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 - JeremyRand

Pages: [1] 2
1
Firmware / Re: Updating Talos II firmware to IBM PNOR V2.18?
« on: June 16, 2023, 02:32:16 am »
The 18c & 22c parts are "paired" meaning 2 SMT4 cores share the same L2 & L3, unlike the 4c and 8c which are "unpaired" meaning each core gets the full L2 and L3 to itself.    This is not the same as "fused" (i.e. SMT8 cores) but it is quite likely that the fix will also work for "paired" cores as presumably the issue is sharing cacheable/non-cacheable pathways.   Good luck!

I think you're right about the terminology, but yeah, I doubt the IBM docs draw a distinction since those docs are exclusively written for SMT8 users.

I've successfully built a PNOR with IBM's patch; it installed without major issues (I accidentally hosed things the first time I installed it due to the BMC running out of RAM -- protip, don't put multiple PNOR images in OpenBMC /tmp/ -- but power-cycling the BMC fixed that). So far it seems stable, I'll be running it for the next month or so to see if any checkstops happen.

Would you be able to provide more details on which patch? Many thanks

This is the hcode branch I used: https://github.com/JeremyRand/hcode/tree/talos-2019-07-25-master-rebased

As you can see, it's simply a copy of Raptor's hcode, rebased against current upstream IBM hcode (there were no rebase conflicts). The specific bugfix commit is this one: https://github.com/JeremyRand/hcode/commit/ca06a0c996e3b48c02cfb3912dddf7ca23ec4202

I've been running it on my DD2.3 2x18-core Talos II for 2 months, and my DD2.2 2x22-core Talos II for 1 month, and I have had no checkstops on either since applying the patch, nor any new issues. At this point I can recommend that Raptor integrate the bugfix into a new PNOR release. Since the rebase had no conflicts, it should be trivially easy for Raptor to do this.

2
Linux v6.3 is now in Fedora stable, and 1920x1200 works fine there. \o/

3
One nice (depending on your point of view) change in Fedora 38 is that SDDM now uses Wayland by default, meaning that DisplayLink GPU's work properly in SDDM without changing settings.

4
Firmware / Re: Updating Talos II firmware to IBM PNOR V2.18?
« on: April 21, 2023, 10:15:14 pm »
The 18c & 22c parts are "paired" meaning 2 SMT4 cores share the same L2 & L3, unlike the 4c and 8c which are "unpaired" meaning each core gets the full L2 and L3 to itself.    This is not the same as "fused" (i.e. SMT8 cores) but it is quite likely that the fix will also work for "paired" cores as presumably the issue is sharing cacheable/non-cacheable pathways.   Good luck!

I think you're right about the terminology, but yeah, I doubt the IBM docs draw a distinction since those docs are exclusively written for SMT8 users.

I've successfully built a PNOR with IBM's patch; it installed without major issues (I accidentally hosed things the first time I installed it due to the BMC running out of RAM -- protip, don't put multiple PNOR images in OpenBMC /tmp/ -- but power-cycling the BMC fixed that). So far it seems stable, I'll be running it for the next month or so to see if any checkstops happen.

5
1080p was never an issue. It's about 1200p.

Kind of. There was an issue with 1200p affecting the AST2500 (which is fixed in Linux 6.3); there's a separate issue with 1080p (among other resolutions) affecting the IT66121 (which AFAIK is not fixed yet, but I haven't tested). Sometimes these issues get conflated because Blackbird users normally don't use the AST2500 or IT66121 without the other.

6
User Zone / Re: Calling for gaming experiences
« on: April 21, 2023, 10:02:51 pm »
Ren'Py (visual novel engine) works fine when built from source.

mkxp-z (more modern implementation of RPG Maker XP / VX / VX Ace) works fine too, but you need to pass in a configure flag that disables Altivec in SDL (to work around an endianness bug).

7
Firmware / Re: Updating Talos II firmware to IBM PNOR V2.18?
« on: April 10, 2023, 01:21:43 am »
I believe that all OpenPOWER/OPAL-based systems use SMT4 cores, not SMT8 (i.e. "fused") cores.    So that may mean you aren't seeing the same problem.

Do you get the same fault callout?

I get the same error "NCU no response to snooped TLBIE". I asked Timothy on IRC and he said that the 18/22-core Raptor CPU's use fused cores. Which tracks with the fact that this started happening to me when I upgraded from 4-core CPU's to 22-core CPU's.

Looks like Raptor only has minimal changes to the IBM repo that contains the fix, and there are no rebase conflicts. So, I'll try building a modified firmware that incorporates IBM's fix and see if it helps here.

8
Linux v6.3 release candidates are in Fedora Rawhide now, and I can confirm that the bug is fixed there (tested with Linux fedora 6.3.0-0.rc5.20230407gitf2afccfefe7b.46.fc39.ppc64le).

9
Firmware / Updating Talos II firmware to IBM PNOR V2.18?
« on: February 09, 2023, 07:07:59 pm »
According to this IBM documentation (thanks awilfox for the pointer), PNOR v2.18 contains the following change:

Quote
HIPER/Pervasive:   A problem was fixed for a processor core checkstop with SRC BC70E540 logged with Signature Description " ex(n0p1c4) (NCUFIR[11]) NCU no response to snooped TLBIE".  This problem is intermittent and random but occurs with relatively high frequency for certain workloads.  The trigger for the failure is one core of a fused core pair going into a stopped state while the other core of the pair continues running.

Is it possible to get the Talos II firmware bumped to upstream IBM v2.18? I believe I may be running into this bug.

10
What's the status on dual-CPU support? According to https://github.com/3mdeb/openpower-coreboot-docs/blob/main/devnotes/documentation.md , it's not supported, but the blogpost at https://web.archive.org/web/20220926061511/https://blog.3mdeb.com/2022/2022-04-12-talos2_2nd_cpu_and_testing/ suggests that it may be working.

Also, can https://blog.3mdeb.com/ please unblock Tor users? Tor users currently get "Your IP has been banned".

11
Is Wayland still limited to 1024x768 on the ASPEED?

I think that's a Blackbird-specific issue; I think DKnoto is on a Talos II.

12
It looks like the fix has successfully advanced from drm-tip to drm-next, and is still on track for Linux 6.3: https://cgit.freedesktop.org/drm/drm/commit/?id=f2fa5a99ca81ce1056539e83c705f3d6bec62e31

I haven't had a chance to test drm-next yet, and probably will not have time to do so before the Linux 6.3 merge window opens in February.

13
I can confirm that the bug is fixed in the drm-tip branch (as of 2022 November 20), as stated in the linked dri-devel thread.  Per the thread, the fix is expected to be released in Linux v6.3.

14
User Zone / Re: Calling for gaming experiences
« on: November 14, 2022, 06:55:38 am »
No, probably not. Ignoring endianness for the sake of discussion, while Cell does have a conventional PowerPC core as the PPE, the SPEs are unique and so is the EIB connecting them.

While code from PowerPC-based game consoles such as GC/Wii/PS3 would almost definitely not run verbatim on POWER9 systems, I think it's likely that JIT-based emulators for such consoles would be much more efficient on POWER9 than on x86, simply because there's a lot less translation work for the JIT to do.  (Still waiting for a POWER9 Dolphin JIT....)

15
Filed a bug with upstream dri-devel at Dan HorĂ¡k's request, feel free to follow along there: https://lists.freedesktop.org/archives/dri-devel/2022-October/377090.html

Pages: [1] 2