Intel® C++ Compiler
Community support and assistance for creating C++ code that runs on platforms based on Intel® processors.

About Sandy Bridge to Ivy bridge

aazue
New Contributor I
1,581 Views
Hi
I read here that new Ivy bridge must use socket LGA 1155
http://en.wikipedia.org/wiki/Ivy_Bridge_%28microarchitecture%29
I am already perfectly satisfied Sandy Bridge but just an question
Mother board that have already for example G620 icore xx etc.... also using LGA 1155 will possible receive an upgrade with new to Ivy Bridge ?
Have you an link where is described details characteristics more complete ?
I have already difficulty to understand side graphic hosted posisition sandy bridge
Example
when i use lspci linux system with sandy bridge processor G620
I show
00:02.0 VGA compatible controler: Intel Corporation Sandy Bridge Integrated Graphics Controler (rev 09)
I have rebuild all source in relation graphic (Mesa and all incuded)
I have rewrite an file xorg.conf correctly appropriated
Xrandr -q
confirm now that i have 1280*1024
video working very well but i use driver vesa .. with intel driver, is always rejected by xorg server...
Your data base about G620 give an warning about graphic could be not ???
Have you also an link where is described details characteristics graphic more complete ?
Regards

0 Kudos
1 Solution
TimP
Honored Contributor III
1,573 Views
If you're asking whether the gcc Ivy Bridge option would generate instructions not supported on Sandy Bridge, you should be checking the gcc source code or asking on the gcc-help mail list. Setting -march for Sandy Bridge and -mtune for IVB should be OK. This may make more use of 256-bit wide loads, which would be slower if not aligned on Sandy Bridge.

View solution in original post

0 Kudos
24 Replies
aazue
New Contributor I
1,491 Views
Hi ...
Here Intel and i observe that no one can answer two simples questions ?
I want use the Intel processors that I bought,correctly with his characteristics documented
and no with way approximately using nose .
Regards
0 Kudos
JenniferJ
Moderator
1,491 Views
Hello,
This is not the right forum for your question. Let me move your thread/question to a Processor related forum to see if you can get some answers there.

thanks,
Jennifer
0 Kudos
jimdempseyatthecove
Honored Contributor III
1,491 Views
Jennifer,

Did you inform the OP of which forum you moved the thread to?

Jim
0 Kudos
JenniferJ
Moderator
1,491 Views
Ihave not moved the thread, but is looking for which forum still. I'll update here soon.

Jennifer
0 Kudos
aazue
New Contributor I
1,492 Views
Hi and thank you for at least an response ..
You site have no section for answers to programing drivers.
Also for me is not problem bug with compiler GCC or ICC,
I have already compile success all package Linux have graphics relation.
I wait only technical characteristics for is rectified some sources for it work better.
Now, in the last version kernel several ,firmwares are separate and is better same for preserving
free Linux kernel independent.
For IVY and SANDY one of my customer have already several machines model SANDY have request to
me if futures his machines
could be upgrade IVY,only after this first confirmation
i must verify if with some flags compiler specifics the binary ready compatible two sides
Here ,is for compiler C/C++ I think , no ?
Regards
0 Kudos
JenniferJ
Moderator
1,492 Views

Hi,
There is a good forum for such questions: http://communities.intel.com/community/tech/processors?view=discussions

But I can not move your thread there. It's from two different forum programs. You can repost the question there.

thanks,
Jennifer

0 Kudos
aazue
New Contributor I
1,492 Views
Hello
Thank you very much for the link Jennifer.
This Site require registering an new account...
I will see later.
Regards
0 Kudos
aazue
New Contributor I
1,492 Views
Hello Jennifer
registering account on your link not working for me
after the procedure registering I don't receive the mail confirmation... ??
(I have tested on three different PPP connections)
I have answer my customer less complex to me , in futures he made the upgrade complete AMD or ARM...
(is just an kidding)
Regards

0 Kudos
JenniferJ
Moderator
1,492 Views
emmmm, I encountered the same problem. I'm looking for helper right now.....

Jennifer
0 Kudos
JenniferJ
Moderator
1,492 Views
There is a bug on the other forum registration. The issue has been escalated. Please hold on ....

Jennifer
0 Kudos
aazue
New Contributor I
1,492 Views
Hello Thank, Jennifer for all your answers Here ,all servers using very complex routing, i have always doubt It's the cause of the problem with server distant. As You having also the problem I am reassured. Regards
0 Kudos
aazue
New Contributor I
1,492 Views
Hi Jennifer
I don't know if my customer already his retirement but mail confirmation of your link work now ....


For help you on the relation of my questions and c/c++ that have also value on this site.

GCC 4.7 is ready now but (frozen for release)
New flag for IVY & Haswell...

Support for new Intel processor codename IvyBridge with RDRND, FSGSBASE and F16C is available through -march=core-avx-i.
Support for the new Intel processor codename Haswell with AVX2, FMA, BMI, BMI2, LZCNT is available through -march=core-avx2.


Fortunately Them at least they are very effective and also are informed actualized for the new hardware.
Thank
Regards
0 Kudos
aazue
New Contributor I
1,492 Views
Hi Jennifer
I don't know if my customer already his retirement but mail confirmation of your link work now ....


For help you on the relation of my questions and c/c++ that have also value on this site.

GCC 4.7 is ready now but (frozen for release)
New flag for IVY & Haswell...

Support for new Intel processor codename IvyBridge with RDRND, FSGSBASE and F16C is available through -march=core-avx-i.
Support for the new Intel processor codename Haswell with AVX2, FMA, BMI, BMI2, LZCNT is available through -march=core-avx2.


Fortunately Them at least they are very effective and also are informed actualized for the new hardware.
Thank
Regards
0 Kudos
JenniferJ
Moderator
1,492 Views
Thanks for this info.

I was checking the log-in this morning with my other email account, just saw your response now.

This is great.

thanks,
Jennifer
0 Kudos
aazue
New Contributor I
1,492 Views
Hello

I have build sources Gcc 4.7 and test with 450 sources
He work very very very well for me... (improved evident)
Extraordinary with Sandy Bridge ...(only this model i have testing)
Congratulation friends Team Gnu compiler, yet an work fabulous ..


If i use flag -march=core-avx-i to Sandy Bridge complier result success but
the program that call external backend not working, probably I must rebuild also all backend with
same options flags aligned.

for the Icc & 4.7 coexisting ... (4.6.2 not this problem)

/usr/include/c++/4.7/ext/atomicity.h(48): error: identifier "__ATOMIC_ACQ_REL" is undefined
{ return __atomic_fetch_add(__mem, __val, __ATOMIC_ACQ_REL); }


root@debian-bustaf:/usr/bin# icc -v
icc version 12.1.0 (gcc version 4.6.0 compatibility)

root@debian-bustaf:/usr/bin# g++ -v
Using built-in specs.
COLLECT_GCC=g++
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-linux-gnu/4.7.0/lto-wrapper
Target: x86_64-linux-gnu
Configured with: ./configure --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --program-suffix=-4.7 --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.7 --libdir=/usr/lib --enable-nls --enable-lto --with-ppl --with-mpfr --with-mpc --with-cloog --enable-cloog-backend=isl --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --with-arch-32=i586 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu
Thread model: posix
gcc version 4.7.0 20120302 (prerelease) (GCC)


Regards..
0 Kudos
TimP
Honored Contributor III
1,492 Views
I submitted a premier issue (659938) on the incompatibility between g++ 4.7 atomicity.h and icpc 5 weeks ago. It was accepted as a feature request, with no indication of when it might be considered.
According to past history, I would not count on the problem being solved until there is a SLES, Ubuntu, or RHEL release using gcc 4.7 or 4.8 which would available for development of a future icpc release. So we would be stuck with removing current gcc from PATH, up through icc 13.0.
The efforts which have gone into making gcc more effective on current Intel platforms conflict with the use of Intel C++ until this problem is solved.
In my version of Levine-Callahan-Dongarra vectors test suite, under what I consider reasonable ground rules, g++ 4.7 out-performs icpc 12.1.3, on Westmere. It's necessary to update with all the special icpc pragmas in order to pull ahead.
0 Kudos
aazue
New Contributor I
1,492 Views
Hi, thank for your informations

It's no problem for me, I can wait the future upgrade..
the rule to multiple path for get around the ICC problem header is the link (ln) version compiler
name complete (/usr/bin/g++-4.7) to short name g++ reflecting (cc) link.
I link dynamically g++-4.6.2 to g++ ( or with an loader using putenv(...)) temporary for having header
are correct to ICC.
Same, the Cloud Sandy Bridge machine using only one instance ,will able serving the two compilers...
(Better that nothing with preserving ICC before an upgrade could be better aligned)

(up through icc 13.0) more prudent is jumps directly 14.0 that is harmless number, all difficulty
existing already sufficing ....

Regards
0 Kudos
aazue
New Contributor I
1,492 Views
Hi
I extend other tests...
I think wrong way to use -march-core-avx-i (the diy for anticipate an compatibility binary with IVY )
to Sandy Bridge..
Apache 2.4 (with apr, pcre etc ..) compile success but when started detected Illegal instruction with this flag.
But without this flag specific persisting largely improved with this new version and also GNU compiler 4.7 ..
An jewel ...
compile Apache 2.4 pass success also with ICC with well results
Regards

0 Kudos
TimP
Honored Contributor III
1,574 Views
If you're asking whether the gcc Ivy Bridge option would generate instructions not supported on Sandy Bridge, you should be checking the gcc source code or asking on the gcc-help mail list. Setting -march for Sandy Bridge and -mtune for IVB should be OK. This may make more use of 256-bit wide loads, which would be slower if not aligned on Sandy Bridge.
0 Kudos
aazue
New Contributor I
1,257 Views
Hi, thanks for your answers
(Setting -march for Sandy Bridge and -mtune for IVB should be OK)

Yes it work , I have make an control with large charge using (ab and also real) it
seem tuned is yet more effective

I have test an new build apr-1.4.6 & apr-util-1.4.1 with 2 flags IVY before an new build Apache 2.4
just when lib apr compiled with 2 flags Ivy apache give fault Illegal instruction.
I am already happy like that , In my life I never had a performance equaled like that with
using machine model range not very expensive.
Also congratulations Team apache, 2.4 is an fabulous result
when I have time i see source Gnu compiler particularity of (core-avx-i) but I'm afraid it 's too
complex for my old head now ...
Thanks lot

Regards

0 Kudos
Reply