[External] : Re: Question about RegMask::is_aligned_sets()

Vladimir Ivanov vladimir.x.ivanov at oracle.com
Fri Mar 12 09:36:22 UTC 2021



On 12.03.2021 01:08, Corey Ashford wrote:
> Thanks for your reply, Vladimir.  A few more questions below when you 
> have the chance :)
> 
> On 3/11/21 12:10 AM, Vladimir Ivanov wrote:
>>
>>
>> On 11.03.2021 06:13, Corey Ashford wrote:
>>> Hello Vladimir,
>>>
>>> Currently I'm looking at the register mask defined for the Power64-LE 
>>> machine.  There are 64 128-bit registers defined via reg_def, e.g.:
>>>
>>> reg_def VSR25 ( SOC, SOC, Op_VecX, 25, NULL);
>>>
>>> But only the 20 of those vector registers that are declared as part 
>>> of "reg_class vs_reg( ... )" end up with a mask in the generated 
>>> ad_ppc_expand.cpp source file, and further, each of those registers 
>>> is allocated just a single bit in the register mask:
>>>
>>> const RegMask _VS_REG_mask( 0x0, 0x0, 0x0, 0x0, 0x0, 0xfffff00, 0x0, 
>>> 0x0, 0x0, 0x0 );
>>>
>>> I would have expected that since Op_VecX is a 128-bit type, it would 
>>> have received four bits per register in the mask.
>>>
>>>
>>> On x86, each of the 512-bit vector registers is declared using 16 
>>> Op_RegF register declarations (which makes sense - 16 x 32 = 512), 
>>> but on aarch64, which can have up to a 1024-bit vector register, 
>>> vector registers are declared using just 8 x Op_RegF (8 x 32 = 256 
>>> bits). There is an extensive comment in the aarch64.ad about this, 
>>> but it seems to imply that the 32-bits-per-slot rule is not rigid 
>>> (just as on PPC64)
>>
>> As you noted, there's no relation between register mask (1 slot per 
>> register definition) and ideal register (VecX et al) chosed. You had 
>> to declare as many registers as there are slots in the value to get it 
>> working.
> 
> So we should be declaring four register slots per vector register, 
> instead of one, right?  I'm a bit worried as to that screwing up the 
> existing implementation for vector register allocation.  I will 
> experiment to see what happens.

Take a look at how x86 handles that:

   https://github.com/openjdk/jdk/blob/master/src/hotspot/cpu/x86/x86.ad#L62

> 
>> But, as part of SVE support, Arm came up with special VecA ideal 
>> register which represents a variable-length vector >
>> Considering what you wrote about Power64-LE, I'd avoid VecA and just 
>> provide additional register definitions.
>>
>>> As to your comment about not needing to use a vector register for the 
>>> boolean vectors, that's quite interesting.  So for all vector types 
>>> except for tiny integers, I should be able to use a 64-bit bit 
>>> general purpose register.  I'm very new to all of this that I'm not 
>>> clear how easy it will be to mix and match register types like this, 
>>> but I will start experimenting with the idea.
>>
>> Can you elaborate how masks are represented in Power64-LE ISA?
>>
> 
> There are no special mask/predicate registers.  Masking is done via the 
> uppermost bit of each corresponding element in a vector register (the 
> other bits are ignored).  Conversion from an array of boolean bytes of 
> 0|1 is simple: the boolean bytes of the mask are first shuffled (via the 
> vperm instr. on ppc64) into each element, then arithmetically negated, 
> producing 0|-1, which means either all 0's or all 1's in each element, 
> which effectively sets or clears the upper bit as desired.

That sounds very similar to how masks are represented in AVX/AVX2 on x86.

>> If there are special predicate registers, then you can rely on what 
>> Intel and Arm folks are working on for AVX-512 and SVE support.
>>
>> On x86 AVX/AVX2 ISA, masks occupy wide vector register. So, a mask of 
>> 4 elements for vector of ints (128-bit) ocupies 128-bit vector. And, 
>> currently, ideal vector nodes follow that representation: mask has the 
>> same ideal type as the value it is applied to (e.g., vectorx[4]{int} 
>> and not a vector of 4 booleans).
> 
> The mask is represented in Java is as a byte array, so there has to be a 
> conversion from the byte array in memory to a vector mask (in this case 
> a VecX). So it seems that conversion can done "locally" within node 
> matching code in ppc.ad, and that the mask representation isn't ever 
> seen as a different type for operand matching.  If that's the case, the 
> fog is lifting a little.

There are special ideal nodes inserted to convert vector masks between 
in-memory and in-register representations: VectorLoadMask and 
VectorStoreMask.

Best regards,
Vladimir Ivanov



> 
> Thank you,
> 
> - Corey
>>
>> Best regards,
>> Vladimir Ivanov
>>
>>
>>> On 3/5/21 3:19 PM, Vladimir Ivanov wrote:
>>>> Hi Corey,
>>>>
>>>>> I'd like to understand the concept of "aligned sets" in RegMask.  I 
>>>>> believe I understand the RegMask idea overall, but I don't 
>>>>> understand the idea of alignment of sets (actually the concept of 
>>>>> sets in this context is also fuzzy).  I've looked at the code that 
>>>>> implements is_aligned_sets, and I just can't yet seem to grok what 
>>>>> requirement it is trying to verify.  I read RegMask.hpp's comments 
>>>>> on the method protoype, and it didn't help me much, I'm afraid.  If 
>>>>> someone could give a paragraph or two of explanation, I'd really 
>>>>> appreciate it.
>>>>
>>>> A register in RegMask is comprised of packed bits each representing 
>>>> a 32-bit slot. So, a VecX register occupies 4 bits (128 = 4 x 32) 
>>>> while VecZ needs 16 (512 = 16 x 32).
>>>>
>>>> Some code relies on the alignment when recovering base register from 
>>>> VMReg:
>>>>
>>>> https://urldefense.com/v3/__https://github.com/openjdk/jdk/blob/e1cad97049642ab201d53ff608937f7e7ef3ff3e/src/hotspot/cpu/x86/registerMap_x86.cpp*L29__;Iw!!GqivPVa7Brio!ONsyUV4W1ucXIXrwfbZN_XYI2Q8iF0_VwN44U7AfWOKj6jGzA9o19MVhnSL9h4n4MN_Kgz0$ 
>>>>
>>>>
>>>> src/hotspot/cpu/x86/registerMap_x86.cpp
>>>>
>>>>      29 address RegisterMap::pd_location(VMReg reg) const {
>>>>      30   if (reg->is_XMMRegister()) {
>>>>      31     int reg_base = reg->value() - 
>>>> ConcreteRegisterImpl::max_fpr;
>>>>      32     int base_reg_enc = (reg_base / 
>>>> XMMRegisterImpl::max_slots_per_register);
>>>>      33     assert(base_reg_enc >= 0 && base_reg_enc < 
>>>> XMMRegisterImpl::number_of_registers, "invalid XMMRegister: %d", 
>>>> base_reg_enc);
>>>>      34     VMReg base_reg = as_XMMRegister(base_reg_enc)->as_VMReg();
>>>>
>>>>> We have started working on adding support to the PPC64-LE hotspot 
>>>>> code for the Vector API.  In order to support Vector Masks, it 
>>>>> seems we need to change our current support for fixed-length, 
>>>>> 128-bit vectors to something that can be as short as two booleans.  
>>>>> To do that we have changed the function min_vector_size in 
>>>>> hotspot/cpu/ppc.ad to return 2 when the type is T_BOOLEAN, 
>>>>> otherwise it still returns 16.
>>>>>
>>>>> My first task was to add support for vector masks, and so I added a 
>>>>> new instruct to cpu/ppc/ppc.ad to match VectorLoadMask, which then 
>>>>> necessitated adding some instructs for LoadVector and StoreVector 
>>>>> of the appropriate lengths.
>>>>
>>>> I don't know much about PPC64-LE, but you don't have to use boolean 
>>>> vectors. FTR masks have the same type as the vectors they are 
>>>> applied to. Until recently (when work on predicated registers 
>>>> started), it was the only mask representation in Ideal IR.
>>>>
>>>> Best regards,
>>>> Vladimir Ivanov
>>>>
>>>>> I have a test case that loads a vector mask for a vector of shorts:
>>>>>
>>>>> import jdk.incubator.vector.ShortVector;
>>>>> import jdk.incubator.vector.VectorSpecies;
>>>>> import jdk.incubator.vector.VectorMask;
>>>>> import java.util.Random;
>>>>>
>>>>>
>>>>> class TestVectorMaskShort {
>>>>>    private static final VectorSpecies<Short> SPECIES = 
>>>>> ShortVector.SPECIES_128;
>>>>>
>>>>>    public static VectorMask<Short> test(boolean[] bary) {
>>>>>        VectorMask<Short> vmask = VectorMask.fromArray(SPECIES, 
>>>>> bary, 0);
>>>>>        return vmask;
>>>>>    }
>>>>>
>>>>>    public static void main(String args[]) {
>>>>>      Random ran = new Random(100);
>>>>>      int counter = 0;
>>>>>      boolean[] bary = new boolean[8];
>>>>>      for (int i = 0; i < 20_000; i++) {
>>>>>        for (int j = 0; j < bary.length; j++) {
>>>>>          bary[j] = ran.nextBoolean();
>>>>>        }
>>>>>        VectorMask<Short> vmask = test(bary);
>>>>>        if (vmask.allTrue()) {
>>>>>          counter++;
>>>>>        }
>>>>>      }
>>>>>      System.out.printf("counter = %d\n", counter);
>>>>>    }
>>>>> }
>>>>>
>>>>>
>>>>> When I run this test case, I get a runtime error:
>>>>>
>>>>> #  Internal Error 
>>>>> (/home/cjashfor/git-trees/jdk/src/hotspot/share/opto/chaitin.cpp:951), 
>>>>> pid=1341588, tid=1341601
>>>>> #  assert(lrgmask.is_aligned_sets(RegMask::SlotsPerVecX)) failed: 
>>>>> vector should be aligned
>>>>>
>>>>>
>>>>> - Corey
>>>>>
>>>>> Corey Ashford
>>>>> Software Engineer
>>>>> IBM Systems, LTC OpenJDK team
>>>>>
>>>>> IBM


More information about the hotspot-compiler-dev mailing list