Hello
I'm running Boinc Manager 7.20.4 (wx widgets 3.1.5) on a M1 MacBook Air.
For some time now, the display in columns is erratic. Most of the time, when I'm switching from the Boinc Manager app to another app and back, the advancement column goes back to 0.000% for most tasks. The elapsed time for the tasks is also reset to 00:00:00.
I tried to observe what's happening while not leaving the app window, but it seems like the app display would be unstable, unpredictable.
I also tried to reset the Options (calculation preferences), to "Use web preferences", but it had no effect.

Here's a copy of my event log.
Any help will be appreciated.

Sam 19 nov 05:54:58 2022 | | cc_config.xml not found - using defaults
Sam 19 nov 05:54:58 2022 | | Starting BOINC client version 7.20.4 for arm64-apple-darwin
Sam 19 nov 05:54:58 2022 | | log flags: file_xfer, sched_ops, task
Sam 19 nov 05:54:58 2022 | | Libraries: libcurl/7.79.1 SecureTransport zlib/1.2.11 c-ares/1.17.2
Sam 19 nov 05:54:58 2022 | | Data directory: /Library/Application Support/BOINC Data
Sam 19 nov 05:54:58 2022 | | OpenCL: Apple M1 0: Apple M1 (driver version 1.2 1.0, device version OpenCL 1.2, 10923MB, 10923MB available, 8 GFLOPS peak)
Sam 19 nov 05:54:58 2022 | | Version change (7.20.2 -> 7.20.4)
Sam 19 nov 05:54:58 2022 | | Host name: Gerards-MacBook-Air-2.local
Sam 19 nov 05:54:58 2022 | | Processor: 8 Apple Apple M1
Sam 19 nov 05:54:58 2022 | | Processor features: amx_version_1 arm64 armv8_1_atomics armv8_2_fhm armv8_2_sha3 armv8_2_sha512 armv8_crc32 floatingpoint neon neon_fp16 neon_hpfp ucnormal_mem fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clfsh ds acpi mmx fxsr sse sse2 ss htt tm pbe pni pclmulqdq dtse64 mon dscpl vmx est tm2 ssse3 cx16 tpr pdcm sse4_1 sse4_2 aes seglim64
Sam 19 nov 05:54:58 2022 | | OS: Mac OS X 13.0 (Darwin 22.1.0)
Sam 19 nov 05:54:58 2022 | | Memory: 16.00 GB physical, 215.77 GB virtual
Sam 19 nov 05:54:58 2022 | | Disk: 460.43 GB total, 215.77 GB free
Sam 19 nov 05:54:58 2022 | | Local time is UTC +1 hours
Sam 19 nov 05:54:58 2022 | World Community Grid | General prefs: from World Community Grid (last modified 15-Sep-2019 11:59:25)
Sam 19 nov 05:54:58 2022 | World Community Grid | Host location: none
Sam 19 nov 05:54:58 2022 | World Community Grid | General prefs: using your defaults
Sam 19 nov 05:54:58 2022 | | Reading preferences override file
Sam 19 nov 05:54:58 2022 | | Preferences:
Sam 19 nov 05:54:58 2022 | | max memory usage when active: 6553.60 MB
Sam 19 nov 05:54:58 2022 | | max memory usage when idle: 9830.40 MB
Sam 19 nov 05:54:58 2022 | | max disk usage: 10.00 GB
Sam 19 nov 05:54:58 2022 | | max CPUs used: 4
Sam 19 nov 05:54:58 2022 | | don't use GPU while active
Sam 19 nov 05:54:58 2022 | | suspend work if non-BOINC CPU load exceeds 30%
Sam 19 nov 05:54:58 2022 | | (to change preferences, visit a project web site or select Preferences in the Manager)
Sam 19 nov 05:54:58 2022 | | Setting up project and slot directories
Sam 19 nov 05:54:58 2022 | | Checking active tasks
Sam 19 nov 05:54:58 2022 | RNA World | URL http://www.rnaworld.de/rnaworld/; Computer ID 120781; resource share 100
Sam 19 nov 05:54:58 2022 | World Community Grid | URL http://www.worldcommunitygrid.org/; Computer ID 8520143; resource share 100
Sam 19 nov 05:54:58 2022 | | Setting up GUI RPC socket
Sam 19 nov 05:54:58 2022 | | Checking presence of 110 project files
Sam 19 nov 05:54:58 2022 | | Running CPU benchmarks
Sam 19 nov 05:54:58 2022 | | Suspending computation - CPU benchmarks in progress
Sam 19 nov 05:55:29 2022 | | Benchmark results:
Sam 19 nov 05:55:29 2022 | | Number of CPUs: 4
Sam 19 nov 05:55:29 2022 | | 5315 floating point MIPS (Whetstone) per CPU
Sam 19 nov 05:55:29 2022 | | 41365 integer MIPS (Dhrystone) per CPU
Sam 19 nov 05:55:31 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 05:55:59 2022 | RNA World | Sending scheduler request: To fetch work.
Sam 19 nov 05:55:59 2022 | RNA World | Requesting new tasks for Apple M1
Sam 19 nov 05:56:00 2022 | RNA World | Scheduler request completed: got 0 new tasks
Sam 19 nov 05:56:00 2022 | RNA World | Project requested delay of 3636 seconds
Sam 19 nov 05:56:05 2022 | World Community Grid | Sending scheduler request: To fetch work.
Sam 19 nov 05:56:05 2022 | World Community Grid | Reporting 1 completed tasks
Sam 19 nov 05:56:05 2022 | World Community Grid | Requesting new tasks for Apple M1
Sam 19 nov 05:56:07 2022 | World Community Grid | Scheduler request completed: got 0 new tasks
Sam 19 nov 05:56:07 2022 | World Community Grid | Project requested delay of 121 seconds
Sam 19 nov 05:56:53 2022 | RNA World | project suspended by user
Sam 19 nov 05:58:00 2022 | | Resuming GPU computation
Sam 19 nov 06:01:51 2022 | World Community Grid | Sending scheduler request: To fetch work.
Sam 19 nov 06:01:51 2022 | World Community Grid | Requesting new tasks for Apple M1
Sam 19 nov 06:01:53 2022 | World Community Grid | Scheduler request completed: got 0 new tasks
Sam 19 nov 06:01:53 2022 | World Community Grid | Project requested delay of 121 seconds
Sam 19 nov 06:26:50 2022 | World Community Grid | Sending scheduler request: To fetch work.
Sam 19 nov 06:26:50 2022 | World Community Grid | Requesting new tasks for Apple M1
Sam 19 nov 06:26:52 2022 | World Community Grid | Scheduler request completed: got 0 new tasks
Sam 19 nov 06:26:52 2022 | World Community Grid | Project requested delay of 121 seconds
Sam 19 nov 07:26:35 2022 | World Community Grid | Sending scheduler request: To fetch work.
Sam 19 nov 07:26:35 2022 | World Community Grid | Requesting new tasks for Apple M1
Sam 19 nov 07:26:37 2022 | World Community Grid | Scheduler request completed: got 0 new tasks
Sam 19 nov 07:26:37 2022 | World Community Grid | Project requested delay of 121 seconds
Sam 19 nov 08:45:43 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 08:46:43 2022 | | Resuming GPU computation
Sam 19 nov 08:57:41 2022 | World Community Grid | Sending scheduler request: To fetch work.
Sam 19 nov 08:57:41 2022 | World Community Grid | Requesting new tasks for Apple M1
Sam 19 nov 08:57:42 2022 | World Community Grid | Scheduler request completed: got 0 new tasks
Sam 19 nov 08:57:42 2022 | World Community Grid | Project requested delay of 121 seconds
Sam 19 nov 09:07:35 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 09:11:54 2022 | | Resuming GPU computation
Sam 19 nov 09:17:26 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 09:22:24 2022 | | Resuming GPU computation
Sam 19 nov 09:25:06 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 09:26:30 2022 | | Suspending computation - CPU is busy
Sam 19 nov 09:26:40 2022 | | Resuming computation
Sam 19 nov 09:30:23 2022 | | Suspending computation - CPU is busy
Sam 19 nov 09:30:33 2022 | | Resuming computation
Sam 19 nov 09:42:31 2022 | | Resuming GPU computation
Sam 19 nov 09:42:49 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 09:54:56 2022 | World Community Grid | General prefs: from World Community Grid (last modified 15-Sep-2019 11:59:25)
Sam 19 nov 09:54:56 2022 | World Community Grid | Host location: none
Sam 19 nov 09:54:56 2022 | World Community Grid | General prefs: using your defaults
Sam 19 nov 09:54:56 2022 | | Reading preferences override file
Sam 19 nov 09:54:56 2022 | | Preferences:
Sam 19 nov 09:54:56 2022 | | max memory usage when active: 6553.60 MB
Sam 19 nov 09:54:56 2022 | | max memory usage when idle: 9830.40 MB
Sam 19 nov 09:54:56 2022 | | max disk usage: 10.00 GB
Sam 19 nov 09:54:56 2022 | | max CPUs used: 4
Sam 19 nov 09:54:56 2022 | | don't use GPU while active
Sam 19 nov 09:54:56 2022 | | suspend work if non-BOINC CPU load exceeds 30%
Sam 19 nov 09:54:56 2022 | | (to change preferences, visit a project web site or select Preferences in the Manager)
Sam 19 nov 09:55:25 2022 | RNA World | project resumed by user
Sam 19 nov 09:55:29 2022 | RNA World | Sending scheduler request: To fetch work.
Sam 19 nov 09:55:29 2022 | RNA World | Requesting new tasks for Apple M1
Sam 19 nov 09:55:31 2022 | RNA World | Scheduler request completed: got 0 new tasks
Sam 19 nov 09:55:31 2022 | RNA World | Project requested delay of 3636 seconds
Sam 19 nov 09:55:37 2022 | RNA World | project suspended by user
Sam 19 nov 09:58:29 2022 | | Resuming GPU computation
Sam 19 nov 10:00:43 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 10:01:44 2022 | | Resuming GPU computation
Sam 19 nov 10:03:59 2022 | | Suspending GPU computation - computer is in use
Sam 19 nov 10:11:58 2022 | | Suspending computation - on batteries
Dim 20 nov 11:31:39 2022 | | Resuming computation
Dim 20 nov 11:31:48 2022 | | Suspending computation - CPU is busy
Dim 20 nov 11:31:59 2022 | | Resuming computation
Dim 20 nov 11:32:19 2022 | | Suspending computation - CPU is busy
Dim 20 nov 11:33:00 2022 | | Resuming computation
Dim 20 nov 11:33:36 2022 | World Community Grid | task MCM1_0192240_4063_0 aborted by user
Dim 20 nov 11:34:38 2022 | World Community Grid | Computation for task MCM1_0192240_4063_0 finished
Dim 20 nov 11:35:37 2022 | World Community Grid | Sending scheduler request: To report completed tasks.
Dim 20 nov 11:35:37 2022 | World Community Grid | Reporting 1 completed tasks
Dim 20 nov 11:35:37 2022 | World Community Grid | Requesting new tasks for CPU and Apple M1
Dim 20 nov 11:35:39 2022 | World Community Grid | Scheduler request completed: got 1 new tasks
Dim 20 nov 11:35:39 2022 | World Community Grid | Project requested delay of 121 seconds
Dim 20 nov 11:35:41 2022 | World Community Grid | Started download of MCM1_0192529_9712_MCM1_0192529_9712.txt
Dim 20 nov 11:35:43 2022 | World Community Grid | Finished download of MCM1_0192529_9712_MCM1_0192529_9712.txt
Dim 20 nov 11:35:43 2022 | World Community Grid | Starting task MCM1_0192529_9712_0
Dim 20 nov 11:37:41 2022 | World Community Grid | Sending scheduler request: To fetch work.
Dim 20 nov 11:37:41 2022 | World Community Grid | Requesting new tasks for Apple M1
Dim 20 nov 11:37:42 2022 | World Community Grid | Scheduler request completed: got 0 new tasks
Dim 20 nov 11:37:42 2022 | World Community Grid | Project requested delay of 121 seconds
Dim 20 nov 11:38:02 2022 | | Resuming GPU computation
Dim 20 nov 11:40:39 2022 | | Suspending GPU computation - computer is in use
Dim 20 nov 12:27:56 2022 | | Resuming GPU computation
Dim 20 nov 12:28:09 2022 | | Suspending GPU computation - computer is in use

Set the checkpoint_debug flag in Event Log Options and make sure that before you switch out apps or quit BOINC and restart it, the tasks have actually checkpointed. Because if they do not, then they'll restart from zero on the next start. That's a project thing, btw, their application should checkpoint tasks at regular intervals.

Also, 7.20.4 is an alpha version.
The BOINC developer has said: "Note: I have noticed that some World Community Grid graphics apps appear to be buggy. They sometime work and sometimes don't, both with this new version 7.20.4 and the previous version 7.20.2, so that problem is not due to these changes in BOINC."

Maybe it's not just the graphics, but the whole apps.

"7.20.4 is an alpha version" ? I just installed it because it appears as the official stable version on the boinc download page !

Thanks for the release management...

I'm using an Intel iMac so I see how see how it goes.

Here's a copy of my event log.
Any help will be appreciated.

A couple of questions:
1. Does this occur with CPU tasks, GPU tasks, or both?
2. Do you have "Leave non-GPU tasks in memory while suspended" enabled or disabled?
3. Do you notice the progress resetting with the BOINC Manager active? That is, it is the active window, and you're moving the mouse (causing activity)?

Thank you!

Bill

Thanks for the suggestiion, Jord
Unfortunately, it didn't solve the issue.

Hello Bill, thanks for your commitment
Some feedback first:
- The suggestion from Jord did not have any effect
- Neither did it work stepping back to version 7.16; I'm again on the latest "stable" release: 7.20.4

Answering your own questions:
1. On a M1 Mac - as far as I understand - there's no difference between CPU/GPU, right?
2. Where should I look for "Leave non-GPU tasks in memory while suspended" ? (Sorry... I didn't find this option)
3. While BOINC Manager is active, mouse moving does not cause any progress resetting. (Anyway, progress was kept for 2 of the tasks on 0.000% until it reached 1%.)
BUT those two tasks DID reset their progress display while resuming activity after having been suspended through computer activity!

Thanks for further help, if possible

Thanks for the suggestiion, Jord
Unfortunately, it didn't solve the issue.

The suggestion to turn on the checkpoint_debug flag will have the effect that it shows in the Event Log (Command+SHIFT+E) whether or not the affected science application(s) checkpoint or not. That way you can see if they do, and post that evidence and we can try to figure things out further. As you have the effect that you start from zero every time, correct?

Hello Bill, thanks for your commitment
Some feedback first:
- The suggestion from Jord did not have any effect
- Neither did it work stepping back to version 7.16; I'm again on the latest "stable" release: 7.20.4

Answering your own questions:
1. On a M1 Mac - as far as I understand - there's no difference between CPU/GPU, right?
2. Where should I look for "Leave non-GPU tasks in memory while suspended" ? (Sorry... I didn't find this option)
3. While BOINC Manager is active, mouse moving does not cause any progress resetting. (Anyway, progress was kept for 2 of the tasks on 0.000% until it reached 1%.)
BUT those two tasks DID reset their progress display while resuming activity after having been suspended through computer activity!

Thanks for further help, if possible

1. I'm not familiar with M1 Mac, but if you don't see "GPU" under the status of the task, then it is likely a CPU task. It would be next to "running" or "ready to start".
2. Look under Options->"Computing preferences...", then click on the "disk and memory" tab, and it is towards the bottom.
3. I think I understand what you're saying, and that is the concern I have. If I hear you correctly, the tasks suspend when you move the mouse, but after the computer returns to an idle state (not moving the mouse), then the tasks suspend. Check the setting I pointed you to in #2 above. It should be enabled. If it is not enabled, please click the box and see if that fixes the problem. If it is already enabled, then we have more troubleshooting to do. A good start would be to know what project and tasks you are having these problems with. As Jord and Tenny pointed out, the checkpoint debug flag would help as well. Let us know if you need more help with getting the debug flag set up.

M1 Mac or M1 Ultra Mac are incapable of running any BOINC gpu application whether OpenCL or CUDA as M1* Mac does not support either graphic compute API.

---

M1 Mac or M1 Ultra Mac are incapable of running any BOINC gpu application whether OpenCL or CUDA as M1* Mac does not support either graphic compute API.

This is horrible to read. I assume it also applies to mac M2 ? and so any Mac in the future will never allow any GPGPU capacity ?

Well it seems PrimeGrid did the job :

Intel ARC GPUs! Apple M1/M2 CPU and GPU!!!

Effective immediately, PrimeGrid has apps for the following sub-projects for the new hardware:

Native Apple M1/M2 ARM CPU apps are available on GFN-17 through GFN-22.

Native M1/M2 GPU apps are available on GFN-16 through GFN-22 as well as DYFL.

Intel ARC GPU apps are available on GFN-16 through GFN-22 as well as DYFL.

AFAIK OpenCL has been deprecated for Macs for some time now, even though most MacOSes still support it, even on their ARM64 Apple Silicon CPUs. But they do want to change over to Metal. You can read about it in #2550.

Thanks for that info. Believe they are the only project to manage that. Is it because those Primegrid applications are working for the older MacOS' where there was still some OpenCL compatibility?
What about the latest MacOS?

---