Fix ld error with TBB_LIB on macos

andrjohns commented 1 month ago

Summary

While helping a user specify the Homebrew TBB for use with CmdStan, I discovered that the linker on macos doesn't support the -Wl,--disable-new-dtags argument:

andrew@Andrews-MacBook-Air ~ % clang -v
Apple clang version 15.0.0 (clang-1500.3.9.4)
Target: arm64-apple-darwin23.5.0
Thread model: posix
InstalledDir: /Library/Developer/CommandLineTools/usr/bin
andrew@Andrews-MacBook-Air ~ % touch foo.cpp
andrew@Andrews-MacBook-Air ~ % clang++ foo.cpp -Wl,--disable-new-dtags
ld: unknown options: --disable-new-dtags 
clang: error: linker command failed with exit code 1 (use -v to see invocation)

It's not an issue with the Linux clang though (checked in a Debian Sid docker image)

Tests

N/A

Side Effects

N/A

Release notes

Fix linker error when using external TBB on MacOS

Checklist

[x] Copyright holder: Andrew Johnson

The copyright holder is typically you or your assignee, such as a university or company. By submitting this pull request, the copyright holder is agreeing to the license the submitted work under the following licenses:
- Code: BSD 3-clause (https://opensource.org/licenses/BSD-3-Clause)
- Documentation: CC-BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
[x] the basic tests are passing
- unit tests pass (to run, use: ./runTests.py test/unit)
- header checks pass, (make test-headers)
- dependencies checks pass, (make test-math-dependencies)
- docs build, (make doxygen)
- code passes the built in C++ standards checks (make cpplint)
[x] the code is written in idiomatic C++ and changes are documented in the doxygen
[x] the new changes are tested

wds15 commented 1 month ago

And we still get the right TBB being linked in on macOS? The flag was introduced on Linux to enforce that no system defaults lead to link against a system TBB rather than the supplied TBB by Stan.

andrjohns commented 1 month ago

And we still get the right TBB being linked in on macOS? The flag was introduced on Linux to enforce that no system defaults lead to link against a system TBB rather than the supplied TBB by Stan.

The flag is only used when the TBB_LIB variable is set - and a system TBB is being linked against

andrjohns commented 1 month ago

Great!

We've actually been patching this out in the conda build for some time, I've just continuously forgotten to upstream it: https://github.com/conda-forge/cmdstan-feedstock/blob/main/recipe/no-dtag-flag.patch

Peak downstream behaviour

stan-buildbot commented 1 month ago

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.19	0.2	0.94	-5.92% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	1.07	6.56% faster
gp_regr/gen_gp_data.stan	0.02	0.02	1.05	4.43% faster
gp_regr/gp_regr.stan	0.11	0.11	1.01	0.53% faster
sir/sir.stan	83.13	84.21	0.99	-1.3% slower
irt_2pl/irt_2pl.stan	4.04	4.0	1.01	0.8% faster
eight_schools/eight_schools.stan	0.06	0.05	1.01	1.28% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.25	0.26	0.99	-0.93% slower
pkpd/one_comp_mm_elim_abs.stan	18.66	19.73	0.95	-5.74% slower
garch/garch.stan	0.52	0.43	1.2	16.89% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.94	2.99	0.98	-1.7% slower
arK/arK.stan	1.76	1.58	1.11	10.09% faster
gp_pois_regr/gp_pois_regr.stan	2.71	2.52	1.07	6.77% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	9.77	8.96	1.09	8.3% faster
performance.compilation	190.04	189.86	1.0	0.1% faster

Mean result: 1.0315659926708851

Jenkins Console Log Blue Ocean Commit hash: 24717e4befa79396a1c98440805cce73129090b6

Machine information

No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 20.04.3 LTS Release: 20.04 Codename: focal CPU: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian Address sizes: 46 bits physical, 48 bits virtual CPU(s): 80 On-line CPU(s) list: 0-79 Thread(s) per core: 2 Core(s) per socket: 20 Socket(s): 2 NUMA node(s): 2 Vendor ID: GenuineIntel CPU family: 6 Model: 85 Model name: Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz Stepping: 4 CPU MHz: 3324.708 CPU max MHz: 3700.0000 CPU min MHz: 1000.0000 BogoMIPS: 4800.00 Virtualization: VT-x L1d cache: 1.3 MiB L1i cache: 1.3 MiB L2 cache: 40 MiB L3 cache: 55 MiB NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78 NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79 Vulnerability Gather data sampling: Mitigation; Microcode Vulnerability Itlb multihit: KVM: Mitigation: VMX disabled Vulnerability L1tf: Mitigation; PTE Inversion; VMX conditional cache flushes, SMT vulnerable Vulnerability Mds: Mitigation; Clear CPU buffers; SMT vulnerable Vulnerability Meltdown: Mitigation; PTI Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable Vulnerability Retbleed: Mitigation; IBRS Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; IBRS, IBPB conditional, STIBP conditional, RSB filling, PBRSB-eIBRS Not affected Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT vulnerable Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single pti intel_ppin ssbd mba ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req pku ospke md_clear flush_l1d arch_capabilities G++: g++ (Ubuntu 9.4.0-1ubuntu1~20.04) 9.4.0 Copyright (C) 2019 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Clang: clang version 10.0.0-4ubuntu1 Target: x86_64-pc-linux-gnu Thread model: posix InstalledDir: /usr/bin

WardBrian commented 1 month ago

This change broke a build I had that was manually setting LDFLAGS_TBB, since now the dtags flag is +='d it doesn't respect previous settings the way ?= does.

The build in question is not on a Mac, but does use clang's lld. We don't really have a good way of detecting this automatically.

I also noticed this flag is only actually being passed if you are using a custom TBB, not one vendored by Stan, so I wonder why we even need it at all?

andrjohns commented 1 month ago

This change broke a build I had that was manually setting LDFLAGS_TBB, since now the dtags flag is +='d it doesn't respect previous settings the way ?= does.

A that makes sense, we could just wrap the += in an ifeq ($(LDFLAGS_TBB),)?

I also noticed this flag is only actually being passed if you are using a custom TBB, not one vendored by Stan, so I wonder why we even need it at all?

Because if you try to use a custom TBB with Stan on Macos then you can't build the math library/cmdstan at all - because macos doesn't support the flag

WardBrian commented 1 month ago

Because if you try to use a custom TBB with Stan on Macos then you can't build the math library/cmdstan at all - because macos doesn't support the flag

I meant "why do we need --disable-new-dtags on any platform". Like what about our vendored TBB means we don't need it on linux etc, but a user-supplied ones means we do?

andrjohns commented 1 month ago

Ah sorry, I thought you were asking why we needed the change.

I don't know enough about the TBB & linking or what the flag does, but @wds15 mentioned above that it was used to stop Stan from unintentionally linking to the wrong TBB

WardBrian commented 1 month ago

I think we can get nice, customizable behavior with


# MacOS ld does not support --disable-new-dtags
ifneq ($(OS),Darwin)
  LDFLAGS_TBB_DTAGS ?= -Wl,--disable-new-dtags
endif

# Windows LLVM/Clang does not support -rpath, but is not needed on Windows anyway
ifneq ($(OS), Windows_NT)
  LDFLAGS_TBB_RPATH ?= -Wl,-rpath,"$(TBB_LIB)"
endif

LDFLAGS_TBB ?= -Wl,-L,"$(TBB_LIB)" $(LDFLAGS_TBB_DTAGS) $(LDFLAGS_TBB_RPATH)

stan-dev / math