* Add AArch64 tests to CI This commit enhances our CI with an AArch64 builder. Currently we have no physical hardware to run on so for now we run all tests in an emulator. The AArch64 build is cross-compiled from x86_64 from Linux. Tests all happen in release mode with a recent version of QEMU (recent version because it's so much faster, and in release mode because debug mode tests take quite a long time in an emulator). The goal here was not to get all tests passing on CI, but rather to get AArch64 running on CI and get it green at the same time. To achieve that goal many tests are now ignored on aarch64 platforms. Many tests fail due to unimplemented functionality in the aarch64 backend (#1521), and all wasmtime tests involving compilation are also disabled due to panicking attempting to generate generate instruction offset information for trap symbolication (#1523). Despite this, though, all Cranelift tests and other wasmtime tests should be runnin on AArch64 through QEMU with this PR. Additionally we'll have an AArch64 binary release of Wasmtime for Linux, although it won't be too useful just yet since it will panic on almost all wasm modules. * Review comments
248 lines
5.1 KiB
Plaintext
248 lines
5.1 KiB
Plaintext
test run
|
|
set enable_simd
|
|
target x86_64
|
|
|
|
function %icmp_eq_i8x16() -> b8 {
|
|
block0:
|
|
v0 = vconst.i8x16 0x00
|
|
v1 = vconst.i8x16 0x00
|
|
v2 = icmp eq v0, v1
|
|
v3 = extractlane v2, 0
|
|
return v3
|
|
}
|
|
; run
|
|
|
|
function %icmp_eq_i64x2() -> b64 {
|
|
block0:
|
|
v0 = vconst.i64x2 0xffffffffffffffffffffffffffffffff
|
|
v1 = vconst.i64x2 0xffffffffffffffffffffffffffffffff
|
|
v2 = icmp eq v0, v1
|
|
v3 = extractlane v2, 1
|
|
return v3
|
|
}
|
|
; run
|
|
|
|
function %icmp_ne_i32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.i32x4 [0 1 2 3]
|
|
v1 = vconst.i32x4 [7 7 7 7]
|
|
v2 = icmp ne v0, v1
|
|
v3 = vall_true v2
|
|
return v3
|
|
}
|
|
; run
|
|
|
|
function %icmp_ne_i16x8() -> b1 {
|
|
block0:
|
|
v0 = vconst.i16x8 [0 1 2 3 4 5 6 7]
|
|
v1 = vconst.i16x8 [0 1 2 3 4 5 6 7]
|
|
v2 = icmp ne v0, v1
|
|
v3 = vall_true v2
|
|
v4 = bint.i32 v3
|
|
v5 = icmp_imm eq v4, 0
|
|
return v5
|
|
}
|
|
; run
|
|
|
|
function %icmp_sgt_i8x16() -> b1 {
|
|
block0:
|
|
v0 = vconst.i8x16 [0 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0]
|
|
v1 = vconst.i8x16 [1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0xff]
|
|
v2 = icmp sgt v0, v1
|
|
v3 = raw_bitcast.i8x16 v2
|
|
v4 = vconst.i8x16 [0 0 0xff 0 0 0 0 0 0 0 0 0 0 0 0 0xff]
|
|
v7 = icmp eq v3, v4
|
|
v8 = vall_true v7
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_sgt_i64x2() -> b1 {
|
|
block0:
|
|
v0 = vconst.i64x2 [0 -42]
|
|
v1 = vconst.i64x2 [-1 -43]
|
|
v2 = icmp sgt v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %maxs_i8x16() -> b1 {
|
|
block0:
|
|
v0 = vconst.i8x16 [1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1] ; 1 will be greater than -1 == 0xff with
|
|
; signed max
|
|
v1 = vconst.i8x16 [0xff 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1]
|
|
v2 = x86_pmaxs v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %maxu_i16x8() -> b1 {
|
|
block0:
|
|
v0 = vconst.i16x8 [0 1 1 1 1 1 1 1]
|
|
v1 = vconst.i16x8 [-1 1 1 1 1 1 1 1] ; -1 == 0xff will be greater with unsigned max
|
|
v2 = x86_pmaxu v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %mins_i32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.i32x4 [0 1 1 1]
|
|
v1 = vconst.i32x4 [-1 1 1 1] ; -1 == 0xff will be less with signed min
|
|
v2 = x86_pmins v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %minu_i8x16() -> b1 {
|
|
block0:
|
|
v0 = vconst.i8x16 [1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1] ; 1 < 2 with unsiged min
|
|
v1 = vconst.i8x16 [2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2]
|
|
v2 = x86_pminu v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_ugt_i8x16() -> b1 {
|
|
block0:
|
|
v0 = vconst.i8x16 [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16]
|
|
v1 = vconst.i8x16 [0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1]
|
|
v2 = icmp ugt v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_sge_i16x8() -> b1 {
|
|
block0:
|
|
v0 = vconst.i16x8 [-1 1 2 3 4 5 6 7]
|
|
v1 = vconst.i16x8 [-1 1 1 1 1 1 1 1]
|
|
v2 = icmp sge v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_uge_i32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.i32x4 [1 2 3 4]
|
|
v1 = vconst.i32x4 [1 1 1 1]
|
|
v2 = icmp uge v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_slt_i32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.i32x4 [-1 1 1 1]
|
|
v1 = vconst.i32x4 [1 2 3 4]
|
|
v2 = icmp slt v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_ult_i32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.i32x4 [1 1 1 1]
|
|
v1 = vconst.i32x4 [-1 2 3 4] ; -1 = 0xffff... will be greater than 1 when unsigned
|
|
v2 = icmp ult v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
|
|
function %icmp_ult_i16x8() -> b1 {
|
|
block0:
|
|
v0 = vconst.i16x8 [-1 -1 -1 -1 -1 -1 -1 -1]
|
|
v1 = vconst.i16x8 [-1 -1 -1 -1 -1 -1 -1 -1]
|
|
v2 = icmp ult v0, v1
|
|
v3 = vconst.i16x8 0x00
|
|
v4 = raw_bitcast.i16x8 v2
|
|
v5 = icmp eq v3, v4
|
|
v8 = vall_true v5
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_sle_i16x8() -> b1 {
|
|
block0:
|
|
v0 = vconst.i16x8 [-1 -1 0 0 0 0 0 0]
|
|
v1 = vconst.i16x8 [-1 0 0 0 0 0 0 0]
|
|
v2 = icmp sle v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %icmp_ule_i16x8() -> b1 {
|
|
block0:
|
|
v0 = vconst.i16x8 [-1 0 0 0 0 0 0 0]
|
|
v1 = vconst.i16x8 [-1 -1 0 0 0 0 0 0]
|
|
v2 = icmp ule v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %fcmp_eq_f32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.f32x4 [0.0 -0x4.2 0x0.33333 -0.0]
|
|
v1 = vconst.f32x4 [0.0 -0x4.2 0x0.33333 -0.0]
|
|
v2 = fcmp eq v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %fcmp_lt_f32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.f32x4 [0.0 -0x4.2 0x0.0 -0.0]
|
|
v1 = vconst.f32x4 [0x0.001 0x4.2 0x0.33333 0x1.0]
|
|
v2 = fcmp lt v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %fcmp_ge_f64x2() -> b1 {
|
|
block0:
|
|
v0 = vconst.f64x2 [0x0.0 0x4.2]
|
|
v1 = vconst.f64x2 [0.0 0x4.1]
|
|
v2 = fcmp ge v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %fcmp_uno_f64x2() -> b1 {
|
|
block0:
|
|
v0 = vconst.f64x2 [0.0 NaN]
|
|
v1 = vconst.f64x2 [NaN 0x4.1]
|
|
v2 = fcmp uno v0, v1
|
|
v8 = vall_true v2
|
|
return v8
|
|
}
|
|
; run
|
|
|
|
function %fcmp_gt_nans_f32x4() -> b1 {
|
|
block0:
|
|
v0 = vconst.f32x4 [NaN 0x42.0 -NaN NaN]
|
|
v1 = vconst.f32x4 [NaN NaN 0x42.0 Inf]
|
|
v2 = fcmp gt v0, v1
|
|
; now check that the result v2 is all zeroes
|
|
v3 = vconst.i32x4 0x00
|
|
v4 = raw_bitcast.i32x4 v2
|
|
v5 = icmp eq v3, v4
|
|
v8 = vall_true v5
|
|
return v8
|
|
}
|
|
; run
|