[Mesa-dev] [PATCH] gallium/radeon: never use staging buffers with AMD_pinned_memory

[Edward O'Callaghan]

Reviewed-by: Edward O'Callaghan <funfunctor at folklore1984.net>

On 04/13/2017 01:56 AM, Marek Olšák wrote:
> Reviewed-by: Marek Olšák <marek.olsak at amd.com>
> 
> Marek
> 
> On Wed, Apr 12, 2017 at 12:44 PM, Nicolai Hähnle <nhaehnle at gmail.com> wrote:
>> From: Nicolai Hähnle <nicolai.haehnle at amd.com>
>>
>> ---
>>  src/gallium/drivers/radeon/r600_buffer_common.c | 18 ++++++++++++++++--
>>  1 file changed, 16 insertions(+), 2 deletions(-)
>>
>> diff --git a/src/gallium/drivers/radeon/r600_buffer_common.c b/src/gallium/drivers/radeon/r600_buffer_common.c
>> index 10036af..0e93404 100644
>> --- a/src/gallium/drivers/radeon/r600_buffer_common.c
>> +++ b/src/gallium/drivers/radeon/r600_buffer_common.c
>> @@ -320,25 +320,39 @@ static bool r600_can_dma_copy_buffer(struct r600_common_context *rctx,
>>
>>  static void *r600_buffer_transfer_map(struct pipe_context *ctx,
>>                                        struct pipe_resource *resource,
>>                                        unsigned level,
>>                                        unsigned usage,
>>                                        const struct pipe_box *box,
>>                                        struct pipe_transfer **ptransfer)
>>  {
>>         struct r600_common_context *rctx = (struct r600_common_context*)ctx;
>>         struct r600_common_screen *rscreen = (struct r600_common_screen*)ctx->screen;
>> -        struct r600_resource *rbuffer = r600_resource(resource);
>> -        uint8_t *data;
>> +       struct r600_resource *rbuffer = r600_resource(resource);
>> +       uint8_t *data;
>>
>>         assert(box->x + box->width <= resource->width0);
>>
>> +       /* From GL_AMD_pinned_memory issues:
>> +        *
>> +        *     4) Is glMapBuffer on a shared buffer guaranteed to return the
>> +        *        same system address which was specified at creation time?
>> +        *
>> +        *        RESOLVED: NO. The GL implementation might return a different
>> +        *        virtual mapping of that memory, although the same physical
>> +        *        page will be used.
>> +        *
>> +        * So don't ever use staging buffers.
>> +        */
>> +       if (rscreen->ws->buffer_is_user_ptr(rbuffer->buf))
>> +               usage |= PIPE_TRANSFER_PERSISTENT;
>> +
>>         /* See if the buffer range being mapped has never been initialized,
>>          * in which case it can be mapped unsynchronized. */
>>         if (!(usage & PIPE_TRANSFER_UNSYNCHRONIZED) &&
>>             usage & PIPE_TRANSFER_WRITE &&
>>             !rbuffer->is_shared &&
>>             !util_ranges_intersect(&rbuffer->valid_buffer_range, box->x, box->x + box->width)) {
>>                 usage |= PIPE_TRANSFER_UNSYNCHRONIZED;
>>         }
>>
>>         /* If discarding the entire range, discard the whole resource instead. */
>> --
>> 2.9.3
>>
>> _______________________________________________
>> mesa-dev mailing list
>> mesa-dev at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20170413/2abbbba4/attachment.sig>