Switch CI to Julia v1.10 #1562

ranocha · 2023-07-07T14:35:11Z

No description provided.

ranocha · 2023-07-15T12:12:26Z

Allocations with the P4estMesh are caused by JuliaLang/julia#50562

sloede · 2023-08-18T04:47:30Z

While on vacation I noticed that --check-bounds=no leads again to crazy long compile times,eg, for the optimized Ranocha turbo Euler setup. Do you get the same issues?

ranocha · 2023-08-18T10:29:11Z

While on vacation I noticed that --check-bounds=no leads again to crazy long compile times,eg, for the optimized Ranocha turbo Euler setup. Do you get the same issues?

I haven't tested it. It would be great if you could report this issue. So far, I have concentrated on the other issues with Julia v1.10:

Type instability (allocations) regression in Julia v1.10-alpha1 JuliaLang/julia#50562
_memcpy! not defined in HDF5.jl, see Fix _typed_load fast path for Julia 1.10 (currently nightly) JuliaIO/HDF5.jl#1075

There is evidence for other problems, e.g., p4est_part1 timed out (https://github.com/trixi-framework/Trixi.jl/actions/runs/5898667882/job/16000054066?pr=1562#step:7:3884).

sloede · 2023-08-18T15:31:01Z

OK, so first evidence that something is wrong with Julia v1.10 and --check-bounds=no (measured on Roci):

julia-1.10 --project=.:

julia> @time using Revise, Trixi, OrdinaryDiffEq
  3.221956 seconds (3.73 M allocations: 248.451 MiB, 7.26% gc time, 8.06% compilation time)
julia> @time trixi_include(default_example())
...
 13.680299 seconds (24.82 M allocations: 1.517 GiB, 4.26% gc time, 99.81% compilation time: <1% of which was recompilation)

julia-1.10 --project=. --check-bounds=no:

julia> @time using Revise, Trixi, OrdinaryDiffEq
  10.682994 seconds (25.34 M allocations: 5.952 GiB, 6.10% gc time, 3.66% compilation time: 10% of which was recompilation)
julia> @time trixi_include(default_example())
...
 69.902309 seconds (239.64 M allocations: 6.916 GiB, 1.20% gc time, 16.05% compilation time: <1% of which was recompilation)
julia> @time trixi_include(default_example())
...
 58.920417 seconds (214.77 M allocations: 5.291 GiB, 0.63% gc time)

And it's not just a compilation issue as one can see, it must be a runtime issue (type instabilities again?) 😢

sloede · 2023-08-19T13:38:18Z

Seems like StaticArrays.jl and/or StrideArrays.jl is involved:

With --check-bounds=no:

@code_warntype Trixi.get_node_vars(u, equations, dg, i, j, element)

MethodInstance for Trixi.get_node_vars(::StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, ::LinearScalarAdvectionEquation2D{Float64}, ::DGSEM{LobattoLegendreBasis{Float64, 3, SVector{3, Float64}, Matrix{Float64}, Matrix{Float64}, Matrix{Float64}}, Trixi.LobattoLegendreMortarL2{Float64, 3, Matrix{Float64}, Matrix{Float64}}, SurfaceIntegralWeakForm{FluxLaxFriedrichs{typeof(max_abs_speed_naive)}}, VolumeIntegralWeakForm}, ::Int64, ::Int64, ::Int64)
  from get_node_vars(u, equations, solver::DG, indices...) @ Trixi ~/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:465
Arguments
  #self#::Core.Const(Trixi.get_node_vars)
  u::StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}
  equations::LinearScalarAdvectionEquation2D{Float64}
  solver::DGSEM{LobattoLegendreBasis{Float64, 3, SVector{3, Float64}, Matrix{Float64}, Matrix{Float64}, Matrix{Float64}}, Trixi.LobattoLegendreMortarL2{Float64, 3, Matrix{Float64}, Matrix{Float64}}, SurfaceIntegralWeakForm{FluxLaxFriedrichs{typeof(max_abs_speed_naive)}}, VolumeIntegralWeakForm}
  indices::Tuple{Int64, Int64, Int64}
Locals
  #307::Trixi.var"#307#320"{StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, Tuple{Int64, Int64, Int64}}
Body::Any
1 ─       nothing
│   %2  = Trixi.:(var"#307#320")::Core.Const(Trixi.var"#307#320")
│   %3  = Core.typeof(u)::Core.Const(StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}})
│   %4  = Core.typeof(indices)::Core.Const(Tuple{Int64, Int64, Int64})
│   %5  = Core.apply_type(%2, %3, %4)::Core.Const(Trixi.var"#307#320"{StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, Tuple{Int64, Int64, Int64}})
│         (#307 = %new(%5, u, indices))
│   %7  = #307::Trixi.var"#307#320"{StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, Tuple{Int64, Int64, Int64}}
│   %8  = Trixi.nvariables(equations)::Core.Const(1)
│   %9  = Trixi.Val(%8)::Core.Const(Val{1}())
│   %10 = Trixi.ntuple(%7, %9)::Tuple{Any}
│   %11 = Trixi.SVector(%10)::Any
└──       return %11

@code_native Trixi.get_node_vars(u, equations, dg, i, j, element)

	.text
	.file	"get_node_vars"
	.globl	julia_get_node_vars_2212        # -- Begin function julia_get_node_vars_2212
	.p2align	4, 0x90
	.type	julia_get_node_vars_2212,@function
julia_get_node_vars_2212:               # @julia_get_node_vars_2212
; ┌ @ /home/mschlott/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:465 within `get_node_vars`
# %bb.0:                                # %top
	push	rbp
	mov	rbp, rsp
	push	r15
	push	r14
	push	r13
	push	r12
	push	rbx
	and	rsp, -32
	sub	rsp, 128
	vxorps	xmm0, xmm0, xmm0
	mov	qword ptr [rsp + 96], rdi       # 8-byte Spill
	mov	qword ptr [rsp + 88], r9        # 8-byte Spill
	#APP
	mov	rax, qword ptr fs:[0]
	#NO_APP
	mov	r13, rcx
	lea	rcx, [rsp + 32]
; │ @ /home/mschlott/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:476 within `get_node_vars`
; │┌ @ ntuple.jl:48 within `ntuple`
; ││┌ @ /home/mschlott/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:476 within `#307`
; │││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:941 within `getindex`
; ││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:930 within `_offset_ptr`
; │││││┌ @ tuple.jl:322 within `map`
	movabs	r12, offset "j_-_2214"
	mov	edi, 1
	mov	rbx, r8
	vmovaps	ymmword ptr [rsp + 32], ymm0
	mov	qword ptr [rsp + 72], 0
	mov	qword ptr [rsp + 64], 0
	mov	r14, qword ptr [rax - 8]
; │││││└
; │││││ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:929 within `_offset_ptr`
; │││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:484 within `pointer`
	mov	qword ptr [rsp + 32], 16
	mov	rax, qword ptr [r14]
	mov	qword ptr [rsp + 40], rax
	mov	qword ptr [r14], rcx
; │││││└
; │││││ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:930 within `_offset_ptr`
; │││││┌ @ tuple.jl:322 within `map`
	vzeroupper
	call	r12
; ││││││ @ tuple.jl:322 within `map` @ tuple.jl:322
	mov	rdi, r13
; ││││││ @ tuple.jl:322 within `map`
	mov	r15, rax
	mov	qword ptr [rsp + 72], rax
; ││││││ @ tuple.jl:322 within `map` @ tuple.jl:322
	call	r12
; ││││││ @ tuple.jl:322 within `map` @ tuple.jl:322 @ tuple.jl:319
	mov	rdi, rbx
; ││││││ @ tuple.jl:322 within `map` @ tuple.jl:322
	mov	r13, rax
	mov	qword ptr [rsp + 64], rax
; ││││││ @ tuple.jl:322 within `map` @ tuple.jl:322 @ tuple.jl:319
	call	r12
	mov	rdi, qword ptr [rsp + 88]       # 8-byte Reload
	mov	rbx, rax
	mov	qword ptr [rsp + 56], rax
	call	r12
; │││││└
; │││││ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:933 within `_offset_ptr`
; │││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:9 within `invpermtuple`
; ││││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:9 within `macro expansion`
	mov	qword ptr [rsp], r15
	mov	qword ptr [rsp + 8], r13
	mov	qword ptr [rsp + 16], rbx
	mov	qword ptr [rsp + 48], rax
	mov	qword ptr [rsp + 24], rax
	movabs	rax, offset jl_f_tuple
	mov	rsi, rsp
	xor	edi, edi
	mov	edx, 4
	call	rax
	mov	r12, qword ptr [rsp + 96]       # 8-byte Reload
; │││││└└
	movabs	rbx, offset ijl_gc_pool_alloc
	mov	esi, 1136
	mov	edx, 16
; │││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:9 within `invpermtuple`
; ││││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:9 within `macro expansion`
	mov	r15, rax
	mov	r13, qword ptr [r12 + 8]
	mov	qword ptr [rsp + 56], rax
; │││││└└
	mov	rdi, qword ptr [r14 + 16]
	call	rbx
	mov	rcx, rbx
	mov	rbx, rax
	movabs	rdi, 140223945836848
	mov	esi, 1136
	mov	edx, 16
	mov	qword ptr [rbx - 8], rdi
	mov	rax, qword ptr [r12]
	mov	qword ptr [rsp + 64], rbx
	mov	qword ptr [rbx], rax
	mov	rdi, qword ptr [r14 + 16]
	call	rcx
	movabs	rcx, 140214323774288
	movabs	rdi, 140214480650128
	mov	edx, 3
	mov	qword ptr [rax - 8], rcx
	mov	qword ptr [rsp], rbx
	mov	rbx, rsp
	mov	qword ptr [rsp + 8], r15
	movabs	r15, offset ijl_apply_generic
	mov	qword ptr [rax], r13
	mov	qword ptr [rsp + 48], rax
	mov	qword ptr [rsp + 16], rax
	mov	rsi, rbx
	call	r15
	movabs	rdi, 140214480669152
; ││││└
	mov	rsi, rbx
	mov	edx, 1
	mov	qword ptr [rsp + 48], rax
	mov	qword ptr [rsp], rax
	call	r15
	mov	qword ptr [rsp + 48], rax
; ││└└
	mov	qword ptr [rsp], rax
	xor	edi, edi
	mov	rsi, rbx
	mov	edx, 1
	movabs	rax, offset jl_f_tuple
	call	rax
	movabs	rdi, 140214497112368
; │└
	mov	rsi, rbx
	mov	edx, 1
	mov	qword ptr [rsp + 48], rax
	mov	qword ptr [rsp], rax
	call	r15
	mov	rcx, qword ptr [rsp + 40]
	mov	qword ptr [r14], rcx
	lea	rsp, [rbp - 40]
	pop	rbx
	pop	r12
	pop	r13
	pop	r14
	pop	r15
	pop	rbp
	ret
.Lfunc_end0:
	.size	julia_get_node_vars_2212, .Lfunc_end0-julia_get_node_vars_2212
; └
                                        # -- End function
	.section	".note.GNU-stack","",@progbits

Without --check-bounds=no:

@code_warntype Trixi.get_node_vars(u, equations, dg, i, j, element)

MethodInstance for Trixi.get_node_vars(::StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, ::LinearScalarAdvectionEquation2D{Float64}, ::DGSEM{LobattoLegendreBasis{Float64, 3, SVector{3, Float64}, Matrix{Float64}, Matrix{Float64}, Matrix{Float64}}, Trixi.LobattoLegendreMortarL2{Float64, 3, Matrix{Float64}, Matrix{Float64}}, SurfaceIntegralWeakForm{FluxLaxFriedrichs{typeof(max_abs_speed_naive)}}, VolumeIntegralWeakForm}, ::Int64, ::Int64, ::Int64)
  from get_node_vars(u, equations, solver::DG, indices...) @ Trixi ~/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:465
Arguments
  #self#::Core.Const(Trixi.get_node_vars)
  u::StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}
  equations::LinearScalarAdvectionEquation2D{Float64}
  solver::DGSEM{LobattoLegendreBasis{Float64, 3, SVector{3, Float64}, Matrix{Float64}, Matrix{Float64}, Matrix{Float64}}, Trixi.LobattoLegendreMortarL2{Float64, 3, Matrix{Float64}, Matrix{Float64}}, SurfaceIntegralWeakForm{FluxLaxFriedrichs{typeof(max_abs_speed_naive)}}, VolumeIntegralWeakForm}
  indices::Tuple{Int64, Int64, Int64}
Locals
  #307::Trixi.var"#307#320"{StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, Tuple{Int64, Int64, Int64}}
Body::SVector{1, Float64}
1 ─       nothing
│   %2  = Trixi.:(var"#307#320")::Core.Const(Trixi.var"#307#320")
│   %3  = Core.typeof(u)::Core.Const(StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}})
│   %4  = Core.typeof(indices)::Core.Const(Tuple{Int64, Int64, Int64})
│   %5  = Core.apply_type(%2, %3, %4)::Core.Const(Trixi.var"#307#320"{StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, Tuple{Int64, Int64, Int64}})
│         (#307 = %new(%5, u, indices))
│   %7  = #307::Trixi.var"#307#320"{StrideArraysCore.PtrArray{Float64, 4, (1, 2, 3, 4), Tuple{Static.StaticInt{1}, Static.StaticInt{3}, Static.StaticInt{3}, Int64}, NTuple{4, Nothing}, NTuple{4, Static.StaticInt{1}}}, Tuple{Int64, Int64, Int64}}
│   %8  = Trixi.nvariables(equations)::Core.Const(1)
│   %9  = Trixi.Val(%8)::Core.Const(Val{1}())
│   %10 = Trixi.ntuple(%7, %9)::Tuple{Float64}
│   %11 = Trixi.SVector(%10)::SVector{1, Float64}
└──       return %11

@code_native Trixi.get_node_vars(u, equations, dg, i, j, element)

	.text
	.file	"get_node_vars"
	.globl	julia_get_node_vars_6674        # -- Begin function julia_get_node_vars_6674
	.p2align	4, 0x90
	.type	julia_get_node_vars_6674,@function
julia_get_node_vars_6674:               # @julia_get_node_vars_6674
; ┌ @ /home/mschlott/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:465 within `get_node_vars`
# %bb.0:                                # %top
	push	rbp
; │ @ /home/mschlott/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:476 within `get_node_vars`
; │┌ @ ntuple.jl:48 within `ntuple`
; ││┌ @ /home/mschlott/.julia/packages/Trixi/CdZpe/src/solvers/dg.jl:476 within `#307`
; │││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:941 within `getindex`
; ││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:933 within `_offset_ptr`
; │││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:911 within `_offset_ptr_dense`
; ││││││┌ @ pointer.jl:282 within `+`
; │││││││┌ @ boot.jl:793 within `UInt64`
	mov	rdx, qword ptr [rdi]
; ││││││└└
; ││││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:902 within `_offset_dense` @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:902 @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:901
; │││││││┌ @ promotion.jl:423 within `*` @ int.jl:88
	lea	rax, [r9 + 2*r9]
	mov	rbp, rsp
; │││││││└
; │││││││ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:902 within `_offset_dense` @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:902
; │││││││┌ @ int.jl:87 within `+`
	add	rax, r8
; │││││││└
; │││││││┌ @ promotion.jl:423 within `*` @ int.jl:88
	lea	rax, [rax + 2*rax]
; │││││││└
; │││││││ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:902 within `_offset_dense`
; │││││││┌ @ int.jl:87 within `+`
	add	rax, rcx
; ││││└└└└
; ││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:719 within `pload`
; │││││┌ @ /home/mschlott/.julia/packages/StrideArraysCore/COJRJ/src/ptr_array.jl:719 within `macro expansion`
; ││││││┌ @ pointer.jl:119 within `unsafe_load` @ pointer.jl:119
	vmovsd	xmm0, qword ptr [rdx + 8*rax - 104] # xmm0 = mem[0],zero
; │└└└└└└
	pop	rbp
	ret
.Lfunc_end0:
	.size	julia_get_node_vars_6674, .Lfunc_end0-julia_get_node_vars_6674
; └
                                        # -- End function
	.section	".note.GNU-stack","",@progbits

sloede · 2023-08-19T13:43:13Z

Yep, seems like StrideArrays.jl is required to break it. If I do something like

u_array = similar(Array{Float64}, u)
u_array .= u

and then call

@code_warntype Trixi.get_node_vars(u_array, equations, dg, i, j, element)

there are no type instabilities anymore.

sloede · 2023-08-19T21:18:13Z

xref JuliaSIMD/StrideArrays.jl#78

sloede · 2023-08-20T10:00:44Z

xref JuliaLang/julia#50985

codecov · 2024-04-23T15:18:02Z

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.08%. Comparing base (73a1513) to head (efa9e95).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1562      +/-   ##
==========================================
+ Coverage   89.59%   96.08%   +6.49%     
==========================================
  Files         450      450              
  Lines       36155    36168      +13     
==========================================
+ Hits        32390    34749    +2359     
+ Misses       3765     1419    -2346

Flag	Coverage Δ
unittests	`96.08% <ø> (+6.49%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

DanielDoehring · 2024-04-24T10:14:05Z

As far as I understand, the remaining errors are caused due to some methods being overwritten.
What is the recommended way of handling this?

test/test_trixi.jl

DanielDoehring

Good to got from my side!

sloede

Thanks a lot! Overall, there are fewer changes than I feared.

I noticed that for some tests the difference in the reference values is quite large, showing up already at the 6-7 significant digits (i.e., well within single precision accuracy). Do you have an idea why those particular cases failed, and did you run maybe one or two of them longer and compare them manually to a v1.9 result to ensure that the simulation remains stable?

One commonality I noticed is that most tests you had to modify are either AMR or shock capturing tests. These are sometimes tricky to test, since if values are close to a cut-off threshold, even a small rounding error (as could happen with different floating optimizations) can result in a non-negligible change: A cell that is refined (or not), an element where a pure DG or a mixed DG/FV solution is used etc.

.github/workflows/Documenter.yml

.github/workflows/ci.yml

test/test_parabolic_1d.jl

test/test_parabolic_2d.jl

test/test_tree_2d_euler.jl

.github/workflows/benchmark.yml

test/test_tree_2d_eulermulti.jl

Co-authored-by: Michael Schlottke-Lakemper <[email protected]>

sloede

Sorry, I missed a few spots in my initial review. They are not directly related to v1.10, but should hopefully not require any other changes and we'd have cleaned up our CI a little 😊

Once the final convo is resolved, this is good to merge from my side

.github/workflows/ci.yml

test/test_tree_2d_eulermulti.jl

Co-authored-by: Michael Schlottke-Lakemper <[email protected]>

sloede

LGTM, thank you very much for tackling this! I'd like to get a second review from our CI wizard though...

sloede · 2024-04-28T06:49:13Z

@ranocha Could you please take a final look as well?

ranocha · 2024-04-28T10:16:30Z

Not everything works on ARM:

LoadError: User-defined reduction operators are currently not supported on non-Intel architectures.
See JuliaParallel/MPI.jl#404 for more details.

see https://github.com/trixi-framework/Trixi.jl/actions/runs/8865649949/job/24342153523?pr=1562#step:7:1124

We need to update our CI settings to reflect this - can we still use older versions/architectures for some MacOS tests?

sloede · 2024-04-28T17:47:17Z

We need to update our CI settings to reflect this - can we still use older versions/architectures for some MacOS tests?

Yes. Since this is unrelated to the purpose of this PR, we should proceed here and fix the failing tests on another PR

.github/workflows/ci.yml

sloede

Thanks again @DanielDoehring for tackling this! 🙏

check CI on Julia v1.10 pre-release version

c9f2419

ranocha added the testing label Jul 7, 2023

ranocha added 2 commits July 7, 2023 16:43

Update Project.toml

36f173a

Merge branch 'main' into hr/ci_julia110

b5ebb41

ranocha mentioned this pull request Jul 15, 2023

Type instability (allocations) regression in Julia v1.10-alpha1 JuliaLang/julia#50562

Closed

ranocha added 3 commits July 28, 2023 06:56

Merge branch 'main' into hr/ci_julia110

8f54cde

Merge branch 'main' into hr/ci_julia110

732aec8

Merge branch 'main' into hr/ci_julia110

72dd18a

ranocha and others added 6 commits September 12, 2023 10:44

Merge branch 'main' into hr/ci_julia110

93ad6c2

Merge branch 'main' into hr/ci_julia110

3430a1f

Merge branch 'main' into hr/ci_julia110

c570be4

Merge branch 'main' into hr/ci_julia110

c1581c3

Merge branch 'main' into hr/ci_julia110

7061535

Merge branch 'main' into hr/ci_julia110

9c51ae0

sloede mentioned this pull request Mar 11, 2024

Test on Julia v1.10 (and possibly 1.9) trixi-framework/TrixiShallowWater.jl#26

Closed

Merge branch 'main' into hr/ci_julia110

32fb7f9

DanielDoehring added 3 commits April 23, 2024 19:30

update test vals

4242333

update tests

8284734

update test vals

782653e

sloede mentioned this pull request Apr 24, 2024

Performance evaluation for Julia v1.11 #1916

Open

ranocha commented Apr 24, 2024

View reviewed changes

test/test_trixi.jl Outdated Show resolved Hide resolved

DanielDoehring added 5 commits April 25, 2024 08:53

shorten

05c16c9

comment

4322976

need prefix r

d0bf9a9

test simplification

c419335

shorten

dc1285d

DanielDoehring self-requested a review April 25, 2024 09:45

DanielDoehring previously approved these changes Apr 25, 2024

View reviewed changes

sloede changed the title ~~WIP: check CI on Julia v1.10 pre-release version~~ Switch CI to Julia v1.10 Apr 25, 2024

sloede requested changes Apr 25, 2024

View reviewed changes

DanielDoehring reviewed Apr 26, 2024

View reviewed changes

test/test_tree_2d_eulermulti.jl Show resolved Hide resolved

DanielDoehring dismissed their stale review via a9e9cf5 April 26, 2024 07:47

Apply suggestions from code review

a9e9cf5

Co-authored-by: Michael Schlottke-Lakemper <[email protected]>

sloede reviewed Apr 28, 2024

View reviewed changes

.github/workflows/ci.yml Outdated Show resolved Hide resolved

.github/workflows/ci.yml Outdated Show resolved Hide resolved

.github/workflows/ci.yml Show resolved Hide resolved

test/test_tree_2d_eulermulti.jl Show resolved Hide resolved

DanielDoehring and others added 3 commits April 28, 2024 08:21

Update ci.yml

a5ca4df

Apply suggestions from code review

48ab658

Co-authored-by: Michael Schlottke-Lakemper <[email protected]>

Merge branch 'main' into hr/ci_julia110

1ad2c32

sloede previously approved these changes Apr 28, 2024

View reviewed changes

sloede reviewed Apr 28, 2024

View reviewed changes

.github/workflows/ci.yml Outdated Show resolved Hide resolved

Update .github/workflows/ci.yml

6bb06ad

sloede dismissed their stale review via 6bb06ad April 28, 2024 17:52

Merge branch 'main' into hr/ci_julia110

efa9e95

sloede enabled auto-merge (squash) April 29, 2024 06:58

sloede mentioned this pull request Apr 29, 2024

Fix MPI tests on Apple Silicon #1922

Open

sloede approved these changes Apr 29, 2024

View reviewed changes

sloede merged commit 1ea9e3e into main Apr 29, 2024
38 checks passed

sloede deleted the hr/ci_julia110 branch April 29, 2024 07:10

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Switch CI to Julia v1.10 #1562

Switch CI to Julia v1.10 #1562

ranocha commented Jul 7, 2023

ranocha commented Jul 15, 2023

sloede commented Aug 18, 2023

ranocha commented Aug 18, 2023

sloede commented Aug 18, 2023

sloede commented Aug 19, 2023

sloede commented Aug 19, 2023

sloede commented Aug 19, 2023

sloede commented Aug 20, 2023

codecov bot commented Apr 23, 2024 •

edited

Loading

DanielDoehring commented Apr 24, 2024

DanielDoehring left a comment

sloede left a comment

sloede left a comment

sloede left a comment

sloede commented Apr 28, 2024

ranocha commented Apr 28, 2024

sloede commented Apr 28, 2024

sloede left a comment

Switch CI to Julia v1.10 #1562

Switch CI to Julia v1.10 #1562

Conversation

ranocha commented Jul 7, 2023

ranocha commented Jul 15, 2023

sloede commented Aug 18, 2023

ranocha commented Aug 18, 2023

sloede commented Aug 18, 2023

sloede commented Aug 19, 2023

sloede commented Aug 19, 2023

sloede commented Aug 19, 2023

sloede commented Aug 20, 2023

codecov bot commented Apr 23, 2024 • edited Loading

Codecov Report

DanielDoehring commented Apr 24, 2024

DanielDoehring left a comment

Choose a reason for hiding this comment

sloede left a comment

Choose a reason for hiding this comment

sloede left a comment

Choose a reason for hiding this comment

sloede left a comment

Choose a reason for hiding this comment

sloede commented Apr 28, 2024

ranocha commented Apr 28, 2024

sloede commented Apr 28, 2024

sloede left a comment

Choose a reason for hiding this comment

codecov bot commented Apr 23, 2024 •

edited

Loading