You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 5, 2022. It is now read-only.
From @exavolt:
"""
I've found an implementation incompatibilities of glGet* functions. From the spec1, a glGet* function might returns one or more values which varies between parameters. For example, glGetIntegerv(GL_ACTIVE_TEXTURE) returns only one value, while something like glGetIntegerv(GL_VIEWPORT) would return four values. The binding always assume that the functions return only one value which, in many cases, it makes the functions practically useless because they returned only the first value.
I think that this could cause a more serious issue, i.e., buffer overrun, because the buffer provided to hold the returned values might be smaller than the number of value asked by the parameter.
The updated implementation should use a map of parameter names with their respective expected number of values as defined in the spec1. On call, the binding function will allocate the appropriately-sized buffer based on the parameter name before calling its respective actual glGet function.
rez
"""
The text was updated successfully, but these errors were encountered:
From @exavolt:
"""
I've found an implementation incompatibilities of glGet* functions. From the spec1, a glGet* function might returns one or more values which varies between parameters. For example, glGetIntegerv(GL_ACTIVE_TEXTURE) returns only one value, while something like glGetIntegerv(GL_VIEWPORT) would return four values. The binding always assume that the functions return only one value which, in many cases, it makes the functions practically useless because they returned only the first value.
I think that this could cause a more serious issue, i.e., buffer overrun, because the buffer provided to hold the returned values might be smaller than the number of value asked by the parameter.
The updated implementation should use a map of parameter names with their respective expected number of values as defined in the spec1. On call, the binding function will allocate the appropriately-sized buffer based on the parameter name before calling its respective actual glGet function.
"""
The text was updated successfully, but these errors were encountered: