Re: Issues with building gcc trunk with ada on cygwin

2011-11-20 Thread Christian Joensson
On 20 November 2011 12:06, Eric Botcazou wrote: >> Now, are there anyones here that recognise the issue and perhaps have >> an idea of how to address it? > > Take a look at gcc/ada/Makefile.in lines 279 and below: > > ifeq ($(strip $(filter-out cygwin32% mingw32% pe,$(osys))),) >  # On Windows nati

Re: Bootstrap fails in 32bit libjava on x86_64 Fedora 16 with undefined reference to `__cxa_call_unexpected'

2011-11-20 Thread Jeff Law
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 11/20/11 16:53, Uros Bizjak wrote: > Hello! > > For some reason bootstrap on Fedora 16 fails when compiling 32bit > libjava with undefined reference to `__cxa_call_unexpected': This is (arguably) a glibc issue. I'm still investigating. Attached

Re: Profiling gcc itself

2011-11-20 Thread Hans-Peter Nilsson
On Sun, 20 Nov 2011, Jeff Evarts wrote: > On Sun, 20 Nov 2011 11:34:27 -0500, Tim Prince wrote: > > On 11/20/2011 11:10 AM, Basile Starynkevitch wrote: > >> On Sun, 20 Nov 2011 03:43:20 -0800 Jeff Evarts wrote: > >>> I posted this question at irc://irc.oftc.net/#gcc and they suggested > >>> tha

RE: Is VRP is too conservative to identify boolean value 0 and 1?

2011-11-20 Thread Jiangning Liu
> -Original Message- > From: Richard Guenther [mailto:richard.guent...@gmail.com] > Sent: Friday, September 02, 2011 5:07 PM > To: Jiangning Liu > Cc: gcc@gcc.gnu.org > Subject: Re: Is VRP is too conservative to identify boolean value 0 and > 1? > > On Fri, Sep 2, 2011 at 7:58 AM, Jiangn

Re: Profiling gcc itself

2011-11-20 Thread Jonathan Wakely
On 21 November 2011 00:50, Jeff Evarts wrote: > Seriously, I meant no offense, and didn't want to cause a stir. I didn't interpret the replies as offended or stirred, I think they were just saying "it's not clear from your question whether you are already familiar with the most rudimentary* tools

Re: Profiling gcc itself

2011-11-20 Thread Diego Novillo
On Sun, Nov 20, 2011 at 08:10, Basile Starynkevitch wrote: > I'm not sure the question belongs to gcc@gcc.gnu.org, perhaps > gcc-h...@gcc.gnu.org might > be a better place. gcc@ is the right forum for this question. Diego.

Re: Profiling gcc itself

2011-11-20 Thread Diego Novillo
On Sun, Nov 20, 2011 at 03:43, Jeff Evarts wrote: > I posted this question at irc://irc.oftc.net/#gcc and they suggested > that I pose it here instead. > > I do some "large-ish" builds (linux, gcc itself, etc) on a too-regular > basis, and I was wondering what could be done to speed things up. A >

Re: Profiling gcc itself

2011-11-20 Thread David Edelsohn
On Sun, Nov 20, 2011 at 7:50 PM, Jeff Evarts wrote: > Apparently my question "Has anyone done X and what were the results" > has been mysteriously translated into "The GCC compiler is way too > slow and what are you lazy developers going to do to speed up my > compiles?!?". I will be more careful

Re: libgcc: why emutls.c in LIB2ADDEH instead of LIB2ADD?

2011-11-20 Thread Richard Henderson
On 11/20/2011 01:28 AM, Hans-Peter Nilsson wrote: > As subject says, in libgcc, why is $(srcdir)/emutls.c in > LIB2ADDEH (and LIB2ADDEHSTATIC and LIB2ADDEHSHARED) instead of > LIB2ADD? > > Emulating TLS has nothing to do with exception-handling, nor is > there something that might throw while call

Re: Profiling gcc itself

2011-11-20 Thread Jeff Evarts
I understand that the noob traffic to gcc@ must be high. Nonetheless... On Sun, Nov 20, 2011 at 8:10 AM, Basile Starynkevitch wrote: > > On Sun, 20 Nov 2011 03:43:20 -0800 > Jeff Evarts wrote: > > I posted this question at irc://irc.oftc.net/#gcc and they suggested > > that I pose it here inste

Bootstrap fails in 32bit libjava on x86_64 Fedora 16 with undefined reference to `__cxa_call_unexpected'

2011-11-20 Thread Uros Bizjak
Hello! For some reason bootstrap on Fedora 16 fails when compiling 32bit libjava with undefined reference to `__cxa_call_unexpected': libtool: link: /home/uros/gcc-build/./gcc/xgcc -shared-libgcc -B/home/uros/gcc-build/./gcc -nostdinc++ -L/home/uros/gcc-build/x86_64-unknown-linux-gnu/32/libstdc++

Re: Profiling gcc itself

2011-11-20 Thread Tim Prince
On 11/20/2011 11:10 AM, Basile Starynkevitch wrote: On Sun, 20 Nov 2011 03:43:20 -0800 Jeff Evarts wrote: I posted this question at irc://irc.oftc.net/#gcc and they suggested that I pose it here instead. I do some "large-ish" builds (linux, gcc itself, etc) on a too-regular basis, and I was

Re: Profiling gcc itself

2011-11-20 Thread Basile Starynkevitch
On Sun, 20 Nov 2011 03:43:20 -0800 Jeff Evarts wrote: > I posted this question at irc://irc.oftc.net/#gcc and they suggested > that I pose it here instead. > > I do some "large-ish" builds (linux, gcc itself, etc) on a too-regular > basis, and I was wondering what could be done to speed things u

Profiling gcc itself

2011-11-20 Thread Jeff Evarts
I posted this question at irc://irc.oftc.net/#gcc and they suggested that I pose it here instead. I do some "large-ish" builds (linux, gcc itself, etc) on a too-regular basis, and I was wondering what could be done to speed things up. A little printf-style checking hints to me that I might be spen

libgcc: why emutls.c in LIB2ADDEH instead of LIB2ADD?

2011-11-20 Thread Hans-Peter Nilsson
As subject says, in libgcc, why is $(srcdir)/emutls.c in LIB2ADDEH (and LIB2ADDEHSTATIC and LIB2ADDEHSHARED) instead of LIB2ADD? Emulating TLS has nothing to do with exception-handling, nor is there something that might throw while calling one of its functions. Ok to fix that? brgds, H-P

Re: Issues with building gcc trunk with ada on cygwin

2011-11-20 Thread Eric Botcazou
> Now, are there anyones here that recognise the issue and perhaps have > an idea of how to address it? Take a look at gcc/ada/Makefile.in lines 279 and below: ifeq ($(strip $(filter-out cygwin32% mingw32% pe,$(osys))),) # On Windows native the tconfig.h files used by C runtime files needs to h

Issues with building gcc trunk with ada on cygwin

2011-11-20 Thread Christian Joensson
I have an issue with building gcc trunk, current and since quite a while. There seems to me to be quite some missing includes while building stage 3 compared with stage 2. For example, the adadecode.c comile in stage 2 for me looks like this: make[5]: Entering directory `/usr/local/src/trunk/obj