Daniel's macos_x11_suma branch continues to make updates for suma on this side, but those SUMA_XErrHandler and GLXBadContext errors from suma are a different matter. Daniel and I were just looking into that, since one of my systems had the same problem, and we seem to have a solution, or a set of them (mostly done by Daniel).
It seems like some settings under X11 can get corrupted, and it might be enough to simply rewrite them. Start by noting your current setting for enable_iglx, and then probably setting it to zero (make note of the output from the first (read) command):
Since we won't have worry about the other GLX issue from this thread, it seems like now we should just go through this together. I will send you a private message about that.
Hi folks, I'm experiencing a similar issue, where (i) SUMA is black when I initially open it, and the brain only appears if I resize the window and (ii, more important to me) the surface controller isn't displaying the color bar even though the surface dset coloration looks normal.
I'm running MacOS Tahoe (26.3) with an M5 Pro chip, and my afni -ver output is: Precompiled binary macos_13_ARM: Apr 28 2026 (Version AFNI_26.1.00 'Balbinus')
We have spent some time looking into various fixes for this new problem with updated versions of mesa causing suma to show blank windows on the MacOS. This is a completely different issue than the resizing issue of Tahoe (MacOS v.26) and can affect earlier Sequoia (MacOS v.15) release also with both Intel and ARM architectures. The problem is limited to suma, which makes use of the OpenGL libraries from mesa.
The longest-term solution we found is to use a homebrew "tap", copy of an older version of mesa. Thanks to @pmolfese, who made this tap for us, and we forked it to our git repository.
Please see the installation instructions here, and let us know if there are any problems.
The
National Institute of Mental Health (NIMH) is part of the National Institutes of
Health (NIH), a component of the U.S. Department of Health and Human
Services.