On Wed, 11 Mar 2015 13:32:09 +0100 poma pomidorabelisima@gmail.com wrote:
- You still haven't provided basic information about the processor,
thus 'lscpu' or a similar command.
$ lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 6 On-line CPU(s) list: 0-5 Thread(s) per core: 2 Core(s) per socket: 3 Socket(s): 1 NUMA node(s): 1 Vendor ID: AuthenticAMD CPU family: 21 Model: 2 Model name: AMD FX(tm)-6300 Six-Core Processor Stepping: 0 CPU MHz: 3500.000 CPU max MHz: 3500.0000 CPU min MHz: 1400.0000 BogoMIPS: 7023.57 Virtualization: AMD-V L1d cache: 16K L1i cache: 64K L2 cache: 2048K L3 cache: 8192K NUMA node0 CPU(s): 0-5
- Moreover you still haven't provided the difference between "I'm running Fedora 21 with a custom compiled kernel,
3.19.0-1.20150211.fc21.x86_64." and stock Fedora kernel - 3.18.8-201.fc21.x86_64, i.e. diff /boot/config-3.18.8-201.fc21.x86_64 /boot/3.19.0-1.20150211.fc21.x86_64
I can do this but I think it is too large for a message. There are 1532 lines of output from that command.
- Also you haven't mentioned whether the same happens with the stock
Fedora kernel.
Yes, it did.
- And you haven't explained why you use a custom kernel, and why
particularly 3.19 kernel.
To remove all the generalizations that have to be in stock kernels so they can work for everyone. It greatly speeds compilation to not do all those drivers that I don't need. Sound cards I don't have
Latest that created packages without errors. I'll be upgrading to 4.0 kernel shortly. https://bugzilla.redhat.com/show_bug.cgi?id=1199312
- Etc.
How anybody can help in this way, man.
I don't see how those have any relevance. All the answers to them would do is be noise or distraction. This is at a meta level above that, about the design and control level, rather than the implementation details. I guess I am assuming that the kernel programmers have done their job, and the kernel functionality is abstracted from the underlying hardware. Especially for x86 architectures.
I think Heinz has put his finger on the issue in his response: a single process is limited to a single core. But, if that single process can spawn other processes, which is what 'make -j6' should be doing, why would that be true? I'll be following up in my response to him.
Thanks for taking an interest, and I'm not trying to be hostile, but I honestly don't see how answering your questions helps.
On 11.03.2015 16:32, stan wrote:
On Wed, 11 Mar 2015 13:32:09 +0100 poma pomidorabelisima@gmail.com wrote:
- You still haven't provided basic information about the processor,
thus 'lscpu' or a similar command.
$ lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian CPU(s): 6 On-line CPU(s) list: 0-5 Thread(s) per core: 2 Core(s) per socket: 3 Socket(s): 1 NUMA node(s): 1 Vendor ID: AuthenticAMD CPU family: 21 Model: 2 Model name: AMD FX(tm)-6300 Six-Core Processor Stepping: 0 CPU MHz: 3500.000 CPU max MHz: 3500.0000 CPU min MHz: 1400.0000 BogoMIPS: 7023.57 Virtualization: AMD-V L1d cache: 16K L1i cache: 64K L2 cache: 2048K L3 cache: 8192K NUMA node0 CPU(s): 0-5
- Moreover you still haven't provided the difference between "I'm running Fedora 21 with a custom compiled kernel,
3.19.0-1.20150211.fc21.x86_64." and stock Fedora kernel - 3.18.8-201.fc21.x86_64, i.e. diff /boot/config-3.18.8-201.fc21.x86_64 /boot/3.19.0-1.20150211.fc21.x86_64
I can do this but I think it is too large for a message. There are 1532 lines of output from that command.
- Also you haven't mentioned whether the same happens with the stock
Fedora kernel.
Yes, it did.
- And you haven't explained why you use a custom kernel, and why
particularly 3.19 kernel.
To remove all the generalizations that have to be in stock kernels so they can work for everyone. It greatly speeds compilation to not do all those drivers that I don't need. Sound cards I don't have
Latest that created packages without errors. I'll be upgrading to 4.0 kernel shortly. https://bugzilla.redhat.com/show_bug.cgi?id=1199312
- Etc.
How anybody can help in this way, man.
I don't see how those have any relevance. All the answers to them would do is be noise or distraction. This is at a meta level above that, about the design and control level, rather than the implementation details. I guess I am assuming that the kernel programmers have done their job, and the kernel functionality is abstracted from the underlying hardware. Especially for x86 architectures.
I think Heinz has put his finger on the issue in his response: a single process is limited to a single core. But, if that single process can spawn other processes, which is what 'make -j6' should be doing, why would that be true? I'll be following up in my response to him.
Thanks for taking an interest, and I'm not trying to be hostile, but I honestly don't see how answering your questions helps.
The Devil is in the detail.
Both: 1. diff -u /boot/config-3.18.8-201.fc21.x86_64 /boot/config-3.19.0-1.20150211.fc21.x86_64 2. dmidecode (as root)
to http://fpaste.org s'il vous plaît.
On Wed, 11 Mar 2015 18:21:06 +0100 poma pomidorabelisima@gmail.com wrote:
Both:
- diff
-u /boot/config-3.18.8-201.fc21.x86_64 /boot/config-3.19.0-1.20150211.fc21.x86_64 2. dmidecode (as root)
to http://fpaste.org s'il vous plaît.
1 week
http://fpaste.org/197315/88936142/
The config is now for kernel-4.0.
Are you sure you don't want to know what I had for breakfast today? :-D
For all this work, I expect *results*, sir. ;-)
On 03/12/2015 03:40 PM, stan wrote:
Are you sure you don't want to know what I had for breakfast today? :-D
No! That's what Facebook is for! B^)
On 03/12/2015 01:54 PM, Kevin Cummings wrote:
On 03/12/2015 03:40 PM, stan wrote:
Are you sure you don't want to know what I had for breakfast today? :-D
No! That's what Facebook is for! B^)
Article tagline from the Orange County Register this morning:
A study co-authored by a CSUF (Cal State University Fullerton) professor finds Facebook 'addicts' show some brain behaviors observed in drug abusers
Watch yourself! :-) ---------------------------------------------------------------------- - Rick Stevens, Systems Engineer, AllDigital ricks@alldigital.com - - AIM/Skype: therps2 ICQ: 22643734 Yahoo: origrps2 - - - - Millihelen (n): The amount of beauty required to launch one ship. - ----------------------------------------------------------------------
On Wed, 11 Mar 2015 18:21:06 +0100 poma pomidorabelisima@gmail.com wrote:
The Devil is in the detail.
Both:
- diff
-u /boot/config-3.18.8-201.fc21.x86_64 /boot/config-3.19.0-1.20150211.fc21.x86_64 2. dmidecode (as root)
to http://fpaste.org s'il vous plaît.
You seem to be the last hope for a solution here, so I hope your analysis succeeds.
"As the clock ticks down, it's a hail Mary pass to poma all alone in front of the goal. The game is in his hands now."
On 12.03.2015 20:40, stan wrote:
On Wed, 11 Mar 2015 18:21:06 +0100 poma pomidorabelisima@gmail.com wrote:
Both:
- diff
-u /boot/config-3.18.8-201.fc21.x86_64 /boot/config-3.19.0-1.20150211.fc21.x86_64 2. dmidecode (as root)
to http://fpaste.org s'il vous plaît.
1 week
http://fpaste.org/197315/88936142/
The config is now for kernel-4.0.
Are you sure you don't want to know what I had for breakfast today? :-D
For all this work, I expect *results*, sir. ;-)
Bon appetite!
~~~~
MOBO:
- dmidecode Base Board Information Manufacturer: ASUSTeK COMPUTER INC. Product Name: M5A97 LE R2.0
- ASUS M5A97 LE R2.0 http://www.asus.com/Motherboards/M5A97_LE_R20
Not bad.
~~~~~
BIOS:
- dmidecode BIOS Information Vendor: American Megatrends Inc. Version: 2501 Release Date: 04/09/2014
- http://www.asus.com/Motherboards/M5A97_LE_R20/HelpDesk_Download Support/Driver & Tools/OS Others/BIOS M5A97 LE R2.0 BIOS 2501 - 2014/05/14
BIOS is the latest version, dmidecode and asus.com BIOS date doesn't match, but the version is the same.
~~~~
CPU:
- dmidecode Processor Information Version: AMD FX(tm)-6300 Six-Core Processor
- FX-6300(FD6300WMW6KHK),3.5GHz,6C,95W,rev.C0,AM3+ http://www.asus.com/support/CPU/1/24/46/6/j02VBoX5bQJxJu1B/C20121019175441
- AMD FX-Series FX-6300 - FD6300WMW6KHK / FD6300WMHKBOX http://www.cpu-world.com/CPUs/Bulldozer/AMD-FX-Series%20FX-6300.html
CPU is compatible.
~~~~
RAM:
I do not find KHX1866C10D3/8G in M5A97 LE R2.0 Motherboard Qualified Vendors Lists (QVL)
- M5A97 LE R2.0 User's Manual (English) - 2013/01/16 http://dlcdnet.asus.com/pub/ASUS/mb/SocketAM3+/M5A97_LE_R2.0/E8047_M5A97_LE_...
nor in the - Memory Search - Kingston http://www.kingston.com/en/memory/search/options
The part number KHX1866C10D3/8G, is it correct?
~~~~~~~~~~~
Conclusion:
To achieve maximal possible CPU performance, you can try adjusting the settings in the BIOS. Refer to the manual.
Although this is a guide to overclocking, it still can serve as a reference for some of the tunings: - AMD FX 6300 Overclock Guide (Overclock to 4.3 ghz) https://www.youtube.com/watch?v=T9uXysmgPi8
p.s. Your kernel config I can comment later, if you're interested.