Kevin Rogovin
Kevin Rogovin

Reputation: 41

What precisely does the warning on M1 GPU "GPU Soft Fault count" mean?

I've seen this message come up, but I do not know what it means exactly; it comes up with a (very) large shader. I first suspected that something was getting software rendered, but I have not seen a performance issue with the shader running (with nearly all content of the frame drawn with that shader). The warning message only appears when running from the debugger.

Anyone know what it exactly means?

Upvotes: 4

Views: 472

Answers (1)

Kaan
Kaan

Reputation: 5794

What precisely does the warning on M1 GPU "GPU Soft Fault count" mean? Anyone know what it exactly means?

A "soft fault" or "soft error" is when some piece of data in memory has been corrupted. Soft errors are transient, present only due to a temporary state of memory while the system is running.

These generally occur due to external factors and have nothing to do with the state of your computer hardware (nothing is broken). Also, it's unlikely that your actions as a user were a factor. There are several causes including cosmic rays; you can read about general causes here.

Soft errors are well known in computing, and the reason behind ECC memory (error correction code). ECC memory has been available for decades, though due to higher cost it generally has been used only for "important" systems (business, corporate settings, etc) and skipped for home-use consumer computer systems.

Here's an excerpt (with a bit of emphasis added by me) from the Wikipedia page for Soft error:

In electronics and computing, a soft error is a type of error where a signal or datum is wrong. Errors may be caused by a defect, usually understood either to be a mistake in design or construction, or a broken component. A soft error is also a signal or datum which is wrong, but is not assumed to imply such a mistake or breakage. After observing a soft error, there is no implication that the system is any less reliable than before. One cause of soft errors is single event upsets from cosmic rays.

In a computer's memory system, a soft error changes an instruction in a program or a data value. Soft errors typically can be remedied by cold booting the computer. A soft error will not damage a system's hardware; the only damage is to the data that is being processed.

To your question specifically about an M1 Mac, I found almost no direct evidence of "GPU soft fault" related to M1 Macs. If the cause of what you observed is due to soft errors, then this would be consistent – a soft error on your system has nothing to do with anyone else observing a soft error. If instead the cause was a hardware problem, we would expect to see other people asking about the same error.

Upvotes: 1

Related Questions