You can subscribe to this list here.
2011 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(3) |
Jul
(1) |
Aug
|
Sep
(2) |
Oct
(1) |
Nov
(1) |
Dec
(1) |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2012 |
Jan
|
Feb
|
Mar
(2) |
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
(3) |
Oct
(4) |
Nov
(2) |
Dec
|
2013 |
Jan
(1) |
Feb
|
Mar
|
Apr
(1) |
May
(1) |
Jun
|
Jul
|
Aug
(1) |
Sep
(1) |
Oct
(2) |
Nov
(1) |
Dec
(5) |
2014 |
Jan
(4) |
Feb
|
Mar
(1) |
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
(1) |
Aug
|
Sep
(1) |
Oct
(3) |
Nov
(3) |
Dec
(1) |
2015 |
Jan
|
Feb
(1) |
Mar
(2) |
Apr
(2) |
May
|
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
(3) |
Nov
(1) |
Dec
(1) |
2016 |
Jan
(2) |
Feb
(2) |
Mar
(2) |
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
(2) |
Nov
(1) |
Dec
(1) |
2017 |
Jan
(2) |
Feb
|
Mar
(1) |
Apr
(1) |
May
(1) |
Jun
(1) |
Jul
|
Aug
|
Sep
|
Oct
(1) |
Nov
|
Dec
(1) |
2018 |
Jan
(4) |
Feb
|
Mar
|
Apr
(2) |
May
(2) |
Jun
(1) |
Jul
|
Aug
|
Sep
(1) |
Oct
|
Nov
(3) |
Dec
(1) |
2019 |
Jan
|
Feb
(1) |
Mar
(3) |
Apr
|
May
(1) |
Jun
(2) |
Jul
(2) |
Aug
(4) |
Sep
(1) |
Oct
|
Nov
(1) |
Dec
(1) |
2020 |
Jan
(2) |
Feb
(1) |
Mar
|
Apr
(2) |
May
(2) |
Jun
(2) |
Jul
|
Aug
|
Sep
(1) |
Oct
(1) |
Nov
|
Dec
(1) |
From: Sung-Eun C. <su...@cr...> - 2013-08-01 17:22:30
|
Hello all, Based on results from this survey as well as the recent survey regarding the Lightning Talks BoF, it seems that IPDPS could be a good venue for a Chapel workshop. Since this would be a new workshop for IPDPS, we would like to include information on possible attendees in our submission to the IPDPS workshop committee. So now it's time for yet another survey! - Would you be interested in attending a Chapel workshop at IPDPS on the Friday following the main conference? Note that the registration fee for IPDPS includes all workshops. - Would you be interested in a sort of "developer code camp" on the Saturday following the workshop? This might not be included with the IPDPS registration fee. For those of you unfamiliar with IPDPS, it is a large, multi-track conference that is run yearly. The conference is surrounded on both ends by single-day workshops. For a sampling of content at IPDPS, see last year's program: http://www.ipdps.org/ipdps2013/2013_advance_program.html. IPDPS 2014 will be held May 19-23 in Phoenix, AZ. Thanks again for your time! -- Sung On Thu, Jan 24, 2013 at 06:51:01PM -0800, Brad Chamberlain wrote: > > Hi Chapel Community -- > > As we head into the new year and a post-HPCS world for Chapel at Cray, one > of the things that we've been kicking around is whether we're in the > position to hold a successful Chapel Users and/or Developers workshop in > the coming year or two. If that's something you'd be interested in, > please send back responses to the following informal survey (to me and/or > cha...@cr... -- please don't Cc: the entire list): > > > 1) Would you (or others in your organization for whom you can speak) be > willing to travel to some sort of Chapel Users/Developers workshop? > How many of you are you speaking for? > > > 2) Would you be most interested in a User's workshop, a Developer's > workshop, or both? If both, would you prefer to see them as > distinct events or held on consecutive days? > > > 3) How many days would you expect you'd enjoy spending at such a workshop? > > > 4) Would you prefer such a workshop to be standalone event, or would you > prefer to see it attached to some other event like SC or CUG (the Cray > User's Group workshop)? > > > 5) If "standalone", what general time of year (season/month) would be > ideal for you to attend the workshop? > > > 6) What activities would you most like to see at such a workshop? (talks > from the Cray team, talks from others in the Chapel community, > talks from outside the community, free discussion on particular > topics, code camps, panels, tutorials, etc.) > > > 7) Any other thoughts on this idea? > > > Thanks very much, > -Brad (on behalf of the Cray Chapel team) > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnnow-d2d > _______________________________________________ > Chapel-announce mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chapel-announce > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnnow-d2d > _______________________________________________ > Chapel-users mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chapel-users ------------------------------------------------------------------------------ Get your SQL database under version control now! Version control is standard for application code, but databases havent caught up. So what steps can you take to put your SQL databases under version control? Why should you start doing it? Read more to find out. http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2013-05-07 21:46:01
|
Hi Chapel community -- For those of you who may be at CUG this week, I wanted to mention that Martha Dumler, manager of the Chapel team at Cray, will be presenting a talk on Thursday entitled "The State of the Chapel Union", recapping the Chapel project under HPCS, where that work puts us today, and what we consider our next steps to be. If you won't be at CUG (or want additional detail), the paper for the talk has been added to the Chapel website here (and slides will follow): http://chapel.cray.com/papers/ChapelCUG13.pdf In my mind, this is the last (I hope!) in a series of writings done over the past year or so that capture Chapel progress and philosophy as we catch our breath at the end of the HPCS program. Here's a quick summary of these articles and their contents: * The State of the Chapel Union (CUG): contains a brief history of Chapel, including a "Harper's Index" style enumeration of project trivia; a very high-level, descriptive overview of the language; and a fairly detailed accounting of status, successes & lessons learned under HPCS, and next steps. http://chapel.cray.com/papers/ChapelCUG13.pdf * A Brief Overview of Chapel (pre-print draft of a book chapter): Compared to the previous paper, this chapter contains a more in-depth history of the project, describes the motivating themes in the language design, and provides a more in-depth introduction to language features. It contains some status and future directions, but in less detail than the CUG paper. http://chapel.cray.com/papers/BriefOverviewChapel.pdf * [10] Myths About Scalable Parallel Programming Languages (IEEE TCSC blog): This was a series of blog articles I wrote last year to serve as something between evangelism and therapy. It captures common objections levied against developing new parallel languages like Chapel and gives my counterarguments. see: http://chapel.cray.com/language.html for a series of links to the individual articles, hosted at: https://www.ieeetcsc.org/activities/blog Eagerly looking forward to looking forward for awhile rather than back, -Brad ------------------------------------------------------------------------------ Learn Graph Databases - Download FREE O'Reilly Book "Graph Databases" is the definitive new guide to graph databases and their applications. This 200-page book is written by three acclaimed leaders in the field. The early access version is available now. Download your free book today! http://p.sf.net/sfu/neotech_d2d_may _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2013-04-18 18:39:29
|
Chapel commmunity -- Cray Inc. and the Chapel developer community are pleased to announce the release of version 1.7.0 of Chapel. This release's highlights include: * Support for regular expression operations on strings and channels. See doc/technotes/README.regexp in the release for more information. * Support for formatted I/O via new readf()/writef() routines. See doc/technotes/README.formattedIO in the release for more information. * Support for injecting C declarations or #include files into Chapel code via 'extern blocks'. See doc/technotes/README.extern in the release for more information. * Support for associative domain and array literals. For example, '{"chapel", "mpi", "openmp"}' is an associative domain of strings while '["chapel" => 1.7, "mpi" => 3.0, "openmp" => 3.1]' is an associative array mapping strings to real floating point values. * A new capability for main() to take arguments (as an alternative to config consts and vars) and return a status value. See doc/technotes/README.main in the release for more details. * Support for iterating over homogenous tuples. For example, 'for numtasks in (1, 2, 4, 8, 16) do ...' will now iterate over the integer values defined by the tuple. * Performance improvements related to stack-allocated and constant variables in the context of multi-locale executions. * Improvements to the emerging bulk transfer optimization to communicate chunks of data at a time in multi-locale executions. * Significant improvements to the Chapel implementation of LULESH. * Numerous bug fixes and improvements to semantic checks and error messages. ...and much more! See below for a more complete list of changes in version 1.7.0, or refer to $CHPL_HOME/CHANGES within the release itself. Contributors to this release include: Rafael Asenjo Brad Chamberlain Sung-Eun Choi Martha Dumler Michael Ferguson Tom Hildebrandt David Iten Jessica Jueckstock John Koenig Vassily Litvinov Alberto Sanz Greg Titus To download the release, visit our SourceForge page at: http://sourceforge.net/projects/chapel At this site, you can also browse the mailing list archives and track our progress day-by-day. Our main project page continues to be hosted at: http://chapel.cray.com and it remains the best place to browse papers, presentations, documents, tutorials, and news items; or to read about collaborations with researchers and educators. As always, we're interested in your feedback on how we can make the Chapel language and implementation more useful to you. On behalf of the Chapel team, -Brad Chamberlain ============= version 1.7.0 ============= Tenth public release of Chapel, April 18, 2013 Highlights (see entries in subsequent categories for details) ------------------------------------------------------------- - added support for regular expression operations on strings and channels - added support for readf()/writef() routines for formatted I/O - added support for injecting C declarations into Chapel via "extern blocks" - added support for associative domain/array literals - added a capability for main() to take arguments and return a status value - added support for iteration over homogenous tuples - performance improvements related to stack-allocated and constant variables - improved bulk transfer optimization (use -s useBulkTransferStride to enable) - significant improvements to the Chapel version of LULESH - enabled GMP using the system-installed version on Cray platforms by default - switched to using processor atomics for reference counting on Cray systems - numerous bug fixes and improvements to semantic checks and error messages Environment Changes ------------------- - renamed the platform 'cray-cascade' to 'cray-xc' - renamed the platforms 'xmt'/'xmt-sim' to 'cray-xmt'/'cray-xmt-sim' - enabled GMP support by default if it is already built in the third-party dir Syntactic/Naming Changes ------------------------ - added support for associative domain/array literals (e.g., {"red", "blue"} is an associative domain of strings) (e.g., ["red" => 0, "blue => 1] is an assoc. array from strings to ints) Semantic Changes / Changes to Chapel Language --------------------------------------------- - added support for injecting C declarations into Chapel via "extern blocks" (see "Support for Extern Blocks" section of doc/technotes/README.extern) - added support for iteration over homogenous tuples (see "Tuples" section of the Chapel spec) - added a capability for main() to take arguments and return a status value (see doc/technotes/README.main) - added a 'requestCapacity()' capability to pre-size associative domains (see "Domains:Predefined Functions" section of the Chapel spec) - added a warning() routine to generate warnings, similar to halt(), assert() (see "Standard Modules" section of the Chapel spec) Newly Implemented Features -------------------------- - added support for the .localSlice() method to take a domain argument (e.g., A.localSlice({loRow..hiRow, loCol..hiCol})) Deprecated Features ------------------- - removed the tuple() function (i.e., instead of tuple(x) or tuple(x,y,z), use (x,) or (x,y,z) instead) - deprecated warning for old-style domain literals by default (i.e., [1..m, 1..n] is an array of ranges; re-enable w/ --warn-domain-literal) Standard Modules ---------------- - added support for regular expression operations on strings and channels (see doc/technotes/README.regexp for details) - added support for readf()/writef() routines for formatted I/O (see doc/technotes/README.formattedIO for details) - refactored communication diagnostics into its own module (CommDiagnostics) (see doc/technotes/README.comm-diagnostics for documentation) - added isDomainType(type t)/isArrayType(type t)/isRecordType(type t) queries Documentation ------------- - added README.formattedIO to describe support for writef()/readf() routines (see doc/technotes/README.formattedIO) - added README.regexp to describe support for regular expression operations (see doc/technotes/README.regexp) - added README.main to describe support for arguments/return values for main() (see doc/technotes/README.main) - updated README.extern to describe C type aliases and extern blocks - changed READMEs that used csh-style syntax to use bash-style consistently - fixed a semantic typo in README.atomics - minor updates and fixes to various chapters in the language specification - minor updates to several READMEs Example Codes ------------- - improvements to the Chapel implementation of LULESH (see examples/benchmarks/lulesh/ for more details) - configuration improvements (see README for details): - added ability to compute the cube input set rather than reading from disk - support for 3D vs. 1D representations via 'use3DRepresentation' - support sparse vs. dense material representations via --sparseMaterials - added a --maxcycles config that can be used to stop the computation early - converted 'dtfixed' into a config const - added a per-iteration timing when doTiming and showProgress are enabled - code improvements: - refactored setup code into a helper module: luleshInit.chpl - changed the representation of nodesets to use a sparse subdomain - switched to using 'ref' intents rather than 'in' or 'inout' - converted vars to consts where appropriate - general code cleanup - performance improvements: - improved performance by avoiding an unnecessary reindexing operation - made some unnecessarily serial loops parallel - replaced error-oriented reductions with forall loops + halt - param-unrolled the loop in localizeNeighborNodes() for performance reasons - removed a redundant zero-initialization of an array - documentation improvements: - updated README to document new configuration options - added the LULESH callgraph to the README - improved the Chapel implementation of RA: - added the ability to use an LCG random number generator via 'useLCG' - added a config const 'verify' that permits verification to be skipped - fixed a typo in HPL that inflated the calculated GFLOPs by 10x - improvements to the Chapel implementation of SSCA2: - fixed a bug in the torus representation - fixed a bug with VALIDATE_BC=true Platform-specific Notes ----------------------- - enabled GMP using the system-installed version on cray-x* platforms by default Portability of code base ------------------------ - fixed an issue related to vsnprintf on OS X Lion - fixed an assumption that platforms defining INT8_MIN will also define INT*_MIN Launcher-specific Notes ----------------------- - improved use of cnselect in launchers that use 'aprun' Compiler Flags -------------- - added --no- variants of the following compiler flags: --count-tokens, --llvm, --print-code-size, --print-commands, --print-passes, --print-search-dirs, - added --[no-]warn-tuple-iteration to warn against old-style zipper iteration (see man page for details) - added --[no-]warn-special as a meta-flag for all old-style syntax uses (see man page for details) - improved support for environment variable setting of '--no-'style flags (see man page for details) Error Message Improvements -------------------------- - improved type mismatch error messages when assigning scalars to arrays - added an error for extern functions with formal arguments of array type - made Ctrl-C and the like result in a user, rather than internal, error - added an error for associative domains with unsupported index types Bug Fixes / New Semantic Checks (for old semantics) --------------------------------------------------- - fixed a bug in which sizeof() had leaked into the user namespace in v1.6 - fixed a regression in v1.6 in which type aliases of arrays were broken - fixed a bug in which 0-element arrays of arrays would free domains prematurely - fixed a bug related to writing out instances of the RandomStream class - I/O bug fixes: - QIO_HINT_PARALLEL now works for new/zero length files - fixed bugs related to bit-oriented reads and writes - fixed a bug w.r.t. EOF and preemptive mmap - fixed portability bug related to Intel compilers - changed channels so that once EOF is hit, it will never read further - fixed issues related to ssize_t/int(64) mismatches on 32-bit platforms - fixed an issue related to reading empty strings - fixed some bugs in the implementation of atomic variables - fixed a bug in which AdvancedIters didn't default 'numTasks' as documented - improved shape checking for array assignment - restricted reshape() to be applied for rectangular domains only - restricted reshape() to only work when the source and destination sizes match - improved support for Cyclic distributions of idxType uint(64) - fixed INFINITY and NAN to be of type 'real' rather than 'real(32)' - fixed a bug with '/*/' within a nested comment - fixed a bug in which uses of ** would dispatch to a user's pow() overload - fixed a few bugs in the Buffers.chpl module - fixed a bug in which memory leak tracking would mislabel some allocations - fixed a bug with recursive iterator inlining - fixed a bug in broadcasting large values at program startup Packaging Changes ----------------- - added Chapel mode support for emacs version 23.x (see etc/emacs/README) Performance Improvements ------------------------ - switched to using processor atomics for reference counting on Cray systems - improved bulk transfer optimization (use -s useBulkTransferStride to enable) - put fewer variables on the heap in the presence of nested functions - enabled remote value forwarding for more constants/arguments than before - made atomic variables use compiler intrinsics for cray-prgenv-intel - made associative domains of 'imag' indices more efficient Runtime Library Changes ----------------------- - split chpl_task_begin() to distinguish between new tasks and migrating ones - improved the ability to build the Chapel runtime with a C++ compiler Third-Party Software Changes ---------------------------- - updated to version 1.20 of GASNet - updated to version 1.9 of Qthreads - made the LLVM checkout script grab version 3.2 to ensure compatability - removed the nanox tasking layer due to lack of support/champion Testing System -------------- - extended the 'skipif' capability to support the ability to skip a subdirectory - improved the robustness of start_test in cross-compiled environments - added support for .no-local.good files for --no-local-specific test output - fixed a bug in which the test system looked at incorrect .future files - removed artificial limiting of stacksize in start_test - refactored test system environment setup into util/test/testEnv - made the test system use $cwd rather than 'pwd' - made 'nightly' compopts/execopts get added to end of the command line - made 'nightly' support a --no-futures option Internal/Developer-oriented --------------------------- - added some code that sets the groundwork for supporting hierarchical locales - fixed portability of bswap.h (for once and for all?) - restructured internal modules corresponding to specific runtime layers - removed uses of alloca() - removed PRIM_CHPL_ALLOC_PERMIT_ZERO - deprecated 'make depend' and made use of gcc -MMD/-MP instead - made 'rm -rf' commands a little safer in Makefile clean/clobber rules - refactored Makefiles for Cray systems/compilers to avoid repetition - added a map_view() function for printing out SymbolMaps - removed some dead and possibly incorrect methods from Map and Vec classes - refactored _cast functions for booleans - made the _isPrimitiveType test more concise - improved the mechanism by which line numbers are assigned to AST nodes - factored sync/single out of ChapelBase.chpl and into ChapelSyncvar.chpl - fixed the --gen-extern-prototypes flag - improved the correspondence of FLAG_* and string forms of AST flags - improved the correspondence of PRIM_* and string forms of primitives - refactored chplrt.h - added initial support for a .localAccess accessor to Block/Default arrays - renamed compiler-generated types for classes in generated code - refactored QIO code in the runtime - converted macros in the compiler to inline functions for sanity/performance - cleaned up the functions that build array/domain literals - refactored and commented lowerIterators.cpp - keep dtObject alive even if unused after function resolution - added EOF to the list of reserved symbol names - protected some C identifiers introduced by PGI when compiling with -Kieee ------------------------------------------------------------------------------ Precog is a next-generation analytics platform capable of advanced analytics on semi-structured data. The platform includes APIs for building apps and a phenomenal toolset for data science. Developers can use our toolset for easy data analysis & visualization. Get a free account! http://www2.precog.com/precogplatform/slashdotnewsletter _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2013-01-25 02:51:13
|
Hi Chapel Community -- As we head into the new year and a post-HPCS world for Chapel at Cray, one of the things that we've been kicking around is whether we're in the position to hold a successful Chapel Users and/or Developers workshop in the coming year or two. If that's something you'd be interested in, please send back responses to the following informal survey (to me and/or cha...@cr... -- please don't Cc: the entire list): 1) Would you (or others in your organization for whom you can speak) be willing to travel to some sort of Chapel Users/Developers workshop? How many of you are you speaking for? 2) Would you be most interested in a User's workshop, a Developer's workshop, or both? If both, would you prefer to see them as distinct events or held on consecutive days? 3) How many days would you expect you'd enjoy spending at such a workshop? 4) Would you prefer such a workshop to be standalone event, or would you prefer to see it attached to some other event like SC or CUG (the Cray User's Group workshop)? 5) If "standalone", what general time of year (season/month) would be ideal for you to attend the workshop? 6) What activities would you most like to see at such a workshop? (talks from the Cray team, talks from others in the Chapel community, talks from outside the community, free discussion on particular topics, code camps, panels, tutorials, etc.) 7) Any other thoughts on this idea? Thanks very much, -Brad (on behalf of the Cray Chapel team) ------------------------------------------------------------------------------ Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft MVPs and experts. ON SALE this month only -- learn more at: http://p.sf.net/sfu/learnnow-d2d _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2012-11-16 23:46:21
|
Hi Chapel Community -- Chapel had a great year at SC12 this year, and as the week wraps up, I wanted to thank everyone who helped work hard to make it so, both within Cray and across the broader community. We've archived the presentation materials from the conference on the Chapel website at: http://chapel.cray.com/presentations.html for those who would like to take a second (or first) look at them. Here's a quick round-up of the week's activities from the Chapel perspective: * We had a new piece of swag to hand out this year -- a snappy drawstring backpack/cinchsack bearing the Chapel logo. We effortlessly handed out about 225 of these at Chapel events as well as the Cray and PGAS booths, and could easily have given out more. We also passed out several dozen leftover Chapel USBs last year loaded with the latest release and some recent highlights from the Chapel website. * We prepared a poster for the PGAS booth and helped staff it during the week. We also contributed content (a subset of the USB content) to the PGAS DVD which was passed out at the booth. * We did a half-day tutorial on Sunday morning that drew about 20 participants (with some extremely positive reactions) in spite of being on the early side of the conference. * The third annual CHUG happy hour drew a small crowd of familiar faces, and a few new ones as well, for a few hours of good conversation (somehow, though, next year I'll have to get placed at the side of the table discussing fun, rather than technical topics). * On Tuesday, we went head-to-head with X10, XcalableMP, and Charm++ for the title of "Most Productive" language, retaining our "most elegant" title for the third time over the past three entries we've submitted. In spite of great leaps forward in performance (including a 500,000x performance improvement on SSCA#2, we were unable to be the language to unite elegance and performance into a single "most productive" award as originally intended, yet never realized in this competition. More disappointingly, many entrants no longer even bother showing code and use library calls, somehow missing the point of the competition. * On Wednesday, we had our second Chapel Lightning Talks BoF, featuring a number of great talks from members of the Chapel community: - Rob Neely from LLNL spoke about interest from the lab in Chapel's future - Sagnak Tasirlar from Rice described recent work to support futures/task graphs in Chapel - Ray Chen from U Maryland described work to apply autotuning techniques to Chapel - Casey Battaglino from Georgia Tech talked about recent work (jointly with Cray's Tom Hildebrandt) to introduce hierarchical locales into Chapel - Kyle Wheeler from Sandia talked about ongoing work to introduce task teams and eurekas into Chapel using the qthreads tasking layer - and Michael Ferguson from LTS described work that he and colleagues have been doing to support LLVM as an alternative back-end to the default C code generator This was followed with a Q&A session featuring questions from the audience and other speakers that started out slow, but gained steam as it went along. * Later that afternoon, Sung-Eun Choi gave an overview Chapel talk at the KISTI booth in support of a workshop that they were holding which was well-received. * That evening, Sung gave a talk describing work that's taken place during the past year to study LULESH within Chapel in a BoF about codesign using proxy-applications -- it used a timely Star Wars analogy to frame the effort. * Meanwhile, I sat in on a BoF that provided a retrospective of the HPCS program that spawned Chapel, serving on a panel that responded to questions, most of which were related to what the program had done well or poorly, along with one specific question about the international standardization of Chapel and X10. * And then yesterday, David Bunde and Kyle Burke gave an afternoon session in the HPC educators forum on teaching with Chapel that was part tutorial, part explanation of how they use Chapel in their classrooms. In spite of being at the end of the technical program, they managed to attract a crowd of 36, a good number of whom stayed until the very end. * I also wanted to mention that a series of blog articles entitled "10 Myths about Scalable Parallel Programming Languaes" that I've been writing over the past several months wrapped up on Monday, freeing up some spare time on my weekends going forward. The series is partly a capture of counterarguments to the traditional dismissals of the likelihood that new languages will ever catch on, and partly a therapeutic way for me to wrap up the HPCS program. In any event, I hope some of you might get something useful out of them for your own use in evangelizing for Chapel and other high-level/productive parallel languages. Thanks again to everyone who helped make SC12 successful from a Chapel perspective -- we look forward to seing you at future events and continuing to work with all of you, on behalf of the Chapel team, -Brad Chamberlain ------------------------------------------------------------------------------ Monitor your physical, virtual and cloud infrastructure from a single web console. Get in-depth insight into apps, servers, databases, vmware, SAP, cloud infrastructure, etc. Download 30-day Free Trial. Pricing starts from $795 for 25 servers or applications! http://p.sf.net/sfu/zoho_dev2dev_nov _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2012-11-06 20:03:37
|
Hi Chapel Community -- Just in time, here's a quick overview of Chapel-related activities at SC12. For those attending, please join us for whichever events you're interested in and able to. And please feel free to pass this list along to colleagues who may be attending and interested. For more specific details, or for updates during the show, please refer to the following webpage: http://chapel.cray.com/events.html * TUTORIAL: On Sunday morning, we'll be presenting a Chapel tutorial that provides an overview of the language and its features from 8:30am - noon. This is a good way to get up-to-speed on Chapel for new users to the language. * HAPPY HOUR: Monday evening, we'll be having the 3rd annual Chapel User's Group (CHUG) Happy Hour/Meeting at Squatter's Pub Brewery, near the convention center. * HPC CHALLENGE: Tuesday @ 12:15pm: Come cheer on the Chapel team as we go head-to-head with X10, Charm++, and XcalableMP for the title of "most productive" programming model. * LIGHTNING TALKS: Wednesday @ 12:15pm will be the second annual Chapel Lightning Talks BoF featuring 5-minute talks from members of the Chapel community on such topics as application studies, autotuning, futures, hierarchical locales, task teams, and LLVM. * INVITED TALK: Wednesday afternoon, around 3pm, Sung-Eun Choi will be giving a Chapel talk at the 1st KIISE-KOCSEA HPC SIG Joint Workshop on High Performance and Throughput Computing at the KISTI Booth in the Exhibition Hall. The precise time hasn't been specified yet, so watch the website for updates. * BOFs: Wednesday evening, Chapel will be represented at two Birds-of-a-Feather sessions: One will be a report on the HPCS program from which Chapel emerged, in which Brad will serve as a panelist during the Q&A session. The second is about using proxy applications to prepare for exascale and Sung-Eun Choi will be giving a brief talk describing our work studying LLNL's LULESH benchmark in Chapel. * EDUCATION: Thursday afternoon from 1:30pm-5:00pm in the HPC Educator forum, professors David Bunde and Kyle Burke will be providing an overview of Chapel and its use in academic courses. Hope to see you at SC12! -Brad Chamberlain, on behalf of the Chapel team ------------------------------------------------------------------------------ LogMeIn Central: Instant, anywhere, Remote PC access and management. Stay in control, update software, and manage PCs from one command center Diagnose problems and improve visibility into emerging IT issues Automate, monitor and manage. Do more in less time with Central http://p.sf.net/sfu/logmein12331_d2d _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Sung-Eun C. <su...@cr...> - 2012-10-30 19:28:49
|
Mark your calendars! SC12 is right around the corner, and it's once again time for CHUG, the annual Chapel Users Group meeting. This year's CHUG will be held at Squatter's Pub Brewery in downtown Salt Lake City (2 blocks from the Convention Center). As in previous years, we will meet for Happy Hour before the SC Opening Gala on Monday November 12th at 5pm. If you think you might attend, please drop me a line so we can make an adequate reservation, but please feel free to stop by even if you don't respond. Hope to see you in Salt Lake City! -- Sung |
From: Brad C. <br...@cr...> - 2012-10-18 17:04:14
|
Chapel commmunity -- Cray Inc. and the Chapel developer community are pleased to announce the release of version 1.6.0 of Chapel. This release's highlights include: * A number of syntactic improvements: - introduction of an explicit 'zip' keyword to express zippered iteration more clearly and permit tuples to support iteration in future releases. - a change in domain literal syntax from square brackets to curly brackets (e.g., [1..m, 1..n] is now written {1..m, 1..n}) in order to better reflect their nature as index sets while also freeing up square brackets for: - support for array literals for the first time; uses square brackets (e.g., [1.0, 2.3, 4.5] is a 3-element array of reals). For more details about these syntax changes, please refer to a new page on our website designed to track language changes that create backwards-compatibility problems: http://chapel.cray.com/language-changes.html * Improved support for atomic variables: new methods and support for floating point atomics. See $CHPL_HOME/doc/technotes/README.atomics in the release for further details. For users of the chapel module on Cray XE (TM), Cray XK (TM), and Cray Cascade (TM) systems, atomics are also implemented using the network by default. * Added support for 'ref' intents to support pass-by-reference argument passing. * Added support for user overloads of 'op=' and '<=>' assignments. * Added a prototype source-based documentation capability, 'chpldoc' that generates HTML documentation based on comments that precede Chapel declarations. See $CHPL_HOME/doc/technotes/README.chpldoc in the release for more information. * Added support for using LLVM as a back-end compiler target rather than C. See $CHPL_HOME/doc/technotes/README.llvm in the release for more information. * Added support for a strided bulk communication optimization that reduces the number of gets/puts/memcpys required to perform assignments between arrays or array slices. In this release, this feature is off by default, but can be enabled by compiling with -s useBulkTransferStride. In future releases, this will be turned on by default. * Added a new tasking runtime that uses University of Tokyo's MassiveThreads library (http://code.google.com/p/massivethreads/). See $CHPL_HOME/doc/README.tasks and $CHPL_HOME/third-party/README in the release for more information. * Made significant improvements to the SSCA2, RA, and LULESH example codes in examples/benchmarks/... * Added improved support for the Tilera chip architecture (see $CHPL_HOME/doc/platforms/README.tilera for more information). * Numerous performance improvements, bug fixes, and documentation updates and improvements. ...and much more! See below for a more complete list of changes in version 1.6.0, or refer to $CHPL_HOME/CHANGES within the release itself. Contributors to this release include: Rafael Asenjo Brad Chamberlain Sung-Eun Choi Lydia Davis Martha Dumler Michael Ferguson Steven Hemmy Tom Hildebrandt David Iten John Koenig Matthew Lentz Vassily Litvinov Juan Lopez Jun Nakashima Maria Angeles Navarro Alberto Sanz Rachel Sobel Greg Titus Joe Yan To download the release, visit our SourceForge page at: http://sourceforge.net/projects/chapel At this site, you can also browse the mailing list archives and track our progress day-by-day. Our main project page continues to be hosted at: http://chapel.cray.com and it remains the best place to browse papers, presentations, documents, and tutorials, or to read about collaborations with researchers and educators. As always, we're interested in your feedback on how we can make the Chapel language and implementation more useful to you. On behalf of the Chapel Team, -Brad Chamberlain ============= version 1.6.0 ============= Ninth public release of Chapel, October 18, 2012 Highlights (see entries in subsequent categories for details) ------------------------------------------------------------- - syntax improvements: - added support for array literals using square brackets - improved domain literals to use curly brackets to suggest set notation - changed zippered iteration to use an explicit 'zip' keyword - improved atomic variables: new methods, support for floating point atomics - added support for 'ref' intents to support pass-by-reference arguments - added support for user overloads of 'op=' and '<=>' assignments - added a prototype of 'chpldoc', a code-based documentation feature - added support for using LLVM as a back-end compiler target rather than C - added support for a strided bulk communication optimization - numerous performance improvements and bug fixes - significant improvements to the SSCA2, RA, and LULESH example codes - added a new tasking runtime that uses U Tokyo's MassiveThreads library - network-based atomic variables for users of the Chapel module on Cray systems - improved support for the Tilera chip architecture Environment Changes ------------------- - made compiler infer CHPL_HOME if unset, or complain if mis-set - made util/setchplenv.* scripts more cleanly re-runnable - added new platforms: 'cray-cascade' and 'cray-xk' (see doc/platforms/README.cray) - renamed 'xe-cle' and 'xt-cle' to 'cray-xe' and 'cray-xt', respectively - renamed Cray PrgEnv compiler settings to be more platform-independent (e.g., 'cray-xe-cray'/'cray-xe-gnu' -> 'cray-prgenv-cray'/'cray-prgenv-gnu') - made CHPL_MEM default to 'tcmalloc' when CHPL_COMM is 'ugni' - made CHPL_ATOMICS default to 'intrinsics' for CHPL_TARGET_COMPILER=PrgEnv-gnu Syntactic/Naming Changes ------------------------ - changed domain literals to use curly brackets rather than square brackets (e.g., '[1..n, 1..n]' would now be written '{1..n, 1..n}') - introduced array literals using square brackets (e.g., '[3, 5, 7, 11]' describes an array of type '[1..4] int') - added a 'zip' keyword to express zippered iterations rather than using tuples (e.g., 'for (a,b) in (A,B)' is now written 'for (a,b) in zip(A,B)') - added a bidirectional I/O operator (<~>) for reading from/writing to channels (e.g., 'myChannel <~> myStr;' will read/write 'myStr' from/to 'myChannel') - added support for creating one-tuples using the syntax (x,) (e.g., '(1.0,)' describes a tuple of type '1*real') - re-established the underscore as the means of ignoring a tuple component (e.g., '(x,,z)' would now be written '(x,_,z)') Semantic Changes / Changes to Chapel Language --------------------------------------------- - added support for 'ref' intents, supporting pass-by-reference arguments (e.g., 'proc foo(ref x: int) ...' passes an int arg to 'foo' by reference) - improved support for atomic variables - added support for floating point atomic types (e.g., 'var x:atomic real;') - added waitFor(val) method (e.g., 'x.waitFor(1.0)' would wait for x to hold the value 1.0) - added non-fetching variations of operations to avoid excess communication (e.g., 'x.add(1)' is like 'x.fetchAdd(1)' but doesn't return a value) - added compareExchange() method as a shorthand for compareExchangeStrong() - added support for user-defined overloads of '<=>' and 'op=' assignments (e.g., 'proc <=>(ref x: t, ref y: t) ...' or 'proc +=(ref lhs:t, rhs:t) ...') - added a '.size' method to arrays, domains, strings, and ranges - made enum types behave more similarly to param ints (or collections thereof) - removed certain expression forms as being valid types or standalone stmts Newly Implemented Features -------------------------- - added a prototype implementation of a code-based documentation feature (see doc/technotes/README.chpldoc) - added an LLVM-based back-end (see doc/technotes/README.llvm) - added support for degenerate reindexing of non-rectangular arrays Deprecated Features ------------------- - removed the '_extern' keyword; use 'extern' instead - removed the ability to use blanks to drop tuple components -- use '_' instead Standard Modules ---------------- - made Dimensional Block Cyclic distributions support non-default index types - added HeapSort() to the standard Sort module - added the ability to clear() timers without stopping them Documentation ------------- - added README.llvm to describe new LLVM back-end option (see doc/technotes/README.llvm) - added README.chpldoc to describe new chpldoc documentation option (see doc/technotes/README.chpldoc) - updated README.atomics to describe new capabilities for atomic variables - added doc/platforms/README.tilera to describe using Chapel with Tilera chips - noted that Python and bc are now required features to build Chapel (see doc/README.prereqs and doc/platforms/README.cygwin) - updated README.extern to utilize 'ref' intents rather than 'inout' - updated language specification to describe new features and syntax changes - minor updates and fixes to several chapters of the language specification - updated the Quick Reference to illustrate new syntax, features - improved command-line help for -d option - minor updates to several READMEs Example Codes ------------- - extensive revisions to SSCA#2 including: - extensive performance and memory improvements - improved ability to 'make' with different options from the command line - changed the neighbor list representation to use 1D arrays of 2-tuples - added ability to read/write graphs from/to files in setting up - added support for distributed/parallel graph construction - fixed a bug in the TEPs computation when starting from a subset of vertices - added the ability to skip various kernels via configs - optimized and fixed a bug in the generation of filtered edge lists - converted uses of sync variables to atomic variables - improvements to RA: - extended ra.chpl to use either on-clauses or direct array accesses - made verification updates lossless using synchronization variables (see examples/benchmarks/hpcc/ra.chpl) - added a lossless version of RA that uses atomic variables (see examples/benchmarks/hpcc/ra-atomics.chpl) - improvements to LULESH - replaced 'sync' variables with 'atomic's - converted a global array into tuples to minimize communication - general code cleanup improvements - examples/primer/ updates: - added a new primer to demonstrate the --docs/chpldoc capability (see examples/primers/chpldoc.chpl) - updated atomic_vars.chpl primer to reflect recent changes - updated arrays.chpl primer to reflect array literal syntax - updated all examples to reflect syntax changes Platform-specific Notes ----------------------- - improved support for the Tilera chip architecture (see doc/platforms/README.tilera) - added support for Cray Gemini- and Aries-based atomics to the ugni comm layer (see doc/platforms/README.cray) Launcher-specific Notes ----------------------- - improved error handling of failed forked processes - for aprun-based launchers: - made '-cc none' the default mode used - added -j option to aprun (for supported versions) - for the pbs-aprun launcher: - fixed a timing bug - added a flag --generate-qsub-script Compiler Flags -------------- - added --docs and related flags for generating documentation for Chapel code (see doc/technotes/README.chpldoc) - added a --gen-extern-prototypes flag to cross-verify extern prototypes - added an --llvm flag for targeting LLVM as the back-end compiler (see doc/technotes/README.llvm) Bug Fixes / New Semantic Checks (for old semantics) --------------------------------------------------- - fixed a bug with type-inferred module-level domains/arrays - fixed a bug relating to strings on distributed memory 32-bit platforms - improved support for uint idxTypes in domains and arrays - improved the ability to use user-defined constructors in derived classes - fixed a bug in which the block-cyclic distribution didn't support 1D domains - fixed several bugs in I/O: - one related to using stdin on Mac OS X - one related to bitwise writing - one related to writing to/reading from a Writer/Reader from a remote locale - one related to issues on Cygwin - fixed a bug in which classes were improperly considered unused and removed - fixed a bug in tcmalloc when computing reallocation growth and shrinkage Packaging Changes ----------------- - simplified the etc/emacs/ directory organization to avoid duplication (emacs users may need to update their .emacs settings -- see etc/emacs/README) - added a 'chpldoc' alias to the 'chpl' compiler to just generate documentation - added setchplenv.fish to the release (intended for 1.5.0, but mis-packaged) - made the subdirectories used to store object files more unique Performance Improvements ------------------------ - added a strided bulk communication optimization, disabled by default (to enable, compile with -s useBulkTransferStride) - replaced internal sync var counters with atomics when appropriate - made initialization of remote array access caches lazy to reduce setup time - removed a level of pointer indirection in the implementation of array data - reduced the runtime cost of creating array aliases - optimized reindexing for the case when the original and new domains match - optimized the performance of equality/inequality on identical domains - made most associative domain/array ops to occur on the owning locale - enabled remote-value forwarding of array descriptors in the presence of syncs Runtime Library Changes ----------------------- - added support for MassiveThreads-based tasking (courtesey U Tokyo) (see doc/README.tasks) - added the ability to implement atomic variables using network-based AMOs - made counters used for communication diagnostics 64 bits to avoid overflow - turned off the cooperative comm/task non-blocking get interface for GASNet Third-Party Software Changes ---------------------------- - updated Qthreads to version 1.8 - updated GASNet to version 1.18.2 - added MassiveThreads 0.3beta to the third-party directory - added a directory to download creoleparser into for use with chpldoc (see doc/technotes/README.chpldoc) Testing System -------------- - reordered the order in which execution options are passed to tests (see #comments at top of start_test for details) - added a 'lastexecopts' capability to require an execution option to come last - improved the 'timedexec' script to kill the child's process group Internal/Developer-oriented --------------------------- - significantly refactored the code generation pass - began the process of replacing homegrown ADTs with STL classes - removed pragma "inline" (use 'inline' keyword instead) - renamed some primitives: 'get ref' -> 'deref' and 'set ref' -> 'addr of' - fixed type-punned pointers in comm-gasnet.c - fixed a case of deleting global type symbols prematurely - fixed a use-before-def warning for default-initialized vars of extern type - added consistency checking to basic block analysis - made use of BLOCK_TYPE tag more consistent - reduced reliance on primitives and locale internals in standard modules ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_sfd2d_oct _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2012-10-15 22:49:31
|
Hi Chapel Community -- As in past years, there will be a PGAS (Partitioned Global Address Space) language/technology booth on the SC12 floor, providing a forum for attendees to learn about UPC, CAF, SHMEM, Chapel, and the like. If you are doing Chapel-related work that you would like to highlight on the SC12 floor, you can propose to create a 22" x 36" poster for inclusion in the booth. Since there is limited space, the PGAS booth organizers will select from the submissions to curate a set of material that spans the PGAS space. Based on past years, I would guess that there could be 1-2 slots for Chapel posters. Each accepted poster is expected to have someone at the PGAS booth for at least one 2-hour shift sometime during the week to answer questions about the work (and more generally to help keep the booth staffed). If you are interested in participating, please reply to me to indicate your interest and include the information requested below by noon Pacific Time on Thursday October 18th. I'll then forward all proposed Chapel posters to the PGAS booth organizers for consideration. You should expect to receive a decision back by Monday the 22nd. For each poster please provide a Title, Point-of-contact email, and brief abstract. If the poster is based on a recent publication, a link to the paper can substitute for emailing the abstract. The abstract will be used by the booth organizers to select posters. For reference, the preferred poster template is available here: https://sites.google.com/a/lbl.gov/pgas-booth-at-sc12/documents/SC11_PGASPosterTemplate_22x36.ppt?attredirects=0 Thanks, -Brad |
From: Sung-Eun C. <su...@cr...> - 2012-10-04 21:32:03
|
Hello Chapel community, Just wanted to remind everyone that the deadline for submitting proposals for the Chapel Lightning Talks session at SC12 is tomorrow, October 5. Please see the original message below for details. Last year's session was great fun, and we're hoping for a similar turn out this year. Hope to see you all in Salt Lake City! -- Sung |
From: Brad C. <br...@cr...> - 2012-09-19 00:42:55
|
Hi Chapel community -- A few quick reminders about upcoming deadlines: * BLURBS: tomorrow is the last day to get blurbs in to me for the first talk in which I'll use them (blurbs coming later may be used in subsequent talks). (for details, see https://sourceforge.net/mailarchive/forum.php?thread_name=alpine.LNX.2.00.1209140916220.32669%40bradc-lnx.us.cray.com&forum_name=chapel-users) * LIGHTNING TALKS: the deadline for proposals for the Chapel Lightning Talks BoF at SC12 is coming up fast -- Oct 5th. (for details, see http://sourceforge.net/mailarchive/forum.php?thread_name=alpine.LNX.2.00.1209111504110.16877%40bradc-lnx.us.cray.com&forum_name=chapel-announce) Thanks! -Brad |
From: Brad C. <br...@cr...> - 2012-09-14 20:12:14
|
Hi Chapel Community -- As we enter the final weeks of the HPCS program, I'm working on putting together some talks to summarize the Chapel project under HPCS. It occurred to me this morning that an interesting slide for these talks would be one containing testimonials (or "blurbs" in the publishing world) from members of the community summarizing their (hopefully positive) reactions to the Chapel language, compiler, and project thus far. (As always, if your reactions are primarily negative, you're encouraged to let us know about those as well :). If you'd be interested in contributing a blurb for consideration, please reply directly to me to spare everyone else the mail. I can collate the responses we receive into a summary email if there's sufficient interest. The first of these talks is due next week, so quick responses are appreciated, but there will also be other talks, so send them whenever... A couple of notes: * Don't feel any need to overstate Chapel's case -- we're under no illusion that Chapel is complete or perfect (yet!) and don't think it's to the project's benefit to exaggerate. * Please let us know how you'd like to be credited/cited (name, title, institution, etc.) Thanks, -Brad PS -- Please feel free to forward this on to anyone else who's likely to be a good source for a Chapel blurb yet is unlikely to be on these lists or paying attention to them. PPS -- To address the obvious question: though HPCS is coming to a close, we fully intend to keep the Chapel project going, at least as long as the community has sufficient interest and we are able to find sufficient funding. |
From: Brad C. <br...@cr...> - 2012-09-11 22:11:51
|
Hi Chapel community -- We're happy to announce that we've had a Birds-of-a-Feather (BoF) session accepted for SC12 entitled "Chapel Lightning Talks 2012". Last year's lightning talks session was very well-attended and fun, and we hope to repeat its success this year. If you will be attending SC12 and have Chapel-related work that you would like to highlight there, please consider proposing a talk for this session (scheduled for Wednesday, November 14th, from 12:15-1:15 -- the official description follows at the end of this mail). Our goal for the session is to generate increased interest in the Chapel language within the HPC community by covering a diverse set of topics from within the broader Chapel community: research collaborations, use of Chapel in education, application studies, supporting technologies, etc. We hope to have all-new content this year, though last year's speakers are definitely encouraged to submit a talk that focuses on new work. We chose the lightning talk format in order to cover a broad spectrum of activities and to keep the session quick and lively. To that end, we anticipate having ~6 talks @ 5 minutes each as the core of the session and will be striving to select engaging topics and speakers. We'll start with a brief overview of the Chapel project to provide context for the talks and will wrap up with a Q&A session across all talks. To propose a talk, please send the following to cha...@cr... by October 5th: * proposed title * a paragraph summarizing what you'd cover in your 5 minutes * a short speaker bio (1 short paragraph) * (optional, but recommended) sample slides from previous talks demonstrating that you can create engaging slides; or that you've presented a short, lightning-style talk before. (Note that these can be from any technical talk or lecture; they don't need to be related to your proposed topic in any way). Key Dates: * Oct 5th: Proposals due * Oct 15th: Speakers notified of decisions * Oct 31st: Speakers submit draft slides for feedback Please note that in order to participate in the BoF, you'll need to have the appropriate SC12 registration (this year "exhibits-only" badges won't be sufficient for BoF attendance). We're looking forward to this BOF and hope you'll submit something and/or join us to hear the talks! -Brad (on behalf of the Chapel team) ----------------------------------------------- Chapel Lightning Talks 2012 ======== Abstract, i.e., announcement Are you a scientist considering a modern high-level language for your research? Are you a language enthusiast who wants to stay on top of new developments? Are you wondering what the future of Chapel looks like after the end of the DARPA HPCS program? Then this is the BOF for you! In this BOF, we will hear "lightning talks" on community activities involving Chapel. We will begin with a talk on the state of the Chapel project, followed by a series of talks from the broad Chapel community, wrapping up with Q&A and discussion. ======== BOF Proposal In this BOF, we will showcase the work being done throughout the broad Chapel community via a series of short lighting talks. The purpose of this BOF is to bring together the Chapel community and would-be enthusiasts to share results, foster collaboration, and spark new research ideas. Following the format of last year's very successful Chapel Lightning Talks BOF, we will begin with a lighting talk from from a Chapel team member on the state of Chapel and its future as the DARPA HPCS program ends. We will follow this with a series of talks on current work being done using Chapel. We will allot 20 minutes at the end of the session for Q&A involving all participants and audience members. ======== Description of the session format As with last year's BOF, we expect to have about 6 short talks, with the remaining time going to speaker transition and Q&A. Each speaker will submit a slide deck to the organizers 2 weeks prior to the start of SC12. We will work with the speakers to ensure that their talks will fit into the allotted time, and collect the final slide decks on to a single computer to minimize speaker transition times. The audience will be asked to save questions for the Q&A session, at which point all speakers may be addressed. |
From: Brad C. <br...@cr...> - 2012-04-20 00:03:44
|
Chapel commmunity -- Cray Inc. and the Chapel developer community are pleased to announce the release of version 1.5.0 of Chapel. This release's highlights include: * A complete revamping and re-implementation of Chapel's I/O features with the aim of making them more general, robust, parallel-capable, and practical. For details, see examples/primers/fileIO.chpl, doc/technotes/README.io, and the I/O chapter of the language specification. * Support for 'atomic' bool, int, and uint variables featuring operations such as reads, writes, compare-and-exchanges, fetch-ops, test-and-set, ... For details, see examples/primers/atomics.chpl and doc/technotes/README.atomics * A Chapel version of the LLNL/DARPA LULESH benchmark for Unstructured Lagrangian Explicit Shock Hydrodynamics. For more details, see examples/benchmarks/lulesh/ * Improved GMP support for integer types. For more details, see doc/technotes/README.gmp * An initial implementation of a 2D Dimensional distribution permitting per-dimension recipes for domain/array distribution. See examples/primers/distributions.chpl and examples/benchmarks/hpcc/hpl.chpl for example use cases. * Optimized tasking/threading/communication runtime layers for users of the Chapel module on Cray XE (TM) systems. For more details, users of the Chapel module on such systems should refer to $CHPL_HOME/doc/platforms/README.xe-cle-system-specific-layers (NOTE: We regret that the module version of this release has been delayed for a week or so due to last-minute packaging problems. Sysadmins of Cray systems should receive notification when it becomes available). * An updated version of the GASNet library, which includes initial support for Cray XE systems via its 'gemini' conduit. * Changed integer and index types to default to 64 bits in most contexts rather than the previous 32 bits. ...and much more! See below for a more complete list of changes in version 1.5.0, or refer to $CHPL_HOME/CHANGES within the release itself. Contributors to this release include: Brad Chamberlain Sung-Eun Choi Joel Denny Martha Dumler Michael Ferguson Roald Frederickx Tom Hildebrandt Vassily Litvinov Greg Titus To download the release, visit our SourceForge page at: http://sourceforge.net/projects/chapel At this site, you can also browse the mailing list archives and track our progress day-by-day. Our main project page continues to be hosted at: http://chapel.cray.com and it remains the best place to browse papers, presentations, documents, and tutorials, or to read about collaborations with researchers and educators. As always, we're interested in your feedback on how we can make the Chapel language and implementation more useful to you. On behalf of the Chapel Team, -Brad Chamberlain ============= version 1.5.0 ============= Eighth public release of Chapel, April 19, 2012 Highlights (see entries in subsequent categories for details) ------------------------------------------------------------- - completely revamped and improved Chapel's I/O features - added support for atomic bool/int/uint variables to Chapel - added a version of the LLNL/DARPA LULESH benchmark - improved GMP support for integer types - added an initial implementation of a 2D Dimensional distribution - added Cray XE (TM)-specific tasking/thread/communication layers for the Cray XE module (binary only) - updated GASNet from 1.16.2 to 1.18.0 which includes initial ugni support - made integers default to 64-bits in most contexts rather than the previous 32-bits - numerous performance improvements, bug fixes, and documentation improvements Environment Changes ------------------- - added support for fish shell users (http://fishshell.com) (see $CHPL_HOME/README and $CHPL_HOME/util/setchplenv.fish) Semantic Changes / Changes to Chapel Language --------------------------------------------- - completely revamped and improved Chapel's I/O features (see examples/primers/fileIO.chpl, doc/technotes/README.io, I/O spec chapter) - added support for atomic bool/int/uint variables to Chapel (see doc/technotes/README.atomics, examples/primers/atomics.chpl) - added support for /* ... */-style comments to be nested (e.g., /* in comment! /* in nested comment! */ still in comment! */) - made integers default to 64-bits in most contexts rather than the previous 32-bits - integer literals are now of type int(64)/uint(64) by default - the default 'int'/'uint' types now correspond to int(64)/uint(64) - the default idxType for ranges, domains, and arrays is now int(64) - a number of locale-oriented capabilities now use/return 64-bit ints: - numLocales, LocaleSpace, locale.id - locale.numCores, locale.callStackSize - numThreadsPerLocale, maxThreadsPerLocale - locale.*Threads(), locale.*tasks() - the 'by' and '#' operators now accept int(w)/uint(w) args for w-bit idxTypes - param arguments now play a role in disambiguation between function overloads (see the 'Function Resolution' section in the Chapel language specification) Newly Implemented Features -------------------------- - added an initial implementation of a 2D Dimensional distribution (see examples/primers/distributions.chpl) - added support for reading arrays, domains, ranges, and tuples (e.g., var A:[1..10] real; read(A); ) - added support for an iterator's arguments to have intents - added a new getCommDiagnosticsHere() routine for querying local communications (see doc/technotes/README.comm-diagnostics) Deprecated Features ------------------- - removed support for tensor iteration from the Chapel compiler (e.g., 'forall (i,j) in [foo(), bar()]' no longer supported; use nested loop) - removed volatile types in favor of atomic types - removed the ability to use 'def' to define functions; use 'proc' or 'iter' now Standard Modules ---------------- - improved GMP support for integer types (see doc/technotes/README.gmp) - changed sgn() in Math.chpl to return an int(8)/uint(8) Documentation ------------- - major updates to the I/O chapter of the spec to reflect new capabilities - added a new technical note describing support for atomic types (doc/technotes/README.atomics) - revamped technical notes for I/O and GMP to reflect revamped features (doc/technotes/README.io, doc/technotes/README.gmp) - minor updates to several other chapters of the language specification - minor fixes and updates to several READMEs Example Codes ------------- - added a version of the LLNL/DARPA LULESH benchmark (see examples/benchmarks/lulesh/README) - numerous updates to the Chapel version of SSCA#2 - store visited children in forward pass to avoid recomputing in backwards - use hand-coded task-private variables to store private variables - replaced arrays of sync vars with arrays of atomics in most instances - fixed a number of on-clauses that referred to a poor choice of locale (see examples/benchmarks/ssca2/README) - updated hpcc/hpl.chpl to use Dimensional distributions and replicated storage - moved ssca2 and hpcc examples into a new benchmarks/ subdirectory - fixed some bugs in hpcc/hpl.chpl - in ra.chpl, copied manual optimizations from timed loop into verification loop - updated the arrays.chpl primer to deprecate tensors and improve explanations Platform-specific Notes ----------------------- - added Cray XE-specific tasking/thread/communication layers to the Cray XE module (available using 'module load chapel', see doc/platforms/README.xe-cle-system-specific-layers) - fixed GASNet executions on Mac OS X Lion - generated C identifier lengths are now capped to avoid PGI warnings Compiler Flags -------------- - added --max-c-ident-len to cap the length of C identifiers in generated code Error Message Improvements -------------------------- - in most contexts, the compiler now prints out a fully-specified type (e.g., where before it printed 'int', it now prints 'int(64)') - line number reporting has been improved in certain cases Bug Fixes / New Semantic Checks (for old semantics) --------------------------------------------------- - added a compile-time error for incorrect domain slicing cases - fixed an issue in which users couldn't use 'this' as an identifier - fixed the ability to delete an object located on a remote locale - fixed a non-deterministic seg fault for parent classes with types/params only - fixed a race condition in the implementation of halt() - fixed an issue in which a .exe filename could not be specified using -o - the --explain-calls flag now properly displays 'param' intents as such - protected a user's ability to name variables j0, j1, jn, y0, yn Performance Improvements ------------------------ - changed array/domain reference counting to use atomics rather than syncs - made Chapel allocation/free routines inline to reduce overheads - optimized local on-clauses to reduce overheads by avoiding chpl_comm_fork() - optimized bulk transfers for Block-distributed arrays that share a domain - optimized the creation of simple Block- and Cyclic- array aliases Runtime Library Changes ----------------------- - no longer count communication threads against numThreadsPerLocale for pthreads - removed assumption that globals in generated code will have identical addrs - added support for keeping builds for multiple GASNet segments in one tree - moved most Chapel module initialization from main() to first task - simplified the tcmalloc memory layer to support use with GASNet - added dlmalloc support for communication layers other than GASNet-fast - broke GASNet AM_signal handler request/reply to use short messages - fixed a bug in which non-blocking gets could seg-fault - simplified the initialization and shutdown of the communication layer - fixed a potential race in broadcasts at program startup Third-Party Software Changes ---------------------------- - updated GASNet from 1.16.2 to 1.18.0 which includes initial ugni support - made third-party packages build automatically when needed by the runtime - updated Qthreads from version 1.7.0 to 1.7.1 - added support for a UTF-8 decoder from Bjoern Hoehrmann for use in the new I/O - added support for Nanos++ tasking to yield when firing off an active message - added support for the 'serial' statement to Nanos++ tasking layer - turned off tcmalloc's tendency to print messages for large allocations Testing System -------------- - added the ability for a skipf file to check for substrings in an env. variable - fixed several bugs in the testing system: - when using named .good files in compopts files - when using multiple compopts files and no execopts files - in the generation of perforamnce data filenames - condensed how compilation/execution options are identified in testing logs - added a --logtmp flag to start_test to support single-node parallelism Internal/Developer-oriented --------------------------- - fixed an issue in which internal modules resulted in too many circularities - changed module initialization code so that modules initialize those they use - added a pragma to control implicit inclusion of ChapelStandard.chpl - added support for automatically mapping external C types to Chapel types - added the ability to dump the compiler IR using a text-file format - added filenames back into the BaseAST class to preserve info during inlining - added chpl__unsignedType(), chpl__maxIntTypeSameSign() utility functions - added a typeToString() overload to catch common error cases - removed debugWrite* capability - fixed a bug in the compiler's basic block analysis framework - made existing tests much less sensitive to ordering of output - added a --[no-]preserve-inlined-line-numbers flag to control inlining behavior - added a "no instantiation limit" pragma/flag to the compiler - added an isParameter() method to the ArgSymbol class - removed strcmps against types to normalize scalar type queries (e.g., int(?w)) - made several refactorings to functionResolution.cpp - refactored common flag queries into helper query functions - refactored creation of wrapper functions - fixed get_int()/get_uint() to avoid reliance on underspecified C int sizes - made external classes less like Chapel classes within the Chapel compiler - cleaned up the memory tracking enums associated with the communication layer - removed chpl_error_noexit primitive - added code to trace through the disambiguation logic of function resolution - refactored Makefiles, replacing CHAPEL_ROOT with CHPL_MAKE_HOME - made third-party/*/Makefile rules more consistent ------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2 _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2012-03-26 17:32:08
|
Hi David -- Thanks for sending this out to the list. I'd be interested in pointing to it from the Chapel education webpage, would that be OK? I haven't had a chance to read through it in detail (and am not sure when I will), but in a quick glance wanted to mention that we've deprecated tensor iteration (in the spec as of the last release; in the implementation as of the upcoming one). There were a number of reasons for this, but the main ones are arguably that the syntax was confusing (and we'd like to co-opt it for something else down the line), and tensors can typically be implemented using nested loops. We've also discussed what it would take to create a tensor iterator in the language, which would be preferable to having it embedded in the syntax. (Ultimately, we intend to pull zippered iteration out of the syntax as well, but are not there yet). Thanks, -Brad On Tue, 20 Mar 2012, David Bunde wrote: > Hi all, > > Just writing to let you know that my students and I have created a > Chapel tutorial (actually a mix of tutorial and reference- we can > probably improve the focus). It's available at > > http://faculty.knox.edu/dbunde/teaching/chapel/ > > I hope that it's useful. The idea was to be between Kyle's short > tutorial (http://www4.wittenberg.edu/academics/mathcomp/kburke/chapelTutorial.html) > and the long presentations on the official Chapel webpages > (http://chapel.cray.com/tutorials.html). The intended audience is > undergrads. As with all webpages, this one is still being worked on. > In particular, I hope to add something about reductions within the > next week or so. If you have any other comments, I'd love to hear > them. > > - David Bunde > Knox College > > ------------------------------------------------------------------------------ > This SF email is sponsosred by: > Try Windows Azure free for 90 days Click Here > http://p.sf.net/sfu/sfd2d-msazure > _______________________________________________ > Chapel-education mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chapel-education > |
From: David B. <db...@kn...> - 2012-03-20 16:24:06
|
Hi all, Just writing to let you know that my students and I have created a Chapel tutorial (actually a mix of tutorial and reference- we can probably improve the focus). It's available at http://faculty.knox.edu/dbunde/teaching/chapel/ I hope that it's useful. The idea was to be between Kyle's short tutorial (http://www4.wittenberg.edu/academics/mathcomp/kburke/chapelTutorial.html) and the long presentations on the official Chapel webpages (http://chapel.cray.com/tutorials.html). The intended audience is undergrads. As with all webpages, this one is still being worked on. In particular, I hope to add something about reductions within the next week or so. If you have any other comments, I'd love to hear them. - David Bunde Knox College |
From: Brad C. <br...@cr...> - 2011-12-08 05:34:00
|
Hi Chapel community -- Three topics in this mail: Chapel's license, new Chapel job listings, and an SC11 report. LICENSE: We are planning to move Chapel from the BSD license that it's currently released under to the Apache license, primarily to move to a more standarized contributor agreement than our current one. If this causes anyone any concerns, please let me know. Here are links to the license and contributor agreements for reference: License: http://www.apache.org/licenses/LICENSE-2.0 Contributor Agreements: individual: http://www.apache.org/licenses/icla.txt corporate: http://www.apache.org/licenses/cla-corporate.txt JOB LISTINGS: The Chapel team at Cray Inc. has three positions open currently, one for a team manager, two for software developers. If you or someone you know would be appropraite for these positions, please send a resume our way. See http://chapel.cray.com/jobs.html for more detail and instructions for finding the official job listings. SC11 REPORT: Chapel had a busy and successful work at SC11 this year -- thanks to everyone who helped make it so. * The Monday and Friday tutorials drew interested and attentive audiences with lots of good questions. * We were happy to see several of you at the 2nd annual CHUG happy hour and hope to make this an ongoing annual event. (and one that eventually scares vendors from holding their parties that night for fear nobody will show). * Our Chapel entry won the "most elegant" award again in the class 2 HPC Challenge competition -- see hpcchallenge.org for more details. (Now to nail down the performance and take the whole prize!) * The Chapel Lightning Talks BoF was a great new event that highlighted work from the Chapel community outside of Cray and drew a strong audience in spite of being scheduled simultaneously with the PGAS BoF. * We handed out dozens of USB sticks throughout the week sporting the new Chapel logo and containing recent publications and presentations. (Hopefully you got yours before they climb to $100+ on ebay...) Thanks for your ongoing interest in Chapel, -Brad ------------------------------------------------------------------------------ Cloud Services Checklist: Pricing and Packaging Optimization This white paper is intended to serve as a reference, checklist and point of discussion for anyone considering optimizing the pricing and packaging model of a cloud services business. Read Now! http://www.accelacomm.com/jaw/sfnl/114/51491232/ _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2011-11-09 23:56:27
|
Chapel Community -- Here's an update on Chapel-related activities at SC11, in approximate priority order. More specific times and places for all events is available at: http://chapel.cray.com/events.html * The 2nd annual CHUG (Chapel User's Group) Happy Hour: will be Monday evening at the Pike Brewing Company from 5:30-7:30 (or later if we're still standing). If you're receiving a copy of this mail, you should consider yourself a CHUG member and come join us. * Chapel Lightning Talks BoF: will be Wednesday at 12:15pm and will feature talks from the broad Chapel community on I/O, Chapel in education, lightweight tasking, GPU computing, and interoperability. Q&A to follow. * Chapel Tutorials: are all-day on Monday or a half-day on Friday. Note that attending tutorials requires paying an additional registration fee. The Friday version is intended for a broader (not-necessarily-HPC) audience and will be somewhat abridged, yet is also cheaper, esp. if you are already attending T-Th. * HPC Challenge BoF: we were selected as finalists in the annual HPC Challenge competition for the "most productive" award and will be giving a 10-minute presentation on our entry, vying against the other finalists for cash and honor. (Wednesday at 12:15). * talk to a live Chapel team member: throughout the week, we'll be taking turns helping staff the PGAS booth (#124). See the webpage above for a schedule of times/days. * Punctuated Equilibrium at Exascale Panel/BoF: I'll be giving a short talk at this panel arguing for looking at more revolutionary programming models as we consider the transition to exascale computing (Thursday at 5:30pm). * SWAG! While supplies last, we'll be handing out a USB stick containing the new 1.4 release, slides, recent papers, etc. This will be the first time the new Chapel logo has appeared on a physical item that we're aware of. Beat the eBay rush! We hope to see many of you at these events and around the show next week! -Brad |
From: Brad C. <br...@cr...> - 2011-10-21 00:47:43
|
Chapel users -- Cray Inc. and the Chapel team are pleased to announce the release of version 1.4.0 of the Chapel compiler. This release's highlights include: * Improved support for the Qthreads and Nanos++ user-level tasking layers, including support for multi-locale execution. See $CHPL_HOME/doc/README.tasks for more information. * The addition of dynamically-scheduled range iterators (similar to those supported by OpenMP's loop schedules). See the 'AdvancedIters' section of the 'Standard Modules' section of the language specification for details. * The addition of an 'export' keyword for marking symbols intended for use outside of the Chapel code (e.g., export'ed functions can be called from an external C routine). In addition, the '_extern' keyword has been promoted to a full-fledged keyword, 'extern'. * New communication optimizations targeted at reducing the number of data transfers required for whole-array assignments and remote array accesses. This optimization also implements local whole-array assignments using memcpy()s rather than element-by-element copying. * Improvements to the range type's semantics and implementation. * The addition of an 'inline' keyword to assert that a procedure should be inlined at its callsites. * Newly-added support for the # operator on dense rectangular domains and arrays. * Associative domains now support parallel-safe additions/removals of indices. * New --static/--dynamic flags to specify linkage. See the man page for details. * New example codes: SSCA2 (an unstructured graph benchmark), and primer examples for sync/single variables and leader-follower iterators. * Updated and improved documentation. * Many bug fixes. ...and much more! See below for a more complete list of changes in version 1.4.0, or refer to $CHPL_HOME/CHANGES within the release itself. To download the new release, visit our SourceForge page at: http://sourceforge.net/projects/chapel Here, you can also browse the mailing list archives and track our progress day-by-day. Our main project page continues to be hosted at: http://chapel.cray.com and it remains the best place to browse papers and presentations, download the language spec, or read about our collaborations. (A number of updates and improvements have been made to it today in support of the release). As always, we're interested in your feedback on how we can make the Chapel language or compiler more useful to you. On behalf of the Chapel Team, -Brad Chamberlain PS -- As mentioned previously, we will be hosting Chapel tutorials and a BoF at SC11 in Seattle this year. Please join us or recommend them to colleagues that you think would be interested in attending. See http://chapel.cray.com/events.html for details. ============= version 1.4.0 ============= Seventh public release of Chapel, October 20, 2011 Highlights (see below for details) ---------------------------------- - added 'extern', 'export', and 'inline' keywords - new optimizations for whole-array assignments and remote array accesses - improved support for Qthreads and Nanos++ tasking, including multi-locale - added dynamic range iterators via an 'AdvancedIters' module - continued improving range semantics and implementation - added support for # operator on dense rectangular domains and arrays - made associative domain index modifications parallel-safe - added --static/--dynamic linking flags - improved communication diagnostics capabilities - new example codes: SSCA2; primers on sync vars, leader-follower iterators - updated documentation - many bug fixes Syntactic/Naming Changes ------------------------ - renamed '_extern' keyword to 'extern' - renamed 'maxThreadsPerLocale' to 'numThreadsPerLocale' (see doc/README.tasks and doc/README.executing for new definitions) - made it illegal for a string literal to contain an unescaped newline - renamed dsiIndexLocale to dsiIndexToLocale in domain map standard interface (see doc/technotes/README.dsi) - reserved 'ref' as a keyword for future use - renamed 'iterator' enumeration to 'iterKind' for leader-follower iterators - renamed 'follower' argument to follower iterators 'followThis' Semantic Changes / Changes to Chapel Language --------------------------------------------- - added 'export' keyword to support calling Chapel functions from C - added 'inline' keyword to force a function to be inlined - added support to # operator for dense rectangular domains and arrays - removed support for tensor iteration from the language definition - removed the realm concept from the language - ambiguous ranges are now == if their representations match - disabled promotion of the '..' operator Newly Implemented Features -------------------------- - improved qthreads and nanox task layers to support multilocale execution (see doc/README.tasks) - adding and removing indices from associative domains is now parallel-safe - updated the communication diagnostics interface (see doc/technotes/README.comm-diagnostics) Standard Modules ---------------- - added an 'AdvancedIters' module with various dynamic scheduling algorithms - added floor/ceiling/mod functions for integers to 'Math' module - added new sorting routines to standard 'Sort' module - added a 'UtilMath' module with other integer ceiling/floor functions - added a param version of abs() for integers to 'Math' module Documentation ------------- - spec updates: - added a new chapter on 'Interoperability' - refreshed 'Data Parallelism', 'Task Parallelism', and 'Conversions' chapters - updated 'Ranges' chapter significantly - minor updates to 'Domains', 'Arrays', and 'Classes' chapters - split 'Memory Consistency Model' into its own chapter (still poorly defined) - documented new Advanced Iterators and Sort routines in 'Standard Modules' - added documentation about the compiler's module search path to man page/technotes (see 'man chpl' and doc/technotes/README.module_search) - added note about parallel safety to doc/technotes/README.extern - generally refreshed README* and other text-based documentation files Example Codes ------------- - added a(n in-progress) implementation of SSCA2 to the release examples (see examples/ssca2) - added a primer for sync and single variables (see examples/primers/syncsingle.chpl) - added a primer on leader-follower iterators (user-defined forall scheduling) (see examples/primers/leaderfollower.chpl) - added an example of creating a private locale view to distributions primer (see examples/primers/distributions.chpl) - improved the ranges primer (see examples/primers/ranges.chpl) - added a description of subdomains to the domains primer (see examples/primers/domains.chpl) - minor wordsmithing for the procedures primer (see examples/primers/procedures.chpl) Platform-specific Notes ----------------------- - improved Makefile settings for IBM/AIX platforms and compilers Feature Improvements -------------------- - several improvements to the implementation of ranges (see 'Ranges' chapter of spec for details) - made casts-to-string produce the same string as write() does Compiler Flags -------------- - added --static/--dynamic flags to specify how the binary should be linked Interoperability Changes ------------------------ - made 'extern' a keyword; '_extern' is deprecated Error Message Improvements -------------------------- - added a compiler warning for param functions with declared return types - improved a few other difficult compiler error messages Bug Fixes / New Semantic Checks (for old semantics) --------------------------------------------------- - fixed the result widths of complex math operations - improved parsing of escaped characters in string literals - deleting a record now generates an error - added a variety of bug fixes for range operations - fixed a bug with passing strings to external functions for --no-local cases - fixed a bug for var functions that returned nothing - fixed a bug related to terminating a CHPL_COMM=gasnet program early - fixed a bug in which ref temps were not inserted for virtual method calls - fixed a bug related to removing a label from the AST but not its goto - fixed bug with applying serial statement to optimized coforall+on loops - fixed bugs regarding single variables not being treated like sync - fixed a bug related to incorrect complaints about ambiguous module paths - fixed a bug related to followers accepting empty ranges - fixed a few reference-after-deallocation bugs in the compiler - fixed a bug related to flattening var functions multiple times Packaging Changes ----------------- - removed the second-level 'chapel/' directory that releases used to have - updated emacs/vim modes to bring syntax up-to-date - util/setchplenv.* scripts now permit $CHPL_HOME to be other than 'chapel' - added the util/ directory to PATH in the setchplenv.* scripts - improved make 'clean' and 'clobber' rules for third-party directories Third-Party Software Changes ---------------------------- - removed requirement to specify LD_LIBRARY_PATHs for qthreads and nanox tasks - updated to GASNet version 1.16.2 (see third-party/gasnet/README) - updated Qthreads to version 1.7 (see third-party/qthreads/README) - added tcmalloc version 1.7 (see third-party/tcmalloc/README) - updated dlmalloc to version 2.8.5 Runtime Library Changes ----------------------- - split the 'none' task layer into 'none' and 'minimal' (see doc/README.tasks) - added a tcmalloc-based memory layer to the runtime - improved program initialization order and sequence - made nanox/qthreads task layers use a #threads==#cores by default - increased the default callstack size for qthreads tasks - removed pthread calls from the gasnet communication layer - began process of unifying runtime file and directory names - standardized our threading layer interface - made tasking layers more robust to early calls during program setup - moved Chapel code in the runtime/ directory into internal modules Compiler Analysis and Optimizations/Performance of Generated Code ----------------------------------------------------------------- - added a block transfer optimization for whole-array assignment - reduced the number of communications for remote Block or Cyclic array access - added an optimization to remove dead code blocks Testing System -------------- - valgrind testing with CHPL_COMM!=none now valgrinds the child processes Internal -------- - the threading layer of the runtime now has its own header file - added a --library flag for creation of a standalone Chapel library - made normalize_returns() able to be called multiple times - fixed a bug related to internal module dependences of depth > 2 - added capabilities to help debug deleted AST nodes (see --break-on-delete-id and --log-deleted-ids-to) - removed --runtime flag from compiler - added a compilerAssert() capability ------------------------------------------------------------------------------ The demand for IT networking professionals continues to grow, and the demand for specialized networking skills is growing even more rapidly. Take a complimentary Learning@Cisco Self-Assessment and learn about Cisco certifications, training, and career opportunities. http://p.sf.net/sfu/cisco-dev2dev _______________________________________________ Chapel-announce mailing list Cha...@li... https://lists.sourceforge.net/lists/listinfo/chapel-announce |
From: Brad C. <br...@cr...> - 2011-09-22 00:30:52
|
Hi Chapel Community -- We wanted to give everyone an early heads-up about some of the Chapel-oriented activities that will be taking place at SC11. Please feel encouraged to forward this information to colleagues and friends who might also be interested in participating in these sessions. * Chapel Lightning Talks BOF (Wednesday, November 16th) http://sc11.supercomputing.org/schedule/event_detail.php?evid=bof184 This Birds-of-a-Feather session will feature talks from members of the broader Chapel community about interesting work they're doing. We're currently accepting proposals for talks (due September 30th), so if you have an idea for one, send it in! See our earlier mail for more information: https://sourceforge.net/mailarchive/forum.php?thread_name=alpine.LNX.2.00.1109021747030.25056%40bradc-lnx.us.cray.com&forum_name=chapel-announce * Full-Day Chapel Tutorial (Monday, November 14th) http://sc11.supercomputing.org/schedule/event_detail.php?evid=tut133 This will be a full-day Chapel tutorial, designed for programmers who are familiar with parallel computing and want to learn about how to do it in Chapel. The tutorial will feature 4 hours of lecture and two 1-hour hands-on sessions with members of the Chapel team on-hand for Q&A. * Half-Day "Outreach" Chapel Tutorial (Friday, November 18th) http://sc11.supercomputing.org/schedule/event_detail.php?evid=tut170 As part of SC11's new Friday Outreach program, this tutorial will be a shorter and higher-level introduction to Chapel, assuming less familiarity with HPC programming models. It will consist of 3 hours of lecture including Q&A with members of the Chapel team. We anticipate that we'll have more events as we get closer to the date (like the second annual CHUG Happy Hour?), so stay tuned for more information, On behalf of the Chapel team, -Brad |
From: Brad C. <br...@cr...> - 2011-09-03 01:21:09
|
Hi Chapel community -- We've had a Birds-of-a-Feather (BoF) session accepted for SC11 in Seattle entitled "Chapel Lightning Talks". If you are doing Chapel-related work that you would like to highlight at SC11, please consider proposing a talk for this session (scheduled for Wednesday, November 16th, from 12:15-1:15 -- the official program description follows at the end of this mail). Our goal for the session is to generate increased interest in Chapel by highlighting related activities from the broader community: research collaborations, use of Chapel in education, code studies, supporting technologies, etc. We chose the lightning talk format in order to cover a broad spectrum of activities and to keep the session quick and lively. To that end, we anticipate having ~6 talks @ 5 minutes each as the core of the session and will be striving to select engaging topics and speakers. We'll start with a brief overview of the Chapel project to provide context for the talks and will wrap up with a Q&A session. To propose a talk, please send the following to cha...@cr...: * title * a paragraph summarizing what you'd cover in 5 minutes * speaker bio * (optional, but recommended) sample slides from previous talks demonstrating that you can create engaging slides; or that you've presented a short, sharp talk before. (Note that these can be from any technical talk or lecture; they don't need to be related to your proposed topic at all). Proposals should be sent to us by Septmeber 30th. We'll plan on announcing the speaker lineup by mid-October to give everyone a month to prepare (because we all know how long it takes to make a smashing 5-minute talk :). Please note that in order to participate in the BoF, you'll need to be registered for SC11 and that we will not be providing funding for travel or registration. We're looking forward to this BOF -- hope you'll submit something and/or join us! -Brad (on behalf of the Chapel team) ----------------------------------------------- Chapel Lightning Talks Are you a scientist wondering what it might be like to use a high-level language for your research programming? Are you an educator looking for better ways to teach parallel programming? Are you a language geek who wants to stay on top of new developments? Have you ever wanted to know more about the hype surrounding Chapel? Then this is the BOF for you! In this BOF, we will hear "lightning talks" on community activities involving Chapel, a new programming language being developed under the DARPA High Productivity Computer Systems Program (HPCS). We will begin the BOF with a brief overview of the Chapel project. Following this, we will turn the stage over to our users and collaborators. There will be a series of curated five minute talks from the broad Chapel community -- HPC programmers, educators, Chapel tinkerers, developers, etc. We will wrap up the BOF with Q&A and discussion involving all participants and audience members. |
From: Kyle B. <pai...@gm...> - 2011-07-08 17:52:11
|
FYI, I've updated the tutorial to use Chapel 1.3.0 (replaced 'def' with 'proc'). I do hope to add more, but that will have to happen another week. -Kyle On Thu, Jun 23, 2011 at 2:23 PM, Brad Chamberlain <br...@cr...> wrote: > > Hi David -- > > > Some students and I are learning Chapel this summer. We've primarily >> been using the SC tutorial and the URL that Kyle Burke mentioned >> (http://www4.wittenberg.edu/**academics/mathcomp/kburke/** >> chapelTutorial.html<http://www4.wittenberg.edu/academics/mathcomp/kburke/chapelTutorial.html> >> ) >> as resources. Our thinking at the moment is to make a tutorial >> somewhere between these- less comprehensive and aimed at less advanced >> students than the SC tutorial, but covering more of the features than >> Kyle's. >> > > w.r.t. this part of your question, a team at Cray has been working on a set > of slides with the notes section filled in for use with some summer interns > that have been learning Chapel. I think this amounts to something very > similar to the Chapel tutorial slides when someone is speaking to them, > though it provides more examples in some cases and maybe wanders less into > esoterics in other cases. I can see about getting you a copy if it sounds > like it would be of interest. It may not fit your "less advanced" > requirement... > > > > Also, what editors do people use with Chapel? I'm an emacs person and >> it's frustrating that emacs doesn't know how to handle indentation for >> Chapel. Putting it into c-mode helps, but this is still not perfect. >> I'm interested in writing a chpl-mode. Has anyone done this or found >> other programming tools that work well? >> > > Just as a follow-up to Sung's note, the current Chapel emacs mode is much > better than nothing (or C), but is somewhat frustrating in other cases. We > had someone contribute it awhile back, and it was much better than the > version we had attempted ourselves before that, but still leaves a lot to be > desired. Most of the challenges relate to Chapel's use of > keyword-/Modula-based declaration styles and keyword-based control flow for > single statements. > > -Brad > > ------------------------------------------------------------------------------ > Simplify data backup and recovery for your virtual environment with > vRanger. > Installation's a snap, and flexible recovery options mean your data is > safe, > secure and there when you need it. Data protection magic? > Nope - It's vRanger. Get your free trial download today. > http://p.sf.net/sfu/quest-sfdev2dev > _______________________________________________ > Chapel-education mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chapel-education > > -- Me: kb...@wi... (or pai...@gm...) http://www4.wittenberg.edu/academics/mathcomp/kburke/ Combinatorial Game Blog: http://combinatorialgametheory.blogspot.com/ |
From: Brad C. <br...@cr...> - 2011-06-23 18:23:12
|
Hi David -- > Some students and I are learning Chapel this summer. We've primarily > been using the SC tutorial and the URL that Kyle Burke mentioned > (http://www4.wittenberg.edu/academics/mathcomp/kburke/chapelTutorial.html) > as resources. Our thinking at the moment is to make a tutorial > somewhere between these- less comprehensive and aimed at less advanced > students than the SC tutorial, but covering more of the features than > Kyle's. w.r.t. this part of your question, a team at Cray has been working on a set of slides with the notes section filled in for use with some summer interns that have been learning Chapel. I think this amounts to something very similar to the Chapel tutorial slides when someone is speaking to them, though it provides more examples in some cases and maybe wanders less into esoterics in other cases. I can see about getting you a copy if it sounds like it would be of interest. It may not fit your "less advanced" requirement... > Also, what editors do people use with Chapel? I'm an emacs person and > it's frustrating that emacs doesn't know how to handle indentation for > Chapel. Putting it into c-mode helps, but this is still not perfect. > I'm interested in writing a chpl-mode. Has anyone done this or found > other programming tools that work well? Just as a follow-up to Sung's note, the current Chapel emacs mode is much better than nothing (or C), but is somewhat frustrating in other cases. We had someone contribute it awhile back, and it was much better than the version we had attempted ourselves before that, but still leaves a lot to be desired. Most of the challenges relate to Chapel's use of keyword-/Modula-based declaration styles and keyword-based control flow for single statements. -Brad |
From: Sung-Eun C. <su...@cr...> - 2011-06-23 16:10:07
|
Hi David, I'm not sure if you have found this yet, but in the release there is a sort of cheat sheet that you describe in $CHPL_HOME/doc/quickReference.pdf. I think it covers most of the bases, but if you have any suggestions for improvement, we'd be happy to hear them. As for editors, we have editor support for emacs and vim in $CHPL_HOME/etc (see the README). They aren't perfect, so feel free to submit improvements back to the team. Good luck.. -- Sung On Wed, Jun 22, 2011 at 10:47:01PM -0500, David Bunde wrote: > Hi all. > > I posed this question to a group of educators interested in Chapel and > it was suggested that I send it here instead. My apologies if you've > already received it. > > ------- > Some students and I are learning Chapel this summer. We've primarily > been using the SC tutorial and the URL that Kyle Burke mentioned > (http://www4.wittenberg.edu/academics/mathcomp/kburke/chapelTutorial.html) > as resources. Our thinking at the moment is to make a tutorial > somewhere between these- less comprehensive and aimed at less advanced > students than the SC tutorial, but covering more of the features than > Kyle's. I'd also like a "cheat sheet" that lists syntax since I have > to keep looking it up while writing programs. Has anyone developed > resources like these? I'd like not to duplicate effort if they've > already been written. > > Also, what editors do people use with Chapel? I'm an emacs person and > it's frustrating that emacs doesn't know how to handle indentation for > Chapel. Putting it into c-mode helps, but this is still not perfect. > I'm interested in writing a chpl-mode. Has anyone done this or found > other programming tools that work well? > > Thanks. I'll share materials we develop. > > - David Bunde > Knox College > > ------------------------------------------------------------------------------ > Simplify data backup and recovery for your virtual environment with vRanger. > Installation's a snap, and flexible recovery options mean your data is safe, > secure and there when you need it. Data protection magic? > Nope - It's vRanger. Get your free trial download today. > http://p.sf.net/sfu/quest-sfdev2dev > _______________________________________________ > Chapel-education mailing list > Cha...@li... > https://lists.sourceforge.net/lists/listinfo/chapel-education |
From: David B. <db...@kn...> - 2011-06-23 03:47:07
|
Hi all. I posed this question to a group of educators interested in Chapel and it was suggested that I send it here instead. My apologies if you've already received it. ------- Some students and I are learning Chapel this summer. We've primarily been using the SC tutorial and the URL that Kyle Burke mentioned (http://www4.wittenberg.edu/academics/mathcomp/kburke/chapelTutorial.html) as resources. Our thinking at the moment is to make a tutorial somewhere between these- less comprehensive and aimed at less advanced students than the SC tutorial, but covering more of the features than Kyle's. I'd also like a "cheat sheet" that lists syntax since I have to keep looking it up while writing programs. Has anyone developed resources like these? I'd like not to duplicate effort if they've already been written. Also, what editors do people use with Chapel? I'm an emacs person and it's frustrating that emacs doesn't know how to handle indentation for Chapel. Putting it into c-mode helps, but this is still not perfect. I'm interested in writing a chpl-mode. Has anyone done this or found other programming tools that work well? Thanks. I'll share materials we develop. - David Bunde Knox College |