Closed Bug 625994 Opened 13 years ago Closed 13 years ago

Resize is jerky on Mac on NVIDIA GeForce 9400M with layer acceleration on

Categories

(Core :: Graphics, defect)

Product:
Core
Component:
Graphics
Platform:
x86
macOS
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED INCOMPLETE

People

(Reporter: pcwalton, Unassigned)

Details

(Whiteboard: [frame-rate-monitor]) 

When I have acceleration on, resize is slow (much slower than Chrome or Safari on google.com). The issue seems to be this:

	0.0%	15.5%	mach_kernel	thread_setuserstack
	0.1%	15.2%	mach_kernel	 mach_msg_overwrite_trap
	0.0%	14.0%	mach_kernel	  ipc_kmsg_send
	0.0%	13.8%	mach_kernel	   ipc_kobject_server
	0.0%	11.4%	mach_kernel	    iokit_server_routine
	0.0%	11.3%	mach_kernel	     is_io_connect_method
	0.0%	11.2%	mach_kernel	      IOUserClient::externalMethod(unsigned int, IOExternalMethodArguments*, IOExternalMethodDispatch*, OSObject*, void*)
	0.0%	7.4%	mach_kernel	       shim_io_connect_method_scalarI_structureO
	0.0%	7.4%	GeForce	        IONVGLContext::submit_command_buffer(unsigned int, sIOGLGetCommandBuffer*)
	0.0%	7.3%	GeForce	         NVGLContext::clientMemoryForType(unsigned long, unsigned long*, IOMemoryDescriptor**)
	0.1%	7.3%	GeForce	          IONVGLContext::clientMemoryForType(unsigned long, unsigned long*, IOMemoryDescriptor**)
	0.0%	5.2%	GeForce	           NVGLContext::processCommandBuffer(VendorCommandDescriptor*)
	0.0%	5.2%	GeForce	            IONVGLContext::processCommandBuffer(VendorCommandDescriptor*)
	0.0%	5.1%	GeForce	             NVGLContext::ProcessTokenEntry(VendorGLStreamInfo&)
	0.0%	5.1%	GeForce	              NVGLContext::ProcessTokenPatchHeapObject(VendorGLStreamInfo&)
	0.0%	5.0%	GeForce	               NVGLContext::make_texture_resident(VendorGLStreamInfo*, NVTextureBuffer**, NvHwSurfaceDescRec*, bool, bool, unsigned long)
	0.0%	4.7%	GeForce	                NVGLContext::get_texture_ext(VendorGLStreamInfo*, NVTextureBuffer*, bool, bool, bool)
	0.0%	4.6%	GeForce	                 nvShared::AllocAndLoadTexture(NVTextureBuffer*, bool, bool, IONVGraphicsClient*, IONVSurface*, NVTextureBuffer**, long, unsigned long)
	0.0%	3.2%	GeForce	                  IONVAccelerator::pageonTexture(IONVGraphicsClient*, IONVSurface*, IONVShared*, NVTextureBuffer*)
	0.0%	1.1%	GeForce	                  NVKernel::force_allocate_texture(NVTextureBuffer*, NVTextureBuffer**, unsigned long, IONVSurface*, bool, IONVSurface*)
	0.0%	0.2%	GeForce	                  IONVShared::mapTransferToGART(NVTransferBuffer*, IONVGraphicsClient*, IONVSurface*)
	0.0%	0.0%	GeForce	                 NVGLContext::SplitCommandBuffer(VendorGLStreamInfo*)
	0.0%	0.3%	GeForce	                IONVShared::delete_texture(NVTextureBuffer*)
	0.0%	0.0%	GeForce	                IONVGraphicsClient::addTextureToStream(NVTextureBuffer*)
	0.0%	0.1%	GeForce	              NVGLContext::ProcessTokenDrawBuffer(VendorGLStreamInfo&)
	0.0%	0.0%	mach_kernel	              memcpy
	0.0%	2.0%	GeForce	           IONVSurface::alloc_surfaces_pageq(unsigned long, unsigned long, bool)
I can't reproduce this problem with a Radeon HD 6750M. Is this still a problem on the GeForce 9400M?
Please re-open if the problem persist with STR.
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.