This is the mail archive of the
cygwin-developers@sourceware.cygnus.com
mailing list for the Cygwin project.
Re[2]: gcc-2.95/ July 5 snapshot
- To: Chris Faylor <cgf@cygnus.com>
- Subject: Re[2]: gcc-2.95/ July 5 snapshot
- From: Egor Duda <deo@logos-m.ru>
- Date: Fri, 16 Jul 1999 12:00:03 +0400
- CC: cygwin-developers <cygwin-developers@sourceware.cygnus.com>
- Organization: DEO
- References: <19990715132926.A1102@cygnus.com>
- Reply-To: Egor Duda <deo@logos-m.ru>
Hi!
15 Èþëü 1999 Chris Faylor cgf@cygnus.com wrote:
Sorry, if i'm giving wrong suggestions. But Suhaib's report looked
pretty much like gcc is using global headers such as dirent.h and
assert.h, instead of local ones from snapshot's newlib and winsup. If
$(updir) is incorrect (in case of building in source directory or any of its
subdirs) -- the result would be exactly as Suhaib's.
CF> What in the world does this have to do with the July 5 snapshot? How
CF> does changing a variable in the Makefile fix a problem with compiling
CF> winsup?
CF> Just to reiterate: Everyone should be building their stuff in a separate
CF> directory, i.e., don't build winsup in the same directory as the winsup
CF> sources.
I think it should read "don't build winsup in the same directory as the
winsup sources and in any of its subdirectories" ? if srcdir == ./..
or ./../.. then updir would be incorrect too, as far as i understand.
Correct me, if i'm wrong.
Egor. mailto:deo@logos-m.ru ICQ 5165414 FidoNet 2:5020/496.19