Re: Is it possible to enhance the vector STG registers(Xmm, Ymm, Zmm) with more information?

The formatting of the function[1] got spoilt in the last mail. It is modified from: globalRegType _ (XmmReg _) = cmmVec 4 (cmmBits W32) to globalRegType _ (XmmReg _ m ty) = let (l,w) = fromMaybe (2, W64) m in case fromMaybe Float ty of Integer -> cmmVec l (cmmBits w) Float -> cmmVec l (cmmFloat w) Thank, Abhiroop [1] https://github.com/Abhiroop/ghc-1/blob/wip/simd-ncg-support/compiler/cmm/Cmm... On Fri, Jul 13, 2018 at 4:26 PM Abhiroop Sarkar wrote:

Sorry I mistyped and forgot to include the register number in the mail but it is present in the patch. So it is:

data GlobalReg = ... | XmmReg !Int (Maybe (Length, Width)) (Maybe GlobalVecRegTy)

The purpose of the GlobalVecRegTy field is to rewrite this function definition : https://github.com/ghc/ghc/blob/master/compiler/cmm/CmmExpr.hs#L585-L587

globalRegType _ (XmmReg _) = cmmVec 4 (cmmBits W32)

and rewrite it to :

globalRegType _ (XmmReg _ m ty) = let (l,w) = fromMaybe (2, W64) m -- a random default value of 2 and W64 chosen in case fromMaybe Float ty of Integer -> cmmVec l (cmmBits w) Float -> cmmVec l (cmmFloat w) Thanks, Abhiroop

On Fri, Jul 13, 2018 at 4:12 PM Simon Peyton Jones wrote:

...

What is the GlobalVecRegTy field doing?

Don’t you need an Int for the register number, like all the rest?

Generally, sounds good though

S

*From:* Abhiroop Sarkar *Sent:* 13 July 2018 14:07 *To:* Simon Peyton Jones *Cc:* ghc-devs@haskell.org *Subject:* Re: Is it possible to enhance the vector STG registers(Xmm, Ymm, Zmm) with more information?

Hello Simon,

Thanks for your response. I had written a patch[1] for this and the approach I took was quite similar to what you pointed out.

data GlobalReg = ... | XmmReg

(Maybe (Length, Width))

(Maybe GlobalVecRegTy)

data GlobalVecRegTy = Integer | Float

-- Width and Length are already defined

data Width = W8 | W16 | W32 .....

type Length = Int

I wrapped the types inside a `Maybe` because when initializing a GlobalReg (in the `activeStgRegs`[2] function), I was not sure what value to initialize the register with, so I used a `Nothing` when initializing.

I see now in the case of `VanillaReg` it is initialized with the `VGcPtr` constructor: VanillaReg 1 VGcPtr etc

I think I should modify my patch as well to remove the Maybe and initialize with some default Length, Width and GlobalRegTy. Thanks for the help.

Abhiroop

[1] https://phabricator.haskell.org/D4922

[2] https://github.com/ghc/ghc/blob/master/includes/CodeGen.Platform.hs#L450-L62... https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fghc%2Fghc%2Fblob%2Fmaster%2Fincludes%2FCodeGen.Platform.hs%23L450&data=02%7C01%7Csimonpj%40microsoft.com%7Ccb18f3003224456cfd5b08d5e8c19b21%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636670840587296823&sdata=quBJxkG7r6a7ylFqrh4VMgP8y0P1i1Xc79Xj10kV9ts%3D&reserved=0

On Fri, Jul 13, 2018 at 10:58 AM Simon Peyton Jones < simonpj@microsoft.com> wrote:

Abhiroop

Did anyone reply?

My instinct is this. You want to use the same register (say Xmm reg 3) in different ways. We already have this for ‘VanillaReg’:

data GlobalReg

= VanillaReg -- pointers, unboxed ints and chars

Int -- its number

VGcPtr

| …

data VGcPtr = VGcPtr | VNonGcPtr

We use VanillaReg for both pointers and non-pointers, so (VanillaReg 3 VGcPtr) is register 3 used as a pointer, and (VanillaReg 3 VNonGcPtr) is register 3 used as a non-pointer. And notice that globalRegType looks at this field to decide what type to return.

I think you can do exactly the same: add a field to Xmm that explains how you are gong to divide it up. Would that work?

Simon

*From:* ghc-devs *On Behalf Of *Abhiroop Sarkar *Sent:* 27 June 2018 22:32 *To:* ghc-devs@haskell.org *Subject:* Is it possible to enhance the vector STG registers(Xmm, Ymm, Zmm) with more information?

Hello all,

I am currently working on adding support for SIMD operations to the native code generator. One of the roadblocks I faced recently was the definition of the `globalRegType` function in "compiler/cmm/CmmExpr.hs". The `globalRegType` function maps the STG registers to the respective `CmmType` datatype.

For Xmm, Ymm, Zmm registers the function defines globalRegType like this: https://github.com/ghc/ghc/blob/master/compiler/cmm/CmmExpr.hs#L585-L587 https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fghc%2Fghc%2Fblob%2Fmaster%2Fcompiler%2Fcmm%2FCmmExpr.hs%23L585-L587&data=02%7C01%7Csimonpj%40microsoft.com%7Ccb18f3003224456cfd5b08d5e8c19b21%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636670840587296823&sdata=Ct2zsaFgEsRTclglezuPAg05zlSWwuyU45KZrrzcZCc%3D&reserved=0

Consider the case for an Xmm register, the above definition limits an Xmm register to hold only vectors of size 4. However we can store 2 64-bit Doubles or 16 Int8s or 8 Int16s and so on

The function `globalRegType` is internally called by the function `cmmRegType` ( https://github.com/ghc/ghc/blob/838b69032566ce6ab3918d70e8d5e098d0bcee02/com... https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fghc%2Fghc%2Fblob%2F838b69032566ce6ab3918d70e8d5e098d0bcee02%2Fcompiler%2Fcmm%2FCmmExpr.hs%23L275&data=02%7C01%7Csimonpj%40microsoft.com%7Ccb18f3003224456cfd5b08d5e8c19b21%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C636670840587296823&sdata=yi3CN5tdL%2FEteEwBRH6Pt%2B3QB7Ji0sGWnnlE02eyMqk%3D&reserved=0) which is itself used in a number of places in the x86 code generator.

In fact depending on the result of the `cmmRegType` function is another important function `cmmTypeFormat` defined in Format.hs whose result is used to print the actual assembly instruction.

I have extended all the other Format types to include VectorFormats, however this definition of the `globalRegType` seems incorrect to me. Looking at the signature of the function itself:

`globalRegType :: DynFlags -> GlobalReg -> CmmType`

its actually difficult to predict the CmmType by just looking at the GlobalReg in case of Xmm, Ymm, Zmm. So thats why my original question how do I go about solving this. Should I modify the GlobalReg type to contain more information like Width and Length(for Xmm, Ymm, Zmm) or do I somehow pass the length and width information to the globalRegType function?

Thanks

Abhiroop Sakar

--

Kloona - Coming Soon!

-- Kloona - Coming Soon!

-- Kloona - Coming Soon!