[Mesa-dev] [PATCH 22/22] r600g: fix uploading non-zero mipmap levels of depth textures

[Marek Olšák]

On Tue, Jul 10, 2012 at 11:44 AM, Michel Dänzer <michel at daenzer.net> wrote:
> On Mon, 2012-07-09 at 21:15 +0200, Marek Olšák wrote:
>> This fixes piglit/depth-level-clamp.
>> ---
>>  src/gallium/drivers/r600/r600_texture.c |    5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/src/gallium/drivers/r600/r600_texture.c b/src/gallium/drivers/r600/r600_texture.c
>> index a6a83ca..d16c252 100644
>> --- a/src/gallium/drivers/r600/r600_texture.c
>> +++ b/src/gallium/drivers/r600/r600_texture.c
>> @@ -926,8 +926,9 @@ void r600_texture_transfer_destroy(struct pipe_context *ctx,
>>
>>                       u_box_origin_2d(texture->width0, texture->height0, &sbox);
>>
>> -                     ctx->resource_copy_region(ctx, texture, 0, 0, 0, 0,
>> -                                               &rtransfer->staging->b.b, 0,
>> +                     ctx->resource_copy_region(ctx, texture, transfer->level,
>> +                                               0, 0, transfer->box.z,
>> +                                               &rtransfer->staging->b.b, transfer->level,
>
> Does this really need to take into account transfer->box.z, but not .x
> and .y?

Yes, it could be optimized more. I'd rather leave the optimization to
some other patch. The read transfers always copy the whole layer too.

Marek