Check that IDirect3DVertexBuffer9 and IDirect3DIndexBuffer9::Lock succeed....

Check that IDirect3DVertexBuffer9 and IDirect3DIndexBuffer9::Lock succeed.

I've been seeing crashes like this on Windows XP:

0x013319aa 	[libglesv2.dll 	- memcpy.asm:188] 	memcpy
0x0130989a 	[libglesv2.dll 	- vertexdatamanager.cpp:164] 	gl::VertexDataManager::preRenderValidate(int,int,gl::TranslatedAttribute *)
0x01304f66 	[libglesv2.dll 	- context.cpp:1996] 	gl::Context::applyVertexBuffer(unsigned int,int,int,bool *,gl::TranslatedIndexData *)
0x013061a7 	[libglesv2.dll 	- context.cpp:2648] 	gl::Context::drawArrays(unsigned int,int,int)
0x012f7721 	[libglesv2.dll 	- libglesv2.cpp:1741] 	glDrawArrays
0x01c54f1e 	[chrome.dll 	- gles2_cmd_decoder.cc:3179] 	gpu::gles2::GLES2DecoderImpl::DoDrawArrays(unsigned int,int,int)
0x01c59122 	[chrome.dll 	- gles2_cmd_decoder_autogen.h:640] 	gpu::gles2::GLES2DecoderImpl::HandleDrawArrays(unsigned int,gpu::gles2::DrawArrays const &)

[apatrick1, 2010-11-12 01:11:25]

Hi Daniel,

I fixed some NULL dereference bugs that have been showing up on XP, I'm guessing
after device lost.

I tested by simulating Lock failing in a debugger and verifying that it did not
crash.

Thanks,

Al

[dgkoch, 2010-11-12 16:45:25]

We should either:
 - include an ERR trace in all the cases where we bail out early on a null
pointer, or
 - include an check for an unsuccessful Lock call and trace out the result code
in an ERR trace.
Actually doing both is probably the best...

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
File src/libGLESv2/geometry/IndexDataManager.cpp (right):

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
src/libGLESv2/geometry/IndexDataManager.cpp:107: {
please log an error here

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
src/libGLESv2/geometry/IndexDataManager.cpp:218: {
and here...

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
src/libGLESv2/geometry/IndexDataManager.cpp:244: {
etc..

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/VertexDat...
File src/libGLESv2/geometry/VertexDataManager.cpp (right):

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/VertexDat...
src/libGLESv2/geometry/VertexDataManager.cpp:130: {
ERR(..)

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/VertexDat...
src/libGLESv2/geometry/VertexDataManager.cpp:181: return
processNonArrayAttributes(attribs, activeAttribs, translated, count);
I'd rather not change the control flow here like this here as it seems a bit
strange.  Declare a local variable in this function, initialize it to
GL_NO_ERROR, and overwrite it here.  Then 2 lines later we can return the
variable.

[apatrick1, 2010-11-15 20:52:11]

All done.

Thanks,

Al

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
File src/libGLESv2/geometry/IndexDataManager.cpp (right):

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
src/libGLESv2/geometry/IndexDataManager.cpp:107: {
On 2010/11/12 16:45:26, dgkoch wrote:
> please log an error here

Done.

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
src/libGLESv2/geometry/IndexDataManager.cpp:218: {
On 2010/11/12 16:45:26, dgkoch wrote:
> and here...

Done.

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/IndexData...
src/libGLESv2/geometry/IndexDataManager.cpp:244: {
On 2010/11/12 16:45:26, dgkoch wrote:
> etc..

Done.

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/VertexDat...
File src/libGLESv2/geometry/VertexDataManager.cpp (right):

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/VertexDat...
src/libGLESv2/geometry/VertexDataManager.cpp:130: {
On 2010/11/12 16:45:26, dgkoch wrote:
> ERR(..)

Done.

http://codereview.appspot.com/3043042/diff/1/src/libGLESv2/geometry/VertexDat...
src/libGLESv2/geometry/VertexDataManager.cpp:181: return
processNonArrayAttributes(attribs, activeAttribs, translated, count);
On 2010/11/12 16:45:26, dgkoch wrote:
> I'd rather not change the control flow here like this here as it seems a bit
> strange.  Declare a local variable in this function, initialize it to
> GL_NO_ERROR, and overwrite it here.  Then 2 lines later we can return the
> variable.

Done.

[dgkoch, 2010-11-16 01:55:27]

LGTM

Daniel