Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

Re: [News] Linux Kernel Space Benchmark: 32 Bit Versus 64 Bit

Hash: SHA1

____/ GreyCloud on Wednesday 30 Dec 2009 23:30 : \____

> Roy Schestowitz wrote:
>> Hash: SHA1
>> ____/ Peter KÃhlmann on Wednesday 30 Dec 2009 20:47 : \____
>>> Roy Schestowitz wrote:
>>>> Hash: SHA1
>>>> Ubuntu 32-bit, 32-bit PAE, 64-bit Kernel Benchmarks
>>>> ,----[ Quote ]
>>>> | For this comparison we used Ubuntu 9.10 on a
>>>> | Lenovo ThinkPad T61 notebook running an Intel
>>>> | Core 2 Duo T9300 processor, 4GB of system
>>>> | memory, a 100GB Hitachi HTS7220 SATA HDD, and a
>>>> | NVIDIA Quadro NVS 140M. We were using the
>>>> | Ubuntu-supplied kernels that are based off the
>>>> | Linux 2.6.31 kernel in Ubuntu Karmic. Other
>>>> | packages that were maintained included GNOME
>>>> | 2.28.1, X Server 1.6.4, NVIDIA 195.22 display
>>>> | driver, GCC 4.4.1, and we were using the
>>>> | default EXT4 file-system with all other
>>>> | defaults. With Ubuntu to properly address 4GB
>>>> | or greater of system memory you need to use a
>>>> | PAE kernel as the Physical Address Extension
>>>> | support through the kernel's high-mem
>>>> | configuration options are not enabled in the
>>>> | default 32-bit kernels.
>>>> `----
>>>> http://www.phoronix.com/scan.php?page=article&item=ubuntu_32_pae&num=1
>>> Oh, oh, now our resident 64bit "specialist" Hadron Snot Quark has some
>>> heavy explanation to do why, totally different from his ascertions, the
>>> 64bit linux trounced the 32bit counterpart on *every* test.
>>> Which was maintained as fact all along from linux users who actually use
>>> linux *and* especially in the 64bit version, and which was all the time
>>> naysaid by Hadron and comrades (after all, 64bit windows is actually
>>> *slower* than 32bit windows, as there are practically no 64bit apps. Can't
>>> have that, can we?)
>>> And here we have *again* a benchmark where 64bit linux was faster, in in
>>> several cases *dramatically* faster than 32bit
>>> I guess Hadrons "explanation" will be that those guys are "lying
>>> freetards"
>> Microsoft has to throw mud at 64-bit because it's still struggling to get it
>> working.
> I'd say that the biggest improvements to performance when moving to
> 64-bit is the utilization of the general purpose registers and the
> elimination of the saving of the BP register and movement of addresses
> on every function call.  All one has to do to verify this is to compile
> a simple hello world program in C and to utilize the generate assembly
> instructions only in 32-bit mode and compare the code generated in
> 64-bit code.  This is the way Intel processors work.  I don't get a
> speed advantage when the same code is compiled on a G5 (IBMs 970FX) in
> 64-bit mode.  I get better speed in 32-bit mode as there are 32 general
> purpose registers available and it these same gp registers are there in
> 64-bit mode as well.

Microsoft recommended 64-bit only because it temporarily renders some viruses
incompatible. Not performance, not memory addressing...

- -- 
		~~ Best of wishes

Roy S. Schestowitz      | Windows O/S: chmod a-x internet; kill -9 internet
http://Schestowitz.com  |  Open Prospects   |     PGP-Key: 0x74572E8E
Tasks: 140 total,   1 running, 139 sleeping,   0 stopped,   0 zombie
      http://iuron.com - knowledge engine, not a search engine
Version: GnuPG v1.4.9 (GNU/Linux)


[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index