Broaden the template bound on the lp argument to constraints by WardBrian · Pull Request #3151 · stan-dev/math · GitHub

Summary

This makes the lp argument of the _constrain functions its own template parameter rather that solely relying on the other arguments. This is useful for stan-dev/stanc3#1436, as for technical reasons it is hard to prevent the user from passing all doubles if these are exposed.

It also changes the order of some overloads and uses some template pack tricks to allow the non-Lp versions of these functions to be vectorized as well.

Tests

Side Effects

Release notes

Checklist

Copyright holder: (fill in copyright holder information)

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/)
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)
the code is written in idiomatic C++ and changes are documented in the doxygen
the new changes are tested

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.35	0.39	0.89	-12.7% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	0.94	-6.08% slower
gp_regr/gen_gp_data.stan	0.03	0.02	1.21	17.36% faster
gp_regr/gp_regr.stan	0.1	0.1	1.04	4.08% faster
sir/sir.stan	70.38	70.91	0.99	-0.75% slower
irt_2pl/irt_2pl.stan	4.12	4.33	0.95	-5.07% slower
eight_schools/eight_schools.stan	0.06	0.06	1.04	3.93% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.25	0.26	0.97	-2.58% slower
pkpd/one_comp_mm_elim_abs.stan	19.62	18.97	1.03	3.29% faster
garch/garch.stan	0.44	0.4	1.09	8.57% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.7	2.6	1.04	3.71% faster
arK/arK.stan	1.79	1.73	1.04	3.57% faster
gp_pois_regr/gp_pois_regr.stan	2.86	2.74	1.04	4.16% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	8.82	8.37	1.05	5.06% faster
performance.compilation	182.45	185.91	0.98	-1.9% slower
Mean result: 1.0216045346939837

Jenkins Console Log
Blue Ocean
Commit hash: 919e8c47e5ad903c5eadc62f9102039c73340a9f

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: 2400.000
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 Reg file data sampling: Not affected
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; BHI 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

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.31	0.31	1.03	2.89% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	1.14	12.04% faster
gp_regr/gen_gp_data.stan	0.02	0.02	1.07	6.88% faster
gp_regr/gp_regr.stan	0.09	0.09	1.04	4.22% faster
sir/sir.stan	69.88	68.11	1.03	2.53% faster
irt_2pl/irt_2pl.stan	4.02	3.94	1.02	1.98% faster
eight_schools/eight_schools.stan	0.06	0.05	1.03	2.79% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.25	0.24	1.03	2.63% faster
pkpd/one_comp_mm_elim_abs.stan	19.44	19.0	1.02	2.24% faster
garch/garch.stan	0.44	0.4	1.08	7.63% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.74	2.59	1.06	5.52% faster
arK/arK.stan	1.8	1.73	1.04	4.16% faster
gp_pois_regr/gp_pois_regr.stan	2.92	2.74	1.07	6.41% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	8.8	8.44	1.04	4.06% faster
performance.compilation	181.81	185.39	0.98	-1.97% slower
Mean result: 1.0456849810292956

Jenkins Console Log
Blue Ocean
Commit hash: 919e8c47e5ad903c5eadc62f9102039c73340a9f

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: 2400.000
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 Reg file data sampling: Not affected
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; BHI 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

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.33	0.3	1.1	9.45% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	0.95	-5.47% slower
gp_regr/gen_gp_data.stan	0.02	0.02	1.04	3.78% faster
gp_regr/gp_regr.stan	0.09	0.09	1.03	2.68% faster
sir/sir.stan	70.39	69.82	1.01	0.81% faster
irt_2pl/irt_2pl.stan	4.16	4.27	0.98	-2.5% slower
eight_schools/eight_schools.stan	0.06	0.06	1.0	-0.4% slower
pkpd/sim_one_comp_mm_elim_abs.stan	0.25	0.26	0.98	-1.91% slower
pkpd/one_comp_mm_elim_abs.stan	19.5	19.06	1.02	2.27% faster
garch/garch.stan	0.43	0.4	1.07	6.33% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.71	2.6	1.04	4.02% faster
arK/arK.stan	1.81	1.71	1.06	5.72% faster
gp_pois_regr/gp_pois_regr.stan	2.91	2.71	1.08	7.0% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	8.69	8.4	1.03	3.33% faster
performance.compilation	183.53	188.78	0.97	-2.86% slower
Mean result: 1.0237096279798086

Jenkins Console Log
Blue Ocean
Commit hash: 919e8c47e5ad903c5eadc62f9102039c73340a9f

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: 2400.000
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 Reg file data sampling: Not affected
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; BHI 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

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.35	0.38	0.93	-7.65% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	0.99	-1.34% slower
gp_regr/gen_gp_data.stan	0.03	0.02	1.08	7.78% faster
gp_regr/gp_regr.stan	0.1	0.1	1.02	1.93% faster
sir/sir.stan	73.71	73.96	1.0	-0.35% slower
irt_2pl/irt_2pl.stan	4.71	4.55	1.04	3.39% faster
eight_schools/eight_schools.stan	0.07	0.06	1.25	20.04% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.27	0.26	1.03	2.98% faster
pkpd/one_comp_mm_elim_abs.stan	20.63	20.31	1.02	1.59% faster
garch/garch.stan	0.49	0.46	1.08	7.58% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.85	2.86	1.0	-0.23% slower
arK/arK.stan	1.86	1.89	0.98	-1.96% slower
gp_pois_regr/gp_pois_regr.stan	3.01	3.05	0.99	-1.19% slower
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	10.96	11.0	1.0	-0.34% slower
performance.compilation	194.67	205.52	0.95	-5.58% slower
Mean result: 1.0227631413541356

Jenkins Console Log
Blue Ocean
Commit hash: 919e8c47e5ad903c5eadc62f9102039c73340a9f

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: 3400.000
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 Reg file data sampling: Not affected
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; BHI 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

So this will allow lp to be var and return_type_t<EigVec> to be double or var.

Do we ever want to increment the jacobian when y just has a scalar type of double?

lp += 0.5 * log1m(sum_sqs);

What are the use types you want to pass into this? Is it just the return_type_t<EigVec> = double and lp = var case? Since the data is considered constant in that case shouldn't we just call the one that does not increment the jacobian?

return_type_t<EigVec> = double and lp = var

Yes, that's the relevant one that comes up during code generation. Based on my conversation with @bob-carpenter, I think we still want to let the user do it, even if it's weird. The implementation in stanc3#1494 gives the user both a _jacobian function and a _constrain function, and I think it's reasonable to view these as similar to the difference between doing a target+= with a _lpdf versus a _lupdf.

This looks like a bug in our api since we only have the lp version for corr_matirix_constrain(std::vector<>). imo I'd rather just have two functions. It's more code but I think cleaner just to have two separate function

You mean two separate overloads that both call apply_vector_unary? I can split them up if you want, it's just a bunch of copy-paste

Split!

Thanks! Yes it's just a bunch of copy pastes but imo I'd rather have duplicate code that is a bit easier to read here

So imo everything is good except that I would like to have a new signature in the places where we are doing LP.... That just looks kind of clunky and having two signatures just makes a bit more sense.

Add back docs for jacobian

I think this is just github showing the diff weirdly because a signature moved around -- there is no jacobian template on this overload

@SteveBronder I pushed the change to split up the Lp... earlier today -- is your latest review based on before that?

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.33	0.32	1.03	3.16% faster
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	1.06	5.66% faster
gp_regr/gen_gp_data.stan	0.03	0.02	1.1	9.12% faster
gp_regr/gp_regr.stan	0.1	0.09	1.07	6.85% faster
sir/sir.stan	70.42	68.22	1.03	3.13% faster
irt_2pl/irt_2pl.stan	4.1	3.98	1.03	3.11% faster
eight_schools/eight_schools.stan	0.06	0.05	1.08	7.14% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.25	0.24	1.03	2.85% faster
pkpd/one_comp_mm_elim_abs.stan	19.63	19.05	1.03	2.97% faster
garch/garch.stan	0.43	0.4	1.07	6.87% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.72	2.63	1.04	3.51% faster
arK/arK.stan	1.8	1.71	1.05	4.97% faster
gp_pois_regr/gp_pois_regr.stan	2.82	2.74	1.03	3.0% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	8.72	8.39	1.04	3.83% faster
performance.compilation	182.97	183.76	1.0	-0.43% slower
Mean result: 1.046440904755975

Jenkins Console Log
Blue Ocean
Commit hash: b3ebfc9cf0b6967e9be77a03696d221959159748

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: 2400.000
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 Reg file data sampling: Not affected
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; BHI 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

Did you want to also do cov_matrix_constrain_lkj? (Is that used?)

It’s not used anywhere by the compiler, so I don’t currently plan on exposing it to the user

@SteveBronder it looks like your latest push broke the accompanying stanc3 PR: https://jenkins.flatironinstitute.org/job/Stan/job/Stanc3/job/PR-1494/14/

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
arma/arma.stan	0.34	0.34	1.0	-0.11% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	0.96	-3.89% slower
gp_regr/gen_gp_data.stan	0.03	0.02	1.07	6.51% faster
gp_regr/gp_regr.stan	0.1	0.09	1.06	5.38% faster
sir/sir.stan	71.14	71.32	1.0	-0.26% slower
irt_2pl/irt_2pl.stan	4.22	4.24	1.0	-0.39% slower
eight_schools/eight_schools.stan	0.06	0.06	1.02	1.95% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.26	0.26	0.99	-0.69% slower
pkpd/one_comp_mm_elim_abs.stan	19.93	19.57	1.02	1.79% faster
garch/garch.stan	0.44	0.43	1.03	2.8% faster
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.84	2.74	1.04	3.63% faster
arK/arK.stan	1.84	1.82	1.01	1.13% faster
gp_pois_regr/gp_pois_regr.stan	2.88	2.86	1.01	0.53% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	9.16	8.92	1.03	2.56% faster
performance.compilation	191.13	202.56	0.94	-5.98% slower
Mean result: 1.01105728269144

Jenkins Console Log
Blue Ocean
Commit hash: 41fb879c2b057722f0b8d4e3462ddede193cb95e

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: 3474.467
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 Reg file data sampling: Not affected
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; BHI 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

Broaden the template bound on the lp argument to constraints

6e340ec

WardBrian requested a review from SteveBronder February 21, 2025 21:25

[Jenkins] auto-formatting by clang-format version 10.0.0-4ubuntu1

1b7424b

WardBrian mentioned this pull request Feb 24, 2025

Expose builtin constraints stan-dev/stanc3#1494

Merged

2 tasks

WardBrian added 2 commits February 24, 2025 16:51

Allow vector overloads to be called without lp argument

b7ef5fa

Missed require on ordered_constrain

5d9cacd

WardBrian and others added 6 commits February 25, 2025 10:28

Expression testing tweaks

a3253a1

Allow ad through cholesky_corr_free

df9d406

Allow ad through cholesky_corr_free

977f96e

More template and expression test tweaks

e92490c

[Jenkins] auto-formatting by clang-format version 10.0.0-4ubuntu1

71a32a1

More expressions

40707de

More expression test set-up

26829c1

SteveBronder reviewed Feb 27, 2025

View reviewed changes

Comment thread stan/math/prim/constraint/cov_matrix_constrain.hpp Outdated

SteveBronder reviewed Feb 27, 2025

View reviewed changes

WardBrian and others added 4 commits February 28, 2025 11:01

Split out apply_scalar overloads

f66acc4

[Jenkins] auto-formatting by clang-format version 10.0.0-4ubuntu1

2d4d039

Typos

acf90a7

Typos

7866a45

WardBrian requested a review from SteveBronder February 28, 2025 17:11

SteveBronder requested changes Feb 28, 2025

View reviewed changes

WardBrian force-pushed the broaden-constraint-lp-template branch from e918b34 to 7866a45 Compare March 6, 2025 14:52

WardBrian mentioned this pull request Mar 6, 2025

Perfect forwarding for constraints #3162

Closed

4 tasks

Merge branch 'develop' into broaden-constraint-lp-template

bf77776

WardBrian requested a review from SteveBronder March 6, 2025 14:54

WardBrian and others added 3 commits March 6, 2025 10:35

Missed one Lp...

e76cca4

Clean up if Jacobian and non-inline auto

d38a1cc

[Jenkins] auto-formatting by clang-format version 10.0.0-4ubuntu1

5aad82a

SteveBronder approved these changes Mar 6, 2025

View reviewed changes

WardBrian merged commit 0f1ad25 into develop Mar 7, 2025

WardBrian deleted the broaden-constraint-lp-template branch March 7, 2025 14:43

This was referenced Mar 7, 2025

Add arm64 to GHA CI and split tests across more runners #3158

Merged

Unexpected error when passing a multi-index to a offset or multiply constraint stan-dev/stan#3330

Closed

-              template <typename EigVec, require_eigen_vector_t<EigVec>* = nullptr>
+              template <typename EigVec, typename Lp,
+                        require_eigen_vector_t<EigVec>* = nullptr,
+                        require_convertible_t<return_type_t<EigVec>, Lp>* = nullptr>

+              inline auto corr_matrix_constrain(const T& y, int K, Lp&... lp) {
+                static_assert(sizeof...(lp) == 0 || sizeof...(lp) == 1,
+                              "corr_matrix_constrain should be called with either "
+                              "two or three arguments");
+                return apply_vector_unary<T>::apply(
+                    y, [&lp..., K](auto&& v) { return corr_matrix_constrain(v, K, lp...); });

+               * from transforming the specified finite vector of size K plus (K choose 2).
+               * This overload handles looping over the elements of a standard vector.
                *
-               * @tparam Jacobian if `true`, increment log density accumulator with log

Broaden the template bound on the lp argument to constraints by WardBrian · Pull Request #3151 · stan-dev/math · GitHub

Uh oh!

Uh oh!

Broaden the template bound on the lp argument to constraints - #3151

Broaden the template bound on the lp argument to constraints#3151WardBrian merged 19 commits intodevelopstan-dev/math:developfrom broaden-constraint-lp-templatestan-dev/math:broaden-constraint-lp-templateCopy head branch name to clipboard

Conversation

WardBrian commented Feb 21, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Tests

Side Effects

Release notes

Checklist

Uh oh!

stan-buildbot commented Feb 22, 2025

Uh oh!

stan-buildbot commented Feb 25, 2025

Uh oh!

stan-buildbot commented Feb 26, 2025

Uh oh!

stan-buildbot commented Feb 26, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

SteveBronder left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

WardBrian commented Feb 28, 2025

Uh oh!

stan-buildbot commented Feb 28, 2025

Uh oh!

SteveBronder commented Mar 4, 2025

Uh oh!

WardBrian commented Mar 4, 2025

Uh oh!

WardBrian commented Mar 4, 2025

Uh oh!

stan-buildbot commented Mar 7, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Broaden the template bound on the lp argument to constraints#3151
WardBrian merged 19 commits into
developfrom
broaden-constraint-lp-template

WardBrian commented Feb 21, 2025 •
edited
Loading