AMD's Graphics Core Next Preview: AMD's New GPU, Architected For Compute
by Ryan Smith on December 21, 2011 9:38 PM ESTNot Just A New Architecture, But New Features Too
So far we’ve talked about Graphics Core Next as a new architecture, how that new architecture works, and what that new architecture does that Cayman and other VLIW architectures could not. But along with the new architecture GCN will bring with it a number of new compute features to further flesh out AMD’s GPU computing capabilities and to cement the GPU’s position as the CPU’s partner rather than a subservient peripheral.
In terms of base features the biggest change will be that GCN will implement the underlying features necessary to support C++ and other advanced languages. As a result GCN will be adding support for pointers, virtual functions, exception support, and even recursion. These underlying features mean that developers will not need to “step down” from higher languages to C to write code for the GPU, allowing them to more easily program for the GPU and CPU within the same application. For end-users the benefit won’t be immediate, but eventually it will allow for more complex and useful programs to be GPU accelerated.
Because the underlying feature set is evolving, the memory subsystem is also evolving to be able to service those features. The chief change here is that the hardware is being adapted to support an ISA that uses unified memory. This goes hand-in-hand with the earlier language features to allow programmers to write code to target both the CPU and the GPU, as programs (or rather compilers) can reference memory anywhere, without the need to explicitly copy memory from one device to the other before working on it. Now there’s still a significant performance impact when accessing off-GPU memory – particularly in the case of dGPUs where on-board memory is many times faster than system memory – so developers and compilers will still be copying data around to keep it close to the processor that’s going to use it, but this essentially becomes abstracted from developers.
Now what’s interesting is that the unified address space that will be used is the x86-64 address space. All instructions sent to a GCN GPU will be relative to the x86-64 address space, at which point the GPU will be responsible for doing address translation to local memory addresses. In fact GCN will even be incorporating an I/O Memory Mapping Unit (IOMMU) to provide this functionality; previously we’ve only seen IOMMUs used for sharing peripherals in a virtual machine environment. GCN will even be able to page fault half-way gracefully by properly stalling until the memory fetch completes. How this will work with the OS remains to be seen though, as the OS needs to be able to address the IOMMU. GCN may not be fully exploitable under Windows 7.
Finally on the memory side, AMD is adding proper ECC support to supplement their existing EDC (Error Detection & Correction) functionality, which is used to ensure the integrity of memory transmissions across the GDDR5 memory bus. Both the SRAM and VRAM memory can be ECC protected. For the SRAM this is a free operation, while for the VRAM there will be a performance overhead. We’re assuming that AMD will be using a virtual ECC scheme like NVIDIA, where ECC data is distributed across VRAM rather than using extra memory chips/controllers.
Elsewhere we’ve already mentioned FP64 support. All GCN GPUs will support FP64 in some form, making FP64 support a standard feature across the entire lineup. The actual FP64 performance is configurable – the architecture supports ½ rate FP64, but ¼ rate and 1/16 rate are also options. We expect AMD to take a page from NVIDIA here and configure lower-end consumer parts to use the slower rates since FP64 is not currently important for consumer uses.
Finally, for programmers some additional hardware changes have been made to improve debug support by allowing debugging tools to tap the GPU at additional points. The new ISA for GCN will already make debugging easier, but this will further that goal. As with other developer features this won’t directly impact end-users, but it will ultimately lead to better software sooner as the features and tools available for debugging GPU programs have been well behind the well-established tools used for debugging CPU programs.
Facebook
Twitter
RSS
83 Comments
View All Comments
haplo602 - Saturday, June 18, 2011 - link
I hope that AMD delivers. This is exactly what I expected them to do once Llano was anounced. GPU as a coprocessor. Actualy I hoped that AMD would implement a HTX capable GPU, so I can just plug it into a C32 socket (for example) along with an Opteron.The future past Trinity looks interesting.
jamescox - Monday, June 20, 2011 - link
It would be interesting if they produced a form factor with CPU+GPU on a separate card with memory. Ever since AMD moved the memory controller on die, I wondered if we would see CPU + memory on a separate card. It seems to make a lot of sense. A 4 socket motherboard is huge, especially where each socket has 4 to 6 memory slots associated with it. If the CPU and memory were on a separate card, then you could pack them a lot denser, like you can run 4 GPUs off an ATX board now. It might be cheaper than a massive 4 socket board also. I don't know how many HT links you can run through a slot, but you could always use extra cables/connectors like they use for multiple graphics cards.With the GPU using the same memory space as the CPU, then why leave the CPU attached to the slow system memory? Just put one of these hybrid chips attached to some high-speed graphics card like memory on a separate board. Move the slow system memory out to the chipset again. The current memory hierarchy is not exactly optimal in my opinion. I am using a slightly older macbook pro, which only supports 3 GB of memory. With all of the stuff I run, it is paging a lot to a super slow laptop hard drive. I have been tempted to get an SSD to speed it up rather than a new laptop.
Anyway, with the way the memory hierarchy works now, system memory is kind of like a cache for the swap space on disk. System memory has gotten a lot faster, but disk have not, so people are using SSDs to fill the gap. If you directly connect the "graphics memory" to a CPU/GPU combo, then you don't need as much total memory in the system because you would not need multiple copies of the data. You would just pass pointers to data back and forth between the CPU and GPU components.
Also, it would be nice to switch to something non-volatile for the memory connected to the chipset; just use disk as mass storage only. "System" memory wouldn't need to be that fast, since you would probably have a GB or two of high-speed memory on each processor board. The "system" memory would be used more like the SSD boot/swap drive in a current system. I don't think flash is quite there yet, and the other types of non-volatile memory (magnetic RAM , phase-shift RAM, etc) that promise much better performance and durability seem to still be all talk with no real products.
With keeping the current form factor, it would be nice if they could put a large amount of memory in with the CPU/GPU package to act as high-speed memory for the GPU and L4 cache for the CPU. This form factor doesn't support scaling up to multiple chips easily (too large of main-board), but it would be very power efficient for laptops and other small form factor systems. It would require very little off-module communication which saves a lot of power. Maybe they could use a low-power, wide-interface dram chip originally meant for mobile devices.
Hopefully Trinity is more than just a meaningless code name...
Quantumboredom - Sunday, June 19, 2011 - link
On page 4 ("Many SIMDs Make One Compute Unit") there are two figures showing wavefront scheduling on VLIW4 versus GCN. As I read it the figures seem to indicate that in VLIW4, one 4-wide VLIW handles operations from four wavefronts in parallel, but that's not how I've understood AMD's VLIW4. Only a single work-item is executing on a VLIW4-core at any point in time, the occupancy problems of VLIW4 come from ILP within a work-item, not across wavefronts.At any one point in time, a Cayman/VLIW4 compute unit is only executing instructions from a single wavefront (though they need at least two wavefronts to switch between on VLIW4). Again at any one point in time only 16 work-items are actually being executed, and it's within those 16 work-items that ILP must be extraced to fill the VLIW4 units. Since each work-item is executing on a VLIW4-processor, a total of 16*4=64 operations can be done in parallel, but that requires ILP within the work-items.
On GCN this is quite different, where the four 16-wide vector units are actually executing 64 work-items at a time (four times as many as in Cayman). However the point is that each of these work-items are basically executing on a scalar processor, there's no need for ILP anymore. So again we are executing 64 operations in parallel, but now without any need for ILP.
At least this is how I understood the presentation (I was at AFDS). Basically I agree with how the GCN scheduling is illustrated in this article, but the Cayman part looks wrong to me. A Cayman CU can only execute one wavefront at a time, and it only needs two wavefronts to switch between to be able to fully utilize the hardware, not four like the figures here seem to suggest.
Now I'm just a programmer, not an architecture guy, so if anyone could clear this up for me it would be greatly appreciated :)
Ryan Smith - Monday, June 20, 2011 - link
Hi Quantum;After further consideration you're basically right. I should have made a distinction in the figures between instructions and wholly distinct wavefronts. While there are some ILP considerations to be had, basically the elements Cayman accepts should all be instructions from the same wavefront rather than different wavefronts. Cayman can't really work on multiple wavefronts at once.
I don't have the original files on me, but we'll get this fixed in the morning to show that Cayman is consuming multiple instructions from the same wavefront.
-Thanks
Ryan Smith
jamescox - Monday, June 20, 2011 - link
Would a CPU/GPU integrated chip only be a replacement for integrated graphics, or does it have the possibility to move a little farther up? With multi-threading, 4 to 8 thread CPUs will be common in the mainstream, but that will not be a very big die on smaller processes. Most PC software doesn't make use of more than 4 compute intensive threads, so how much room does that leave for GPU hardware? If they solve the memory speed problem by integrating some high-speed memory into the socket (multi-chip module), or something, then it seems like they could possibly get more mainstream performance out of an integrated chip.
If the integrated GPU isn't being used for graphics, then I really don't see that much software that would use it for compute in the PC space. One of the main things mentioned was usually video encode/decode, but it seems that the best solution is to include specific media encode/decode hardware like sandy bridge does. It seems to be just as fast and much more power efficient. If AMD doesn't include a media processing engine, then that could still be a reason to go with Intel. What other PC software could use the compute power?
There is plenty of software that could use it in professional/HPC markets, so it makes sense to make a GPU that can be used for both if it doesn't sacrifice the graphics performance. The newest generations of GPUs have some things in common with Larrabee and Sony's Cell processors, except both of those tried to move too much of the graphics processing abilities into software. AMD didn't make that mistake, but talk of compute abilities for GPUs in the PC/consumer space seems a bit premature without any real applications to take advantage of it.
GaMEChld - Monday, June 20, 2011 - link
Llano already has low level discrete GPU performance, and that's just the tip of the iceberg. You are correct that on smaller processes they will be able to allocate more space to the GPU while maintaining CPU performance. I believe the successor to Trinity (which is the Bulldozer based successor to Llano) is supposed to be on 28nm. If everything goes exactly right, you could potentially have some kind of monster that has i5-2500K CPU performance with Radeon 6800 GPU performance in some maintstream laptop chip a year or two down the road. (Those numbers are all pure speculation)I encourage everyone to take a moment and remember the first computer you ever used, just to pay homage to what we are capable of as a species in just a few short years.
I remember an IBM computer flipped on by a big red toggle that took 2 minutes to boot to a dos prompt...
Targon - Monday, June 20, 2011 - link
I remember the Timex Sinclaire, with 2KB of memory standard hooked up to a black and white TV and cassette tapes to save/load programs. Z80 running at 1MHz...the old 5.25 inch floppies were MUCH better, at least you could get a list of what was on the storage medium without having to load it.jabber - Monday, June 20, 2011 - link
If only our attitudes to each other and other issues had advanced as much as well.GaMEChld - Monday, June 20, 2011 - link
"Because in the end, aren't all religions the same? They tell us what to eat, when to pray, that this lump of clay called Man can somehow shape himself to resemble the divine. But we can never attain that perfect grace if we have hatred in our hearts. So let us celebrate our commonalites. Some of us don't eat pork. Some of us don't eat shellfish. But we all eat chicken. So spread the word: peace and chicken!"~HOMER SIMPSON
:-D
Cyber.Angel - Saturday, October 15, 2011 - link
off-topic?7th day Adventist don't eat meat, yes, not even chicken
AND
in Christian religion it's God who sacrifices, not human
PLUS
there is a requirement of TOTAL change according to Jesus
That is, the "ME" is buried, forgotten and God lives inside of you
meaning a total change in life
God bless America - but...where is the change?