Re: pkgsrc DragonFly 1.13/i386 2008-06-18 09:18
:Total number of packages: 7678 : Successfully built: 6766 : Failed to build: 357 : Depending on failed package: 207 : Explicitly broken or masked: 282 : Depending on masked package:66 : :Packages breaking the most other packages : :Package Breaks Maintainer Speaking of pkgsrc, what should we be making available for the release? The latest quarterly or the pkgsrc HEAD ? Anyone have any ideas? We need to get started on the build in the next week in order to make the 2.0 release (which will be the middle of this month, in two weeks). -Matt
Re: pkgsrc DragonFly 1.13/i386 2008-06-18 09:18
On Tue, July 1, 2008 1:40 pm, Joerg Sonnenberger wrote: On Tue, Jul 01, 2008 at 10:33:30AM -0700, Matthew Dillon wrote: Speaking of pkgsrc, what should we be making available for the release? The latest quarterly or the pkgsrc HEAD ? I'd go with HEAD, given that it is frozen already. Plus the most recent previous release, 2008Q1, had a lot of failed builds on DragonFly. Not a huge amount, but it was certainly a regression from previous releases. Hasso's been putting in a lot of patches, so depending on which ones are committed (Joerg, can you commit them?), there should be less breakage. I'll try to get a bulk build restarted on pkgbox, though that shouldn't stop anyone else from building too.
prepare for thousands of processing cores
Have a look at http://news.cnet.com/8301-13924_3-9981760-64.html and http://blogs.intel.com/research/2008/06/unwelcome_advice.php ... developers should start thinking about ... thousands of cores now in their algorithmic development and deployment pipeline. I am curious: what about DragonFly and thousands of cores?
Re: pkgsrc DragonFly 1.13/i386 2008-06-18 09:18
Matthew Dillon wrote: Speaking of pkgsrc, what should we be making available for the release? The latest quarterly or the pkgsrc HEAD ? Anyone have any ideas? We need to get started on the build in the next week in order to make the 2.0 release (which will be the middle of this month, in two weeks). Latest quarterly was disaster if speaking about DragonFly. Although there is still great amount of patches not committed yet (sitting in GNATS), HEAD is in much better shape. And 2008Q2 should be released in some weeks as well. -- Hasso Tepper
cvs trouble
I had some trouble trying to track down when traceroute stopped working: (hasso just commited fix ;-) My plan was to use vkernel to find the date it stopped working: Checking out full sources for given date and building and installing world and VKERNEL. But it failed: using cvs checkout, sources checked out w/ -D2008-06-07 fails buildworld. It turns out reason is that checked out src/contrib/libarchive-2/libarchive/archive.h.in is wrong version: It is revision 1.2, but newer revision in vendor branch 1.1.1.X exists (the right one ;-). (reason seems to be that handling vendor branches is difficult for most of us) This seems to be as advetised: cvs checkout -D is documented to give latest revision, checked out to given date. cvs checkout without -D does get right (ie latest) file revision (1.3 dead in this case). Can cvs checkout the right file version for given date (same as checkout (without -D) on date)? -thomas
Re: prepare for thousands of processing cores
:Have a look at http://news.cnet.com/8301-13924_3-9981760-64.html and
:http://blogs.intel.com/research/2008/06/unwelcome_advice.php
:
:... developers should start thinking about ... thousands of cores now in
:their algorithmic development and deployment pipeline.
:
:I am curious: what about DragonFly and thousands of cores?
Kinda what Sun is doing.
The problem we face as developers is not so much in developing new
algorithms but instead in needing a wholely new programming language
which naturally operates, even assumes, a many-threads model.
And here I'm not talking about Java and its piss-poor structural
locking.
I've experimented with this sort of thing by building simple
interpreted languages, for example a new procedural model that is
naturally threaded and works something like this:
...
x += fubar(x);
x += fubar(x);
x += fubar(x);
__thread
int
fubar(int x)
{
... do some stuff synchronously ...
/* return a synchronous result... */
result(x + 50);
/* then continue as a thread ... */
blah blah
blah
blah
blah
return; /* (or fall off the end of the procedure) */
}
The idea being that to take proper advantage of a many-cores system
the programming language must not only be naturally threaded, but
also be able to efficiently and dynamically allocate and free cpu cores
for threading as well as seemlessly combine timeshare scheduling with
threading.
The actual implementation is not too hard to do. The programming
language must do away with the standard notion of a fixed-sized stack
and instead allocate a procedure's stack resources dynamically,
allowing the program to split and collapse many execution paths
as part of its natural operation.
In anycase, a programming language that efficiently does the above as
part of its core is an absolute requirement for being able to work
in a many-cores environment. The developer has to not be afraid of
calling a __thread procedure in a critical path. We will not be able
to take advantage of such an environment until threads are thought of
as low cost, natural entities with overheads on the same order as
an unthreaded procedure call.
Ultimately that means the cpu core needs to have instructions to
dynamically create and destroy threads, which automatically fall back
to time-share if no hardware threads are available. And it has to
be able to do it in less then a few nanoseconds to be viable, and
not break the instruction pipeline.
-Matt
Matthew Dillon
[EMAIL PROTECTED]
Re: cvs trouble
Thomas Nikolajsen wrote: I had some trouble trying to track down when traceroute stopped working: (hasso just commited fix ;-) My plan was to use vkernel to find the date it stopped working: Checking out full sources for given date and building and installing world and VKERNEL. But it failed: using cvs checkout, sources checked out w/ -D2008-06-07 fails buildworld. It turns out reason is that checked out src/contrib/libarchive-2/libarchive/archive.h.in is wrong version: It is revision 1.2, but newer revision in vendor branch 1.1.1.X exists (the right one ;-). I don't have an archive.h.in, and the archive.h doesn't have a 1.2, so I can't really follow the problem. Can cvs checkout the right file version for given date (same as checkout (without -D) on date)? CVS should work correctly with -D as well (to some extent) cheers simon -- Serve - BSD +++ RENT this banner advert +++ASCII Ribbon /\ Work - Mac +++ space for low €€€ NOW!1 +++ Campaign \ / Party Enjoy Relax | http://dragonflybsd.org Against HTML \ Dude 2c 2 the max ! http://golden-apple.biz Mail + News / \
EHCI working?
Hey, could it be that EHCI is not working correctly? On my desktop I get irq 3 interrupt livelocks when loading EHCI (actually it is on/off livelocking). On my laptop it seems to load okay, but then transferring data to my new mp3 player is slow, basically around 1MB/sec. In dmesg, cam writes something about Down reving Protocol Version from 2 to 0?, but I don't know what that means. Does anybody have a working EHCI setup? cheers simon -- Serve - BSD +++ RENT this banner advert +++ASCII Ribbon /\ Work - Mac +++ space for low €€€ NOW!1 +++ Campaign \ / Party Enjoy Relax | http://dragonflybsd.org Against HTML \ Dude 2c 2 the max ! http://golden-apple.biz Mail + News / \
Re: cvs trouble
Simon 'corecode' Schubert [EMAIL PROTECTED] wrote: Thomas Nikolajsen wrote: I had some trouble trying to track down when traceroute stopped working: (hasso just commited fix ;-) My plan was to use vkernel to find the date it stopped working: Checking out full sources for given date and building and installing world and VKERNEL. But it failed: using cvs checkout, sources checked out w/ -D2008-06-07 fails buildworld. It turns out reason is that checked out src/contrib/libarchive-2/libarchive/archive.h.in is wrong version: It is revision 1.2, but newer revision in vendor branch 1.1.1.X exists (the right one ;-). I don't have an archive.h.in, and the archive.h doesn't have a 1.2, so I can't really follow the problem. archive.h.in did exist at June 6th, it was deleted later: http://www.dragonflybsd.org/cvsweb/src/contrib/libarchive-2/libarchive/Attic/archive.h.in (http://leaf.dragonflybsd.org/mailarchive/commits/2008-06/msg00115.html) Can cvs checkout the right file version for given date (same as checkout (without -D) on date)? CVS should work correctly with -D as well (to some extent) Yes, I it works as advertised; but I need a slightly other behavior, as described; do I have to use other / additional flags for checkout? (another option could be using cvsup, but I do expect cvs to be able to do this) -thomas
Re: cvs trouble
Thomas Nikolajsen wrote: Yes, I it works as advertised; but I need a slightly other behavior, as described; do I have to use other / additional flags for checkout? (another option could be using cvsup, but I do expect cvs to be able to do this) No, this is an example of a classical CVS problem - a mixture of CVS import and add was used and thus CVS has no means of reconstructing the real history. Poof, gone. cheers simon -- Serve - BSD +++ RENT this banner advert +++ASCII Ribbon /\ Work - Mac +++ space for low €€€ NOW!1 +++ Campaign \ / Party Enjoy Relax | http://dragonflybsd.org Against HTML \ Dude 2c 2 the max ! http://golden-apple.biz Mail + News / \
