[Intel-gfx] [PATCH] intel: When re-use cache is full, wait instead of allocating

[Keith Packard]

On Mon, 2008-12-15 at 17:40 -0800, Eric Anholt wrote:
> On Mon, 2008-12-15 at 15:08 -0800, Keith Packard wrote:
> > This limits application memory usage by waiting for the GPU to free memory
> > rather than simply continuing to allocate more memory.
> 
> I think I theoretically prefer the current behavior in the presence of
> max_entries where you still alloc a new buffer (so you're not syncing)
> but you free the buffer being unreferenced instead of adding it to the
> cache.

The goal here was to limit 'extra' memory allocated by an application to
a fixed (huge) amount.

> Of course, I haven't run with max_entries yet since we shouldn't need
> it.

Right. Before your kernel patch to return flushing buffers as not-busy,
I was using this to avoid blowing up the X server.

I'm OK with not including this if we can get the kernel patch landed.

-- 
keith.packard at intel.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20081215/9d60ed65/attachment.sig>