Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

rust: use {build,host}Platform.{gcc.arch,rustc.flags} #211868

Closed
wants to merge 2 commits into from
Closed

rust: use {build,host}Platform.{gcc.arch,rustc.flags} #211868

wants to merge 2 commits into from

Conversation

ghost
Copy link

@ghost ghost commented Jan 21, 2023 •
edited by ghost
Loading

Description of changes

This commit causes rustc to use stdenv.hostPlatform.gcc.arch, if present, to set -C target-cpu, and do similarly for the buildPlatform. Commented-out lines to do similarly for gcc.tune are included, to be uncommented once -Z tune-cpu is stabilized.

This commit also allows passing arbitrary rustc flags through {build,host}Platform.rustc.flags.

Things done
  • Built on platform(s)
    • x86_64-linux
    • powerpc64le-linux
    • aarch64-linux
  • Tested basic functionality of all binary files (usually in ./result/bin/)
  • Fits CONTRIBUTING.md.

@ghost ghost changed the title rust: use {build,host}Platform.{gcc.arch,rust.flags} rust: use {build,host}Platform.{gcc.arch,rustc.flags} Jan 21, 2023
@ofborg ofborg bot added 10.rebuild-darwin: 0 This PR does not cause any packages to rebuild on Darwin 10.rebuild-linux: 0 This PR does not cause any packages to rebuild on Linux labels Jan 21, 2023
@ghost ghost marked this pull request as draft January 24, 2023 07:23
@ghost
Copy link
Author

ghost commented Jan 24, 2023 •
edited by ghost
Loading

Hrm, the situation on aarch64 is a mess. Over there, LLVM uses specific product names for target-cpu and a mishmash of architectures, features, and (again) specific product names for target-feature.

I think we're going to end up with some kind of rosetta stone.

Also, rustc tells you to use rustc -C target-cpu=help, which is not helpful:

$rustc -C target-cpu=help
Target CPU help is not supported by this LLVM version.

... apparently "this LLVM version" means "all LLVM versions (unless you apply the rustc-specific patches, which nixpkgs doesn't and shouldn't)".

Rustc should be telling people to use `llc -march=aarch64 -mattr=help`: Available CPUs for this target:

a64fx - Select the a64fx processor.
apple-a10 - Select the apple-a10 processor.
apple-a11 - Select the apple-a11 processor.
apple-a12 - Select the apple-a12 processor.
apple-a13 - Select the apple-a13 processor.
apple-a7 - Select the apple-a7 processor.
apple-a8 - Select the apple-a8 processor.
apple-a9 - Select the apple-a9 processor.
apple-latest - Select the apple-latest processor.
apple-s4 - Select the apple-s4 processor.
apple-s5 - Select the apple-s5 processor.
carmel - Select the carmel processor.
cortex-a34 - Select the cortex-a34 processor.
cortex-a35 - Select the cortex-a35 processor.
cortex-a53 - Select the cortex-a53 processor.
cortex-a55 - Select the cortex-a55 processor.
cortex-a57 - Select the cortex-a57 processor.
cortex-a65 - Select the cortex-a65 processor.
cortex-a65ae - Select the cortex-a65ae processor.
cortex-a72 - Select the cortex-a72 processor.
cortex-a73 - Select the cortex-a73 processor.
cortex-a75 - Select the cortex-a75 processor.
cortex-a76 - Select the cortex-a76 processor.
cortex-a76ae - Select the cortex-a76ae processor.
cortex-a77 - Select the cortex-a77 processor.
cortex-a78 - Select the cortex-a78 processor.
cortex-x1 - Select the cortex-x1 processor.
cyclone - Select the cyclone processor.
exynos-m3 - Select the exynos-m3 processor.
exynos-m4 - Select the exynos-m4 processor.
exynos-m5 - Select the exynos-m5 processor.
falkor - Select the falkor processor.
generic - Select the generic processor.
kryo - Select the kryo processor.
neoverse-e1 - Select the neoverse-e1 processor.
neoverse-n1 - Select the neoverse-n1 processor.
saphira - Select the saphira processor.
thunderx - Select the thunderx processor.
thunderx2t99 - Select the thunderx2t99 processor.
thunderx3t110 - Select the thunderx3t110 processor.
thunderxt81 - Select the thunderxt81 processor.
thunderxt83 - Select the thunderxt83 processor.
thunderxt88 - Select the thunderxt88 processor.
tsv110 - Select the tsv110 processor.

Available features for this target:

a35 - Cortex-A35 ARM processors.
a53 - Cortex-A53 ARM processors.
a55 - Cortex-A55 ARM processors.
a57 - Cortex-A57 ARM processors.
a64fx - Fujitsu A64FX processors.
a65 - Cortex-A65 ARM processors.
a72 - Cortex-A72 ARM processors.
a73 - Cortex-A73 ARM processors.
a75 - Cortex-A75 ARM processors.
a76 - Cortex-A76 ARM processors.
a77 - Cortex-A77 ARM processors.
aes - Enable AES support.
aggressive-fma - Enable Aggressive FMA for floating-point..
alternate-sextload-cvt-f32-pattern - Use alternative pattern for sextload convert to f32.
altnzcv - Enable alternative NZCV format for floating point comparisons.
am - Enable v8.4-A Activity Monitors extension.
amvs - Enable v8.6-A Activity Monitors Virtualization support.
apple-a10 - Apple A10.
apple-a11 - Apple A11.
apple-a12 - Apple A12.
apple-a13 - Apple A13.
apple-a7 - Apple A7 (the CPU formerly known as Cyclone).
arith-bcc-fusion - CPU fuses arithmetic+bcc operations.
arith-cbz-fusion - CPU fuses arithmetic + cbz/cbnz operations.
balance-fp-ops - balance mix of odd and even D-registers for fp multiply(-accumulate) ops.
bf16 - Enable BFloat16 Extension.
bti - Enable Branch Target Identification.
call-saved-x10 - Make X10 callee saved..
call-saved-x11 - Make X11 callee saved..
call-saved-x12 - Make X12 callee saved..
call-saved-x13 - Make X13 callee saved..
call-saved-x14 - Make X14 callee saved..
call-saved-x15 - Make X15 callee saved..
call-saved-x18 - Make X18 callee saved..
call-saved-x8 - Make X8 callee saved..
call-saved-x9 - Make X9 callee saved..
carmel - Nvidia Carmel processors.
ccdp - Enable v8.5 Cache Clean to Point of Deep Persistence.
ccidx - Enable v8.3-A Extend of the CCSIDR number of sets.
ccpp - Enable v8.2 data Cache Clean to Point of Persistence.
complxnum - Enable v8.3-A Floating-point complex number support.
cortex-a78 - Cortex-A78 ARM processors.
cortex-x1 - Cortex-X1 ARM processors.
crc - Enable ARMv8 CRC-32 checksum instructions.
crypto - Enable cryptographic instructions.
custom-cheap-as-move - Use custom handling of cheap instructions.
disable-latency-sched-heuristic - Disable latency scheduling heuristic.
dit - Enable v8.4-A Data Independent Timing instructions.
dotprod - Enable dot product support.
ecv - Enable enhanced counter virtualization extension.
ete - Enable Embedded Trace Extension.
exynos-cheap-as-move - Use Exynos specific handling of cheap instructions.
exynosm3 - Samsung Exynos-M3 processors.
exynosm4 - Samsung Exynos-M4 processors.
f32mm - Enable Matrix Multiply FP32 Extension.
f64mm - Enable Matrix Multiply FP64 Extension.
falkor - Qualcomm Falkor processors.
fgt - Enable fine grained virtualization traps extension.
fmi - Enable v8.4-A Flag Manipulation Instructions.
force-32bit-jump-tables - Force jump table entries to be 32-bits wide except at MinSize.
fp-armv8 - Enable ARMv8 FP.
fp16fml - Enable FP16 FML instructions.
fptoint - Enable FRInt[32|64][Z|X] instructions that round a floating-point number to an integer (in FP format) forcing it to fit into a 32- or 64-bit int.
fullfp16 - Full FP16.
fuse-address - CPU fuses address generation and memory operations.
fuse-aes - CPU fuses AES crypto operations.
fuse-arith-logic - CPU fuses arithmetic and logic operations.
fuse-crypto-eor - CPU fuses AES/PMULL and EOR operations.
fuse-csel - CPU fuses conditional select operations.
fuse-literals - CPU fuses literal generation operations.
harden-sls-blr - Harden against straight line speculation across BLR instructions.
harden-sls-retbr - Harden against straight line speculation across RET and BR instructions.
i8mm - Enable Matrix Multiply Int8 Extension.
jsconv - Enable v8.3-A JavaScript FP conversion instructions.
kryo - Qualcomm Kryo processors.
lor - Enables ARM v8.1 Limited Ordering Regions extension.
lse - Enable ARMv8.1 Large System Extension (LSE) atomic instructions.
lsl-fast - CPU has a fastpath logical shift of up to 3 places.
mpam - Enable v8.4-A Memory system Partitioning and Monitoring extension.
mte - Enable Memory Tagging Extension.
neon - Enable Advanced SIMD instructions.
neoversee1 - Neoverse E1 ARM processors.
neoversen1 - Neoverse N1 ARM processors.
no-neg-immediates - Convert immediates and instructions to their negated or complemented equivalent when the immediate does not fit in the encoding..
nv - Enable v8.4-A Nested Virtualization Enchancement.
pa - Enable v8.3-A Pointer Authentication extension.
pan - Enables ARM v8.1 Privileged Access-Never extension.
pan-rwv - Enable v8.2 PAN s1e1R and s1e1W Variants.
perfmon - Enable ARMv8 PMUv3 Performance Monitors extension.
pmu - Enable v8.4-A PMU extension.
predictable-select-expensive - Prefer likely predicted branches over selects.
predres - Enable v8.5a execution and data prediction invalidation instructions.
rand - Enable Random Number generation instructions.
ras - Enable ARMv8 Reliability, Availability and Serviceability Extensions.
rasv8_4 - Enable v8.4-A Reliability, Availability and Serviceability extension.
rcpc - Enable support for RCPC extension.
rcpc-immo - Enable v8.4-A RCPC instructions with Immediate Offsets.
rdm - Enable ARMv8.1 Rounding Double Multiply Add/Subtract instructions.
reserve-x1 - Reserve X1, making it unavailable as a GPR.
reserve-x10 - Reserve X10, making it unavailable as a GPR.
reserve-x11 - Reserve X11, making it unavailable as a GPR.
reserve-x12 - Reserve X12, making it unavailable as a GPR.
reserve-x13 - Reserve X13, making it unavailable as a GPR.
reserve-x14 - Reserve X14, making it unavailable as a GPR.
reserve-x15 - Reserve X15, making it unavailable as a GPR.
reserve-x18 - Reserve X18, making it unavailable as a GPR.
reserve-x2 - Reserve X2, making it unavailable as a GPR.
reserve-x20 - Reserve X20, making it unavailable as a GPR.
reserve-x21 - Reserve X21, making it unavailable as a GPR.
reserve-x22 - Reserve X22, making it unavailable as a GPR.
reserve-x23 - Reserve X23, making it unavailable as a GPR.
reserve-x24 - Reserve X24, making it unavailable as a GPR.
reserve-x25 - Reserve X25, making it unavailable as a GPR.
reserve-x26 - Reserve X26, making it unavailable as a GPR.
reserve-x27 - Reserve X27, making it unavailable as a GPR.
reserve-x28 - Reserve X28, making it unavailable as a GPR.
reserve-x3 - Reserve X3, making it unavailable as a GPR.
reserve-x30 - Reserve X30, making it unavailable as a GPR.
reserve-x4 - Reserve X4, making it unavailable as a GPR.
reserve-x5 - Reserve X5, making it unavailable as a GPR.
reserve-x6 - Reserve X6, making it unavailable as a GPR.
reserve-x7 - Reserve X7, making it unavailable as a GPR.
reserve-x9 - Reserve X9, making it unavailable as a GPR.
saphira - Qualcomm Saphira processors.
sb - Enable v8.5 Speculation Barrier.
sel2 - Enable v8.4-A Secure Exception Level 2 extension.
sha2 - Enable SHA1 and SHA256 support.
sha3 - Enable SHA512 and SHA3 support.
slow-misaligned-128store - Misaligned 128 bit stores are slow.
slow-paired-128 - Paired 128 bit loads and stores are slow.
slow-strqro-store - STR of Q register with register offset is slow.
sm4 - Enable SM3 and SM4 support.
spe - Enable Statistical Profiling extension.
specrestrict - Enable architectural speculation restriction.
ssbs - Enable Speculative Store Bypass Safe bit.
strict-align - Disallow all unaligned memory access.
sve - Enable Scalable Vector Extension (SVE) instructions.
sve2 - Enable Scalable Vector Extension 2 (SVE2) instructions.
sve2-aes - Enable AES SVE2 instructions.
sve2-bitperm - Enable bit permutation SVE2 instructions.
sve2-sha3 - Enable SHA3 SVE2 instructions.
sve2-sm4 - Enable SM4 SVE2 instructions.
tagged-globals - Use an instruction sequence for taking the address of a global that allows a memory tag in the upper address bits.
thunderx - Cavium ThunderX processors.
thunderx2t99 - Cavium ThunderX2 processors.
thunderx3t110 - Marvell ThunderX3 processors.
thunderxt81 - Cavium ThunderX processors.
thunderxt83 - Cavium ThunderX processors.
thunderxt88 - Cavium ThunderX processors.
tlb-rmi - Enable v8.4-A TLB Range and Maintenance Instructions.
tme - Enable Transactional Memory Extension.
tpidr-el1 - Permit use of TPIDR_EL1 for the TLS base.
tpidr-el2 - Permit use of TPIDR_EL2 for the TLS base.
tpidr-el3 - Permit use of TPIDR_EL3 for the TLS base.
tracev8.4 - Enable v8.4-A Trace extension.
trbe - Enable Trace Buffer Extension.
tsv110 - HiSilicon TS-V110 processors.
uaops - Enable v8.2 UAO PState.
use-aa - Use alias analysis during codegen.
use-experimental-zeroing-pseudos - Hint to the compiler that the MOVPRFX instruction is merged with destructive operations.
use-postra-scheduler - Schedule again after register allocation.
use-reciprocal-square-root - Use the reciprocal square root approximation.
v8.1a - Support ARM v8.1a instructions.
v8.2a - Support ARM v8.2a instructions.
v8.3a - Support ARM v8.3a instructions.
v8.4a - Support ARM v8.4a instructions.
v8.5a - Support ARM v8.5a instructions.
v8.6a - Support ARM v8.6a instructions.
vh - Enables ARM v8.1 Virtual Host extension.
zcm - Has zero-cycle register moves.
zcz - Has zero-cycle zeroing instructions.
zcz-fp - Has zero-cycle zeroing instructions for FP registers.
zcz-fp-workaround - The zero-cycle floating-point zeroing instruction has a bug.
zcz-gp - Has zero-cycle zeroing instructions for generic registers.

Use +feature to enable a feature, or -feature to disable it.
For example, llc -mcpu=mycpu -mattr=+feature1,-feature2

@ghost
Copy link
Author

ghost commented Jan 24, 2023

More details

@ghost ghost changed the base branch from master to staging April 6, 2023 07:37
@ofborg ofborg bot added 10.rebuild-darwin: 501+ 10.rebuild-darwin: 5001+ 10.rebuild-linux: 501+ 10.rebuild-linux: 5001+ and removed 10.rebuild-darwin: 0 This PR does not cause any packages to rebuild on Darwin 10.rebuild-linux: 0 This PR does not cause any packages to rebuild on Linux labels Apr 6, 2023
Adam Joseph added 2 commits April 8, 2023 17:34
Partially revert "rust: fix overriding rust flags on musl"
98b3c2a 
This partially reverts commit 470e613.
rust: use {build,host}Platform.{gcc.arch,rust.flags}
b9d4aa5 
This commit causes `rustc` to use `stdenv.hostPlatform.gcc.arch`, if
present, to set `-C target-cpu`, and do similarly for the
`buildPlatform`.  Commented-out lines to do similarly for `gcc.tune`
are included, to be uncommented once `-Z tune-cpu` is stabilized.

This commit also allows passing arbitrary `rustc` flags through
`{build,host}Platform.rust.flags`.
@ghost ghost mentioned this pull request Apr 9, 2023
Merged
12 tasks
@ghost ghost closed this Jan 23, 2024
@ghost ghost deleted the pr/rust/gcc.arch branch January 23, 2024 06:49
@Artturin Artturin mentioned this pull request Jan 29, 2024
Open
This pull request was closed.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
6.topic: rust 10.rebuild-darwin: 501+ 10.rebuild-darwin: 5001+ 10.rebuild-linux: 501+ 10.rebuild-linux: 5001+
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
0 participants