sablevm-developer Mailing List for SableVM (Page 6)
Brought to you by:
egagnon
Menu
▾
▴
sablevm-developer — SableVM developer mailing list
You can subscribe to this list here.
| 2000 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(27) |
Aug
(22) |
Sep
(1) |
Oct
|
Nov
(1) |
Dec
(30) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2001 |
Jan
(2) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(2) |
Sep
(32) |
Oct
|
Nov
|
Dec
|
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
(69) |
Sep
(10) |
Oct
(31) |
Nov
(15) |
Dec
(58) |
| 2003 |
Jan
(33) |
Feb
(81) |
Mar
(85) |
Apr
(24) |
May
(15) |
Jun
(14) |
Jul
(6) |
Aug
(9) |
Sep
(101) |
Oct
(59) |
Nov
(142) |
Dec
(34) |
| 2004 |
Jan
(107) |
Feb
(164) |
Mar
(181) |
Apr
(96) |
May
(81) |
Jun
(71) |
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: David <db...@cs...> - 2004-05-05 22:47:52
|
On Tue, May 04, 2004 at 10:58:58AM +0200, Fabien Renaud wrote: > Soon, I=B4ll be able to run QT embedded applications but ... I have a=20 > little problem. > I set -p "sablevm.boot.class.path=3Dmy_path" and -c=20 > "my_path_where_there_is_helloworld.class" > There is no problem but it doen=B4t check in=20 > "my_path_where_there_is_helloworld.class" so he cannot find helloworld.= class > If i move my helloworld.class in "my_path" that=B4s works well. >=20 Hi, I cannot see what could be wrong unless the path is not specified correctly. If you don't specify the boot class path and use the default one, do you get the same error? Please copy and paste the full command line to report problems. Sometimes it happens to me what I think is correct but I don't do what I think I am doing and then it doesn't work... David --- David B=E9langer Graduate Student School of Computer Science McGill University Office: MC226 Web page: http://www.cs.mcgill.ca/~dbelan2/ Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt |
|
From: David <db...@cs...> - 2004-05-05 22:44:09
|
On Tue, May 04, 2004 at 03:22:49PM +0200, Fabien Renaud wrote: > Hello, >=20 > I=B4d like to have more informations about how to use qt lib. > I know there is libqtjava but if I want to use awt or swing, how can I=20 > do ? There is an implemtation for kaffe (kaffe-awt-qt) but for sableVM = ? > Can I adapt kaffe-awt-qt easily for sablevm ? > Sorry for all these questions but I don=B4t have IRC for the moment :( >=20 Hi, The interface between SableVM and Classpath is relatively small. SableVM does not really care how the awt is implemented. So, I suspect all the changes you will have to do will be in Classpath. You will have = to replace the current awt-gtk implementation with awt-qt and rebuild Classpath. I don't know what it actually involves. David --- David B=E9langer Graduate Student School of Computer Science McGill University Office: MC226 Web page: http://www.cs.mcgill.ca/~dbelan2/ Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt |
|
From: Etienne G. <gag...@uq...> - 2004-05-05 18:02:21
|
Grzegorz B. Prokopski wrote:
> -lrt (and others of that kind) is on my todo list. I haven't looked at
> M4 yet but I am open for suggestions how it should look like. I have
> a few other small items for M4 ex. the last architecture that still
> bothers me: ARM :-)
To summarize:
* I'd like configure to do something like:
## Pseudo-code!
if (-lrt is required) ## This is a configure test that tries to
## compile a little C program witout -lrt
LIBSVMINCLUDE = -lrt
else
LIBSVMINCLUDE =
endif
Then use @LIBSVMINCLUDE@ in Makefile.am.
* I'd like configure to also do something like:
# Pseudo-code
if ($M4 is not defined)
if (m4 is GNU m4) ## This should actually be a configure test
## that executes m4 and looks for GNU m4 specific
## behavior
M4 = m4
else if (gm4 is GNU m4)
M4 = gm4
endif
endif
Etienne
--
Etienne M. Gagnon, Ph.D. http://www.info.uqam.ca/~egagnon/
SableVM: http://www.sablevm.org/
SableCC: http://www.sablecc.org/
|
|
From: Chris P. <chr...@ma...> - 2004-05-04 16:42:45
|
Chris Pickett wrote:
> As it stands, it's a pretty benign
> patch, and I don't really see any problem with it (otherwise I would not
> have suggested it for staging).
I just found that on tofu, a 2-processor SMP machine, that this change
can cause a NullPointerException if you run a simple Incrementer.java
benchmark enough times (attached to list some time ago). I did not
observe this on other machines.
However, I do not think this is a problem with my change, because all it
does is inflate an already held lock. Please don't interpret this as
arrogance; I honestly don't see why the change itself would be
responsible for this behaviour. All I can think of is that it is
exposing an existing problem with the locking implementation (and indeed
I was seeing several NullPointerException problems earlier on this
year). Putting memory barriers in the JSR-133 cookbook-specified places
does not fix it. Relevant changes from the tag are attached for those
interested without easy repository access (e.g. Clark).
I suppose we probably don't want to commit it now ... but I would
appreciate at some point trying to nail this problem down (possibly
meeting in person).
Cheers,
Chris
Index: src/libsablevm/thread.c
===================================================================
--- src/libsablevm/thread.c (.../sablevm/branches/staging)
(revision 2150)
+++ src/libsablevm/thread.c
(.../developers/chris/tags/inflate_thin_locks) (revision 2150)
@@ -949,6 +949,12 @@
_svmt_JavaVM *vm = env->vm;
_svmt_word old_lockword;
+#ifdef _SABLEVM_FULL_LOCK_INFLATION
+
+ retry:
+
+#endif /* _SABLEVM_FULL_LOCK_INFLATION */
+
assert (instance != NULL);
old_lockword = instance->lockword;
@@ -971,9 +977,39 @@
if (recursive_count == 0)
{
- /* we're releasing the thin lock */
- instance->lockword = _svmf_lockword_get_extra_bits (old_lockword);
- goto handle_contention;
+
+#ifdef _SABLEVM_FULL_LOCK_INFLATION
+
+ /* CJFP
+ if the thin lock being released is under contention,
+ we now inflate it to avoid taking the slow inflation
+ path again.
+ */
+
+ if (env->contention.owner.flag)
+ {
+ if (_svmf_inflate_lock_no_exception (env, instance) != JNI_OK)
+ {
+ _svmf_error_OutOfMemoryError (env);
+ return JNI_ERR;
+ }
+ goto retry;
+ }
+ else
+
+#endif /* _SABLEVM_FULL_LOCK_INFLATION */
+
+ /* EMG
+ we're releasing the thin lock
+
+ CJFP
+ equivalent to line 22 in Onodera's paper
+ */
+
+ {
+ instance->lockword = _svmf_lockword_get_extra_bits
(old_lockword);
+ }
+ goto handle_contention;
}
else
{
|
|
From: Chris P. <chr...@ma...> - 2004-05-04 15:36:46
|
Etienne Gagnon wrote: > Hi, > > Grzegorz B. Prokopski wrote: > >>Secondly, has it been agreed that this stuff is to be merged into >>staging? I'd like to hear from Etienne before we merge. >>(honestly, I wish I could see the diff before asking ;-) > > > I'd like to see a comparison of this proposal with the elimination of > fat locks counter proposal, on performance, before we put this code (or > the counter proposal) in staging. I guess we need some benchmarks before doing that (JGF?), and also a working fat lock elimination implementation. For me this change is interesting mostly from a basic research perspective, just because it brings your algorithm a bit closer to Onodera's. In my experience, it does improve performance, but not all of the time (sometimes it decreases performance), and the gains are not terribly significant. Processor configuration has an impact, and presumably so does kernel version. The other (more involved) change that would make them really comparable would be to implement fat lock deflation. I expect that deflation would also enhance the performance benefit of this tiny "full inflation" change; obviously if you inflate locks more aggressively, while this saves you from taking the expensive inflation path some of the time, you do end up with more fat locks, and so you only benefit if these locks are under a lot of contention. If you only want to commit this once deflation is implemented (assuming that happens), that's fine with me. As it stands, it's a pretty benign patch, and I don't really see any problem with it (otherwise I would not have suggested it for staging). I guess I'm in favour of multiple different algorithms being selectable at compile-time in SableVM, just because I think that's neat and because it shows off the suitability of the VM for doing research. I do think that a paper comparing 1) Onodera's original algorithm (if SableVM can be modified to use it) 2) your original algorithm 3) your original algorithm with full lock inflation + deflation (to bring it closer to Onodera's) 4) this new "fat lock elimination" algorithm (Greg talked to me briefly about it, but I'm not clear on the details) would be interesting. Cheers, Chris |
|
From: Chris P. <chr...@ma...> - 2004-05-04 15:09:36
|
Grzegorz B. Prokopski wrote: > On (30/04/04 16:22), Chris Pickett wrote: > >>Hi, >> >>As an experiment, I am preparing a tag for Greg or David to >>merge into staging. It will contain a compile-time option to enable >>inflating all contended thin locks (currently this is not done; see back >>in the list about how in some situations it might give you a small >>benefit). The option is called --enable-full-lock-inflation. > > > Firstly, how do I get the diff of your changes? I tried: > > svn diff > svn+ssh://svn.sablevm.org/public/developers/chris/tags/inflate_thin_locks > -r 2146 svn+ssh://svn.sablevm.org/public/sablevm/branches/staging |less > > to compare your tag to the last revision of sablevm-staging you've > merged but it seems to be showing lots of other stuff too, like > lt_dlopen stuff Etienne merged lately from Melanie's sandbox, > and I can't seem to find *your* changes. Hi Greg, One way that works to get a diff is this: $ svn log $CHRIS/tags ------------------------------------------------------------------------ r2150 | chris | 2004-04-30 19:07:22 -0400 (Fri, 30 Apr 2004) | 2 lines Create tag with --enable-full-lock-inflation for merging into staging. ------------------------------------------------------------------------ r2148 | chris | 2004-04-30 18:41:45 -0400 (Fri, 30 Apr 2004) | 2 lines Destroy bad tag. ------------------------------------------------------------------------ r2139 | chris | 2004-04-29 20:08:04 -0400 (Thu, 29 Apr 2004) | 2 lines Create staging branch to prepare inflate_thin_locks tag. ------------------------------------------------------------------------ r820 | egagnon | 2003-10-23 20:10:05 -0400 (Thu, 23 Oct 2003) | 2 lines Add chris developer. ------------------------------------------------------------------------ $ svn diff $STAGING@2150 $CHRIS/tags/inflate_thin_locks@2150 (i find it's very helpful to set environment variables for commonly used directories in your .bashrc or .profile) Possibly there is a better way to find the last changed revision. A checkout followed by svn info will also tell you. To merge them: $ cd staging $ svn merge $STAGING@2150 $CHRIS/tags/inflate_thin_locks@2150 $ svn diff [ output should be the same as previous diff, except for the headers ] Cheers, Chris |
|
From: Fabien R. <re...@ne...> - 2004-05-04 13:22:53
|
Hello, I´d like to have more informations about how to use qt lib. I know there is libqtjava but if I want to use awt or swing, how can I do ? There is an implemtation for kaffe (kaffe-awt-qt) but for sableVM ? Can I adapt kaffe-awt-qt easily for sablevm ? Sorry for all these questions but I don´t have IRC for the moment :( Cheers, Fabien |
|
From: Fabien R. <re...@ne...> - 2004-05-04 08:58:58
|
Soon, I´ll be able to run QT embedded applications but ... I have a little problem. I set -p "sablevm.boot.class.path=my_path" and -c "my_path_where_there_is_helloworld.class" There is no problem but it doen´t check in "my_path_where_there_is_helloworld.class" so he cannot find helloworld.class If i move my helloworld.class in "my_path" that´s works well. I´m sure that´s a newbie problem :( Cheers, Fabien |
|
From: Grzegorz B. P. <ga...@de...> - 2004-05-04 07:12:08
|
On (04/05/04 02:42), Etienne Gagnon wrote: > Grzegorz B. Prokopski wrote: > >would it be possible to merge latest GNU CP CVS into our > >staging (or at least vendor branch)? > > Done. > > Ideally, I'd like to see the $(M4) and the -lrt problems solved before > we release 1.1.4. I won't have time to solve them, so have fun doing > it. :-) -lrt (and others of that kind) is on my todo list. I haven't looked at M4 yet but I am open for suggestions how it should look like. I have a few other small items for M4 ex. the last architecture that still bothers me: ARM :-) GBP -- Grzegorz B. Prokopski <ga...@de...> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org Why SableVM ?!? http://devel.sablevm.org/wiki/WhySableVM |
|
From: Etienne G. <gag...@uq...> - 2004-05-04 06:48:08
|
Grzegorz B. Prokopski wrote: > would it be possible to merge latest GNU CP CVS into our > staging (or at least vendor branch)? Done. Ideally, I'd like to see the $(M4) and the -lrt problems solved before we release 1.1.4. I won't have time to solve them, so have fun doing it. :-) Etienne -- Etienne M. Gagnon, Ph.D. http://www.info.uqam.ca/~egagnon/ SableVM: http://www.sablevm.org/ SableCC: http://www.sablecc.org/ |
|
From: Etienne G. <gag...@uq...> - 2004-05-04 06:30:16
|
Hi, Grzegorz B. Prokopski wrote: > Secondly, has it been agreed that this stuff is to be merged into > staging? I'd like to hear from Etienne before we merge. > (honestly, I wish I could see the diff before asking ;-) I'd like to see a comparison of this proposal with the elimination of fat locks counter proposal, on performance, before we put this code (or the counter proposal) in staging. Etienne -- Etienne M. Gagnon, Ph.D. http://www.info.uqam.ca/~egagnon/ SableVM: http://www.sablevm.org/ SableCC: http://www.sablecc.org/ |
|
From: Grzegorz B. P. <ga...@de...> - 2004-05-04 06:14:16
|
On (30/04/04 16:22), Chris Pickett wrote: > Hi, > > As an experiment, I am preparing a tag for Greg or David to > merge into staging. It will contain a compile-time option to enable > inflating all contended thin locks (currently this is not done; see back > in the list about how in some situations it might give you a small > benefit). The option is called --enable-full-lock-inflation. Firstly, how do I get the diff of your changes? I tried: svn diff svn+ssh://svn.sablevm.org/public/developers/chris/tags/inflate_thin_locks -r 2146 svn+ssh://svn.sablevm.org/public/sablevm/branches/staging |less to compare your tag to the last revision of sablevm-staging you've merged but it seems to be showing lots of other stuff too, like lt_dlopen stuff Etienne merged lately from Melanie's sandbox, and I can't seem to find *your* changes. Secondly, has it been agreed that this stuff is to be merged into staging? I'd like to hear from Etienne before we merge. (honestly, I wish I could see the diff before asking ;-) Cheers, GBP -- Grzegorz B. Prokopski <ga...@de...> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org Why SableVM ?!? http://devel.sablevm.org/wiki/WhySableVM |
|
From: Grzegorz B. P. <ga...@de...> - 2004-05-04 06:11:11
|
Hi Etienne, would it be possible to merge latest GNU CP CVS into our staging (or at least vendor branch)? I'd like to see 1.1.4 soon, maybe this weekend even. GBP -- Grzegorz B. Prokopski <ga...@de...> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org Why SableVM ?!? http://devel.sablevm.org/wiki/WhySableVM |
|
From: Etienne G. <gag...@uq...> - 2004-05-04 03:13:04
|
Hi David,
David B=E9langer wrote:
> I don't know how it is on OpenBSD but with FreeBSD, I have to use
> gm4.
It's the same.
> In configure.ac in my sandbox, I have this line:
> AC_CHECK_PROGS([M4],[gm4 m4],[none],[$PATH])
> It does not check the GNUness of the m4 program but it will used gm4 if=
> found, otherwise check for m4. If neither gm4 nor m4 is found, will se=
t
> M4 to none. I guess ideally would need some if stmt for the none case.=
We need 2 things:
1- Check for GNU M4's name and set $(M4) to it. If it's not found,
configure should fail. I guess, configure should also be able
to accept a pre-set $(M4) environment variable as override.
2- Do something similar with "-lrt", which is giving me a hell of a
time. Configure should figure by itself whether it is needed or
not.
Any idea how tom implement these? :-)
Etienne
--=20
Etienne M. Gagnon, Ph.D. http://www.info.uqam.ca/~egagnon/
SableVM: http://www.sablevm.org/
SableCC: http://www.sablecc.org/
|
|
From: David <db...@cs...> - 2004-05-03 21:33:42
|
Hi Etienne,
I noticed that you replaced the m4 command with a ${M4} variable.
However it is currently defined in Makefile.am.
I don't know how it is on OpenBSD but with FreeBSD, I have to use
gm4. In configure.ac in my sandbox, I have this line:
AC_CHECK_PROGS([M4],[gm4 m4],[none],[$PATH])
It does not check the GNUness of the m4 program but it will used gm4 if
found, otherwise check for m4. If neither gm4 nor m4 is found, will set
M4 to none. I guess ideally would need some if stmt for the none case.
What do you think of adding this to staging?
David
---
David B=E9langer
Graduate Student
School of Computer Science
McGill University
Office: MC226
Web page: http://www.cs.mcgill.ca/~dbelan2/
Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt
|
|
From: Fabien R. <re...@ne...> - 2004-05-03 08:22:21
|
I link /usr/lib/libltdl.so on my good lib ( I know that´s not really good)
Now I have some libtool warnings ( libxxxx was moved ) and finally :
sablevm.o(.text+0x5c): In function `displayFileContent':
: undefined reference to `rpl_malloc'
sablevm.o(.text+0x4cc): In function `getLine':
: undefined reference to `rpl_malloc'
sablevm.o(.text+0x10d4): In function `getProperties':
: undefined reference to `rpl_malloc'
sablevm.o(.text+0x11b4): In function `getProperties':
: undefined reference to `rpl_malloc'
sablevm.o(.text+0x11e8): In function `getProperties':
: undefined reference to `rpl_malloc'
sablevm.o(.text+0x14e8): more undefined references to `rpl_malloc' follow
collect2: ld returned 1 exit status
Fabien Renaud wrote:
> Ok actually the lib was not installed ( I only had libffi.la and
> thought that was ok )
> I set prefix to another directory than
> /home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin and now it-s installed.
> I onnly have to set the differents paths but that should be ok.
>
> Thanks you Chris & Grzegorz
>
> Cheers,
>
> Fabien
>
>
> Chris Pickett wrote:
>
>> Fabien Renaud wrote:
>>
>>> Ok set -v ^^
>>>
>>> Well I get a libffi from gcc 3.4 tarball
>>> I apply this patch :
>>> libffi/configure.in~ 2003-10-22 01:02:59.000000000 +0200
>>> +++ libffi/configure.in 2003-11-11 12:33:40.000000000 +0100
>>> @@ -29,6 +29,7 @@
>>> else
>>> libffi_basedir=
>>> fi
>>> +libffi_basedir=
>>> AC_SUBST(libffi_basedir)
>>> AC_CONFIG_AUX_DIR(${libffi_basedir}..)
>>>
>>> Then I mkdir libffi_bin at the top of gcc
>>> cd libffi_bin ;
>>> ../libffi/configure
>>> --prefix=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>> --host=arm-linux --build=i386
>>> make
>>> export LDFLAGS= -L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>> export LDFLAGS= -L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin/include
>>> export LD_LIBRARY_PATH=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>
>>
>>
>> Please try
>>
>> export LDFLAGS=-L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>> export CFLAGS=-I/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin/include
>> export LD_LIBRARY_PATH=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>
>> which is the closest thing to what's written in the wiki and see if
>> that helps. If that doesn't work, please try something like
>>
>> cd libffi_bin
>> ../libffi/configure --prefix=${HOME} --host=arm-linux --build=i386
>> make
>> make install
>> export LDFLAGS=-L/home/commgw/renaud/lib
>> export CFLAGS=-I/home/commgw/renaud/include
>> export LD_LIBRARY_PATH=/home/commgw/renaud/lib
>>
>> Cheers,
>> Chris
>>
>>
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: Oracle 10g
> Get certified on the hottest thing ever to hit the market... Oracle
> 10g. Take an Oracle 10g class now, and we'll give you the exam FREE.
> http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
> _______________________________________________
> Sablevm-developer mailing list
> Sab...@li...
> https://lists.sourceforge.net/lists/listinfo/sablevm-developer
>
>
|
|
From: Fabien R. <re...@ne...> - 2004-05-03 08:08:13
|
I have now another problem.
I set all my flags but the make wants to take a lib in /usr/lib ( it´s
x86 lib). I´ve put the lib into libffi directory ( no problem with libffi )
The LD_LIBRARY_PATH ans LDFLAGS points on this directory.
Can I exclude /usr/lib ?
Cheers,
Fabien
Fabien Renaud wrote:
> Ok actually the lib was not installed ( I only had libffi.la and
> thought that was ok )
> I set prefix to another directory than
> /home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin and now it-s installed.
> I onnly have to set the differents paths but that should be ok.
>
> Thanks you Chris & Grzegorz
>
> Cheers,
>
> Fabien
>
>
> Chris Pickett wrote:
>
>> Fabien Renaud wrote:
>>
>>> Ok set -v ^^
>>>
>>> Well I get a libffi from gcc 3.4 tarball
>>> I apply this patch :
>>> libffi/configure.in~ 2003-10-22 01:02:59.000000000 +0200
>>> +++ libffi/configure.in 2003-11-11 12:33:40.000000000 +0100
>>> @@ -29,6 +29,7 @@
>>> else
>>> libffi_basedir=
>>> fi
>>> +libffi_basedir=
>>> AC_SUBST(libffi_basedir)
>>> AC_CONFIG_AUX_DIR(${libffi_basedir}..)
>>>
>>> Then I mkdir libffi_bin at the top of gcc
>>> cd libffi_bin ;
>>> ../libffi/configure
>>> --prefix=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>> --host=arm-linux --build=i386
>>> make
>>> export LDFLAGS= -L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>> export LDFLAGS= -L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin/include
>>> export LD_LIBRARY_PATH=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>
>>
>>
>> Please try
>>
>> export LDFLAGS=-L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>> export CFLAGS=-I/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin/include
>> export LD_LIBRARY_PATH=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>>
>> which is the closest thing to what's written in the wiki and see if
>> that helps. If that doesn't work, please try something like
>>
>> cd libffi_bin
>> ../libffi/configure --prefix=${HOME} --host=arm-linux --build=i386
>> make
>> make install
>> export LDFLAGS=-L/home/commgw/renaud/lib
>> export CFLAGS=-I/home/commgw/renaud/include
>> export LD_LIBRARY_PATH=/home/commgw/renaud/lib
>>
>> Cheers,
>> Chris
>>
>>
>
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by: Oracle 10g
> Get certified on the hottest thing ever to hit the market... Oracle
> 10g. Take an Oracle 10g class now, and we'll give you the exam FREE.
> http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
> _______________________________________________
> Sablevm-developer mailing list
> Sab...@li...
> https://lists.sourceforge.net/lists/listinfo/sablevm-developer
>
>
|
|
From: Fabien R. <re...@ne...> - 2004-05-03 07:10:17
|
Ok actually the lib was not installed ( I only had libffi.la and thought
that was ok )
I set prefix to another directory than
/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin and now it-s installed.
I onnly have to set the differents paths but that should be ok.
Thanks you Chris & Grzegorz
Cheers,
Fabien
Chris Pickett wrote:
> Fabien Renaud wrote:
>
>> Ok set -v ^^
>>
>> Well I get a libffi from gcc 3.4 tarball
>> I apply this patch :
>> libffi/configure.in~ 2003-10-22 01:02:59.000000000 +0200
>> +++ libffi/configure.in 2003-11-11 12:33:40.000000000 +0100
>> @@ -29,6 +29,7 @@
>> else
>> libffi_basedir=
>> fi
>> +libffi_basedir=
>> AC_SUBST(libffi_basedir)
>> AC_CONFIG_AUX_DIR(${libffi_basedir}..)
>>
>> Then I mkdir libffi_bin at the top of gcc
>> cd libffi_bin ;
>> ../libffi/configure
>> --prefix=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>> --host=arm-linux --build=i386
>> make
>> export LDFLAGS= -L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>> export LDFLAGS= -L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin/include
>> export LD_LIBRARY_PATH=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>
>
> Please try
>
> export LDFLAGS=-L/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
> export CFLAGS=-I/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin/include
> export LD_LIBRARY_PATH=/home/commgw/renaud/gcc/gcc-3.4.0/libffi_bin
>
> which is the closest thing to what's written in the wiki and see if
> that helps. If that doesn't work, please try something like
>
> cd libffi_bin
> ../libffi/configure --prefix=${HOME} --host=arm-linux --build=i386
> make
> make install
> export LDFLAGS=-L/home/commgw/renaud/lib
> export CFLAGS=-I/home/commgw/renaud/include
> export LD_LIBRARY_PATH=/home/commgw/renaud/lib
>
> Cheers,
> Chris
>
>
|
|
From: Etienne G. <gag...@uq...> - 2004-05-02 14:12:27
|
The email registered in mailman is: jewel at users.sourceforge.net. I did give you special access to be able to post from your pixie address. :-) Etienne John Leuner wrote: > I tried to view my mailing list subscription options at > https://lists.sourceforge.net/lists/listinfo/sablevm-developer > > but it always tells me that my address is not subscribed. I wonder if I > could be subscribed under another address? Looking at these headers it > seems that my address is correct. > > Has anybody experienced similar problems with the SF mailing lists? > > John Leuner > > > Return-path: <sab...@li...> > Envelope-to: jewel@localhost > Received: from cmalu ([127.0.0.1] helo=localhost) by cmalu with esmtp > (Exim > 3.36 #1 (Debian)) id 1BK2Gz-0006I1-00 for <jewel@localhost>; > Sat, 01 May > 2004 23:43:49 +0200 > Received: from pixie.co.za [196.2.48.239] by localhost with POP3 > (fetchmail-5.9.11) for jewel@localhost (single-drop); Sat, 01 > May 2004 > 23:43:49 +0200 (SAST) > Received: from postwall04.mweb.co.za by cpt-mailhost2.mweb.co.za (Sun > Internet Mail Server sims.3.5.2000.03.23.18.03.p10) with ESMTP > id > <0HX...@cp...> for > jl25@sims-ms-daemon; Sat, 1 > May 2004 23:43:19 +0200 (GMT-2) > Received: from viruswall04.mweb.co.za (viruswall04.mweb.co.za > [196.2.50.230]) by pwfilter02.mweb.co.za (Postfix) with ESMTP id > D939FB5F > for <je...@pi...>; Sat, 01 May 2004 23:43:19 +0200 (SAST) > Received: from postwall07.mweb.co.za (localhost.localdomain [127.0.0.1]) > by > localhost.localdomain (VaMailArmor-2.0.2-4) id 08225-29B65A8D; > Sat, 01 May > 2004 23:43:19 +0200 > Received: from sc8-sf-mx2.sourceforge.net (lists.sourceforge.net > [66.35.250.206]) by postwall07.mweb.co.za (Postfix) with ESMTP > id AB0FF2B79 > for <je...@pi...>; Sat, 01 May 2004 23:43:18 +0200 (SAST) > Received: from sc8-sf-list1-b.sourceforge.net ([10.3.1.7] > helo=sc8-sf-list1.sourceforge.net) by sc8-sf-mx2.sourceforge.net > with esmtp > (TLSv1:AES256-SHA:256) (Exim 4.30) id 1BK2GP-0001sR-I9; Sat, 01 > May 2004 > 14:43:13 -0700 > Received: from localhost ([127.0.0.1] helo=projects.sourceforge.net) by > sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id > 1BK2GF-0003GC-Ou; > Sat, 01 May 2004 14:43:03 -0700 > Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] > helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net > with esmtp > (Exim 4.30) id 1BK2Fy-0003Ci-ON for > sab...@li...; Sat, 01 May 2004 > 14:42:46 -0700 > Received: from anis.telecom.uqam.ca ([132.208.250.6]) by > sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.30) id > 1BK2Fy-0000nJ-2j for > sab...@li...; Sat, 01 May 2004 > 14:42:46 -0700 > Received: from anis4.telecom.uqam.ca (anis4.telecom.uqam.ca > [132.208.250.236]) by sortant.uqam.ca (8.12.10/8.12.1) with SMTP > id > i41LUAUf024998 for <sab...@li...>; > Sat, 01 May > 2004 17:30:25 -0400 (EDT) > Received: from antivirus.uqam.ca ([132.208.250.6]) by > anis4.telecom.uqam.ca > (SAVSMTP 3.1.1.32) with SMTP id M2004050117315007654 for > <sab...@li...>; Sat, 01 May 2004 > 17:31:50 -0400 > Received: from perseval.info.uqam.ca (perseval.info.uqam.ca > [132.208.137.96]) by intrant.uqam.ca > (8.12.10/8.12.2/uqam-filtres) with > ESMTP id i41LQZY3019106 for > <sab...@li...>; Sat, > 01 May 2004 17:26:35 -0400 (EDT) > Received: from localhost ([127.0.0.1] helo=uqam.ca) by > perseval.info.uqam.ca with esmtp (Exim 4.32) id 1BK20J-0004hI-JW > for > sab...@li...; Sat, 01 May 2004 > 17:26:35 -0400 > Date: Sat, 01 May 2004 17:26:27 -0400 > From: Etienne Gagnon <gag...@uq...> > Subject: [Sablevm-developer] lt_dlopen(NULL) dependency eliminated > Sender: sab...@li... > To: SableVM-devel ML <sab...@li...> > Errors-to: sab...@li... > Message-id: <409...@uq...> > Organization: UQAM > MIME-version: 1.0 > Content-type: text/plain; charset=us-ascii; format=flowed > Content-transfer-encoding: 7bit > Precedence: bulk > X-Spam-Filter: Filtre-Uqam re: ab...@uq... > User-Agent: Mozilla/5.0 (X11; U; Linux ppc; en-US; rv:1.6) > Gecko/20040414 > Debian/1.6-5 > X-Accept-Language: en > X-Spam-Score: 0.0 (/) > X-Spam-Score: 0.0 (/) > X-Spam-Report: Spam Filtering performed by sourceforge.net. See > http://spamassassin.org/tag/ for more details. Report problems > to > http://sf.net/tracker/?func=add&group_id=1&atid=200001 > X-Spam-Report: Spam Filtering performed by sourceforge.net. See > http://spamassassin.org/tag/ for more details. Report problems > to > http://sf.net/tracker/?func=add&group_id=1&atid=200001 > X-BeenThere: sab...@li... > X-Mailman-Version: 2.0.9-sf.net > List-Unsubscribe: > <https://lists.sourceforge.net/lists/listinfo/sablevm-developer>, > <mailto:sab...@li...?subject=unsubscribe> > List-Id: SableVM Developer Mailing List > <sablevm-developer.lists.sourceforge.net> > List-Post: <mailto:sab...@li...> > List-Help: > <mailto:sab...@li...?subject=help> > List-Subscribe: > <https://lists.sourceforge.net/lists/listinfo/sablevm-developer>, > <mailto:sab...@li...?subject=subscribe> > List-Archive: > <http://sourceforge.net/mailarchive/forum.php?forum=sablevm-developer> > X-Original-Date: Sat, 01 May 2004 17:26:27 -0400 > X-AntiVirus: checked by Vexira MailArmor (version: 2.0.2-4; VAE: > 6.25.0.3; > VDF: 6.25.0.42; host: viruswall04.mweb.co.za) > X-Evolution-Source: mbox:/var/mail/jewel > > Hi all. > > I have checked in staging my adaptation of Melanie's code to eliminate > dependencies on a working lt_dlopen(NULL)/lt_dlsym() pair. > > Hopefully, this change will greatly simplify porting SableVM to new > platforms. Why not a "real Windows" port, soon, using Microsoft > compiler? > > In order for this code to work, it is *imperative* that all internal > native methods (e.g. Java_*) to be defined as: > > JNIEXPORT ... > Java_... (...) > { > ... > > Where "Java_" starts on the first character of a line, and the "(" of > formal parameters is on the same line. If you do not follow this > requirement, you method won't be found at runtime... You've been > warned;-) > > Anyway, this should be true of *all* functions in SableVM. You're > welcome to fix any function that does not abide by this rule. :-) > > Etienne > -- Etienne M. Gagnon, Ph.D. http://www.info.uqam.ca/~egagnon/ SableVM: http://www.sablevm.org/ SableCC: http://www.sablecc.org/ |
|
From: John L. <je...@pi...> - 2004-05-02 12:20:50
|
On Sat, 2004-05-01 at 23:26, Etienne Gagnon wrote:
> Hi all.
>
> I have checked in staging my adaptation of Melanie's code to eliminate
> dependencies on a working lt_dlopen(NULL)/lt_dlsym() pair.
>
> Hopefully, this change will greatly simplify porting SableVM to new
> platforms. Why not a "real Windows" port, soon, using Microsoft compiler?
>
> In order for this code to work, it is *imperative* that all internal
> native methods (e.g. Java_*) to be defined as:
>
> JNIEXPORT ...
> Java_... (...)
> {
> ...
>
> Where "Java_" starts on the first character of a line, and the "(" of
> formal parameters is on the same line. If you do not follow this
> requirement, you method won't be found at runtime... You've been warned;-)
>
> Anyway, this should be true of *all* functions in SableVM. You're
> welcome to fix any function that does not abide by this rule. :-)
>
> Etienne
--
John Leuner <je...@pi...>
|
|
From: John L. <je...@pi...> - 2004-05-02 12:04:44
|
I tried to view my mailing list subscription options at https://lists.sourceforge.net/lists/listinfo/sablevm-developer but it always tells me that my address is not subscribed. I wonder if I could be subscribed under another address? Looking at these headers it seems that my address is correct. Has anybody experienced similar problems with the SF mailing lists? John Leuner Return-path: <sab...@li...> Envelope-to: jewel@localhost Received: from cmalu ([127.0.0.1] helo=localhost) by cmalu with esmtp (Exim 3.36 #1 (Debian)) id 1BK2Gz-0006I1-00 for <jewel@localhost>; Sat, 01 May 2004 23:43:49 +0200 Received: from pixie.co.za [196.2.48.239] by localhost with POP3 (fetchmail-5.9.11) for jewel@localhost (single-drop); Sat, 01 May 2004 23:43:49 +0200 (SAST) Received: from postwall04.mweb.co.za by cpt-mailhost2.mweb.co.za (Sun Internet Mail Server sims.3.5.2000.03.23.18.03.p10) with ESMTP id <0HX...@cp...> for jl25@sims-ms-daemon; Sat, 1 May 2004 23:43:19 +0200 (GMT-2) Received: from viruswall04.mweb.co.za (viruswall04.mweb.co.za [196.2.50.230]) by pwfilter02.mweb.co.za (Postfix) with ESMTP id D939FB5F for <je...@pi...>; Sat, 01 May 2004 23:43:19 +0200 (SAST) Received: from postwall07.mweb.co.za (localhost.localdomain [127.0.0.1]) by localhost.localdomain (VaMailArmor-2.0.2-4) id 08225-29B65A8D; Sat, 01 May 2004 23:43:19 +0200 Received: from sc8-sf-mx2.sourceforge.net (lists.sourceforge.net [66.35.250.206]) by postwall07.mweb.co.za (Postfix) with ESMTP id AB0FF2B79 for <je...@pi...>; Sat, 01 May 2004 23:43:18 +0200 (SAST) Received: from sc8-sf-list1-b.sourceforge.net ([10.3.1.7] helo=sc8-sf-list1.sourceforge.net) by sc8-sf-mx2.sourceforge.net with esmtp (TLSv1:AES256-SHA:256) (Exim 4.30) id 1BK2GP-0001sR-I9; Sat, 01 May 2004 14:43:13 -0700 Received: from localhost ([127.0.0.1] helo=projects.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1BK2GF-0003GC-Ou; Sat, 01 May 2004 14:43:03 -0700 Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Exim 4.30) id 1BK2Fy-0003Ci-ON for sab...@li...; Sat, 01 May 2004 14:42:46 -0700 Received: from anis.telecom.uqam.ca ([132.208.250.6]) by sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.30) id 1BK2Fy-0000nJ-2j for sab...@li...; Sat, 01 May 2004 14:42:46 -0700 Received: from anis4.telecom.uqam.ca (anis4.telecom.uqam.ca [132.208.250.236]) by sortant.uqam.ca (8.12.10/8.12.1) with SMTP id i41LUAUf024998 for <sab...@li...>; Sat, 01 May 2004 17:30:25 -0400 (EDT) Received: from antivirus.uqam.ca ([132.208.250.6]) by anis4.telecom.uqam.ca (SAVSMTP 3.1.1.32) with SMTP id M2004050117315007654 for <sab...@li...>; Sat, 01 May 2004 17:31:50 -0400 Received: from perseval.info.uqam.ca (perseval.info.uqam.ca [132.208.137.96]) by intrant.uqam.ca (8.12.10/8.12.2/uqam-filtres) with ESMTP id i41LQZY3019106 for <sab...@li...>; Sat, 01 May 2004 17:26:35 -0400 (EDT) Received: from localhost ([127.0.0.1] helo=uqam.ca) by perseval.info.uqam.ca with esmtp (Exim 4.32) id 1BK20J-0004hI-JW for sab...@li...; Sat, 01 May 2004 17:26:35 -0400 Date: Sat, 01 May 2004 17:26:27 -0400 From: Etienne Gagnon <gag...@uq...> Subject: [Sablevm-developer] lt_dlopen(NULL) dependency eliminated Sender: sab...@li... To: SableVM-devel ML <sab...@li...> Errors-to: sab...@li... Message-id: <409...@uq...> Organization: UQAM MIME-version: 1.0 Content-type: text/plain; charset=us-ascii; format=flowed Content-transfer-encoding: 7bit Precedence: bulk X-Spam-Filter: Filtre-Uqam re: ab...@uq... User-Agent: Mozilla/5.0 (X11; U; Linux ppc; en-US; rv:1.6) Gecko/20040414 Debian/1.6-5 X-Accept-Language: en X-Spam-Score: 0.0 (/) X-Spam-Score: 0.0 (/) X-Spam-Report: Spam Filtering performed by sourceforge.net. See http://spamassassin.org/tag/ for more details. Report problems to http://sf.net/tracker/?func=add&group_id=1&atid=200001 X-Spam-Report: Spam Filtering performed by sourceforge.net. See http://spamassassin.org/tag/ for more details. Report problems to http://sf.net/tracker/?func=add&group_id=1&atid=200001 X-BeenThere: sab...@li... X-Mailman-Version: 2.0.9-sf.net List-Unsubscribe: <https://lists.sourceforge.net/lists/listinfo/sablevm-developer>, <mailto:sab...@li...?subject=unsubscribe> List-Id: SableVM Developer Mailing List <sablevm-developer.lists.sourceforge.net> List-Post: <mailto:sab...@li...> List-Help: <mailto:sab...@li...?subject=help> List-Subscribe: <https://lists.sourceforge.net/lists/listinfo/sablevm-developer>, <mailto:sab...@li...?subject=subscribe> List-Archive: <http://sourceforge.net/mailarchive/forum.php?forum=sablevm-developer> X-Original-Date: Sat, 01 May 2004 17:26:27 -0400 X-AntiVirus: checked by Vexira MailArmor (version: 2.0.2-4; VAE: 6.25.0.3; VDF: 6.25.0.42; host: viruswall04.mweb.co.za) X-Evolution-Source: mbox:/var/mail/jewel Hi all. I have checked in staging my adaptation of Melanie's code to eliminate dependencies on a working lt_dlopen(NULL)/lt_dlsym() pair. Hopefully, this change will greatly simplify porting SableVM to new platforms. Why not a "real Windows" port, soon, using Microsoft compiler? In order for this code to work, it is *imperative* that all internal native methods (e.g. Java_*) to be defined as: JNIEXPORT ... Java_... (...) { ... Where "Java_" starts on the first character of a line, and the "(" of formal parameters is on the same line. If you do not follow this requirement, you method won't be found at runtime... You've been warned;-) Anyway, this should be true of *all* functions in SableVM. You're welcome to fix any function that does not abide by this rule. :-) Etienne -- Etienne M. Gagnon, Ph.D. http://www.info.uqam.ca/~egagnon/ SableVM: http://www.sablevm.org/ SableCC: http://www.sablecc.org/ ------------------------------------------------------- This SF.Net email is sponsored by: Oracle 10g Get certified on the hottest thing ever to hit the market... Oracle 10g. Take an Oracle 10g class now, and we'll give you the exam FREE. http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click _______________________________________________ Sablevm-developer mailing list Sab...@li... https://lists.sourceforge.net/lists/listinfo/sablevm-developer -- John Leuner <je...@pi...> |
|
From: Andreas J. <aj...@an...> - 2004-05-02 07:38:11
|
On 04-May-01 21:17, Grzegorz B. Prokopski wrote: > Just one thing: are we talking about x86-64 in 64bit mode? Yes, we are talking about x86_64 in 64bit native mode. > If yes, then did you have any other problems w/ porting? I needed a small patch to build the sablevm-classlib package for the amd64 architecture (see below). A few other packages needed similar small patches. I did not find any serious obstacles for a x86_64 native 64bit port. > If no, then what would be needed to make the 64bit port? > ex. is there libffi for x86-64 in 64bit mode? There is an almost complete port of the Debian system to x86_64 in native 64bit mode (including a libffi). However, there is no 'official' archive for this port yet. I hope this will change soon. Regards Andreas Jochens diff -urN ../tmp-orig/sablevm-classlib-1.1.3/native/fdlibm/ieeefp.h ./native/fdlibm/ieeefp.h --- ../tmp-orig/sablevm-classlib-1.1.3/native/fdlibm/ieeefp.h 2004-04-12 20:33:25.000000000 +0000 +++ ./native/fdlibm/ieeefp.h 2004-05-01 22:05:54.104662224 +0000 @@ -59,6 +59,10 @@ #define __IEEE_LITTLE_ENDIAN #endif +#ifdef __x86_64__ +#define __IEEE_LITTLE_ENDIAN +#endif + #ifdef __i960__ #define __IEEE_LITTLE_ENDIAN #endif |
|
From: Grzegorz B. P. <ga...@de...> - 2004-05-02 01:38:10
|
On (02/05/04 00:19), Andreas Jochens wrote: > tags 245834 + patch > > On 04-Apr-25 19:42, Kurt Roeckx wrote: > > It currently doesn't seem to have support for amd64. Adding > > support for it seems mostly adding || defined(__x86_64__) where > > __i386__ is used, except for system.h where the word/allignment > > needs to be chagned from 32bit/4byte to 64bit/8byte. > > > > I just don't know enough assembler/hardware details for the > > things in system.c > > The attached patch seems to work for me. > Thanks. Just one thing: are we talking about x86-64 in 64bit mode? If yes, then did you have any other problems w/ porting? If no, then what would be needed to make the 64bit port? ex. is there libffi for x86-64 in 64bit mode? GBP PS: And how do I distinguish between x86-64 in 32 and 64bit mode from within C program source? -- Grzegorz B. Prokopski <ga...@de...> Debian GNU/Linux http://www.debian.org SableVM - LGPLed JVM http://www.sablevm.org Why SableVM ?!? http://devel.sablevm.org/wiki/WhySableVM |
|
From: David <db...@cs...> - 2004-05-02 01:31:05
|
Hello, CP now has the Process/VMProcess implementation done by Archie. Currently, our implementation is dead code. So, unless there is any specific issue, I think we should adopt CP version. I will proceed to delete our implementation some time in the next few days. David --- David B=E9langer Graduate Student School of Computer Science McGill University Office: MC226 Web page: http://www.cs.mcgill.ca/~dbelan2/ Public key: http://www.cs.mcgill.ca/~dbelan2/public_key.txt |
|
From: Andreas J. <aj...@an...> - 2004-05-01 22:38:10
|
tags 245834 + patch
On 04-Apr-25 19:42, Kurt Roeckx wrote:
> It currently doesn't seem to have support for amd64. Adding
> support for it seems mostly adding || defined(__x86_64__) where
> __i386__ is used, except for system.h where the word/allignment
> needs to be chagned from 32bit/4byte to 64bit/8byte.
>
> I just don't know enough assembler/hardware details for the
> things in system.c
The attached patch seems to work for me.
Regards
Andreas Jochens
diff -urN ../tmp-orig/sablevm-1.1.3/src/libsablevm/include/jni_system_specific.h ./src/libsablevm/include/jni_system_specific.h
--- ../tmp-orig/sablevm-1.1.3/src/libsablevm/include/jni_system_specific.h 2004-04-12 20:29:07.000000000 +0000
+++ ./src/libsablevm/include/jni_system_specific.h 2004-05-01 20:42:28.402644600 +0000
@@ -20,6 +20,7 @@
|| defined (__i386__) || defined (__powerpc__) || defined (__s390__) \
|| defined (__hppa__) || defined (__arm__) || defined (__m68k__) \
|| defined (__mc68000__) || defined (__mips__) || defined (__mipsel__) \
+ || defined (__x86_64__) \
) && defined (__GNUC__))
#define JNICALL
diff -urN ../tmp-orig/sablevm-1.1.3/src/libsablevm/system.c ./src/libsablevm/system.c
--- ../tmp-orig/sablevm-1.1.3/src/libsablevm/system.c 2004-04-12 20:29:08.000000000 +0000
+++ ./src/libsablevm/system.c 2004-05-01 20:54:17.985771568 +0000
@@ -18,6 +18,7 @@
|| defined (__i386__) || defined (__powerpc__) || defined (__s390__) \
|| defined (__hppa__) || defined (__arm__) || defined (__m68k__) \
|| defined (__mc68000__) || defined (__mips__) || defined (__mipsel__) \
+ || defined (__x86_64__) \
) && defined (__GNUC__))
/*
@@ -346,6 +347,20 @@
: "memory");
/* *INDENT-ON* */
+#elif defined (__x86_64__)
+
+ _svmt_word current_value;
+ _svmt_u8 result;
+
+/* *INDENT-OFF* */
+ __asm__ __volatile__ ("lock\n\t"
+ "cmpxchgq %3, %1\n\t"
+ "sete %0"
+ :"=q" (result), "=m" (*pword), "=a" (current_value)
+ :"r" (new_value), "m" (*pword), "a" (old_value)
+ :"memory");
+/* *INDENT-ON* */
+
#else
#error "SableVM is not prepared to run on this kind of a system; no atomic compare&swap defined."
#endif
diff -urN ../tmp-orig/sablevm-1.1.3/src/libsablevm/system.h ./src/libsablevm/system.h
--- ../tmp-orig/sablevm-1.1.3/src/libsablevm/system.h 2004-04-12 20:29:08.000000000 +0000
+++ ./src/libsablevm/system.h 2004-05-01 20:53:12.493727872 +0000
@@ -54,6 +54,7 @@
|| defined (__i386__) || defined (__powerpc__) || defined (__s390__) \
|| defined (__hppa__) || defined (__arm__) || defined (__m68k__) \
|| defined (__mc68000__) || defined (__mips__) || defined (__mipsel__) \
+ || defined (__x86_64__) \
) && defined (__GNUC__))
/* "inline" is now an official keyword since the latest C standard (1999).
@@ -231,6 +232,25 @@
/* #include <inlined_i386-gcc-2.95.h> */
+#elif defined (__x86_64__)
+
+typedef _svmt_u64 _svmt_word;
+typedef _svmt_d64 _svmt_float_word;
+
+#define SVM_WORD_SIZE 8 /* size in bytes */
+#define SVM_WORD_BIT_COUNT 64 /* size in bits */
+
+/* see comments at the head of this file */
+#define SVM_ALIGNMENT 8
+#define SVM_ALIGNMENT_POWER 3 /* 2 ^^ SVM_ALIGNMENT_POWER == SVM_ALIGNMENT */
+#define SVM_PAGE_SIZE 4096
+
+#if (GCC_VERSION >= 2095) && (GCC_VERSION <3000)
+#include <inlined_i386-gcc-2.95.h>
+#else
+#include <inlined_i386-gcc-3.3.h>
+#endif /* (GCC_VERSION >= 2095) && (GCC_VERSION <3000) */
+
#else
#error "unknown system"
#endif
|
6 messages has been excluded from this view by a project administrator.
