Folding Forum

mwroggenbuck wrote:Unfortunately, the one thread limit does not work 100%. I have other posts on this forum about this suggestion. I ran it and still had issues.

The one core workaround is for this error only 0xc0000374, so if you have other error codes then it is something else.

As far as I have seen this 0xc0000374 only happens on rigs (in Windows) where the CPU cores are multithreaded, so typically 4 core 8 thread, 8 core 16 thread etc., which indicate heap issues in the end process of the software and not memory issues. A workaround is to limit each FAHCore_22 process to...

How are you giving the gpu work units more cores? I am using a process manager https://www.bill2-software.com/processmanager/download.shtml to automatically reduce the number cores a FAHcore_22 process gets access to in the first place, so unless you have reduced the amount of cores the process get...

Both WUs ran on its own RX580 GPU for like 8-10 hours, reaching like 3%, when the remaining ETA was recorded, before giving each of the WUs more cores to play with, resulting in a more normal runtime and ETA of like 3h, and this was the only two WUs running at the time, since everything else was tur...

PRCG numbers for some abnormally long runtime WUs In my case this is based on 1 x CPU core running 1 x FAHCore_22 13421 - 4444, 0, 0 - ETA 13 days 13421 – 3142, 11, 0 – ETA 11 days Tested what makes them move forward at some normal speed, and giving each of them 2 x CPU cores makes them rather happy...

GPUSpecies 4 is supposed to be supported. I don't know what happened. It'll take be a bit to undo a lot of today's work. Well, according to your list in the other thread the Radeon RX 470/480/570/580/590 is technically more Species 5, since all but the RX 470 is above 5 TFLOPS, while the RX 470 is ...

The fact that it just started to happen on several already running systems should probably tell you something. 23:27:43:Trying to access database... 23:27:43:Successfully acquired database lock 23:27:43:Read GPUs.txt 23:27:46:Enabled folding slot 01: READY gpu:0:Ellesmere XT [Radeon RX 470/480/570/5...

The log tells you where it is failing, the slots asks for a new WU assignment after finishing the previous one, then gets a WU assignment for 192.0.2.1, makes the directory in the work folder, but there is nothing to download, since the IP is invalid.

These setups have been running for months, and all the other, exactly the same GPUs, as can be seen in the log, are folding fine on the same machines at the same time in the other slots, so since this 192 thing suddenly happened on several different machines when requesting a new WU in the last few ...

The Assignment Server is sending out Work Server requests for 192.0.2.1, which is for obvious reasons going to be an issue. 23:19:08:WU02:FS01:Connecting to assign1.foldingathome.org:80 23:19:08:WU02:FS01:Assigned to work server 192.0.2.1 23:19:08:WU02:FS01:Requesting new work unit for slot 01: READ...

Let us just collect the PRCG numbers in one thread, since these abnormally long runtime WUs seem to arrive more and more over the last day. In my case this is 1 x CPU core running 1 x FAHCore_22 13416 – 807, 270, 1 – ETA 14 days 13416 - 1289, 20, 4 - ETA 9 days 13416 - 1226, 216, 1 - ETA 11 days 134...

bruce wrote:We need to be talking about a single issue.

I am only talking about FAHCore_22 issues in Windows, so I don’t know if it is the same with Linux.

The 100% issue is a rather obvious heap thing, since doing the following affinity changes removes the issue completely:

FAHCore_22 does sometimes use more than one CPU thread per GPU. Studies have shown that FAHCore_22 can speed up the throughput of it's processing by using more CPU resources. Nobody has yet studied how much a CPU thread that's freed up from CPU assignments will add to the GPU PPD compared to the lo...

It depends on whay you mean by "issues" What I mean by issues is that the software isn’t handling the multithread distribution over several different cores very gracefully, so when releasing heap addresses on 1 core, the software thread running on another core, since there are several sof...

Something is seriously fishy with the 13416 WUs and how they run on GPU systems, since one FULL CORE isn’t enough to keep them happy and running smoothly, they want to grab even more. The following is a combination of 4 WUs from a configuration with Affinity set, to show what is going on for each WU...