Re: [RFC] Introduce -Ofast

On 06/05/2010 21:25, Ian Lance Taylor wrote: > Richard Guenther writes: > >> This is a proposal to introduce an optimization level -Ofast >> that can collect (target specific) optimization flags that >> can affect runtime behavior such as -funsafe-math-optimizations >> or -mrecip. > > Sounds lik

gcc-4.5-20100506 is now available

Snapshot gcc-4.5-20100506 is now available on ftp://gcc.gnu.org/pub/gcc/snapshots/4.5-20100506/ and on various mirrors, see http://gcc.gnu.org/mirrors.html for details. This snapshot has been generated from the GCC 4.5 SVN branch with the following options: svn://gcc.gnu.org/svn/gcc/branches

Re: Broken GCC mirror sites

Dnia czwartek, 6 maja 2010 o 22:23:47 Holger Weiß napisał(a): > * Krzysztof Zelechowski [2010-05-06 22:00]: > > Regarding your page "GCC mirror sites" > > http://gcc.gnu.org/mirrors.html>: > > > > * ftp://ftp.fu-berlin.de/unix/languages/gcc/> is empty. > > It's not. > > > * ftp://ftp.gwdg.de/pub

[Fwd: Cron nice ionice -n6 /home/bernie/bin/git-update-toolchain.sh]

Can we kill this empty branch from svn? - Mensaje reenviado De: Cron Daemon Para: ber...@infradead.org Asunto: Cron nice ionice -n6 /home/bernie/bin/git-update-toolchain.sh Fecha: Thu, 06 May 2010 19:41:42 + W: -empty_dir: branches/ibm/power7-meissner -- // Bernie Inn

Re: C++0x Memory model and gcc

Ian Lance Taylor wrote: Andrew MacLeod writes: I've been working for a while on understanding how the new memory model and Atomics work, and what the impacts are on GCC. Thanks for looking at this. One issue I didn't see clearly was how to actually implement this in the compiler. F

Re: C++0x Memory model and gcc

Andrew MacLeod writes: > I've been working for a while on understanding how the new memory > model and Atomics work, and what the impacts are on GCC. Thanks for looking at this. One issue I didn't see clearly was how to actually implement this in the compiler. For example, speculated stores ar

Re: role of "register" C keyword?

On Thu, May 06, 2010 at 01:22:27PM -0700, Ian Lance Taylor wrote: > Basile Starynkevitch writes: > > > Long time ago (probably in the GCC 2.95 & 3.2 time frame) I would > > imagine that the "register" keyword indeed affected register > > allocation, in the sense that variables declared with regis

Re: role of "register" C keyword?

On Thu, May 06, 2010 at 01:22:27PM -0700, Ian Lance Taylor wrote: > Basile Starynkevitch writes: > > > Long time ago (probably in the GCC 2.95 & 3.2 time frame) I would > > imagine that the "register" keyword indeed affected register > > allocation, in the sense that variables declared with regis

Re: [RFC] Introduce -Ofast

Richard Guenther writes: > This is a proposal to introduce an optimization level -Ofast > that can collect (target specific) optimization flags that > can affect runtime behavior such as -funsafe-math-optimizations > or -mrecip. Sounds like a good idea to me. I don't like the name -Ofast, which

Re: Broken GCC mirror sites

* Krzysztof Zelechowski [2010-05-06 22:00]: > Regarding your page "GCC mirror sites" http://gcc.gnu.org/mirrors.html>: > > * ftp://ftp.fu-berlin.de/unix/languages/gcc/> is empty. It's not. > * ftp://ftp.gwdg.de/pub/misc/gcc/> is broken. In what way? > Please remove these entries. Or just ask

Re: role of "register" C keyword?

Basile Starynkevitch writes: > Long time ago (probably in the GCC 2.95 & 3.2 time frame) I would > imagine that the "register" keyword indeed affected register > allocation, in the sense that variables declared with register where > indeed & preferentially put in a machine register. I don't thin

Re: Question about Machine Description

yazdanbakhsh writes: > (define_expand "cbranchsi4" > [(set (pc) > (if_then_else: SI (le:SI (match_operand:SI 0 "register_operand" "=d,d") > (match_operand:SI 1 "register_operand" "=d,d")) > (label_ref (match_operand 2 "" "")) > (pc)))] > "" > "

Re: Broken GCC mirror sites

Hi, On Thu, 6 May 2010, Krzysztof ?elechowski wrote: Regarding your page "GCC mirror sites" http://gcc.gnu.org/mirrors.html>: * ftp://ftp.fu-berlin.de/unix/languages/gcc/> is empty. * ftp://ftp.gwdg.de/pub/misc/gcc/> is broken. * http://www.mpi-inf.mpg.de/pub/gnu/mirror/gcc.gnu.org/pub/gcc> is

Broken GCC mirror sites

Regarding your page "GCC mirror sites" http://gcc.gnu.org/mirrors.html>: * ftp://ftp.fu-berlin.de/unix/languages/gcc/> is empty. * ftp://ftp.gwdg.de/pub/misc/gcc/> is broken. * http://www.mpi-inf.mpg.de/pub/gnu/mirror/gcc.gnu.org/pub/gcc> is broken. Please remove these entries. Cheers, Chris

Re: role of "register" C keyword?

On 05/06/2010 08:35 PM, Jonathan Wakely wrote: On 06/05/2010, Basile Starynkevitch wrote: [ register variable ] In C++ it's only a hint, which the compiler is free to ignore. Its use will be deprecated in C++1x. Indeed, g++ won't even complain about taking its address (much to my surpris

Re: role of "register" C keyword?

On 06/05/2010, Basile Starynkevitch wrote: > > IIRC, (from my human rusty memory, I don't have the documents on hand) > the current standard only requires that variable declared with the > register class should not have their adress taken thru the unary & > operator, and might have some imprecise p

Re: Question about Machine Description

Hi, I changed the .md and .c file: I add these lines to the .md file: - (define_insn "ble1" [(set (pc) (if_then_else: SI (le:SI (match_operand:SI 0 "register_operand" "=d,d") (match_operand:SI 1 "register_operand" "=d,d")

Re: C++0x Memory model and gcc

Richard Guenther wrote: On Thu, May 6, 2010 at 5:50 PM, Richard Guenther wrote: First let me say that the C++ memory model is crap when it forces data-races to be avoided for unannotated data like the examples for packed data. And it isn't consistent across the board, since neighbouri

Re: C++0x Memory model and gcc

Joseph S. Myers wrote: On Thu, 6 May 2010, Andrew MacLeod wrote: - Implementation of atomic variables and synchronization modes as well as a new infrastructure to test these types of things. I presume you've read the long thread starting at

Re: C++0x Memory model and gcc

On Thu, May 6, 2010 at 5:50 PM, Richard Guenther wrote: > On Thu, May 6, 2010 at 5:43 PM, Andrew MacLeod wrote: >> I've been working for a while on understanding how the new memory model and >> Atomics work, and what the impacts are on GCC. >> >> It would be ideal to get as many of these changes

Re: C++0x Memory model and gcc

On Thu, 6 May 2010, Andrew MacLeod wrote: > - Implementation of atomic variables and synchronization modes > as well as a new infrastructure to test these types of things. I presume you've read the long thread starting at regarding the issues in

Re: C++0x Memory model and gcc

On Thu, May 6, 2010 at 5:43 PM, Andrew MacLeod wrote: > I've been working for a while on understanding how the new memory model and > Atomics work, and what the impacts are on GCC. > > It would be ideal to get as many of these changes into GCC 4.6 as possible. > I've started work on some of the mo

C++0x Memory model and gcc

I've been working for a while on understanding how the new memory model and Atomics work, and what the impacts are on GCC. It would be ideal to get as many of these changes into GCC 4.6 as possible. I've started work on some of the modifications and testing, and the overall impact on GCC shou

Re: [RFC] Introduce -Ofast

On 05/06/2010 04:24 PM, Richard Guenther wrote: > I would initially propose to make -Ofast be -O3 -ffast-math > and would strongly encourage target maintainers that consider > adding to -Ofast make sure that popular benchmarks for > their target still behave correctly when using -Ofast. I am in fa

[RFC] Introduce -Ofast

This is a proposal to introduce an optimization level -Ofast that can collect (target specific) optimization flags that can affect runtime behavior such as -funsafe-math-optimizations or -mrecip. Currently none of the standard optimization levels have this kind of affect and we should not change

role of "register" C keyword?

Hello All, What is the role of the "register" keyword [=storage class specifier] in C (& C++) in the current GCC compiler? IIRC, (from my human rusty memory, I don't have the documents on hand) the current standard only requires that variable declared with the register class should not have their

Re: Build report for gcc-4.5.0 - i686-pc-cygwin m68k-unknown-elf

On 05/06/2010 04:46 AM, Alexandr Yeremenko wrote: $ ../../src/gcc-4.5.0/config. guess i686-pc-cygwin $ m68k-elf-gcc -v Using built-in specs. COLLECT_GCC=m68k-elf-gcc COLLECT_LTO_WRAPPER=/usr/gcc-m68k/libexec/gcc/m68k-elf/4.5.0/lto-wrapper.exe Target: m68k-elf Configured with: ../../src/gcc-4.5.0

Re: a peculiar fpload problem on an inferior processor

On 06/05/2010 10:00, Amker.Cheng wrote: > here is the question : Is it possible to replace all direct fpload > with "load_into_gpr ; move_gpr_into_fpr" > sequence. I doubt it since the reload pass might generate direct > fpload insn for spilling fpu register. It is possible. Your expander can

Re: scheduling on VLIW architecture

On Thu, May 6, 2010 at 11:47 AM, roy rosen wrote: > Hi all. > > I work on a VLIW architecture. > The sched2 pass adds a TImode to insns which should start a new issue group. > But, after this pass, other passes change the insns, so the sched2 > work that was done is not correct anymore (the groups

scheduling on VLIW architecture

Hi all. I work on a VLIW architecture. The sched2 pass adds a TImode to insns which should start a new issue group. But, after this pass, other passes change the insns, so the sched2 work that was done is not correct anymore (the groups of insns might be invalid). In particular I see that the com

Build report for gcc-4.5.0 - i686-pc-cygwin m68k-unknown-elf

$ ../../src/gcc-4.5.0/config. guess i686-pc-cygwin $ m68k-elf-gcc -v Using built-in specs. COLLECT_GCC=m68k-elf-gcc COLLECT_LTO_WRAPPER=/usr/gcc-m68k/libexec/gcc/m68k-elf/4.5.0/lto-wrapper.exe Target: m68k-elf Configured with: ../../src/gcc-4.5.0/configure --target=m68k-elf --prefix=/usr/g cc-m68k

Re: Question about Machine Description

Hi, As I told I want to add "ble" intruction in MIPS that works like "beq". I used from the available branch_equality instruction that shows in the following paragraph: -- (define_insn "branch_equal

a peculiar fpload problem on an inferior processor

Hi : Our processor has an errata that the direct fpu load cannot work right, so I have to substitute instruction sequence "load_into_gpr ; move_gpr_into_fpr" for direct fpload insn. Currently I thought of two potential methods as following: method 1: step1 : keep a scratch register when e

Re: Building GCC & CFLAGS settings

Steve Ellcey writes: > HJ seems to have the way to make it work, set CC to 'gcc -mlp64' instead > of setting CC to 'gcc' and CFLAGS to '-mlp64' but that seems a bit > counter-intuitive to the meaning of the flags. '-mlp64' is an ABI-changing flag, so it is better to put it in CC, since you are,

gcc/resource.h conflicts with sysroot/usr/include/resource.h (alpha-dec-vms)

lpha-dec-vms-gcc -c   -DIN_GCC   -W -Wall -Wwrite-strings -Wcast-qual -Wstrict-prototypes -Wmissing-prototypes -Wmissing-format-attri bute -pedantic -Wno-long-long -Wno-variadic-macros -Wno-overlength-strings -Wold-style-definition -Wc++-compat   -DHAVE_CONFIG_H -I. - I. -I/src/gcc-4.5.0/gcc -I/