Comme=
nt # 12
on bug 10002=
4
from Julien Isorce
Hi Michel. Thx you for your comment. In a way it is a good and=
only my attached
mesa will be needed to fix the test. I will send it to the mailing list.
But I still do not properly understand why there is not enough graphic memo=
ry
because lspci reports 2048M. And there are some contradictions depending on=
the
subSize for the test sequence which is:
for i in 0 1:
glGenTextures
glTexImage2D maxSize
glTexSubImage2D subSize =3D maxSize / 2.
glDeleteTextures
A: The kernel trace ENOMEM appears with:
subSize =3D maxSize - 1 -> "radeon: Failed to allocate a buffer&q=
uot;
subSize =3D 3 * maxSize / 4 -> "radeon: Failed to allocate a buff=
er"
subSize =3D maxSize / 2 -> "radeon: Not enough memory for command=
submission"
B: It works with:
subSize =3D maxSize
subSize =3D maxSize / 4
Isn't A and B a contradiction ? For me A should succeeds because
glDeteteTextures should . Also note that on other=20
Also by calling glFush() within an additional loop iteration i =3D 2, it can
recover so it will fail for iteration i =3D 1 but succeeds for i =3D 2.=20
So what could the explanation here ? Should the driver just flush the cs
automatically in that case like it does for subSize =3D maxSize ? Sorry I a=
m just
trying to understand :). Thx!