cb48ea406e
This PR switches the default backend on x86, for both the `cranelift-codegen` crate and for Wasmtime, to the new (`MachInst`-style, `VCode`-based) backend that has been under development and testing for some time now. The old backend is still available by default in builds with the `old-x86-backend` feature, or by requesting `BackendVariant::Legacy` from the appropriate APIs. As part of that switch, it adds some more runtime-configurable plumbing to the testing infrastructure so that tests can be run using the appropriate backend. `clif-util test` is now capable of parsing a backend selector option from filetests and instantiating the correct backend. CI has been updated so that the old x86 backend continues to run its tests, just as we used to run the new x64 backend separately. At some point, we will remove the old x86 backend entirely, once we are satisfied that the new backend has not caused any unforeseen issues and we do not need to revert.
52 lines
1.6 KiB
Plaintext
52 lines
1.6 KiB
Plaintext
test regalloc
|
|
target x86_64 legacy haswell
|
|
|
|
; Test combinations of constraints.
|
|
;
|
|
; The x86 ushr instruction requires its second operand to be passed in %rcx and its output is
|
|
; tied to the first input operand.
|
|
;
|
|
; If we pass the same value to both operands, both constraints must be satisfied.
|
|
|
|
; Found by the Binaryen fuzzer in PR221.
|
|
;
|
|
; Conditions triggering the problem:
|
|
;
|
|
; - The same value used for a tied operand and a fixed operand.
|
|
; - The common value is already in %rcx.
|
|
; - The tied output value is live outside the block.
|
|
;
|
|
; Under these conditions, Solver::add_tied_input() would create a variable for the tied input
|
|
; without considering the fixed constraint.
|
|
function %pr221(i64 [%rdi], i64 [%rsi], i64 [%rdx], i64 [%rcx]) -> i64 [%rax] {
|
|
block0(v0: i64, v1: i64, v2: i64, v3: i64):
|
|
v4 = ushr v3, v3
|
|
jump block1
|
|
|
|
block1:
|
|
return v4
|
|
}
|
|
|
|
; Found by the Binaryen fuzzer in PR218.
|
|
;
|
|
; This is a similar situation involving combined constraints on the ushr instruction:
|
|
;
|
|
; - The %rcx register is already in use by a globally live value.
|
|
; - The ushr x, x result is also a globally live value.
|
|
;
|
|
; Since the ushr x, x result is forced to be placed in %rcx, we must set the replace_global_defines
|
|
; flag so it can be reassigned to a different global register.
|
|
function %pr218(i64 [%rdi], i64 [%rsi], i64 [%rdx], i64 [%rcx]) -> i64 [%rax] {
|
|
block0(v0: i64, v1: i64, v2: i64, v3: i64):
|
|
; check: regmove v3, %rcx ->
|
|
v4 = ushr v0, v0
|
|
; check: v4 = copy
|
|
jump block1
|
|
|
|
block1:
|
|
; v3 is globally live in %rcx.
|
|
; v4 is also globally live. Needs to be assigned something else for the trip across the CFG edge.
|
|
v5 = iadd v3, v4
|
|
return v5
|
|
}
|