Files
avo/reg/collection.go
Michael McLoughlin b76e849b5c all: AVX-512 (#217)
Extends avo to support most AVX-512 instruction sets.

The instruction type is extended to support suffixes. The K family of opmask
registers is added to the register package, and the operand package is updated
to support the new operand types. Move instruction deduction in `Load` and
`Store` is extended to support KMOV* and VMOV* forms.

Internal code generation packages were overhauled. Instruction database loading
required various messy changes to account for the additional complexities of the
AVX-512 instruction sets. The internal/api package was added to introduce a
separation between instruction forms in the database, and the functions avo
provides to create them. This was required since with instruction suffixes there
is no longer a one-to-one mapping between instruction constructors and opcodes.

AVX-512 bloated generated source code size substantially, initially increasing
compilation and CI test times to an unacceptable level. Two changes were made to
address this:

1.  Instruction constructors in the `x86` package moved to an optab-based
    approach. This compiles substantially faster than the verbose code
    generation we had before.

2.  The most verbose code-generated tests are moved under build tags and
    limited to a stress test mode. Stress test builds are run on
    schedule but not in regular CI.

An example of AVX-512 accelerated 16-lane MD5 is provided to demonstrate and
test the new functionality.

Updates #20 #163 #229

Co-authored-by: Vaughn Iverson <vsivsi@yahoo.com>
2021-11-12 19:02:39 -08:00

58 lines
2.1 KiB
Go

package reg
// Collection represents a collection of virtual registers. This is primarily
// useful for allocating virtual registers with distinct IDs.
type Collection struct {
idx map[Kind]Index
}
// NewCollection builds an empty register collection.
func NewCollection() *Collection {
return &Collection{
idx: map[Kind]Index{},
}
}
// VirtualRegister allocates and returns a new virtual register of the given kind and width.
func (c *Collection) VirtualRegister(k Kind, s Spec) Virtual {
idx := c.idx[k]
c.idx[k]++
return NewVirtual(idx, k, s)
}
// GP8L allocates and returns a general-purpose 8-bit register (low byte).
func (c *Collection) GP8L() GPVirtual { return c.GP(S8L) }
// GP8H allocates and returns a general-purpose 8-bit register (high byte).
func (c *Collection) GP8H() GPVirtual { return c.GP(S8H) }
// GP8 allocates and returns a general-purpose 8-bit register (low byte).
func (c *Collection) GP8() GPVirtual { return c.GP8L() }
// GP16 allocates and returns a general-purpose 16-bit register.
func (c *Collection) GP16() GPVirtual { return c.GP(S16) }
// GP32 allocates and returns a general-purpose 32-bit register.
func (c *Collection) GP32() GPVirtual { return c.GP(S32) }
// GP64 allocates and returns a general-purpose 64-bit register.
func (c *Collection) GP64() GPVirtual { return c.GP(S64) }
// GP allocates and returns a general-purpose register of the given width.
func (c *Collection) GP(s Spec) GPVirtual { return newgpv(c.VirtualRegister(KindGP, s)) }
// XMM allocates and returns a 128-bit vector register.
func (c *Collection) XMM() VecVirtual { return c.Vec(S128) }
// YMM allocates and returns a 256-bit vector register.
func (c *Collection) YMM() VecVirtual { return c.Vec(S256) }
// ZMM allocates and returns a 512-bit vector register.
func (c *Collection) ZMM() VecVirtual { return c.Vec(S512) }
// Vec allocates and returns a vector register of the given width.
func (c *Collection) Vec(s Spec) VecVirtual { return newvecv(c.VirtualRegister(KindVector, s)) }
// K allocates and returns an opmask register.
func (c *Collection) K() OpmaskVirtual { return newopmaskv(c.VirtualRegister(KindOpmask, S64)) }