sparky

16 ms·

91.
by sparky 15y ago
Most companies only make a small fraction of the tools you need for a full fab; for examples of heavy hitters in key areas like lithography, chemical mechanical polishing, and ion implant, see ASML, Nikon, Canon, Applied Materials. For any
92.
by sparky 15y ago
These days, it's not really up to a company whether it wants to be fabless or an IDM. Apart from the several billion dollars needed to build a competitive fab, you'll also need to contend with the 20+-year head start existing IDMs and foun
93.
by sparky 15y ago
I'm guessing he's referring to either the disk space needed to complete a build (~4GB on my machine) or the size of the generated binaries and libraries (304MB on my machine). In either case, I think your point holds that LLVM isn't substa
94.
by sparky 15y ago
Hypercube?
95.
by sparky 15y ago
A more charitable interpretation is "What do you guys need the most help with?" In that respect, the Github page is not immediately helpful.
96.
by sparky 15y ago
Hi all, We've just opened up a large fraction of the tools we've developed as part of the Rigel project[1] at UIUC. The goal of the project is to investigate several aspects of efficient many-core processors, including microarchitecture, ar
97.
1000-core Processor Simulator, Compiler, Libraries, and Applications(rigelproject.github.com)
16 points by sparky 15y ago|1 comments
98.
by sparky 15y ago
Did you try measuring the power at the wall end of the cord while the heater was running? Most space heaters aren't very smart, and it wouldn't surprise me if the draw varied wildly depending on the power supply and how warmed up the heate
99.
by sparky 15y ago
Thanks for the pointer, I look forward to playing around with wrpme. For context, my work on lock-free algorithms and data structures is in the context of processors with 128-1024 cores. I'm curious about your take on reference counting;
100.
by sparky 15y ago
Did what, exactly? The 'impossible' bit may be meant in more of a practical sense than theoretical, since obviously you can just implement your own garbage collector if it comes to that, but it's well-established in the lock-free algorithm
101.
by sparky 15y ago
Would be interesting to see how Julia perf compares to C++ compiled or JITed with LLVM via G++ 4.6+Dragonegg or Clang 3.0/svn. Would be more apples-to-apples, since both would use the same middle- and back-end. G++ 4.2.1 is a bit obsolete
102.
by sparky 15y ago
The last REPL I wrote used linenoise ( https://github.com/antirez/linenoise ) instead of readline, and it worked great for us. You may need more of readline's features than we did, but it's worth a look.
103.
by sparky 15y ago
LLVM has had a PowerPC backend for years http://llvm.org/docs/CodeGenerator.html#targetfeatures . Do you mean something more specific?
104.
by sparky 15y ago
I'm guessing the name is a play on 'swoll', which colloquially means 'muscular'. http://www.urbandictionary.com/define.php?term=swoll
105.
by sparky 15y ago
I think it's more directly derived from the French 'Mille Mille ', or 'Thousand Thousand ', used in finance.
106.
by sparky 15y ago
There's a good chance the LLVM backend will emit PTX, not machine code. PTX is well documented [1]. Under such a system, the generated PTX would be JITed at runtime by the driver. Note that LLVM already has a (very experimental and not co
107.
by sparky 15y ago
The first 10 amendments to the Constitution are known as the Bill of Rights.
108.
by sparky 15y ago
Fix SConscript.curl to look in /usr/lib32 for zlib (package name: lib32z1-dev ), and to pass -m32 to your compiler. /media/d2/sw/doom3.gpl/neo$ git diff diff --git a/neo/sys/scons/SConscript.curl b/neo/sys/scons/SConscript.curl inde
109.
by sparky 15y ago
Going lock-free has the benefit on any platform (well, any platform with atomic primitives) that you don't have to worry about a thread getting preempted while holding a lock and blocking others indefinitely. However, I think the scalabili
110.
by sparky 15y ago
It is a more complex world, but I think that's orthogonal to whether you prefer return codes or exceptions. In fact, I think you can argue that keeping error-handling local to the call site (return codes) encapsulates and abstracts the erro
111.
by sparky 15y ago
SoftWare EnGineeRing.
112.
by sparky 15y ago
Software transactional memory.
113.
by sparky 15y ago
Anyone know if there's a self-contained way to spool up a dyno sporadically to complete scheduled tasks? I have a webapp now that's hosted on a VPS and uses Celery to schedule tasks. The tasks themselves only take about a second, the dyno
114.
by sparky 15y ago
Err.. I would argue that (+1)^2+(-1)^2 would not yield a false positive. You are right that the explanation is incomplete, though. The squaring provides an 'absolute value' effect as well as disproportionately weighting larger alphabetical
115.
by sparky 15y ago
http://www.blogcdn.com/www.engadget.com/media/2011/09/amazon... Now, the picture could be incorrect or stretching the truth, but there was some indication.
116.
by sparky 15y ago
This looks like great work. Does anyone know: * Approximately how much work it would be to port this to another Python framework (e.g., Flask+WTForms) from Django? * If something like this already exists in Python in a more framework
117.
by sparky 15y ago
I mocked up the first figure in PPT 2007 in about 5 minutes: http://mattj.me/llvm_ra_copycat.pptx http://mattj.me/llvm_ra_copycat.png http://mattj.me/llvm_ra_copycat.pdf The default font, drop shadow, and line spacing are not exactly
118.
by sparky 15y ago
My money's on powerpoint 2007 or later. Could also be visio.
119.
by sparky 15y ago
Yes, but a (big?) part of what makes a team fundable is some ability to come up with a viable idea. The market may not always agree with the founders' and PG's idea of viability, so good teams can pivot, but if you don't have any ideas or
120.
by sparky 15y ago
understood :) just putting it out there in an attempt to avoid dupes. thanks joshua!

More ›