Re: Major bootstrap time regression on March 30

2005-04-07 Thread Arnaud Charlet
> I was doing full builds, except for Ada. I should have mentioned that. > Ada doesn't get configured by default, and I haven't bothered to check > why. I probably forgot to install the Ada compiler package. Ada is not enabled by default, you need to include it explicitely, e.g. --enable-langua

Re: How to specify customized base addr?

2005-04-07 Thread Zhenyu Guo
Ok, I've got it now. But still some stupid problems. http://sourceware.org/ml/binutils/2005-04/msg00193.html Thanks, Zhenyu. -邮件原件- 发件人: Zagorodnev, Grigory [mailto:[EMAIL PROTECTED] 发送时间: 2005年4月8日 4:32 收件人: [EMAIL PROTECTED]; gcc@gcc.gnu.org 主题: RE: How to specify customized base addr?

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Eric Christopher
> You mean this? > > /* For each hard register, the widest mode object that it can contain. >This will be a MODE_INT mode if the register can hold integers. Otherwise >it will be a MODE_FLOAT or a MODE_CC mode, whichever is valid for the >register. */ > > But take a look at the hea

Re: Does anyone use -fprofile-use with C++?

2005-04-07 Thread Gabriel Dos Reis
Joe Buck <[EMAIL PROTECTED]> writes: | On Thu, Apr 07, 2005 at 06:32:26PM -0400, Andrew Pinski wrote: | > | > On Apr 7, 2005, at 6:01 PM, Daniel Kegel wrote: | > | > >Judging by http://gcc.gnu.org/PR20815, I get the feeling | > >not many people are using the -fprofile-generate | > >and -fprofile

Re: Major bootstrap time regression on March 30

2005-04-07 Thread Diego Novillo
On Thu, Apr 07, 2005 at 05:57:06PM -0700, James E Wilson wrote: > So I think this is a possibility only if the Ada library got suddenly > bigger and slower to compile, and if you are including Ada in your > builds. > No Ada. I only build default languages. Oh, well, we'll see what the timings lo

Re: Major bootstrap time regression on March 30

2005-04-07 Thread James E Wilson
On Thu, 2005-04-07 at 17:34, Diego Novillo wrote: > Another thing, has our library code base (libjava, libstdc++) > grown significantly lately? I was doing full builds, except for Ada. I should have mentioned that. Ada doesn't get configured by default, and I haven't bothered to check why. I pr

Re: Major bootstrap time regression on March 30

2005-04-07 Thread Diego Novillo
On Thu, Apr 07, 2005 at 03:08:50PM -0700, James E Wilson wrote: > Is it possible that something changed on Diego's machine? Such as a new > binutils, or a new kernel, etc? > Yes. I rebooted the machine into a new kernel on 2005-03-31 (2.6.10-1.770_FC3). The slowdown coincided with the box bei

Re: Does anyone use -fprofile-use with C++?

2005-04-07 Thread Joe Buck
On Thu, Apr 07, 2005 at 06:32:26PM -0400, Andrew Pinski wrote: > > On Apr 7, 2005, at 6:01 PM, Daniel Kegel wrote: > > >Judging by http://gcc.gnu.org/PR20815, I get the feeling > >not many people are using the -fprofile-generate > >and -fprofile-use options yet, at least not with > >C++, since it

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Eric Botcazou
> Personally, I think that builtin_apply_args should go away, but what are > your thoughts on the rest of the thread. Richard didn't respond to my > last mail in the thread here: > > http://gcc.gnu.org/ml/gcc-patches/2004-03/msg02184.html I'm with Richard on this, I think reg_raw_mode is not appro

Re: Obsoleting c4x last minute for 4.0

2005-04-07 Thread Björn Haase
Am Freitag, 8. April 2005 01:06 schrieb Janis Johnson: > > I should have done that, I must have missed seeing your patch. I'll look > for it now in the archives. > > Janis I just had a look at the archives and found that the subject of the mail I have been sending was not very clear either :-) (a

Re: Obsoleting c4x last minute for 4.0

2005-04-07 Thread Janis Johnson
On Thu, Apr 07, 2005 at 11:20:46PM +0200, Björn Haase wrote: > > The reason why I have stopped posting the test results is that we are > currently having 481 failures for the AVR target and the existing real bugs > are completely hidden behind the huge number of failures due to issues like > "t

Re: about alias analysis

2005-04-07 Thread James E Wilson
zouq wrote: > i wonder whether there exists or not alias analysis for scalar variable, > array variable, even pointers. > thank you. Yes. See alias.c for alias analysis on RTL. See also tree-ssa-alias.c for alias analysis on trees. The latter is much more sophisticated. The latter hasn't been

Re: Use normal section names for comdat group?

2005-04-07 Thread James E Wilson
H. J. Lu wrote: Gcc 4,0 generates .section .gnu.linkonce.t._ZN3FooC1Ev,"axG",@progbits,_ZN3FooC1Ev,comdat for comdat group. Can gcc use .section .text._ZN3FooC1Ev,"axG",@progbits,_ZN3FooC1Ev,comdat instead? At least, there will be less characters for linker to process. I know of no reason why we mu

Re: Does anyone use -fprofile-use with C++?

2005-04-07 Thread Andrew Pinski
On Apr 7, 2005, at 6:01 PM, Daniel Kegel wrote: Judging by http://gcc.gnu.org/PR20815, I get the feeling not many people are using the -fprofile-generate and -fprofile-use options yet, at least not with C++, since it appears that namespaces make those options fall over... It'd be nice to get this

Re: Obsoleting c4x last minute for 4.0

2005-04-07 Thread Björn Haase
References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> Joseph Myers wrote: >One possible way of assessing activity would be to say that after 4.1 maintained CPU ports should have test results for mainline regularly

Re: ia64 bootstrap failure with the reload-branch

2005-04-07 Thread James E Wilson
On Thu, 2005-04-07 at 04:35, Bernd Schmidt wrote: > Try the patch below - does it fix the problem? OK, I will try it. This may take about 2 days. My lone IA-64 machine is busy doing other stuff at the moment. -- Jim Wilson, GNU Tools Support, http://www.SpecifixInc.com

Re: Major bootstrap time regression on March 30

2005-04-07 Thread James E Wilson
Steven Bosscher wrote: We have a bootstrap time regression since March 30. Bootstrap times on Diego Novillo's SPEC box went up from (an already high) 5500s to almost 8000s I tried looking at this, but I haven't been able to find any bootstrap time regressions. I've got a mainline tree checked o

Does anyone use -fprofile-use with C++?

2005-04-07 Thread Daniel Kegel
Judging by http://gcc.gnu.org/PR20815, I get the feeling not many people are using the -fprofile-generate and -fprofile-use options yet, at least not with C++, since it appears that namespaces make those options fall over... It'd be nice to get this fixed for gcc-4.0 (assuming it's a real bug and

Re: How to "disable" register allocation?

2005-04-07 Thread Øyvind Harboe
> If you fix all but 4 registers, then the compiler will only use the > remaining 4. Extra credit if your code compiles with just 2 registers > left! Fixing registers causes the compiler to fight for its life to put those values into the remaining registers => code bloat, and I guess you are re

Re: 2 suggestions

2005-04-07 Thread Hugh Sasse Staff Elec Eng
On Thu, 7 Apr 2005, Joe Buck wrote: On Thu, Apr 07, 2005 at 03:11:25PM -0400, Kaveh R. Ghazi wrote: I don't know about the utility of example scripts in general, but for this specific case, I strongly feel autoconf should automatically detect this and reexec the configure script under /bin/ksh. Is

Re: How to "disable" register allocation?

2005-04-07 Thread Mike Stump
On Thursday, April 7, 2005, at 09:12 AM, Øyvind Harboe wrote: Is there an option(compile/build time?) to tell GCC to use as few registers as possible? There's always the manual: @item [EMAIL PROTECTED] @opindex ffixed Treat the register named @var{reg} as a fixed register; generated code should ne

Re: How to specify customized base addr?

2005-04-07 Thread Mike Stump
On Thursday, April 7, 2005, at 09:25 AM, [EMAIL PROTECTED] wrote: Any idea about how to set the options of gcc to produce a static linked elf with customized base addr? I've googled, but found nothing worth:( Wrong list. Hint, go read the _linker_ documentation. The usual linker is part of bi

RE: How to specify customized base addr?

2005-04-07 Thread Zagorodnev, Grigory
You may extract default linker script "gcc ... -Wl,--verbose", change the base address and rebuild application, overriding default linker script by the new-one. SECTIONS { /* Read-only sections, merged into text segment: */ . = 0x40 + SIZEOF_HEADERS; - Grigory >-Original Message

Re: ERROR : pls help

2005-04-07 Thread Devang Patel
On Apr 7, 2005, at 12:36 PM, Virender Kashyap wrote: On Thu, 7 Apr 2005, Dale Johannesen wrote: The command line you show is the built compiler trying to build gcc's library. It doesn't work, which means there is a bug in your changes. actually it was working fine before even with changes. but n

Re: 2 suggestions

2005-04-07 Thread Gabriel Dos Reis
Toon Moene <[EMAIL PROTECTED]> writes: | Kaveh R. Ghazi wrote: | | > If people would simply follow the directions here: | > by setting | > CONFIG_SHELL to /bin/ksh before configure;make bootstrap, | | Indeed - I stopped counting how many h

Re: ERROR : pls help

2005-04-07 Thread Virender Kashyap
On Thu, 7 Apr 2005, Dale Johannesen wrote: The command line you show is the built compiler trying to build gcc's library. It doesn't work, which means there is a bug in your changes. actually it was working fine before even with changes. but how do i locate where is the problem . this does not

Re: Inline round for IA64

2005-04-07 Thread Geert Bosch
On Apr 7, 2005, at 13:54, Steve Kargl wrote: I missed that part of the output. The exceeding long string of digits caught my attention. Can you submit a PR? These routines should really be done as builtins, as almost all front ends need this facility and we'd fit in with the common frameworks for

Re: 2 suggestions

2005-04-07 Thread Joe Buck
On Thu, Apr 07, 2005 at 03:11:25PM -0400, Kaveh R. Ghazi wrote: > I don't know about the utility of example scripts in general, but for > this specific case, I strongly feel autoconf should automatically > detect this and reexec the configure script under /bin/ksh. Is there a specific test we can

Re: ERROR : pls help

2005-04-07 Thread Dale Johannesen
On Apr 7, 2005, at 11:58 AM, Virender Kashyap wrote: hi i made some changes in gcc code. when i try to comile it using make , i get the following error ( last few lines from output ). Pls help me in removing this error. The command line you show is the built compiler trying to build gcc's l

Re: 2 suggestions

2005-04-07 Thread Kaveh R. Ghazi
> > CONFIG_SHELL to /bin/ksh before configure;make bootstrap, they > > wouldn't have such insane build times. I bet it cuts the 48 hours > > to single digits. > > The trouble is that *people* are building this. Googling turns up: > "Freemans rule: Nothing is so simple that it cannot be mi

Re: HEAD regression: All java tests are failing with an ICE when optimized

2005-04-07 Thread Andrew Pinski
On Apr 7, 2005, at 1:59 PM, Andrew Haley wrote: So maybe it is a front end bug. I'm working on it now. Note I really doubt this is a front-end bug as exception handling is lowered on the tree and then expanded from the tree tables to the rtl table during expand time. -- Pinski

ERROR : pls help

2005-04-07 Thread Virender Kashyap
hi i made some changes in gcc code. when i try to comile it using make , i get the following error ( last few lines from output ). Pls help me in removing this error. -f libgcc.mk all make[2]: Entering directory `/home/virenk/obj_gcc/gcc' /home/virenk/obj_gcc/gcc/xgcc -B/home/virenk/obj_gc

Re: 2 suggestions

2005-04-07 Thread Kaveh R. Ghazi
> > Not necessary. If people would simply follow the directions here: > > by setting > > I think this docoument could be made more useful by having the more > specific cases refer people to the applicable more general cases. If by t

Re: 2 suggestions

2005-04-07 Thread Joe Buck
On Thu, Apr 07, 2005 at 01:14:58PM -0400, Ray Holme wrote: > Perhaps this is why I use /bin/sh for all scripts I write - tis leaner and > meaner by far. But the implementation provided by several vendors (Solaris, AIX) is *extremely* slow for some operations, so slow as to add 24 hours to the time

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Eric Christopher
On Thu, 2005-04-07 at 16:48 +0200, Eric Botcazou wrote: > > That was my thoughts too. You could take a look at how I fixed it on > > ARM. > > Thanks. So basically you bypass the apply_result_mode array entirely, which > is still wrong as it is on SPARC? > Personally, I think that builtin_appl

Re: 2 suggestions

2005-04-07 Thread Toon Moene
Kaveh R. Ghazi wrote: If people would simply follow the directions here: by setting CONFIG_SHELL to /bin/ksh before configure;make bootstrap, Indeed - I stopped counting how many hours of debugging shell scripts on our Sun server I saved (fo

Re: Inline round for IA64

2005-04-07 Thread Paul Brook
On Thursday 07 April 2005 18:54, Steve Kargl wrote: > On Thu, Apr 07, 2005 at 01:41:21PM -0400, Geert Bosch wrote: > > On Apr 7, 2005, at 13:27, Steve Kargl wrote: > > >Try -fdump-parse-tree. You've given more digits in y than > > >its precision. This is permitted by the standard. It appears > >

Re: HEAD regression: All java tests are failing with an ICE when optimized

2005-04-07 Thread Andrew Haley
Ranjit Mathew writes: > Andrew Haley wrote: > > HEAD, clean build this morning. i686-linux-gnu. > > > > With the libgcj "make check", there are many identical failures. > > > > These are of the form > > > > PR4766.java: In class 'PR4766': > > PR4766.java: In method 'PR4766.myfunction()

Re: Inline round for IA64

2005-04-07 Thread Steve Kargl
On Thu, Apr 07, 2005 at 01:41:21PM -0400, Geert Bosch wrote: > On Apr 7, 2005, at 13:27, Steve Kargl wrote: > >Try -fdump-parse-tree. You've given more digits in y than > >its precision. This is permitted by the standard. It appears > >the gfortran frontend is taking y = 0.49 and the closest

Re: Inline round for IA64

2005-04-07 Thread Geert Bosch
On Apr 7, 2005, at 13:27, Steve Kargl wrote: Try -fdump-parse-tree. You've given more digits in y than its precision. This is permitted by the standard. It appears the gfortran frontend is taking y = 0.49 and the closest representable nubmer is y = 0.5. So, why does the test y < 0.5 yield tr

Re: 2 suggestions

2005-04-07 Thread E. Weddington
Hugh Sasse Staff Elec Eng wrote: So, I'd like to know if the variations in how to build GCC are so numerous that having a collection of example build scripts is a stupid idea. I think that examples are valuable in aiding understanding. Examples often seem clearer than descriptive text, though the

Re: 2 suggestions

2005-04-07 Thread Hugh Sasse Staff Elec Eng
On Thu, 7 Apr 2005, Kaveh R. Ghazi wrote: > > 1) years ago GCC took about 2 hours to compile, last year it was 26 > > hours for me, this year I just surpassed 48 hours and it is still [...] > > Configure with --disable-libgcj. I even considered making this the [...] Not necessary.

Re: How to "disable" register allocation?

2005-04-07 Thread Vladimir Makarov
Øyvind Harboe wrote: Is there an option(compile/build time?) to tell GCC to use as few registers as possible? Basically I want to run some tests where the assumption is that stack slots generate faster/smaller code than registers. Any pointers to where I should look in the GCC source, would be grea

Re: Inline round for IA64

2005-04-07 Thread Richard Kenner
> print *, 'nint ( y ) =', nint (y), ', where y < 0.5 = ', y < 0.5 > end > > output is > nint ( 8388609. ) = 8388610 > nint ( y ) = 1 , where y < 0.5 = T Try -fdump-parse-tree. You've given more digits in y than its precision. T

Re: Inline round for IA64

2005-04-07 Thread Steve Kargl
On Thu, Apr 07, 2005 at 01:10:49PM -0400, Geert Bosch wrote: > > On Apr 7, 2005, at 10:12, Steve Kargl wrote: > > >On Thu, Apr 07, 2005 at 08:08:15AM -0400, Geert Bosch wrote: > >>As far as I can seem from this patch, it rounds incorrectly. > >>This is a problem with the library version as well,

Re: 2 suggestions

2005-04-07 Thread Ray Holme
This is why I suggested several parameters to the question. Assumedly if the parameters were the same, so would the answer be. Note that lots of scripts do the same style confuring and a dictionary or param-set responses was what I was suggesting. It was merely a simple suggestion and I believe it

Re: 2 suggestions

2005-04-07 Thread Ray Holme
>Also it helps a lot to remove paths to directories over the network (like >NFS) >from PATH and LD_LIBRARY_PATH if this is possible. Saves me half of >bootstrap time with our crappy network setup here at university. > Old trick and highly relevent - but this sun mounts nothing NFS wise. Thanks,

Re: 2 suggestions

2005-04-07 Thread Ray Holme
Perhaps this is why I use /bin/sh for all scripts I write - tis leaner and meaner by far. Course it cannot do some things - but so far that has never been a problem. Ray Eric Botcazou <[EMAIL PROTECTED]> on Thursday, April 7, 2005 at 12:56 PM wrote: >> Not necessary. If people would simply f

Re: Inline round for IA64

2005-04-07 Thread Geert Bosch
On Apr 7, 2005, at 10:12, Steve Kargl wrote: On Thu, Apr 07, 2005 at 08:08:15AM -0400, Geert Bosch wrote: As far as I can seem from this patch, it rounds incorrectly. This is a problem with the library version as well, I believe. Which library? libgfortran, or whatever is used to implement NINT and

Re: ARM EABI Exception Handling

2005-04-07 Thread Paul Brook
On Thursday 07 April 2005 17:46, Josh Conner wrote: > Can anyone tell me if there are plans to incorporate the ARM EABI > exception handling (a la CSL) into the mainline gcc ARM port? Yes. It's on our (CodeSourcery's) list of things to do. Paul

Re: 2 suggestions

2005-04-07 Thread Eric Botcazou
> Not necessary. If people would simply follow the directions here: > by setting > CONFIG_SHELL to /bin/ksh before configure;make bootstrap, they > wouldn't have such insane build times. I bet it cuts the 48 hours to > single digits. Sure,

ARM EABI Exception Handling

2005-04-07 Thread Josh Conner
Can anyone tell me if there are plans to incorporate the ARM EABI exception handling (a la CSL) into the mainline gcc ARM port? Thanks - Josh Josh Conner Apple Computer

Re: building GCC 4.0 for arm-elf target on mingw host

2005-04-07 Thread E. Weddington
Dave Murphy wrote: E. Weddington wrote: Dave Murphy wrote: copying the compile line and removing the spurious -I and the -I../../../gcc-4.0-20050319-new/gcc/ results in no errors. I'm having a little trouble finding where this line is built up in the makefiles, can anyone point me in the right d

Re: 2 suggestions

2005-04-07 Thread Kaveh R. Ghazi
> > 1) years ago GCC took about 2 hours to compile, last year it was 26 > > hours for me, this year I just surpassed 48 hours and it is still > > going - it would be very nice if one could compile the compiler and > > what it needs without having to build the entire java set (yes I > > know it

How to specify customized base addr?

2005-04-07 Thread guozy03
Hi there, Any idea about how to set the options of gcc to produce a static linked elf with customized base addr? I've googled, but found nothing worth:( Thanks, Zhenyu

Re: 2 suggestions

2005-04-07 Thread Christian Joensson
On Apr 7, 2005 6:12 PM, Eric Botcazou <[EMAIL PROTECTED]> wrote: > > 1) years ago GCC took about 2 hours to compile, last year it was 26 hours > > for me, this year I just surpassed 48 hours and it is still going - it > > would be very nice if one could compile the compiler and what it needs > > wi

Re: 2 suggestions

2005-04-07 Thread Eric Botcazou
> 1) years ago GCC took about 2 hours to compile, last year it was 26 hours > for me, this year I just surpassed 48 hours and it is still going - it > would be very nice if one could compile the compiler and what it needs > without having to build the entire java set (yes I know it is bigger and >

Re: Input and print statements for Front End?

2005-04-07 Thread Paolo Bonzini
I can't seem to find any info regarding an input or print statement, so i can read integers(my language only deals with integers) from the stdio and return integer results to stdio. You need to map these to printf/scanf calls. Paolo

How to "disable" register allocation?

2005-04-07 Thread Øyvind Harboe
Is there an option(compile/build time?) to tell GCC to use as few registers as possible? Basically I want to run some tests where the assumption is that stack slots generate faster/smaller code than registers. Any pointers to where I should look in the GCC source, would be greatly appreciated. -

Re: 2 suggestions

2005-04-07 Thread Richard Guenther
On Apr 7, 2005 5:54 PM, David Edelsohn <[EMAIL PROTECTED]> wrote: > > Ray Holme writes: > > Ray> 2) Much of the time is spent in the several iterations of building a > Ray> product doing the convfigure steps. These are repeated ad nauseum with > the > Ray> results being obtained the hard way

Re: 2 suggestions

2005-04-07 Thread David Edelsohn
> Ray Holme writes: Ray> 2) Much of the time is spent in the several iterations of building a Ray> product doing the convfigure steps. These are repeated ad nauseum with the Ray> results being obtained the hard way each time. As a database person, it Ray> seems to me that by perhaps having a s

Re: 2 suggestions

2005-04-07 Thread Andrew Pinski
> > Firstly - I love your product and have used it for years. > > I run on the Solaris Sparc platform mostly but use Linux and flavors of > Windows at times. My Sun is getting older so both of these suggestions > would help developers on less-than-super machines. > > 1) years ago GCC took about

2 suggestions

2005-04-07 Thread Ray Holme
Firstly - I love your product and have used it for years. I run on the Solaris Sparc platform mostly but use Linux and flavors of Windows at times. My Sun is getting older so both of these suggestions would help developers on less-than-super machines. 1) years ago GCC took about 2 hours to compil

Re: Inline round for IA64

2005-04-07 Thread Vincent Lefevre
On 2005-04-07 08:08:15 -0400, Geert Bosch wrote: > On platforms such as x86, where GCC implicitly widens intermediate > results for IEEE double, This depends on the FPU configuration, so... > the rounding to integer should be performed entirely in long double > mode, using the long double predece

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Richard Earnshaw
On Thu, 2005-04-07 at 15:51, David Edelsohn wrote: > > Richard Earnshaw writes: > > >> I think the change breaks __builtin_apply/__builtin_return on every > >> platform > >> that can return values in several contiguous registers, because the entry > >> for > >> FUNCTION_VALUE_REGNO_P in t

Re: Status of conversions to predicates.md

2005-04-07 Thread Eric Botcazou
> sparc > > Eric Botcazou says he is working on predicates.md himself. > > http://gcc.gnu.org/ml/gcc-patches/2005-03/msg01694.html Confirmed. :-) However don't hold your breath, my priority is 4.0.0 until after it is released. -- Eric Botcazou

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Eric Botcazou
> That was my thoughts too. You could take a look at how I fixed it on > ARM. Thanks. So basically you bypass the apply_result_mode array entirely, which is still wrong as it is on SPARC? -- Eric Botcazou

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread David Edelsohn
> Richard Earnshaw writes: >> I think the change breaks __builtin_apply/__builtin_return on every >> platform >> that can return values in several contiguous registers, because the entry >> for >> FUNCTION_VALUE_REGNO_P in the manual reads: Richard> That was my thoughts too. You could t

Status of conversions to predicates.md

2005-04-07 Thread Kazu Hirata
Hi, c4x The port doesn't build. cris A patch has been posted. http://gcc.gnu.org/ml/gcc-patches/2005-03/msg01705.html Hans-Peter Nilsson says he wants to modify the port, so the predicate conversion patch may need to be adjusted. i860 Obsoleted. ip2k Obsoleted. mmix A

Re: apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Richard Earnshaw
On Thu, 2005-04-07 at 14:49, Eric Botcazou wrote: > Hi Eric, > > The __builtin_apply/__builtin_return mechanism is broken on SPARC in 4.x > because of > > 2004-03-17 Eric Christopher <[EMAIL PROTECTED]> > > * builtins.c (apply_args_size): Use reg_raw_mode. > (apply_result_size): D

Re: Inline round for IA64

2005-04-07 Thread Steve Kargl
On Thu, Apr 07, 2005 at 08:08:15AM -0400, Geert Bosch wrote: > As far as I can seem from this patch, it rounds incorrectly. > This is a problem with the library version as well, I believe. > Which library? -- Steve

apply_result_size vs FUNCTION_VALUE_REGNO_P

2005-04-07 Thread Eric Botcazou
Hi Eric, The __builtin_apply/__builtin_return mechanism is broken on SPARC in 4.x because of 2004-03-17 Eric Christopher <[EMAIL PROTECTED]> * builtins.c (apply_args_size): Use reg_raw_mode. (apply_result_size): Ditto. http://gcc.gnu.org/ml/gcc-patches/2004-03/msg01422.html

Re: gcc-3.4-20050401 BUG? generates illegal instruction in X11R6.4.2/mkfontscale/freetypemacro

2005-04-07 Thread Clemens Koller
Hello, Dan! I did some more debugging on that problem. This bug was also reported on this list: http://gcc.gnu.org/ml/gcc/2004-07/msg00861.html To isolate it for the first step it's sufficient to only build mkfontscale within <...>/xc/programs/mkfontscale and then call it with mkfontscale Type1 (as

Re: Inline round for IA64

2005-04-07 Thread Geert Bosch
As far as I can seem from this patch, it rounds incorrectly. This is a problem with the library version as well, I believe. The issue is that one cannot round a positive float to int by adding 0.5 and truncating. (Same issues with negative values and subtracting 0.5, of course). This gives an error

Re: ia64 bootstrap failure with the reload-branch

2005-04-07 Thread Bernd Schmidt
(slowly catching up with vacation backlog) James E Wilson wrote: Reload is using registers without setting regs_ever_live. The IA-64 prologue code needs a temp register to save/restore b0, and it uses regs_ever_live to find one. Also, because we have variable size register windows, we need the

Re: 3.4.3 on Solaris9, boehm-gc probs.

2005-04-07 Thread Hugh Sasse Staff Elec Eng
On Thu, 7 Apr 2005, Eric Botcazou wrote: But to get the definitions we are after it should be looking in sys, No, is not directly included, only . Oh, I see. And I think I now understand what the --with-local-prefix does, so I'll let you know the results in a few hours. I've put --with-local-pr

Re: 3.4.3 on Solaris9, boehm-gc probs.

2005-04-07 Thread Eric Botcazou
> But to get the definitions we are after it should be looking in sys, > so it should look for /usr/local/include/sys/link.h rather than > /usr/local/include/link.h. > > /usr/local/include/sys/link.h: No such file or directory > > so then it should look in /usr/include/sys No, is not directly in

Re: 3.4.3 on Solaris9, boehm-gc probs.

2005-04-07 Thread Hugh Sasse Staff Elec Eng
On Thu, 7 Apr 2005, Eric Botcazou wrote: This is totally unlike /usr/include/link.h, but I don't see why it And unlike /usr/include/sys/link.h should be picked up by what must be a directive such as #include So I think this file isn't relevant. It is relevant, the local-prefix directory is searche

Re: 3.4.3 on Solaris9, boehm-gc probs.

2005-04-07 Thread Eric Botcazou
> This is totally unlike /usr/include/link.h, but I don't see why it > should be picked up by what must be a directive such as > > #include > > So I think this file isn't relevant. It is relevant, the local-prefix directory is searched first. > I'm not sure how to proceed though. Remove that fi

Re: 3.4.3 on Solaris9, boehm-gc probs.

2005-04-07 Thread Hugh Sasse Staff Elec Eng
On Wed, 6 Apr 2005, Eric Botcazou wrote: Here's what I get with -H: Sorry? -H applied to what command? The GCC command line you pasted. Oh ... "list included files"... sh-2.05$ chmod u+x !$ chmod u+x tmp-gcc-cmd.bash bash-2.05$ ./!$ ./tmp-gcc-cmd.bash + /export/home/Scratch/hgs/gcc-build/gcc/xgcc

Inline round for IA64

2005-04-07 Thread Canqun Yang
Hi, all Gfortran translates the Fortran 95 intrinsic DNINT to round operation with double precision type argument and return value. Inline round operation will speed up the SPEC CFP2000 benchmark 189.lucas which contains function calls of intrinsic DNINT from 706 (SPEC ratio) to 783 on IA64 1