This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Fix mingw32 bootstrap regression.
- From: Christopher Faylor <me at cgf dot cx>
- To: cgf at gcc dot gnu dot org, 'Mark Mitchell' <mark at codesourcery dot com>, Danny Smith <dannysmith at clear dot net dot nz>, 'Bob Rossi' <bob_rossi at cox dot net>, gcc-patches at gcc dot gnu dot org, 'Carlos O'Donell' <carlos at codesourcery dot com>
- Date: Thu, 21 Dec 2006 07:03:04 -0500
- Subject: Re: [PATCH] Fix mingw32 bootstrap regression.
- References: <20061220163920.GF14845@lios> <000001c724d7$6fee5f50$3b6d65da@anykey>
On Thu, Dec 21, 2006 at 09:10:05PM +1300, Danny Smith wrote:
>Carlos O'Donell wrote:
>> The following patch fixes the boostrap regression for mingw32.
>> We only do the substitution if the path prefix matches the configured
>> prefix, otherwise we add the path as before (instead of asserting).
>> To bootstrap a mingw32 toolchain you use either msys or
>> cygwin. Both of
>> these environments build gcc with cygwin style paths.
>I did not notice the regression.
>Using cygwin bash, I do this:
>mount c:/mingw /mingw
>Then, using cygwin bash, "configure --prefix=/mingw ... " in a dir on
>C: drive, just works, because the native path is identical to the cygwin
>> Therefore --prefix
>> is "/c/mingw" while the compiler, a native mingw32 executable, thinks
>> the prefix is "c:/mingw". A simple string comparison does not work.
>> Danny, Christopher, Do you see anything else that might be a
>> problem on
>> 2006-12-19 Carlos O'Donell <email@example.com>
>> PR bootstrap/30242
>> * gcc/c-incpath.c (add_standard_paths): Only relocate
>> paths that
>> begin with the configured prefix.
>mingw bootstraps for me before the patch and after. The patch is OK
I'll defer to Danny on this one. If he says it is ok, then go ahead
and check it in.