Skip to content

Revert "scripts: NULL empty strings array"#380

Closed
spencer-lunarg wants to merge 1 commit into
KhronosGroup:mainfrom
spencer-lunarg:spencer-lunarg-we-spent-so-much-time-on-this-dumb-field-that-wasnt-used-for-10-years
Closed

Revert "scripts: NULL empty strings array"#380
spencer-lunarg wants to merge 1 commit into
KhronosGroup:mainfrom
spencer-lunarg:spencer-lunarg-we-spent-so-much-time-on-this-dumb-field-that-wasnt-used-for-10-years

Conversation

@spencer-lunarg
Copy link
Copy Markdown
Contributor

@spencer-lunarg spencer-lunarg commented May 4, 2026

This reverts commit a96bd49.

see KhronosGroup/Vulkan-ValidationLayers#12185 (review)

@arno-lunarg which test was giving you a self validation error... that test might be just wrong now, I will not merge this yet, but want to have this ready as I don't agree with the previous change

(sorry, I'm already pissed the working group decided to mess around with ppEnabledLayerNames after it was not broken for 10 years and now we are all having to play stupid code gen dances around a fake problem)

@arno-lunarg
Copy link
Copy Markdown
Contributor

arno-lunarg commented May 4, 2026

Pretty much any test should trigger that issue with self validation, as long as the VkDeviceCreateInfo the test uses has 0 and NULL ppEnabledLayerNames and corresponding count. Self validation will intercept the safe struct made from it and thus trigger validation error

@arno-lunarg
Copy link
Copy Markdown
Contributor

Also this fix aside, we should not try to allocate memory for empty arrays, so I would vote for keeping those changes anyway

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants