[Mesa-dev] [PATCH] mesa: Don't check (offset + size <= bufObj->Size) in BindBufferRange.

[Ian Romanick]

On 02/07/2013 07:01 PM, Ian Romanick wrote:
> On 02/07/2013 05:54 PM, Paul Berry wrote:
>> In the documentation for BindBufferRange, OpenGL specs from 3.0
>> through 4.1 contain this language:
>>
>>      "The error INVALID_VALUE is generated if size is less than or
>>      equal to zero or if offset + size is greater than the value of
>>      BUFFER_SIZE."
>>
>> This text was dropped from OpenGL 4.2, and it does not appear in the
>> GLES 3.0 spec.
>>
>> Presumably the reason for the change is because come clients change
>> the size of the buffer after calling BindBufferRange.  We don't want
>> to generate an error at the time of the BindBufferRange call just
>> because the old size of the buffer was too small, when the buffer is
>> about to be resized.
>>
>> Since this is a deliberate relaxation of error conditions in order to
>> allow clients to work, it seems sensible to apply it to all versions
>> of GL, not just GL 4.2 and above.
>>
>> (Note that there is no danger of this change allowing a client to
>> access data beyond the end of a buffer.  We already have code to
>> ensure that that doesn't happen in the case where the client shrinks
>> the buffer after calling BindBufferRange).
>>
>> Eliminates a spurious error message in the gles3 conformance test
>> "transform_feedback_offset_size".
>> ---
>>   src/mesa/main/bufferobj.c | 7 -------
>>   1 file changed, 7 deletions(-)
>>
>> diff --git a/src/mesa/main/bufferobj.c b/src/mesa/main/bufferobj.c
>> index 7c06938..b82ba7b 100644
>> --- a/src/mesa/main/bufferobj.c
>> +++ b/src/mesa/main/bufferobj.c
>> @@ -2152,13 +2152,6 @@ _mesa_BindBufferRange(GLenum target, GLuint index,
>>                        (int) size);
>>            return;
>>         }
>> -
>> -      if (offset + size > bufObj->Size) {
>> -         _mesa_error(ctx, GL_INVALID_VALUE,
>> -                     "glBindBufferRange(offset + size %d > buffer
>> size %d)",
>> -                     (int) (offset + size), (int) (bufObj->Size));
>> -         return;
>> -      }
>>      }
>>
>>      switch (target) {
>>
>
> For some reason, this change makes one of the UBO tests fail... while
> making the transform feedback tests not get an error.  Sigh... I'll look
> in to it...

Stupid test... it looks like that test fails because we don't generate 
an error for an offset = 0 when our GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT 
is 1.  However, I think GL_UNIFORM_BUFFER_OFFSET_ALIGNMENT should not be 
1.  It looks like the Windows driver sets 16 (vec4 alignment, makes 
sense).  I don't even know where to look for this in the bspec.