search for: embarrassed

A few random observations: 1. Clang could do better with large but boring switches like this: http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/E8/E88C5111.shtml Performance of clang's output will be fine but this is a major code size lose. 2. Destruction of stupid loops is incomplete, sometimes due to phase ordering problems:

>> Repetitive code with lots of bitwise operations is compiled by LLVM into >> much larger code than the other compilers: >> >> http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/ED/ED37DAF5.shtml >> http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/1F/1F4003C7.shtml >> >> Note that this is straight-line code, so LLVM's output will

> Umm, can you find one that isn't a popcount implementation? Ok. MMX psadbw instruction: http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/CE/CE3DA132.shtml Position of first set bit: http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/1F/1F4003C7.shtml Log2 floor: http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/83/837A80E9.shtml Pixel format

On Tue, Jan 26, 2010 at 12:36 PM, John Regehr <regehr at cs.utah.edu> wrote: > 2. > Sometimes not: > > http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/EC/ECC74C0C.shtml The primary issue here is that scalar evolution doesn't know how to deal with loops using "sle" for the exit condition. Shouldn't be too hard to fix now that we have overflow flags

I have an "embarrassingly parallel" routine that I need to run 24000^2/2 times (based on some microarray data). All I really need to do is parallelize a nested for-loop. But I haven't found a clear list of what packages/commands I'd need to do this. I've got a dual quad core xeon system running RHEL5, so if I could use hyperthreading to increase the number of (virtual)

I have a dataset which I am trying to smooth, using locally weighted regression. The y values are count data, integers with Poisson distribution, and it is important for the regression function to know this, since assuming a Gaussian distribution will lead to substantial errors. It is a time series; the x values have equal five minute intervals. Here is the problem: I have an embarrassment

On Tue, Jan 26, 2010 at 5:55 PM, John Regehr <regehr at cs.utah.edu> wrote: >>> Repetitive code with lots of bitwise operations is compiled by LLVM into >>> much larger code than the other compilers: >>> >>> >>> http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/ED/ED37DAF5.shtml >>> >>>

We have some results that are somewhat entertaining and that relate to the size/speed discussion. The basic idea is exhaustive generation of C functions where "exhaustive" is qualified by some structural restrictions (depth of AST, node type, etc.). For one particular set of restrictions we ended up with about 7 million C functions. We then compiled each of these functions with 7

Hi folks, I've posted an updated code size comparison between LLVM, GCC, and others here: http://embed.cs.utah.edu/embarrassing/ New in this version: - much larger collection of harvested functions: more than 360,000 - bug fixes and UI improvements - added the x86 Open64 compiler John

Dear R helpers, I have what another member on this forum described as an embarrassingly parallel problem. I am trying to fit models on subsets of some data based on unique combinations of two id factors in the dataset. Total number of combinations is 30^5, and this takes a long time. So, I would like fit models for each of the datasets produced by subsetting on the unique combinations, splitting

Dear list, could someone point me to the right command to subtract 2 columns in a data.frame. Might be a bit embarrassing question. But I cannot figure out how to do this simple command in R. Thanks, Ulrich -- __________________________________________________ Ulrich Leopold MSc. Department of Physical Geography Institute for Biodiversity and Ecosystem Dynamics Faculty of Science University

I am having trouble saving graphs. Using the Aqua interface (which is not my preferred interface), I have no problems plotting a graph, adding additional lines, points, references, etc., and then saving it to a file using, for example, the dev2bitmap command. I have found that, running R with Xemacs+ESS under X11 (which I prefer over Aqua), this is not possible. I can either send the graph to a

2009/11/22 Jon Harrop <jon at ffconsultancy.com>: > What about parallelization? I thought about that for a while, but if you keep your classes/files small, intra-unit parallelization gains are probably not worth the time invested. Compiling multiple files is embarrassingly parallel. [1] MHO is that, though inter-unit optimizations can take much longer, the benefits are worthwhile.

Hi, Until last week, I could install a CentOS 7 based desktop using the following approach: 1. Install minimal system. 2. yum groupinstall "X Window System" 3. yum install gdm gnome-classic-session gnome-terminal liberation-fonts 4. Install applications as needed. This morning, the package group "X Window System" seems to have disappeared. This is embarrassing. What

[cross-posting to the GCC and LLVM lists] I've updated the code size results here: http://embed.cs.utah.edu/embarrassing/dec_09/ The changes for this run were: - delete a number of testcases that contained use of uninitialized local variables - turn off frame pointer emission for all compilers - ask all compilers to target x86 + SSE3 - ask all compilers to not emit stack protector

On Tue, Jan 26, 2010 at 7:42 PM, John Regehr <regehr at cs.utah.edu> wrote: >> Umm, can you find one that isn't a popcount implementation? > > Ok. > > MMX psadbw instruction: > > http://embed.cs.utah.edu/embarrassing/jan_10/harvest/source/CE/CE3DA132.shtml > > Position of first set bit: > >

Renato Golin wrote: > 2009/11/22 Jon Harrop<jon at ffconsultancy.com>: >> What about parallelization? > > I thought about that for a while, but if you keep your classes/files > small, intra-unit parallelization gains are probably not worth the > time invested. Compiling multiple files is embarrassingly parallel. > [1] Compiling multiple files is embarrassingly

I'm planning on releasing 6.01 next week. The list of changes (so far) are appended below. This is going to be a quick bug fix release, mainly because of the rather embarrassing bug in the 6.00 that prevents booting a Linux kernel on BIOS. The idea is to get a fixed release into everyone's hands ASAP. Thanks to everyone who has helped with debugging. --- Matt Fleming (9): efi:

...e, while it was being written to, could subsequently be opened by another client for reading. I used a DVR with chasing play as an example. Didn't seem like that difficult of a question, but maybe it isn't geeky enough for some. (Or perhaps the answer is "No" and people are too embarrassed to admit Samba can't do it.)

So it would appear that llvm-gcc and clang are both slower than gcc4 which is infamous for being slow at compiling code, and yes this is with a release build/--enable-optimizations. This seems to go against notes such as http://clang.llvm.org/features.html#performance which claim clang is signifcantly faster than gcc. Below are some times and the larger object files when compiling an i386