On 01/02/2013 02:31 PM, Terje Bergstr?m wrote:
> On 02.01.2013 04:44, Mark Zhang wrote:
>> Agree. If we are able to do something dynamically, normally that'll be
>> better.
>>
>> Terje, we can get the Tegra version in FUSE. I think we don't need this
>> kind of try-catch logics.
>
> We'd need to h
On 12/31/2012 02:22 PM, Terje Bergstr?m wrote:
> On 28.12.2012 22:48, Thierry Reding wrote:
>> I disagree. We shouldn't be hiding this kind of detail behind an #ifdef.
>> Instead it should be detected at runtime. Otherwise you'll need to build
>> different versions of libdrm for every generation of
On 02.01.2013 04:44, Mark Zhang wrote:
> Agree. If we are able to do something dynamically, normally that'll be
> better.
>
> Terje, we can get the Tegra version in FUSE. I think we don't need this
> kind of try-catch logics.
We'd need to have in user space a map of SoC version -> number of sync
On 01/02/2013 02:31 PM, Terje Bergström wrote:
> On 02.01.2013 04:44, Mark Zhang wrote:
>> Agree. If we are able to do something dynamically, normally that'll be
>> better.
>>
>> Terje, we can get the Tegra version in FUSE. I think we don't need this
>> kind of try-catch logics.
>
> We'd need to h
On 02.01.2013 04:44, Mark Zhang wrote:
> Agree. If we are able to do something dynamically, normally that'll be
> better.
>
> Terje, we can get the Tegra version in FUSE. I think we don't need this
> kind of try-catch logics.
We'd need to have in user space a map of SoC version -> number of sync
On 12/31/2012 02:22 PM, Terje Bergström wrote:
> On 28.12.2012 22:48, Thierry Reding wrote:
>> I disagree. We shouldn't be hiding this kind of detail behind an #ifdef.
>> Instead it should be detected at runtime. Otherwise you'll need to build
>> different versions of libdrm for every generation of
On 28.12.2012 22:48, Thierry Reding wrote:
> I disagree. We shouldn't be hiding this kind of detail behind an #ifdef.
> Instead it should be detected at runtime. Otherwise you'll need to build
> different versions of libdrm for every generation of Tegra. That may be
> fine for NVIDIA provided BSPs,
On 28.12.2012 22:48, Thierry Reding wrote:
> I disagree. We shouldn't be hiding this kind of detail behind an #ifdef.
> Instead it should be detected at runtime. Otherwise you'll need to build
> different versions of libdrm for every generation of Tegra. That may be
> fine for NVIDIA provided BSPs,
On Fri, Dec 28, 2012 at 09:45:48AM +0200, Arto Merilainen wrote:
> On 12/28/2012 08:47 AM, Mark Zhang wrote:
> >>+int tegra_fence_is_valid(const struct tegra_fence *fence)
> >>+{
> >>+int valid = fence ? 1 : 0;
> >>+valid = valid && fence->id != (uint32_t) -1;
> >>+valid = valid && fenc
On 12/28/2012 04:50 PM, Arto Merilainen wrote:
> On 12/28/2012 09:57 AM, Mark Zhang wrote:
>> On 12/28/2012 03:45 PM, Arto Merilainen wrote:
>>> On 12/28/2012 08:47 AM, Mark Zhang wrote:
> +
> +/* Add fences */
> +if (num_fences) {
> +
> +tegra_stream_push(stream
On 12/28/2012 03:45 PM, Arto Merilainen wrote:
> On 12/28/2012 08:47 AM, Mark Zhang wrote:
>>> +int tegra_fence_is_valid(const struct tegra_fence *fence)
>>> +{
>>> +int valid = fence ? 1 : 0;
>>> +valid = valid && fence->id != (uint32_t) -1;
>>> +valid = valid && fence->id < 32;
>>
>>
I just skimmed the code of libdrm while I'm trying to understand the
host1x driver. So below is what I found.
Mark
On 12/13/2012 10:01 PM, Arto Meril?inen wrote:
> From: Arto Merilainen
>
> This patch introduces tegra stream library. The library is used for
> buffer management, command stream co
On Fri, Dec 28, 2012 at 09:45:48AM +0200, Arto Merilainen wrote:
> On 12/28/2012 08:47 AM, Mark Zhang wrote:
> >>+int tegra_fence_is_valid(const struct tegra_fence *fence)
> >>+{
> >>+int valid = fence ? 1 : 0;
> >>+valid = valid && fence->id != (uint32_t) -1;
> >>+valid = valid && fenc
On 12/28/2012 11:04 AM, Mark Zhang wrote:
> On 12/28/2012 04:50 PM, Arto Merilainen wrote:
>>
>> In my opinion asking tegra_stream_begin() to put a bad fence into the
>> stream is a case we should never be. assert() kills the application
>> immediately (in debug builds) and usually this helps the p
On 12/28/2012 09:57 AM, Mark Zhang wrote:
> On 12/28/2012 03:45 PM, Arto Merilainen wrote:
>> On 12/28/2012 08:47 AM, Mark Zhang wrote:
+
+/* Add fences */
+if (num_fences) {
+
+tegra_stream_push(stream,
+nvhost_opcode_setclass(NV_HOST1X_CLA
On 12/28/2012 08:47 AM, Mark Zhang wrote:
>> +int tegra_fence_is_valid(const struct tegra_fence *fence)
>> +{
>> +int valid = fence ? 1 : 0;
>> +valid = valid && fence->id != (uint32_t) -1;
>> +valid = valid && fence->id < 32;
>
> Hardcode here. Assume always has 32 syncpts.
> Change to
On 12/28/2012 11:04 AM, Mark Zhang wrote:
On 12/28/2012 04:50 PM, Arto Merilainen wrote:
In my opinion asking tegra_stream_begin() to put a bad fence into the
stream is a case we should never be. assert() kills the application
immediately (in debug builds) and usually this helps the programmer
On 12/28/2012 04:50 PM, Arto Merilainen wrote:
> On 12/28/2012 09:57 AM, Mark Zhang wrote:
>> On 12/28/2012 03:45 PM, Arto Merilainen wrote:
>>> On 12/28/2012 08:47 AM, Mark Zhang wrote:
> +
> +/* Add fences */
> +if (num_fences) {
> +
> +tegra_stream_push(stream
On 12/28/2012 09:57 AM, Mark Zhang wrote:
On 12/28/2012 03:45 PM, Arto Merilainen wrote:
On 12/28/2012 08:47 AM, Mark Zhang wrote:
+
+/* Add fences */
+if (num_fences) {
+
+tegra_stream_push(stream,
+nvhost_opcode_setclass(NV_HOST1X_CLASS_ID,
+host1x_ucla
On 12/28/2012 03:45 PM, Arto Merilainen wrote:
> On 12/28/2012 08:47 AM, Mark Zhang wrote:
>>> +int tegra_fence_is_valid(const struct tegra_fence *fence)
>>> +{
>>> +int valid = fence ? 1 : 0;
>>> +valid = valid && fence->id != (uint32_t) -1;
>>> +valid = valid && fence->id < 32;
>>
>>
On 12/28/2012 08:47 AM, Mark Zhang wrote:
+int tegra_fence_is_valid(const struct tegra_fence *fence)
+{
+int valid = fence ? 1 : 0;
+valid = valid && fence->id != (uint32_t) -1;
+valid = valid && fence->id < 32;
Hardcode here. Assume always has 32 syncpts.
Change to a micro wrapped
I just skimmed the code of libdrm while I'm trying to understand the
host1x driver. So below is what I found.
Mark
On 12/13/2012 10:01 PM, Arto Meriläinen wrote:
> From: Arto Merilainen
>
> This patch introduces tegra stream library. The library is used for
> buffer management, command stream co
From: Arto Merilainen
This patch introduces tegra stream library. The library is used for
buffer management, command stream construction and work
synchronization.
Signed-off-by: Arto Merilainen
---
Makefile.am|6 +-
configure.ac | 13 +
tegra/Makefile.am
From: Arto Merilainen
This patch introduces tegra stream library. The library is used for
buffer management, command stream construction and work
synchronization.
Signed-off-by: Arto Merilainen
---
Makefile.am|6 +-
configure.ac | 13 +
tegra/Makefile.am
24 matches
Mail list logo
