On Sun, Jan 27, 2013 at 1:14 PM, Charles Elliott <[email protected]> wrote:
> But VS 2012 Express sure produces fast code.  I was experimenting with a
> sqrt routine that uses Newton's method.  It was so fast that using the API
> routine GetSystemTimeAsFiletime measured 0 secs elapsed time.  Compiled with
> VS .NET 2003, the same routine took about 2 ms.  The compiler now intermixes
> FPU and MMX instructions, which may be the reason 2012 is faster.
This might be relevant here: Microsoft usually favors small code, and
typically leans towards -Os to keep the caches hot. Boinc uses -O2 or
-O3, which does things like unroll loops, which often results in
larger code.

Jeffrey Richter (et al) recommends -Os due to fears of paging code.
Richter is one of those guys who wrote the book on Windows [0].

I only say its relevant because I'm not familiar enough with GCC (and
other compilers, like ICC). Plus, Android, Apple, RIM, and Windows
devices don't have page files, so the Richter's concern of paging are
a non-sequitur.

OT: the fastest code I have seen is produced by Intel's compiler.
Sometimes, its the slowest for political reasons [1].

Intel's ICC and ICPC are a free download for non-commercial endeavors
[2]. I would encourage Boinc to use it during building/testing because
additional compilers add additional static analysis. The additional
analysis ensure conforming programs and help with reliable and secure
programs. I require it in my engineering processes for the same
reasons.

Each compiler has a personality, and by the time you get through VS,
ICC, GCC and Comeau, you have something portable, reliable, and
conforming. "Secure" is an emergent property of "correct" or
"reliable", so you have a chance to be secure.

Jeff

[0] 
http://www.amazon.com/s/ref=nb_sb_noss_1?url=search-alias%3Dstripbooks&field-keywords=richter+windows
[1] 
http://www.theinquirer.net/inquirer/news/1567108/intel-compiler-cripples-code-amd-via-chips
[2] http://software.intel.com/en-us/non-commercial-software-development

>> -----Original Message-----
>> From: [email protected] [mailto:boinc_dev-
>> [email protected]] On Behalf Of Daniel Carrion
>> Sent: Sunday, January 27, 2013 8:49 AM
>> To: Bernd Machenschalk
>> Cc: BOINC Developers Mailing List
>> Subject: Re: [boinc_dev] MinGW boinc_zip updates
>>
>> Hey Bernd
>>
>> Sorry been off on some other things.
>>
>> I can't get something together that plays cleanly. I'm trying to put
>> together a patch but I'm running into all sorts of problems during
>> building
>> (particularly on what c files to include). I'm sure I'll get there
>> eventually.
>>
>> The only reason I want to use mingw for boinc_zip is to build the
>> sample
>> wrapper from Dave's recent updates in my existing mingw build
>> environment.
>> I'm stubbornly refusing to install gigabytes worth of Visual Studio
>> clunkiness.
>>
>> I should probably just move over to Visual Studio to building it if
>> most of
>> the Windows build support is for Visual Studio. Argghh, it's just so
>> clunky. :)
>>
>> Regards
>>
>> Daniel
>>
>> On Sun, Jan 27, 2013 at 4:37 AM, Bernd Machenschalk <
>> [email protected]> wrote:
>>
>> > Hi Daniel!
>> >
>> > Thanks for the update.
>> >
>> > The boinc_zip stuff in the Makefile.mingw is left over from an old
>> version
>> > of boinc_zip that was removed from the boinc source tree. As nobody
>> > complained so far I think that no one (else) is actually compiling
>> > boinc_zip with MinGW. We, who originally put that in the
>> Makefile.mingw,
>> > have by now switched to using zlib, and thus don't need it anymore.
>> >
>> > If you need it and want to adapt the Makefile to the newly added
>> > boinc_zip, I'd happily incorporate your updates or patches.
>> >
>> > Best,
>> > Bernd
>> >
>> >
>> > Daniel Carrion wrote, On 26.01.13 14:27:
>> >
>> >> Just a quick one just to check if "lib/Makefile.mingw" is due for an
>> >> update
>> >> to accommodate recent zip changes? No issues compiling the main
>> libraries,
>> >> just boinc_zip. I'm working through it now to get it working and
>> noticed a
>> >> few things. E.g.  ZIP_OBJ doesn't reflect what exists, error finding
>> >> zipmain def in boinc_zip.cpp. End result is undefined reference
>> errors
>> >> when
>> >> going to compile boinc_zip.
>> >>
>> >> Not a biggie as I can get around it but thought I should give a
>> heads up
>> >> in
>> >> case someone wants to update it for the mingw users that want to
>> compile
>> >> boinc_zip lib.
_______________________________________________
boinc_dev mailing list
[email protected]
http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev
To unsubscribe, visit the above URL and
(near bottom of page) enter your email address.

Reply via email to