Hello
I ponder about writing a "i386 16bit realmode" gcc backend as my master
thesis - which would be usefull for generating 16-bit bios code needed
by the virtual machine developed at my university.
Having never programmed any compiler-related stuff and having a strict
deadline (after 6 months
Mark Mitchell <[EMAIL PROTECTED]> writes:
| Gabriel Dos Reis wrote:
|
| > Actually, we do have machinery to print statement-expression. It is
| > just that calling dump_expr() is wrong.
|
| Why do we want to print them? When does that help the user debug the
| problem? The only situation I ca
I think the decision to force the
user to specify -lmudflap should be revisited.
I agree.
The fixincludes build failed with link errors for undefined mudflap
functions. The fixincludes Makefile.in does not use CFLAGS when linking. I
added $(CFLAGS) to the 3 rules that contain a link command.
I th
Hi,
I want to know if its possible to get warnings for unused variables that have a
user defined constructor and/or destructor.
I know that they are side-effect free and want to mark them for the compiler.
But neither attribute const nor pure helped me here.
Btw: std::string would be IMO a good
Hi,
During expression hash table construction in gcse pass(gcc vercion 3.4.1),
expressions like a*b does not get included into the expression hash table.
Such expressions occur in PARALLEL along with clobbers. This means that
such expression are not being subjected to PRE. Isn't it surprising? C
hello there :
I have a piece of code which will cause SIGBUS ( unaligned access ) under
HPUX-11.23
It will be optimized with +u1 option if I use HP C/C++ compiler and
the bus error will never appear.
But when swtich to using gcc for hpux , there is no compiler option
can op
Today lilo (the FreeNode network owner) has decided to make one step away in a
direction opposite of freedom, and banned all Tor users from the FreeNode
network.
Tor ( http://tor.eff.org ) is an open source anonymous gateway system. Many
users who are not in the position to be able to use IRC o
Patrick McFarland writes:
> Today lilo (the FreeNode network owner) has decided to make one step away in
> a
> direction opposite of freedom, and banned all Tor users from the FreeNode
> network.
I can't find a statement from FreeNode.
This would be a more appropriate discussion for GNU, w
Hi Dai,
If you choose not to change the bad code that is causing a SIGBUS, I
believe you cannot use GCC since (as far as I am aware) GCC does not
provide a mechanism *IN THE COMPILE/LINK* to catch the SIGBUS signal and do
the byte-by-byte manipulation to complete the action.
Such handling of a
Gabriel Dos Reis wrote:-
> That statement is factually false as can be verified with EDG-3.5:
Oh come on Gaby, that's not printing an expression, it prints
the raw source line, comments and trigraphs included.
Are you proposing your pretty printer do that too?
Neil.
"Dylan Cuthbert" <[EMAIL PROTECTED]> writes:
>emit_move_insn( gen_rtx_SUBREG (V4SFmode, op0, 0 ), gen_rtx_MEM(
> V4SFmode, src_reg ) );
>emit_move_insn( gen_rtx_SUBREG (V4SFmode, op0, 16 ), gen_rtx_MEM(
> V4SFmode, plus_constant( src_reg, 16 ) ) );
>emit_move_insn( gen_rtx_SUBREG (V4SF
On Tue, 1 Feb 2005, Steven Bosscher wrote:
> On Feb 01, 2005 05:10 PM, Richard Guenther <[EMAIL PROTECTED]> wrote:
>
> > Hi!
> >
> > I get the following error during bootstrap:
> >
> > ./xgcc -B./ -B/usr/local/x86_64-unknown-linux-gnu/bin/ -isystem
> > /usr/local/x86_64-unknown-linux-gnu/include -
Neil Booth <[EMAIL PROTECTED]> writes:
| Gabriel Dos Reis wrote:-
|
| > That statement is factually false as can be verified with EDG-3.5:
|
| Oh come on Gaby, that's not printing an expression, it prints
Please, the statement was that EDG does not print expression outside
declarations. But th
* Andrew Haley:
> Patrick McFarland writes:
> > Today lilo (the FreeNode network owner) has decided to make one step away
> in a
> > direction opposite of freedom, and banned all Tor users from the FreeNode
> > network.
>
> I can't find a statement from FreeNode.
There is one, but it's fair
> I ponder about writing a "i386 16bit realmode" gcc backend as my master
> thesis - which would be usefull for generating 16-bit bios code needed
> by the virtual machine developed at my university.
It's been done a couple of times already, first by me, and later my
code was extended by a couple
Gabriel Dos Reis wrote:
Lower forms appear because currently we do not do a good in the
front-end and pretty-printer telling which level of abstraction is
preferred. I noted initial effort in that direction has already been
undermined by you. Since, no technical reason was given, I must
assume
Thanks for the info...
I was worried about the aliasing problems - adjust_address fits the ticket
perfectly.
The information for simplify_gen_subreg is a little sparse, what does it do
differently to gen_rtx_subreg?
Regards
Dylan
"Richard Sandiford" <[EMAIL PROTECTED]> wrote in message
news:[E
well , Tru64 can catch & handle unaligned access by OS itself . the
unaligned access only cause a system warning , but the code can be running
only with performance penalty . there is a command to turn off the
unaligned access message by OS ( I can forgot the name ).
HPUX can not handle unalign
Ok, I think I found out why gen_subreg crashes here: (with gcc 3.3.3)
if (byte % GET_MODE_SIZE (outermode)
|| byte >= GET_MODE_SIZE (innermode))
abort ();
This check doesn't seem right to me ;-)
I'll see what's in the latest cvs for this function.
Regards
Dylan
"Richard Sandiford" <[EMAIL
I tried simplify_gen_subreg but it crashes with a compiler error. Maybe
because V4SF isn't really thought of as a subreg of a V16SF at the moment?
I am using gcc 3.3.3 right now so it might be just that it works in a later
version of the compiler?
"Richard Sandiford" <[EMAIL PROTECTED]> wrote
On Wed, Feb 23, 2005 at 10:14:22AM +0100, "Richter, Jörg" wrote:
> I want to know if its possible to get warnings for unused variables that
> have a user defined constructor and/or destructor.
> I know that they are side-effect free and want to mark them for the
> compiler. But neither attribute
Unless in gcc-world outermode has the meaning of innermode? (and vice
versa)
which.. from looking at some other source... perhaps it does.. :-/
"Dylan Cuthbert" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
Ok, I think I found out why gen_subreg crashes here: (with gcc 3.3.3)
if
On 23 Feb 2005 16:49:51 +0100, Gabriel Dos Reis <[EMAIL PROTECTED]> wrote:
> Neil Booth <[EMAIL PROTECTED]> writes:
>
> | Gabriel Dos Reis wrote:-
> |
> | > That statement is factually false as can be verified with EDG-3.5:
> |
> | Oh come on Gaby, that's not printing an expression, it prints
>
On Wed, Feb 23, 2005 at 12:14:23PM -0500, Jason Merrill wrote:
> On 23 Feb 2005 16:49:51 +0100, Gabriel Dos Reis <[EMAIL PROTECTED]> wrote:
>
> > Neil Booth <[EMAIL PROTECTED]> writes:
> >
> > | Gabriel Dos Reis wrote:-
> > |
> > | > That statement is factually false as can be verified with EDG-3
On Wednesday 23 February 2005 06:58 am, Andrew Haley wrote:
> Patrick McFarland writes:
> > Today lilo (the FreeNode network owner) has decided to make one step
> > away in a direction opposite of freedom, and banned all Tor users from
> > the FreeNode network.
>
> I can't find a statement from
Mark Mitchell <[EMAIL PROTECTED]> writes:
| Gabriel Dos Reis wrote:
|
| > Lower forms appear because currently we do not do a good in the
| > front-end and pretty-printer telling which level of abstraction is
| > preferred. I noted initial effort in that direction has already
| > been undermine
Hi, Jim -
> A customer expressed interest in mudflap, so I tried to see if I
> could use it compile something large. [...]
Thanks for giving it a try.
For what it's worth, I've run entire gcc bootstraps on Linux with the
instrumentation running (using BOOT_CFLAGS rather than CFLAGS). Each
ti
Joe Buck <[EMAIL PROTECTED]> writes:
| On Wed, Feb 23, 2005 at 12:14:23PM -0500, Jason Merrill wrote:
| > On 23 Feb 2005 16:49:51 +0100, Gabriel Dos Reis <[EMAIL PROTECTED]> wrote:
| >
| > > Neil Booth <[EMAIL PROTECTED]> writes:
| > >
| > > | Gabriel Dos Reis wrote:-
| > > |
| > > | > That stat
On Feb 22, 2005, at 6:36 PM, Joseph S. Myers wrote:
On Tue, 22 Feb 2005, Devang Patel wrote:
Would it be OK, if this warnings are disabled for system headers ?
What is the built-in function involved and what (and why) is the
different
system header type? Is this a case where a system reuses a non
On Wednesday 23 February 2005 11:03 am, Florian Weimer wrote:
> * Andrew Haley:
> > Patrick McFarland writes:
> > > Today lilo (the FreeNode network owner) has decided to make one step
> > > away in a direction opposite of freedom, and banned all Tor users from
> > > the FreeNode network.
> >
>
On Wed, Feb 23, 2005 at 12:49:41PM -0500, Frank Ch. Eigler wrote:
> Regarding -fmudflap => -lmudflap, it used to do that. The problem was
> that the simplest use of specs machinery creates a final sequence of
> "-lFOO" options that sometimes cannot work. libmudflap must be in a
> particular spot
On Wed, Feb 23, 2005 at 06:51:46PM +0100, Gabriel Dos Reis wrote:
> | I think that the best solution for the long term is the caret approach,
> | printing out the original source line that the user typed. Trying to
> | re-generate the expression from the tree is likely to generate something
> | co
Gabriel Dos Reis wrote:
I don't think it makes sense or it serves any useful purpose removing
you from maintainership, although I believe you should exercise it
with more openness.
Are we talking here about my role as a C++ maintainer? What kind of
additional openness would you like to see? I h
On 2005-02-23, at 18:40, Gabriel Dos Reis wrote:
ndards for error messages, etc.)
That certainly would require changing many things, e.g. Emacs support
and like. That is a reason why I approach this issue conservatively.
Factually the support for error handling by integration in to external
tools
On Mon, 21 Feb 2005, Steve Kargl wrote:
> If there's a WITH_FORTRAN optin, then I'm fine with how you
> intend to maintain the gcc4.0 port. One other item to keep
> in mind, gfortran is the *only* Fortran 95 available on
> non-i386 FreeBSD.
That (there is a WITH_FORTRAN option), and there is al
Hi,
I have a C++ program that runs slower under 4.0 CVS than 3.4. So, I am
trying to make some test-cases that might help deduce the reason.
However, when I reduced this testcase sufficiently, it began behaving
badly under BOTH 3.4 and 4.0 but I guess I should start with the
most reduced
Oh, I forgot to note that the compiler is
* Feb 22 4.0 CVS / i686-pc-linux-gnu
And the compilation flags I used were:
* -march=pentium4 -O3
The times come from running the software on a
* P4 2.8 GHz
-BenRI
The deadline for submitting proposals for the GCC & GNU Toolchain
Developers' Summit is March 1 (http:www.gccsummit.org). Besides
thinking about submitting a proposal yourself, consider what you'd
like to hear about or discuss at the Summit and encourage others to
submit proposals to cover those a
On Wed, Feb 23, 2005 at 12:49:41PM -0500, Frank Ch. Eigler wrote:
> Regarding memory consumption, perhaps libmudflap's default backtrace
> parameter should be set to zero, for both speed and space reasons.
If it's storing all the backtraces that is burning up all the memory,
another approach might
Mark Mitchell <[EMAIL PROTECTED]> writes:
| Gabriel Dos Reis wrote:
|
| > I don't think it makes sense or it serves any useful purpose removing
| > you from maintainership, although I believe you should exercise it
| > with more openness.
|
| Are we talking here about my role as a C++ maintainer
> "Gabriel" == Gabriel Dos Reis <[EMAIL PROTECTED]> writes:
Mark> (However, I've never had the time or energy to
Mark> work through the process of implementing the caret approach, which is
Mark> definitely a lot of work, and would necessarily include working
Mark> through issues about the GNU
Marcin Dalecki <[EMAIL PROTECTED]> writes:
| On 2005-02-23, at 18:40, Gabriel Dos Reis wrote:
| > ndards for error messages, etc.)
| >
| > That certainly would require changing many things, e.g. Emacs support
| > and like. That is a reason why I approach this issue conservatively.
| >
| Factually
Jakub Jelinek <[EMAIL PROTECTED]> writes:
| On Wed, Feb 23, 2005 at 06:51:46PM +0100, Gabriel Dos Reis wrote:
| > | I think that the best solution for the long term is the caret approach,
| > | printing out the original source line that the user typed. Trying to
| > | re-generate the expression f
On Wed, 2005-02-23 at 01:00, Paolo Bonzini wrote:
> > The fixincludes Makefile.in does not use CFLAGS when linking.
> I think this can be committed as obvious, especially by a GWP person as
> you are...
I'll be taking care of this today. I don't like to rely on the obvious
rule, and it was rathe
Tom Tromey <[EMAIL PROTECTED]> writes:
| > "Gabriel" == Gabriel Dos Reis <[EMAIL PROTECTED]> writes:
|
| Mark> (However, I've never had the time or energy to
| Mark> work through the process of implementing the caret approach, which is
| Mark> definitely a lot of work, and would necessarily i
On Wed, 2005-02-23 at 09:49, Frank Ch. Eigler wrote:
> You should not have to invent any further static counters
> for this stuff: the __mf_state variable should be able to do the right
> thing.
Thanks for the helpful comments. The __mf_state variable does indeed
look like it can cleanly solve th
Gabriel Dos Reis <[EMAIL PROTECTED]> writes:
> I've seen cpplib uses that format for a long time now (I think it was
> Neil's work), but I do not seem to see emacs actively take advantage
> of it.
Emacs 22 will. Compilation mode has been largely rewritten.
Andreas.
--
Andreas Schwab, SuSE Lab
On Tue, 22 Feb 2005, Michael Eager wrote:
> The DWARF Workgroup of the Free Standards Group
> (http://www.freestandards.org) is pleased to announce the creation of a
> website for the DWARF debugging format (http://dwarf.freestandards.org).
Thanks for the heads up. I just applied to following p
Marcin Dalecki wrote:
this. Every time I see gcc reporting tens of k errors after discovering a
serious parser error for no good reason running out of every xterm
scroll back
The -Wfatal-errors option will make gcc exit after the first error,
instead of trying to recover and continue. I suppose
I think it is a bug, or a "missing feature".
I tried to simpify the testcase below and ended up with a comlete
different testcase, but it causes the same problem:
it seems to be about FPU registers, if anything causes the compiler to
store the value to memory, it treats it as it would be volatil
> I think the decision to force the
> user to specify -lmudflap should be revisited.
This is already in bugzilla see:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18885
#Michael Cieslinski
This is a target bug as it does not effect any reasonable processor.
With -mfpmath=sse -msse2 I get:
.L2:
decl%eax
addsd %xmm1, %xmm0
jne .L2
-- Pinski
On 2005-02-23, at 20:25, Gabriel Dos Reis wrote:
of source code in the diagnostic. That is based on the GNU standard
for diagnostics. Other people seem to have built tools on top of that
too.
Please note that there exists only a GNU *convention* for error
reporting.
A not particularly well tough
Andrew Pinski schrieb:
This is a target bug as it does not effect any reasonable processor.
With -mfpmath=sse -msse2 I get:
.L2:
decl%eax
addsd %xmm1, %xmm0
jne .L2
my example was about version 3.4.4, which still has this problem with
sse options:
.L5:
movsd -8
The documentation for "-fno-weak" still states that:
Do not use weak symbol support, even if it is provided by the linker.
By default, G++ will use weak symbols if they are available. This
option exists only for testing, and should not be used by end-users;
it will result in inferior code and has
On Wed, Feb 23, 2005 at 02:33:33PM -0700, Davda, Bhavesh P (Bhavesh) wrote:
> The documentation for "-fno-weak" still states that:
>
> Do not use weak symbol support, even if it is provided by the linker.
> By default, G++ will use weak symbols if they are available. This
> option exists only for
Marcin Dalecki <[EMAIL PROTECTED]> writes:
> Please note that there exists only a GNU *convention* for error
> reporting. A not particularly well tough out one. One would for example
> welcome if the error reporting would refer to particular points in the
> ANSI papers by reference like TenDRA an
Well, yes and no.
In general what you state is quite true.
But there are always strange applications, and Avaya has one, where a
final executable is really a "merge" of what used to be independent
processes, with methods that also shared the same name, and data
structures and heaps that shared t
On Wed, Feb 23, 2005 at 02:50:13PM -0700, Davda, Bhavesh P (Bhavesh) wrote:
> Well, yes and no.
>
> In general what you state is quite true.
>
> But there are always strange applications, and Avaya has one, where a
> final executable is really a "merge" of what used to be independent
> processes
>
> If you are munging the symbol table, then why on earth aren't
> you just removing STB_WEAK at the same time? When you abuse
> ELF like this, you've got to be prepared to run into its
> stranger corners; I don't think it's GCC's job to assist with this.
>
> > Is there anything in the C++ s
On Wed, Feb 23, 2005 at 03:00:12PM -0700, Davda, Bhavesh P (Bhavesh) wrote:
> >
> > If you are munging the symbol table, then why on earth aren't
> > you just removing STB_WEAK at the same time? When you abuse
> > ELF like this, you've got to be prepared to run into its
> > stranger corners; I
I would like a clear distinction between correctness tests and optimization
tests. At the moment they are being intermixed, often without comment. :-(
This makes the testsuite somewhat less useful than it should be.
Any suggestions on a good policy for this?
--
This space intentionally left bl
Gabriel Dos Reis wrote:-
> Please, the statement was that EDG does not print expression outside
> declarations. But the fact is it does not just print declarations. It
> prints also statements and expressions part of those statements.
And the fact is it Mark is right - it doesn't. It prints ra
On Wed, 23 Feb 2005, Nathanael Nerode wrote:
> I would like a clear distinction between correctness tests and optimization
> tests. At the moment they are being intermixed, often without comment. :-(
> This makes the testsuite somewhat less useful than it should be.
>
> Any suggestions on a goo
James E Wilson <[EMAIL PROTECTED]> writes:
| Marcin Dalecki wrote:
| > this. Every time I see gcc reporting tens of k errors after discovering a
| > serious parser error for no good reason running out of every xterm
| > scroll back
|
| The -Wfatal-errors option will make gcc exit after the first
Andreas Schwab <[EMAIL PROTECTED]> writes:
| Gabriel Dos Reis <[EMAIL PROTECTED]> writes:
|
| > I've seen cpplib uses that format for a long time now (I think it was
| > Neil's work), but I do not seem to see emacs actively take advantage
| > of it.
|
| Emacs 22 will. Compilation mode has been
Neil Booth <[EMAIL PROTECTED]> writes:
| I also suggest you stop viewing everything as a personal attack.
Thanks for the suggestion, but it would be valuable if its predicate
hold. I do not view anything as a personal attack.
-- Gaby
Snapshot gcc-3.3-20050223 is now available on
ftp://gcc.gnu.org/pub/gcc/snapshots/3.3-20050223/
and on various mirrors, see http://gcc.gnu.org/mirrors.html for details.
This snapshot has been generated from the GCC 3.3 CVS branch
with the following options: -rgcc-ss-3_3-20050223
You'll
Gabriel Dos Reis <[EMAIL PROTECTED]> writes:
> Great. The emacs that comes with the suse I'm running is 21.3.1.
> Do you have emacs 22 in your latest release?
It'll probably take a few months until Emacs 22 is released.
Andreas.
--
Andreas Schwab, SuSE Labs, [EMAIL PROTECTED]
SuSE Linux Produ
Tarun Kawatra wrote:
During expression hash table construction in gcse pass(gcc vercion
3.4.1), expressions like a*b does not get included into the expression
hash table. Such expressions occur in PARALLEL along with clobbers.
You didn't mention the target, or exactly what the mult looks like.
Ho
> Hello
>
> I ponder about writing a "i386 16bit realmode" gcc backend as my master
> thesis - which would be usefull for generating 16-bit bios code needed
> by the virtual machine developed at my university.
>
> Having never programmed any compiler-related stuff and having a strict
> deadline
> Complex. RM places a 64K stack segment limit. As far as I
> know, GCC requires more than that. Also, GCC was only
> written for 32-bit machines. My suggestions:
You're confusing hosts with targets. GCC *runs on* 32 bit (or more)
hosts, but it can *target* smaller machines. We've got lot
On Wed, 23 Feb 2005, Doug Graham wrote:
Regarding memory consumption, perhaps libmudflap's default backtrace
parameter should be set to zero, for both speed and space reasons.
If it's storing all the backtraces that is burning up all the memory,
another approach might be to keep a separate hash tab
Joe Buck <[EMAIL PROTECTED]> writes:
| Also, we maintain a standard of civility on this list. I've been known
| to violate it occasionally, but when I do I promptly apologize. Let's
| try to express our disagreements without treating each other with
| disrespect.
I realize I should have express
> I ponder about writing a "i386 16bit realmode" gcc backend as my master
> thesis - which would be usefull for generating 16-bit bios code needed
> by the virtual machine developed at my university.
I do not know the virtual machine at your university, but there is
two different project you ma
75 matches
Mail list logo
