RStudio 1.2 crash in macOS related to GPU acceleration

The latest versions of RStudio 1.2 branch crash constantly after some time. It appears it has something to do with the GPU acceleration.

  • While running RStudio 1.1.463, my computer uses the integrated Intel Iris Pro GPU without issues.
  • If I uncheck the options: "Use GPU blacklist" and "Use GPU driver bug workarounds" I can use RStudio 1.2.
  • With these two options checked, the application crash.
Process:               RStudio [8115]
Path:                  /Users/USER/Desktop/RStudio_1.2.app/Contents/MacOS/RStudio
Identifier:            org.rstudio.RStudio
Version:               1.2.1268 (1.2.1268)
Code Type:             X86-64 (Native)
Parent Process:        ??? [1]
Responsible:           RStudio [8115]
User ID:               501

Date/Time:             2019-02-09 12:10:40.659 -0300
OS Version:            Mac OS X 10.14.3 (18D109)
Report Version:        12
Anonymous UUID:        8E679F31-DC1D-0CB0-E713-D4AC838EA90D

Sleep/Wake UUID:       31AE5748-92D4-44CB-831E-C8180C2923A5

Time Awake Since Boot: 48000 seconds
Time Since Wake:       3000 seconds

System Integrity Protection: enabled

Crashed Thread:        19  Chrome_InProcGpuThread

Exception Type:        EXC_CRASH (SIGABRT)
Exception Codes:       0x0000000000000000, 0x0000000000000000
Exception Note:        EXC_CORPSE_NOTIFY

Application Specific Information:
abort() called

Application Specific Signatures:
Graphics kernel error: 0xfffffffc

Model: MacBookPro11,3, BootROM 149.0.0.0.0, 4 processors, Intel Core i7, 2,5 GHz, 16 GB, SMC 2.19f12
Graphics: Intel Iris Pro, Intel Iris Pro, Built-In
Graphics: NVIDIA GeForce GT 750M, NVIDIA GeForce GT 750M, PCIe

We're tracking this here: https://github.com/rstudio/rstudio/issues/4034.

Interesting to note that unchecking the two GPU-related options makes a difference in your case, though. I'll copy this as a suggestion for other users.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.