-21 3:16 ` stl at caltech dot edu
2004-03-21 5:04 ` dannysmith at gcc dot gnu dot org
` (6 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: stl at caltech dot edu @ 2004-03-21 3:16 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From stl at caltech dot edu 2004-03-21 03:16 -------
Changing summary to reflect my new understanding of the bug.
I guess that this is not MinGW-specific, but people on GNU/Linux systems have
no interest in moving their compiler around (since it's usually dumped in with
a bunch of other programs). On Windows, the compiler is usually put into its
own directory (e.g. C:\MinGW), except that not everyone agrees where that
directory should be.
--
What |Removed |Added
----------------------------------------------------------------------------
Summary|MinGW/MSYS bootstrapped |Bootstrapped compiler tied
|compiler doesn't work |to original directory while
| |crossed native isn't
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
2004-03-20 9:40 ` [Bug driver/14601] " stl at caltech dot edu
2004-03-21 3:16 ` [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't stl at caltech dot edu
@ 2004-03-21 5:04 ` dannysmith at gcc dot gnu dot org
2004-03-21 5:06 ` stl at caltech dot edu
` (5 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: dannysmith at gcc dot gnu dot org @ 2004-03-21 5:04 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From dannysmith at gcc dot gnu dot org 2004-03-21 05:04 -------
I routinely bootstrap gcc with native (msvcrt.dll-dependant) gcc.exe using
cygwin's bash as shell and cygwin1.dll dependent tools. and have not seen this
bug.
The binaries built this way and distributed at mingw.org can be relocated into
different directories, different drives and still find correct system includes
and libs as long as PATH environment variable contains the gcc driver directory
and the builtin relative GCC_EXEC_PREFIX hasn't been overriden by an env setting
Danny
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
` (2 preceding siblings ...)
2004-03-21 5:04 ` dannysmith at gcc dot gnu dot org
@ 2004-03-21 5:06 ` stl at caltech dot edu
2004-04-04 7:57 ` wintermute2k4 at ntlworld dot com
` (4 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: stl at caltech dot edu @ 2004-03-21 5:06 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From stl at caltech dot edu 2004-03-21 05:06 -------
> I routinely bootstrap gcc with native (msvcrt.dll-dependant) gcc.exe using
> cygwin's bash as shell and cygwin1.dll dependent tools. and have not seen
> this bug.
Okay - then this must be related to bootstrapping gcc in MSYS.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
` (3 preceding siblings ...)
2004-03-21 5:06 ` stl at caltech dot edu
@ 2004-04-04 7:57 ` wintermute2k4 at ntlworld dot com
2004-04-04 20:46 ` stl at caltech dot edu
` (3 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: wintermute2k4 at ntlworld dot com @ 2004-04-04 7:57 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From wintermute2k4 at ntlworld dot com 2004-04-04 07:57 -------
(In reply to comment #4)
> > I routinely bootstrap gcc with native (msvcrt.dll-dependant) gcc.exe using
> > cygwin's bash as shell and cygwin1.dll dependent tools. and have not seen
> > this bug.
> Okay - then this must be related to bootstrapping gcc in MSYS.
I believe this one has been fixed with MSYS 1.10.0
on earlier versions if you configure with a full drive/pathname it works correctly i.e. --prefix=c:/crosstools
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
` (4 preceding siblings ...)
2004-04-04 7:57 ` wintermute2k4 at ntlworld dot com
@ 2004-04-04 20:46 ` stl at caltech dot edu
2004-04-05 0:16 ` ian at wasabisystems dot com
` (2 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: stl at caltech dot edu @ 2004-04-04 20:46 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From stl at caltech dot edu 2004-04-04 20:46 -------
MSYS 1.0.10 (not a Release Candidate) and --prefix=/c/temp/gcc/dest does not
fix the problem.
I will try --prefix=c:/temp/gcc/dest next.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
` (5 preceding siblings ...)
2004-04-04 20:46 ` stl at caltech dot edu
@ 2004-04-05 0:16 ` ian at wasabisystems dot com
2004-04-05 0:19 ` stl at caltech dot edu
2004-04-07 3:08 ` [Bug target/14601] " pinskia at gcc dot gnu dot org
8 siblings, 0 replies; 10+ messages in thread
From: ian at wasabisystems dot com @ 2004-04-05 0:16 UTC (permalink / raw)
To: gcc-bugs
--
What |Removed |Added
----------------------------------------------------------------------------
CC| |ian at wasabisystems dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
` (6 preceding siblings ...)
2004-04-05 0:16 ` ian at wasabisystems dot com
@ 2004-04-05 0:19 ` stl at caltech dot edu
2004-04-07 3:08 ` [Bug target/14601] " pinskia at gcc dot gnu dot org
8 siblings, 0 replies; 10+ messages in thread
From: stl at caltech dot edu @ 2004-04-05 0:19 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From stl at caltech dot edu 2004-04-05 00:19 -------
I can confirm that --prefix=c:/temp/gcc/dest produces a directory-independent
bootstrapped compiler!
I don't know whether this bug should be left open or not; it might be a gcc
bug, or it might be an MSYS bug. It's fine with me if the bug is closed
because a simple workaround exists. Probably this should be documented
somewhere though.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug target/14601] Bootstrapped compiler tied to original directory while crossed native isn't
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
` (7 preceding siblings ...)
2004-04-05 0:19 ` stl at caltech dot edu
@ 2004-04-07 3:08 ` pinskia at gcc dot gnu dot org
8 siblings, 0 replies; 10+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-04-07 3:08 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2004-04-07 03:08 -------
Closing as this was a bug in the runtime and not in GCC.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Component|driver |target
Resolution| |INVALID
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14601
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2004-04-07 3:08 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-03-16 9:54 [Bug driver/14601] New: MinGW/MSYS bootstrapped compiler doesn't work stl at caltech dot edu
2004-03-20 9:40 ` [Bug driver/14601] " stl at caltech dot edu
2004-03-21 3:16 ` [Bug driver/14601] Bootstrapped compiler tied to original directory while crossed native isn't stl at caltech dot edu
2004-03-21 5:04 ` dannysmith at gcc dot gnu dot org
2004-03-21 5:06 ` stl at caltech dot edu
2004-04-04 7:57 ` wintermute2k4 at ntlworld dot com
2004-04-04 20:46 ` stl at caltech dot edu
2004-04-05 0:16 ` ian at wasabisystems dot com
2004-04-05 0:19 ` stl at caltech dot edu
2004-04-07 3:08 ` [Bug target/14601] " pinskia at gcc dot gnu dot org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).