diff --git a/Cargo.lock b/Cargo.lock
index 223ffabc91..8c73ff9549 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -2397,9 +2397,9 @@ dependencies = [
 
 [[package]]
 name = "regalloc"
-version = "0.0.31"
+version = "0.0.32"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "571f7f397d61c4755285cd37853fe8e03271c243424a907415909379659381c5"
+checksum = "a6304468554ed921da3d32c355ea107b8d13d7b8996c3adfb7aab48d3bc321f4"
 dependencies = [
  "log",
  "rustc-hash",
diff --git a/cranelift/codegen/Cargo.toml b/cranelift/codegen/Cargo.toml
index a9d8097e66..db677ac45d 100644
--- a/cranelift/codegen/Cargo.toml
+++ b/cranelift/codegen/Cargo.toml
@@ -26,7 +26,7 @@ smallvec = { version = "1.6.1" }
 peepmatic = { path = "../peepmatic", optional = true, version = "0.77.0" }
 peepmatic-traits = { path = "../peepmatic/crates/traits", optional = true, version = "0.77.0" }
 peepmatic-runtime = { path = "../peepmatic/crates/runtime", optional = true, version = "0.77.0" }
-regalloc = { version = "0.0.31" }
+regalloc = { version = "0.0.32" }
 souper-ir = { version = "2.1.0", optional = true }
 wast = { version = "38.0.0", optional = true }
 # It is a goal of the cranelift-codegen crate to have minimal external dependencies.
diff --git a/cranelift/filetests/filetests/isa/s390x/shift-rotate.clif b/cranelift/filetests/filetests/isa/s390x/shift-rotate.clif
index a275b997c8..bb42cfcad5 100644
--- a/cranelift/filetests/filetests/isa/s390x/shift-rotate.clif
+++ b/cranelift/filetests/filetests/isa/s390x/shift-rotate.clif
@@ -52,11 +52,11 @@ block0(v0: i16, v1: i16):
 }
 
 ; check:  llhr %r2, %r2
-; nextln: lr %r5, %r3
-; nextln: lcr %r4, %r3
-; nextln: nill %r5, 15
+; nextln: lr %r3, %r4
+; nextln: lcr %r4, %r4
+; nextln: nill %r3, 15
 ; nextln: nill %r4, 15
-; nextln: sllk %r3, %r2, 0(%r5)
+; nextln: sllk %r3, %r2, 0(%r3)
 ; nextln: srlk %r2, %r2, 0(%r4)
 ; nextln: ork %r2, %r3, %r2
 ; nextln: br %r14
@@ -81,11 +81,11 @@ block0(v0: i8, v1: i8):
 }
 
 ; check:  llcr %r2, %r2
-; nextln: lr %r5, %r3
-; nextln: lcr %r4, %r3
-; nextln: nill %r5, 7
+; nextln: lr %r3, %r4
+; nextln: lcr %r4, %r4
+; nextln: nill %r3, 7
 ; nextln: nill %r4, 7
-; nextln: sllk %r3, %r2, 0(%r5)
+; nextln: sllk %r3, %r2, 0(%r3)
 ; nextln: srlk %r2, %r2, 0(%r4)
 ; nextln: ork %r2, %r3, %r2
 ; nextln: br %r14
diff --git a/cranelift/filetests/filetests/isa/x64/fastcall.clif b/cranelift/filetests/filetests/isa/x64/fastcall.clif
index 99397044d9..2081bf5ebb 100644
--- a/cranelift/filetests/filetests/isa/x64/fastcall.clif
+++ b/cranelift/filetests/filetests/isa/x64/fastcall.clif
@@ -238,35 +238,34 @@ block0(v0: i64):
 ; nextln: unwind PushFrameRegs { offset_upward_to_caller_sp: 16 }
 ; nextln: movq    %rsp, %rbp
 ; nextln: unwind DefineNewFrame { offset_upward_to_caller_sp: 16, offset_downward_to_clobbers: 160 }
-; nextln: subq    $$208, %rsp
-; nextln: movdqu  %xmm6, 48(%rsp)
+; nextln: subq    $$192, %rsp
+; nextln: movdqu  %xmm6, 32(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 0, reg: r6V }
-; nextln: movdqu  %xmm7, 64(%rsp)
+; nextln: movdqu  %xmm7, 48(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 16, reg: r7V }
-; nextln: movdqu  %xmm8, 80(%rsp)
+; nextln: movdqu  %xmm8, 64(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 32, reg: r8V }
-; nextln: movdqu  %xmm9, 96(%rsp)
+; nextln: movdqu  %xmm9, 80(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 48, reg: r9V }
-; nextln: movdqu  %xmm10, 112(%rsp)
+; nextln: movdqu  %xmm10, 96(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 64, reg: r10V }
-; nextln: movdqu  %xmm11, 128(%rsp)
+; nextln: movdqu  %xmm11, 112(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 80, reg: r11V }
-; nextln: movdqu  %xmm12, 144(%rsp)
+; nextln: movdqu  %xmm12, 128(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 96, reg: r12V }
-; nextln: movdqu  %xmm13, 160(%rsp)
+; nextln: movdqu  %xmm13, 144(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 112, reg: r13V }
-; nextln: movdqu  %xmm14, 176(%rsp)
+; nextln: movdqu  %xmm14, 160(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 128, reg: r14V }
-; nextln: movdqu  %xmm15, 192(%rsp)
+; nextln: movdqu  %xmm15, 176(%rsp)
 ; nextln: unwind SaveReg { clobber_offset: 144, reg: r15V }
-; nextln: movsd   0(%rcx), %xmm0
-; nextln: movsd   %xmm0, rsp(16 + virtual offset)
+; nextln: movsd   0(%rcx), %xmm4
 ; nextln: movsd   8(%rcx), %xmm1
 ; nextln: movsd   16(%rcx), %xmm0
-; nextln: movsd   %xmm0, rsp(24 + virtual offset)
+; nextln: movsd   %xmm0, rsp(16 + virtual offset)
 ; nextln: movsd   24(%rcx), %xmm3
 ; nextln: movsd   32(%rcx), %xmm0
-; nextln: movsd   %xmm0, rsp(32 + virtual offset)
+; nextln: movsd   %xmm0, rsp(24 + virtual offset)
 ; nextln: movsd   40(%rcx), %xmm5
 ; nextln: movsd   48(%rcx), %xmm6
 ; nextln: movsd   56(%rcx), %xmm7
@@ -284,45 +283,41 @@ block0(v0: i64):
 ; nextln: movsd   144(%rcx), %xmm2
 ; nextln: movsd   %xmm2, rsp(8 + virtual offset)
 ; nextln: movsd   152(%rcx), %xmm2
-; nextln: nop     len=0
-; nextln: movsd   rsp(16 + virtual offset), %xmm4
 ; nextln: addsd   %xmm1, %xmm4
-; nextln: movsd   %xmm4, rsp(16 + virtual offset)
-; nextln: movsd   rsp(24 + virtual offset), %xmm1
+; nextln: movsd   rsp(16 + virtual offset), %xmm1
 ; nextln: addsd   %xmm3, %xmm1
-; nextln: movsd   rsp(32 + virtual offset), %xmm4
-; nextln: addsd   %xmm5, %xmm4
+; nextln: movsd   rsp(24 + virtual offset), %xmm3
+; nextln: addsd   %xmm5, %xmm3
 ; nextln: addsd   %xmm7, %xmm6
 ; nextln: addsd   %xmm9, %xmm8
 ; nextln: addsd   %xmm11, %xmm10
 ; nextln: addsd   %xmm13, %xmm12
 ; nextln: addsd   %xmm15, %xmm14
-; nextln: movsd   rsp(0 + virtual offset), %xmm3
-; nextln: addsd   %xmm0, %xmm3
+; nextln: movsd   rsp(0 + virtual offset), %xmm5
+; nextln: addsd   %xmm0, %xmm5
 ; nextln: movsd   rsp(8 + virtual offset), %xmm0
 ; nextln: addsd   %xmm2, %xmm0
-; nextln: movsd   rsp(16 + virtual offset), %xmm2
-; nextln: addsd   %xmm1, %xmm2
-; nextln: addsd   %xmm6, %xmm4
+; nextln: addsd   %xmm1, %xmm4
+; nextln: addsd   %xmm6, %xmm3
 ; nextln: addsd   %xmm10, %xmm8
 ; nextln: addsd   %xmm14, %xmm12
-; nextln: addsd   %xmm0, %xmm3
-; nextln: addsd   %xmm4, %xmm2
+; nextln: addsd   %xmm0, %xmm5
+; nextln: addsd   %xmm3, %xmm4
 ; nextln: addsd   %xmm12, %xmm8
-; nextln: addsd   %xmm8, %xmm2
-; nextln: addsd   %xmm3, %xmm2
-; nextln: movaps  %xmm2, %xmm0
-; nextln: movdqu  48(%rsp), %xmm6
-; nextln: movdqu  64(%rsp), %xmm7
-; nextln: movdqu  80(%rsp), %xmm8
-; nextln: movdqu  96(%rsp), %xmm9
-; nextln: movdqu  112(%rsp), %xmm10
-; nextln: movdqu  128(%rsp), %xmm11
-; nextln: movdqu  144(%rsp), %xmm12
-; nextln: movdqu  160(%rsp), %xmm13
-; nextln: movdqu  176(%rsp), %xmm14
-; nextln: movdqu  192(%rsp), %xmm15
-; nextln: addq    $$208, %rsp
+; nextln: addsd   %xmm8, %xmm4
+; nextln: addsd   %xmm5, %xmm4
+; nextln: movaps  %xmm4, %xmm0
+; nextln: movdqu  32(%rsp), %xmm6
+; nextln: movdqu  48(%rsp), %xmm7
+; nextln: movdqu  64(%rsp), %xmm8
+; nextln: movdqu  80(%rsp), %xmm9
+; nextln: movdqu  96(%rsp), %xmm10
+; nextln: movdqu  112(%rsp), %xmm11
+; nextln: movdqu  128(%rsp), %xmm12
+; nextln: movdqu  144(%rsp), %xmm13
+; nextln: movdqu  160(%rsp), %xmm14
+; nextln: movdqu  176(%rsp), %xmm15
+; nextln: addq    $$192, %rsp
 ; nextln: movq    %rbp, %rsp
 ; nextln: popq    %rbp
 ; nextln: ret
diff --git a/cranelift/filetests/filetests/isa/x64/i128.clif b/cranelift/filetests/filetests/isa/x64/i128.clif
index 75013a8170..96520af906 100644
--- a/cranelift/filetests/filetests/isa/x64/i128.clif
+++ b/cranelift/filetests/filetests/isa/x64/i128.clif
@@ -99,12 +99,15 @@ function %f5(i128) -> i128 {
 block0(v0: i128):
 
     v1 = bnot v0
-; nextln:  notq    %rdi
+; nextln:  movq    %rsi, %rax
+; nextln:  movq    %rdi, %rsi
 ; nextln:  notq    %rsi
+; nextln:  movq    %rax, %rdi
+; nextln:  notq    %rdi
 
     return v1
-; nextln:  movq    %rdi, %rax
-; nextln:  movq    %rsi, %rdx
+; nextln:  movq    %rsi, %rax
+; nextln:  movq    %rdi, %rdx
 ; nextln:  movq    %rbp, %rsp
 ; nextln:  popq    %rbp
 ; nextln:  ret
@@ -741,30 +744,34 @@ block0(v0: i128, v1: i128, v2: i64, v3: i128, v4: i128, v5: i128):
 
 ; check:  pushq   %rbp
 ; nextln: movq    %rsp, %rbp
-; nextln: subq    $$16, %rsp
+; nextln: subq    $$32, %rsp
 ; nextln: movq    %r12, 0(%rsp)
 ; nextln: movq    %r13, 8(%rsp)
+; nextln: movq    %r14, 16(%rsp)
+; nextln: movq    %r8, %r14
 ; nextln: movq    16(%rbp), %r10
 ; nextln: movq    24(%rbp), %r12
 ; nextln: movq    32(%rbp), %r11
 ; nextln: movq    40(%rbp), %rax
 ; nextln: movq    48(%rbp), %r13
+; nextln: movq    %rsi, %r8
 ; nextln: addq    %rdx, %rdi
-; nextln: adcq    %rcx, %rsi
-; nextln: xorq    %rcx, %rcx
-; nextln: addq    %r8, %r9
-; nextln: adcq    %rcx, %r10
+; nextln: adcq    %rcx, %r8
+; nextln: xorq    %rsi, %rsi
+; nextln: addq    %r14, %r9
+; nextln: adcq    %rsi, %r10
 ; nextln: addq    %rax, %r12
 ; nextln: adcq    %r13, %r11
 ; nextln: addq    %r9, %rdi
-; nextln: adcq    %r10, %rsi
+; nextln: adcq    %r10, %r8
 ; nextln: addq    %rdi, %r12
-; nextln: adcq    %rsi, %r11
+; nextln: adcq    %r8, %r11
 ; nextln: movq    %r12, %rax
 ; nextln: movq    %r11, %rdx
 ; nextln: movq    0(%rsp), %r12
 ; nextln: movq    8(%rsp), %r13
-; nextln: addq    $$16, %rsp
+; nextln: movq    16(%rsp), %r14
+; nextln: addq    $$32, %rsp
 ; nextln: movq    %rbp, %rsp
 ; nextln: popq    %rbp
 ; nextln: ret
@@ -1059,25 +1066,26 @@ block0(v0: i128, v1: i128):
 
 ; check:  pushq   %rbp
 ; nextln: movq    %rsp, %rbp
-; nextln: movq    %rsi, %r9
+; nextln: movq    %rsi, %rax
+; nextln: movq    %rax, %r9
 ; nextln: movq    %rdx, %rcx
 ; nextln: shrq    %cl, %r9
-; nextln: movq    %rdi, %rax
+; nextln: movq    %rdi, %rsi
 ; nextln: movq    %rdx, %rcx
-; nextln: shrq    %cl, %rax
+; nextln: shrq    %cl, %rsi
 ; nextln: movl    $$64, %ecx
 ; nextln: subq    %rdx, %rcx
-; nextln: movq    %rsi, %r10
+; nextln: movq    %rax, %r10
 ; nextln: shlq    %cl, %r10
 ; nextln: xorq    %rcx, %rcx
 ; nextln: testq   $$127, %rdx
 ; nextln: cmovzq  %rcx, %r10
-; nextln: orq     %rax, %r10
-; nextln: xorq    %rax, %rax
+; nextln: orq     %rsi, %r10
+; nextln: xorq    %rsi, %rsi
 ; nextln: xorq    %r8, %r8
 ; nextln: movq    %rdx, %rcx
 ; nextln: andq    $$64, %rcx
-; nextln: cmovzq  %r9, %rax
+; nextln: cmovzq  %r9, %rsi
 ; nextln: cmovzq  %r10, %r8
 ; nextln: cmovnzq %r9, %r8
 ; nextln: movl    $$128, %r9d
@@ -1085,25 +1093,24 @@ block0(v0: i128, v1: i128):
 ; nextln: movq    %rdi, %rdx
 ; nextln: movq    %r9, %rcx
 ; nextln: shlq    %cl, %rdx
-; nextln: movq    %rsi, %r10
 ; nextln: movq    %r9, %rcx
-; nextln: shlq    %cl, %r10
+; nextln: shlq    %cl, %rax
 ; nextln: movl    $$64, %ecx
 ; nextln: subq    %r9, %rcx
 ; nextln: shrq    %cl, %rdi
-; nextln: xorq    %rsi, %rsi
-; nextln: testq   $$127, %r9
-; nextln: cmovzq  %rsi, %rdi
-; nextln: orq     %r10, %rdi
 ; nextln: xorq    %rcx, %rcx
+; nextln: testq   $$127, %r9
+; nextln: cmovzq  %rcx, %rdi
+; nextln: orq     %rax, %rdi
+; nextln: xorq    %rax, %rax
 ; nextln: andq    $$64, %r9
-; nextln: cmovzq  %rdi, %rcx
-; nextln: cmovzq  %rdx, %rsi
-; nextln: cmovnzq %rdx, %rcx
-; nextln: orq     %r8, %rsi
-; nextln: orq     %rax, %rcx
-; nextln: movq    %rsi, %rax
-; nextln: movq    %rcx, %rdx
+; nextln: cmovzq  %rdi, %rax
+; nextln: cmovzq  %rdx, %rcx
+; nextln: cmovnzq %rdx, %rax
+; nextln: orq     %r8, %rcx
+; nextln: orq     %rsi, %rax
+; nextln: movq    %rax, %rdx
+; nextln: movq    %rcx, %rax
 ; nextln: movq    %rbp, %rsp
 ; nextln: popq    %rbp
 ; nextln: ret