Forget what I wrote the -ENOMEM was a type in my local change :)
Sorry for the noise,
Christian.
Am 05.08.20 um 13:55 schrieb Christian König:
Dave, do you have a branch with the latest version somewhere?
I've tested your ttm-refactor-mem-manager branch from fdo today a bit
and that pretty quickly results in an -ENOMEM.
Looks like a leak somewhere to me.
Christian.
Am 04.08.20 um 04:55 schrieb Dave Airlie:
I've decided to repost the whole queue this time, it has a few driver
patches up front that are just cleanups.
I've reorodered things a little since the first posting, but not much.
misc core/driver cleanups
mem type manager debug callback cleanup (reordered)
new mem type manager init path
new mem type manager cleanup path
new wrapper to access managers
driver managed mem type allocations
cleanup after all of that
rename ttm_bo_manager to ttm_range_manager (new)
rename ttm_mem_type_manager to ttm_resource_manager (new)
rename ttm_mem_reg to ttm_resource (new)
I've marked most things with v2 that have changed, again I'd really
like to land this in drm-misc sooner rather than later, so it would be
good to get driver acks from qxl, vmwgfx and nouveau. (all cc'ed).
I've also commented on the list on any outstanding comments from
previous
review that I disagree with or need more discussion.
Dave.
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/dri-devel
