derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
45 lines
2.3 KiB
LLVM
45 lines
2.3 KiB
LLVM
; RUN: llc -mtriple=armv7-none-linux-gnueabi -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECK-WITH-LDRD
|
|
; RUN: llc -mtriple=armv4-none-linux-gnueabi -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECK-WITHOUT-LDRD
|
|
; RUN: llc -mtriple=thumbv7-none-linux-gnueabi -verify-machineinstrs < %s | FileCheck %s --check-prefix=CHECK-WITH-LDRD
|
|
|
|
define void @foo(ptr %addr) {
|
|
%val1 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
%val2 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
%val3 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
%val4 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
%val5 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
%val6 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
%val7 = tail call i64 asm sideeffect "ldrexd $0, ${0:H}, [r0]", "=&r,r"(ptr %addr)
|
|
|
|
; Key point is that enough 64-bit paired GPR values are live that
|
|
; one of them has to be spilled. This used to cause an abort because
|
|
; an LDMIA was created with both a FrameIndex and an offset, which
|
|
; is not allowed.
|
|
|
|
; CHECK-WITH-LDRD-DAG: strd {{r[0-9]+}}, {{r[0-9]+}}, [sp, #8]
|
|
; CHECK-WITH-LDRD-DAG: strd {{r[0-9]+}}, {{r[0-9]+}}, [sp]
|
|
|
|
; CHECK-WITH-LDRD-DAG: ldrd {{r[0-9]+}}, {{r[0-9]+}}, [sp, #8]
|
|
; CHECK-WITH-LDRD-DAG: ldrd {{r[0-9]+}}, {{r[0-9]+}}, [sp]
|
|
|
|
; We also want to ensure the register scavenger is working (i.e. an
|
|
; offset from sp can be generated), so we need two spills.
|
|
; CHECK-WITHOUT-LDRD-DAG: add [[ADDRREG:[a-z0-9]+]], sp, #{{[0-9]+}}
|
|
; CHECK-WITHOUT-LDRD-DAG: stm [[ADDRREG]], {r{{[0-9]+}}, r{{[0-9]+}}}
|
|
; CHECK-WITHOUT-LDRD-DAG: stm sp, {r{{[0-9]+}}, r{{[0-9]+}}}
|
|
|
|
; In principle LLVM may have to recalculate the offset. At the moment
|
|
; it reuses the original though.
|
|
; CHECK-WITHOUT-LDRD-DAG: ldm [[ADDRREG]], {r{{[0-9]+}}, r{{[0-9]+}}}
|
|
; CHECK-WITHOUT-LDRD-DAG: ldm sp, {r{{[0-9]+}}, r{{[0-9]+}}}
|
|
|
|
store volatile i64 %val1, ptr %addr
|
|
store volatile i64 %val2, ptr %addr
|
|
store volatile i64 %val3, ptr %addr
|
|
store volatile i64 %val4, ptr %addr
|
|
store volatile i64 %val5, ptr %addr
|
|
store volatile i64 %val6, ptr %addr
|
|
store volatile i64 %val7, ptr %addr
|
|
ret void
|
|
}
|