[Mesa-dev] [RFC PATCH 00/16] A new IR for Mesa

Connor Abbott cwabbott0 at gmail.com
Wed Aug 20 05:12:51 PDT 2014


On Tue, Aug 19, 2014 at 11:33 PM, Francisco Jerez <currojerez at riseup.net> wrote:
> Connor Abbott <cwabbott0 at gmail.com> writes:
>
>> On Tue, Aug 19, 2014 at 11:40 AM, Francisco Jerez <currojerez at riseup.net> wrote:
>>> Tom Stellard <tom at stellard.net> writes:
>>>
>>>> On Tue, Aug 19, 2014 at 11:04:59AM -0400, Connor Abbott wrote:
>>>>> On Mon, Aug 18, 2014 at 8:52 PM, Michel Dänzer <michel at daenzer.net> wrote:
>>>>> > On 19.08.2014 01:28, Connor Abbott wrote:
>>>>> >> On Mon, Aug 18, 2014 at 4:32 AM, Michel Dänzer <michel at daenzer.net> wrote:
>>>>> >>> On 16.08.2014 09:12, Connor Abbott wrote:
>>>>> >>>> I know what you might be thinking right now. "Wait, *another* IR? Don't
>>>>> >>>> we already have like 5 of those, not counting all the driver-specific
>>>>> >>>> ones? Isn't this stuff complicated enough already?" Well, there are some
>>>>> >>>> pretty good reasons to start afresh (again...). In the years we've been
>>>>> >>>> using GLSL IR, we've come to realize that, in fact, it's not what we
>>>>> >>>> want *at all* to do optimizations on.
>>>>> >>>
>>>>> >>> Did you evaluate using LLVM IR instead of inventing yet another one?
>>>>> >>>
>>>>> >>>
>>>>> >>> --
>>>>> >>> Earthling Michel Dänzer            |                  http://www.amd.com
>>>>> >>> Libre software enthusiast          |                Mesa and X developer
>>>>> >>
>>>>> >> Yes. See
>>>>> >>
>>>>> >> http://lists.freedesktop.org/archives/mesa-dev/2014-February/053502.html
>>>>> >>
>>>>> >> and
>>>>> >>
>>>>> >> http://lists.freedesktop.org/archives/mesa-dev/2014-February/053522.html
>>>>> >
>>>>> > I know Ian can't deal with LLVM for some reason. I was wondering if
>>>>> > *you* evaluated it, and if so, why you rejected it.
>>>>> >
>>>>> >
>>>>> > --
>>>>> > Earthling Michel Dänzer            |                  http://www.amd.com
>>>>> > Libre software enthusiast          |                Mesa and X developer
>>>>>
>>>>>
>>>>> Well, first of all, the fact that Ian and Ken don't want to use it
>>>>> means that any plan to use LLVM for the Intel driver is dead in the
>>>>> water anyways - you can translate NIR into LLVM if you want, but for
>>>>> i965 we want to share optimizations between our 2 backends (FS and
>>>>> vec4) that we can't do today in GLSL IR so this is what we want to use
>>>>> for that, and since nobody else does anything with the core GLSL
>>>>> compiler except when they have to, when we start moving things out of
>>>>> GLSL IR this will probably replace GLSL IR as the infrastructure that
>>>>> all Mesa drivers use. But with that in mind, here are a few reasons
>>>>> why we wouldn't want to use LLVM:
>>>>>
>>>>> * LLVM wasn't built to understand structured CFG's, meaning that you
>>>>> need to re-structurize it using a pass that's fragile and prone to
>>>>> break if some other pass "optimizes" the shader in a way that makes it
>>>>> non-structured (i.e. not expressible in terms of loops and if
>>>>> statements). This loss of information also means that passes that need
>>>>> to know things like, for example, the loop nesting depth need to do an
>>>>> analysis pass whereas with NIR you can just walk up the control flow
>>>>> tree and count the number of loops we hit.
>>>>>
>>>>
>>>> LLVM has a pass to structurize the CFG.  We use it in the radeon
>>>> drivers, and it is run after all of the other LLVM optimizations which have
>>>> no concept of structured CFG.  It's not bug free, but it works really
>>>> well even with all of the complex OpenCL kernels we throw at it.
>>>>
>>>> Your point about losing information when the CFG is de-structurized is
>>>> valid, but for things like loop depth, I'm not sure why we couldn't write an
>>>> LLVM analysis pass for this (if one doesn't already exist).
>>>>
>>>
>>> I don't think this is such a big deal either.  At least the
>>> structurization pass used on newer AMD hardware isn't "fragile" in the
>>> way you seem to imply -- AFAIK (unlike the old AMDIL heuristic
>>> algorithm) it's guaranteed to give you a valid structurized output no
>>> matter what the previous optimization passes have done to the CFG,
>>> modulo bugs.  I admit that the situation is nevertheless suboptimal.
>>> Ideally this information wouldn't get lost along the way.  For the long
>>> term we may want to represent structured control flow directly in the IR
>>> as you say, I just don't see how reinventing the IR saves us any work if
>>> we could just fix the existing one.
>>
>> It seems to me that something like how we represent control flow is a
>> pretty fundamental part of the IR - it affects any optimization pass
>> that needs to do anything beyond adding and removing instructions. How
>> would you fix that, especially given that LLVM is primarily designed
>> for CPU's where you don't want to be restricted to structured control
>> flow at all? It seems like our goals (preserve the structure) conflict
>> with the way LLVM has been designed.
>>
> I think we can fix this by introducing new structured variants of the
> branch instruction in a way that doesn't alter the fundamental structure
> of the IR.  E.g. an if branch could look like:
>
> ifbr i1 <cond>, label <iftrue>, label <iffalse>, label <join>
>
> Where both branches are guaranteed to converge at <join>.  Sure, this
> will require fixing many assumptions, but on the one hand it's not
> immediately required (as we can address this problem for the time being
> using the same solution AMD uses) and on the other hand it's still less
> work than starting from scratch.

I disagree with the "less work than starting from scratch" part,
especially since it involves modifying it in a pretty invasive way,
when we won't even need half of the things that it does for us. LLVM
just isn't a solution to everything - there is no one-size-fits-all
compiler.

>
>>>
>>>>> * LLVM doesn't do modifiers, meaning that we can't do optimizations
>>>>> like "clamp(x, 0.0, 1.0) => mov.sat x" and "clamp(x, 0.25, 1.0) =>
>>>>> max.sat(x, .25)" in a generic fashion.
>>>>>
>>>>
>>>> The way to handle this with LLVM would be to add intrinsics to represent
>>>> the various modifiers and then fold them into instructions during
>>>> instruction selection.
>>>>
>>>
>>> IMHO this is a feature.  One of the things I don't like about NIR is
>>> that it's still vec4-centric.  Most drivers are going to want something
>>> else and different to each other, we cannot please all of them with one
>>> single vector addressing model built into the core instruction set, so
>>> I'd rather have modifiers, writemasks and swizzles represented as the
>>> composition of separate instructions/intrinsics with simple and
>>> well-defined semantics, which can be coalesced back into the real
>>> instruction as Tom says (easy even if you don't use LLVM's instruction
>>> selector as long as it's SSA form).
>>
>> While NIR is vec4-centric, nothing's stopping you from splitting up
>> instructions and doing optimizations at the scalar level for scalar
>> ISA's - in fact, that's what I expect to happen. And for backends that
>> really do need to have swizzles and writemasks, coalescing these
>> things back into the original instruction is not at all trivial
>
> It's a simple peephole optimization AFAICT:
>
> val2 = alu-op(modifier(val1)) -> hardware-specific-extended-alu-op(val)
> val2 = shuffle(val2, alu-op(val1)) -> hardware-specific-alu-op-with-writemask(val2, val1)

No, it's not. Imagine something like:

vec4 foo = ...
vec4 bar = ...
vec4 baz = vec4(foo.xy, bar.zw)
... = foo
... = bar
... = baz

where the vec4() is the shuffle instruction. In this case, you can't
eliminate the shuffle - you need to insert writemasked moves when you
come out of SSA:

vec4 foo = ...
vec4 bar = ...
baz.xy = foo.xy
baz.zw = bar.zw

This basically comes down to something analogous to a register
allocation problem, where in this case the scalar components that we
want to put into a single vec4 (foo, bar, and baz) can't fit - we need
to "spill" by inserting copies. Then, once we've done this, we have to
convert it into a non-SSA form with registers, writemasks, and
swizzles - something that would be easy to do in the IR -> backend
translation, if it really were just a simple peephole, but in this
case it's not and so you either have to consult the result of your
analysis during the translation or have an IR that can represent
swizzles, writemasks, and non-SSA registers for you like NIR does. Of
course, LLVM will help with none of this because it's vectorization
model is built around CPU vector processors like SSE, NEON, etc. and
so AFAIK it has no concept of per-component liveness, and even if it
did, this stuff is intimately tied to the out-of-SSA process itself so
we would basically have to write it from scratch anyways.

>
>> - in fact, going into and out of SSA without introducing extra copies
>> even in situations like:
>>
>> foo.xyz = ...
>> ... = foo
>> foo.x = ...
>>
>> is a problem that hasn't been solved yet publicly (it seems doable,
>> but difficult).
>
> This problem is orthogonal to the mechanism you use to represent
> swizzles and writemasks AFAICT.  How could having these modifiers built
> into the core ISA help you with transforming vector ops in and out of
> SSA?
>
>> So while we might not need swizzles and writemasks for most backends,
>> for the few that do need it (like, for example, the i965 vec4 backend)
>> it will be very nice to have one common lowering pass that solves this
>> hard problem, which would be impossible to do without having swizzles
>> and writemasks in the IR.
>
> I disagree.  It would be possible if the IR is extensible enough for
> back-ends to be able to represent their exotic vector addressing modes
> as driver-defined machine instructions in a way that generic
> optimization passes can still deal with them.
>
>> And it's very likely that these backends, which probably aren't using
>> SSA due to the aforementioned difficulties, will also benefit from
>> having modifiers already folded for them - this is something that's
>> already a problem for i965 vec4 backend and that NIR will help a lot.
>>
> Well, I have the impression that much of the reason why the i965 vec4
> backend has lagged behind so much in comparison with the fs backend is
> precisely because it's so annoying to optimize vec4 code.  It seems
> painful to me that you have this built into the core instruction set so
> generic optimization passes will have to be explicitly aware of it.  I
> wouldn't be surprised if the i965 vec4 benefited at least as much from
> scalarizing the code, performing optimizations there, and re-vectorizing
> afterwards.
>
>>[...]


More information about the mesa-dev mailing list