Tamarin:WeeklyUpdates: Difference between revisions

From MozillaWiki
Jump to navigation Jump to search
No edit summary
Line 12: Line 12:




= 1st July 2008 =
= July 15th 2008 =


== Attendees ==
== Attendees ==


*Adobe: Scott, Steven, Edwin, Jeff, Dan, Mason, Vladimir
*Adobe: Scott, Steven, Edwin, Jeff, Mason, Jennifer, Rob B., Ed, Rob W.
*Intel: Moh, Carmen, Jungwoo, Shengnan
*Intel: Carmen, Jungwoo, Ming
*Mozilla: Brendan, Jordan, Andreas, Rob, Jason, Aaron, Ben, Joel, Jacob, Dave M., Dave A.
*Mozilla: Brendan, Jordan, Andreas, Jason, Ben, Joel, Dave M., Dave A., Jim, Benjamin


== Updates ==
== Updates ==
 
*(Jungwoo) Update on parallel JIT'ing in TT. 
*(Dave M) Introduce Joel Galenson,
** Is trying moving all patches to interpreter side.
** Joel will be an intern working through August
** Headsup on Moh's vectorization of hot search loop of MMgc::PinStackObjects using SSE2. Both Sunspider's string benchmarks (string-fasta & string-validate-input) show 8+% speedup.
** regexp tracing (generating method IL on the fly, fast access to string chars)
** Moh will explain more in next meeting.
*(David A + Andreas) Explicit field initialization in constructors?
*(Andreas) SSE2 calling conventions. SSE2-only JIT? Shall we specialize with SSE2-only?  
*(Andreas) Reorganize Interpreter.h/cpp? TraceMonitor/TraceRecorder split?  
** This may make slower processers even slower.  This would affect Flash.  This should be investigated though.  Would compile-time be enough? Need to look into this.  SSE2 or later machines would makes sense.
*(Andreas) Register allocator discussion (loop-spanning register use, tree-spanning register use.)
** More investigations needed for SSE2.  Andreas will put a patch together for Mac and Linux.
*(Moh) 13% performance for 13% size increase: space/speed tradeoff in configuring TT: https://bugzilla.mozilla.org/show_bug.cgi?id=442804
*(Benjamin) More GC discussion - jemalloc+gc proposal
*(Jungwoo) Protection problem in Windows 2003 Server: https://bugzilla.mozilla.org/show_bug.cgi?id=442961
**semi-precise GC
*(Ed) Adobe to have sandbox available for everyone to push to prior to the final repository
**shared heap space with malloced memory
** How do we do this?  When is the right time to do this? TBD...
**threadsafety, even with write barriers
**at the cost of DRC
***Lars will be looking into this
***Flash Player benefits from string-like objects.
***MMGC could just do GC and drop fix mallloc.  Tom is going to formulate a response to the email thread going around.
*(Ed) Worked up new patch for 8-byte instructions.  Made parallel patch for 4-byte.  In comparison, 4-byte won!  Will land another patch for this.
**RP and SP pointers: need to special case these pointers.  This will help keep us under the 8-byte instructions.
*(Steven) Is tracking down performance issues for a Flex-based app for Tamarin Tracing.  New issues are coming up with TT vs. TC issues in doing this.
**Adobe is building continuous benchmarking is coming along and should be public in a few weeks. 
**Considering open tri-server so folks can push public patches. Might make sense to have this hosted via Mozilla. Ed to open a bug on this.
**Mozilla is also doing macro benchmarking that is being worked on.
*(Andreas) TraceMonkay- 64-bit backend is now working! Will get pushed into Tamarin soon. Ed can help push patch if it gets too big funky.
**Carmen at Intel has also done some work with microbytes for SpiderMonkey.
*(Mingqui) Submitted 2 patches a few weeks ago.  Any update on these? Steven at Adobe to review these…
**https://bugzilla.mozilla.org/show_bug.cgi?id=438889
**https://bugzilla.mozilla.org/show_bug.cgi?id=438881


= Older meetings =
= Older meetings =

Revision as of 21:51, 15 July 2008

These updates concern Tamarin and related projects only.

Meeting Details

  • 2:00pm Pacific Time (21:00 UTC) on Tuesdays
    • (5PM Eastern US, 11PM Oslo, 6AM (Wed) Seoul, 7AM (Wed) Melbourne)
  • Location: Tel: 866 705 2554 (us), 913 227 1201 (int’l)
    • Passcode: 914008
  • Duration: 60 minutes
  • join irc.mozilla.org #tamarin for attendence taking and questions

Next meeting's Agenda Items:


July 15th 2008

Attendees

  • Adobe: Scott, Steven, Edwin, Jeff, Mason, Jennifer, Rob B., Ed, Rob W.
  • Intel: Carmen, Jungwoo, Ming
  • Mozilla: Brendan, Jordan, Andreas, Jason, Ben, Joel, Dave M., Dave A., Jim, Benjamin

Updates

  • (Jungwoo) Update on parallel JIT'ing in TT.
    • Is trying moving all patches to interpreter side.
    • Headsup on Moh's vectorization of hot search loop of MMgc::PinStackObjects using SSE2. Both Sunspider's string benchmarks (string-fasta & string-validate-input) show 8+% speedup.
    • Moh will explain more in next meeting.
  • (Andreas) SSE2 calling conventions. SSE2-only JIT? Shall we specialize with SSE2-only?
    • This may make slower processers even slower. This would affect Flash. This should be investigated though. Would compile-time be enough? Need to look into this. SSE2 or later machines would makes sense.
    • More investigations needed for SSE2. Andreas will put a patch together for Mac and Linux.
  • (Benjamin) More GC discussion - jemalloc+gc proposal
    • semi-precise GC
    • shared heap space with malloced memory
    • threadsafety, even with write barriers
    • at the cost of DRC
      • Lars will be looking into this
      • Flash Player benefits from string-like objects.
      • MMGC could just do GC and drop fix mallloc. Tom is going to formulate a response to the email thread going around.
  • (Ed) Worked up new patch for 8-byte instructions. Made parallel patch for 4-byte. In comparison, 4-byte won! Will land another patch for this.
    • RP and SP pointers: need to special case these pointers. This will help keep us under the 8-byte instructions.
  • (Steven) Is tracking down performance issues for a Flex-based app for Tamarin Tracing. New issues are coming up with TT vs. TC issues in doing this.
    • Adobe is building continuous benchmarking is coming along and should be public in a few weeks.
    • Considering open tri-server so folks can push public patches. Might make sense to have this hosted via Mozilla. Ed to open a bug on this.
    • Mozilla is also doing macro benchmarking that is being worked on.
  • (Andreas) TraceMonkay- 64-bit backend is now working! Will get pushed into Tamarin soon. Ed can help push patch if it gets too big funky.
    • Carmen at Intel has also done some work with microbytes for SpiderMonkey.
  • (Mingqui) Submitted 2 patches a few weeks ago. Any update on these? Steven at Adobe to review these…

Older meetings

Retrieved from "https://wiki.mozilla.org/index.php?title=Tamarin:WeeklyUpdates&oldid=100546"