Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - FlyingBlackbird

Pages: [1] 2 3 ... 7
1
That are good news! And in their release notes they mention it too (and it is part of the free edition):

https://about.gitlab.com/releases/2021/09/22/gitlab-14-3-released/#gitlab-runner-on-ibm-power9-linux-os

2
I just saw some interesting presentations at the FOSDEM 2021 conference (available online), see:

- https://fosdem.org/2021/schedule/track/openpower/
- https://fosdem.org/2021/schedule/track/open_source_firmware_bmc_and_bootloader/

The conference takes place right now, sorry for posting last-minute here!

3
Quote
The big question though, I have had my blackbird since I think August 2019 and never did try connecting any kind of optical drive other than via USB. Could it be that it's just my board, its production run or a design flaw in the controller itself?

It was me who opened this thread and my Blackbird was RMA'ed and is still waiting to be reassembled again (shame on me but I simply had no time so far).

I had quite similar problems but my system did finally not even boot without any drive attached.

I will try to reassemble my system in the next three weeks with the same configuration as in my first post in this thread and see if I can boot with an optical drive attached via SATA (Asus).

Hopefully this is not a design flaw. I will report my results here...

BTW: Did you already try to enable the "cold restart" setting (I don't remember the exact name but it re-initializes the complete hardware when restarting like a real power-on boot phase). There is some documentation in somewhere here and in the wiki

4
Operating Systems and Porting / Re: kernel config: page size 4k vs 64k
« on: September 15, 2020, 01:16:15 am »
Most distributions have chosen the 4k page size for their kernels on Intel architectures.

The btrfs developers recently started a patch that allows the systems with 64k page size to read a btrfs volume with 4k sectorsize.  It only works in one direction and it is read-only.

You are mentioning the kernel "memory" page size (`getconf PAGESIZE") and the `btrfs` file system page size.

Just to understand this issue right:

How is this related? Does `btrfs` by default (or always?) use the same file system page size as the kernel does for the memory page size?

5
I tried putting a Quadro K2200 into my Talos II.
...
Xorg.0.log shows that the nouveau driver loads but it fails to initialize

May this problem possibly be related to the 4K page size issue you mentioned in another post (https://forums.raptorcs.com/index.php/topic,200.0.html)?


6
Did you have any luck with RStudio?

I also use both R and RStudio and I'm likely to try it in on this platform very shortly

I had no time so far to re-assemble my new workstation with the new planar :-(
So I still need some more time to test RStudio (Server)... (the bug was fixed BTW)

7
Operating Systems and Porting / Re: testing a new machine, memory test, etc
« on: September 07, 2020, 02:05:26 pm »
What burn-in test routines do people use when making a new machine or modifying the machine substantially?
essible in petitboot.

I personally love the `stress-ng` tool (you can vary the args as needed):

Code: [Select]
# CPU, sync and memory load simultanously:
sudo stress --cpu 4 --io 3 --vm 2 --vm-bytes 256M --timeout 20s

stress-ng – is an updated version of the stress workload generator tool which tests your
system for following features:

1. CPU compute
2. drive stress
3. I/O syncs
4. Pipe I/O
5. cache thrashing
6. VM stress
7. socket stressing
8. process creation and termination
9. context switching properties

# https://wiki.ubuntu.com/Kernel/Reference/stress-ng
# https://www.cyberciti.biz/faq/stress-test-linux-unix-server-with-stress-ng/

# sudo apt install stress
# sudo apt install stress-ng

# use stress-ng instead of stress for a newer stress test

sudo stress --cpu 32

# to messure the avg load use
uptime
sudo stress --cpu 32 --timeout 60s
uptime

# use -v for verbose output

# Memory/RAM:
# Watch out: Will swap if vm-bytes >= RAM size (very slow)
# "verify" also checks if the memory can be read
sudo stress-ng --vm 1 --vm-bytes 64G --timeout 90s --verify -v --metrics-brief

# Should I run stess-ng with root access?
# Running stress-ng with root privileges will adjust out of memory settings on Linux systems to make the stressors unkillable
# in low memory situations, so use this judiciously.
# -> may be required to test the RAM completely

8
Firmware / Re: swapping flash chips, ordering spares
« on: September 07, 2020, 01:52:48 pm »

Has anybody seen any useful video that shows the correct way to remove the chip?

I have found a video here but never tried it:

How to remove a SPI flash chip ("BIOS") from the motherboard:
https://www.youtube.com/watch?v=JEcvVWcdXk0

If you look at my pic in the prev post you can see that the planar uses sockets for the chips which support easily replace them...

9
Firmware / Re: swapping flash chips, ordering spares
« on: September 07, 2020, 01:50:36 pm »
Before my RMA of the blackbird planar I wanted to change the flash chips too and searched the web a lot. Here are my findings (without any guarantees ;-)

Flash Socket Type is: G6179T011-003H (2 x 8 pin layout)

Flash chip types are

1. BMC Firmware ROM (J7500 on the planar): MX25L25635FMI-10G
   Data Sheet: https://media.digikey.com/pdf/Data%20Sheets/Macronix/MX25L25635F.pdf
   2,7 - 3,6 Volts !!!!
   256 MB
   16-SOIC

2. Boot Flash ROM/PNOR Flash ROM (J7501): MT25QL512ABB8ESF-0SIT (on my chip I can read: 7RAI5 RWI87 258 => FBGA Code RW187 is the same chip!)
   Datasheet: https://www.digikey.com/product-detail/en/micron-technology-inc/MT25QL512ABB8ESF-0SIT-TR/557-1987-1-ND/9673968
   512 MB
   2,7 - 3,6 V -> 3 V is safe
   16-SOIC

=> you can order the chips here (eg.)
   https://www.digikey.de

HIH

PS: See the attached pic of my (RMA'ed) planar

10
Applications and Porting / Re: Wine patch to support PPC64
« on: August 30, 2020, 05:54:49 pm »
Fascinating!

Can somebody explain how it will be possible to execute Windows binaries via Wine on a ppc64le architecture (which has a different CPU instruction set than x86)? QEMU?

I thought Wine is "just" offering the Windows API as a wrapper around kernel/Linux calls...

11
Applications and Porting / Re: Raspberry Pie
« on: August 20, 2020, 03:30:23 pm »
Could you please add links to the "emulator system" and "retro pie application" that you mention (it is not clear to me at least even though I use Raspberries too)

12
The german computer magazine "ct" (www.ct.de) did test the Blender raytracing performance with different graphics cards including a Radeon RX 5700 XT und Radeon Pro W5700 in ct 01/2020 pages 100 - 101.

One finding was (my personal translation):

Quote
During the test AMD's Radeon Pro W5700 showed incorrect lighting and streaking artifacts using the pro driver 19Q4 - with the standard driver and an RX 5700 or a Radeon VII (Vega) this was not the case.

IMHO this is just a "spotlight" test and may be caused by a bug in the pro driver but it shows that the standard driver and the cheaper RX 5700 (XT) are working quite well...

13
Great news :-)

What I still do not understand is how Windows binaries for Intel x86 CPU instructions sets will execute with Wine on ppc64le which has another CPU instruction set.
I thought that Wine mainly adds the Windows API as compatibility layer to Linux...

Does somebody know this and could explain it here? THX :-)

14
Is there possibly a risk of "overclocking"? Any documentation available that says wrong modelines do not destroy the AST2500?

15
General CPU Discussion / Re: CPU Performance DD2.2 vs DD2.3 (v2)
« on: May 05, 2020, 06:31:33 am »
AFIKR `cat /proc/cpuinfo` also shows the stepping ("revision: 2.3" for the "v2" CPU)

Since the stepping is specific to the chip (Nimbus in this case), I think it would make sense to add your findings to the Steppings secton of the POWER9 page; if we start to get too-much Nimbus-specific information, it can always be split off into a separate page.

As an aside, the CP9M## is not actually a part number, it is (as far as I can tell) just a SKU for Raptor to manage their inventory; according to ClassicHasClass you can see the actual part number in lshw. We are actually missing the parts number for the CP9M31 SKU in the decoder table on the Sforza page, if you could add what yours is that would be excellent.

Pages: [1] 2 3 ... 7