Invicid two-way fluid-structure interaction #1075
Conversation
…is accelerating. Two is it is going uupward randomly. And three is it has a very large pressure interpolation.
…sectino to another in an array
|
Note Other AI code review bot(s) detectedCodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review. WalkthroughAdds two-way immersed-boundary coupling: new mass/moment/force/torque fields, MPI vector all-reduce and broadcast of mass, pressure-based force/torque accumulation, moment-of-inertia estimation, RK vectorized IB state updates, an example case, docs and tooling updates, and new golden test data. Changes
Sequence Diagram(s)sequenceDiagram
participant TS as TimeStepper
participant IBM as m_ibm
participant Fluid as FluidState(pressure)
participant MPI as MPI Layer
participant Patch as patch_ib
TS->>IBM: request force/torque computation
IBM->>Fluid: sample pressure at ib_markers
IBM->>IBM: integrate pressure → local force & torque
IBM->>MPI: s_mpi_allreduce_vectors_sum(force_loc, force_glb, nvec, 3)
MPI->>Patch: reduce/broadcast global force & torque (mass included)
alt moving_ibm == 2 (two-way)
IBM->>IBM: s_compute_moment_of_inertia(geometry)
IBM->>Patch: vel += (force / mass) * dt
IBM->>Patch: angular_vel += (torque / moment) * dt
else moving_ibm == 1 (one-way)
IBM->>Patch: update state without momentum feedback
end
IBM->>Patch: angles += angular_vel * dt
Patch->>TS: updated IB state
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes
Suggested labels
Suggested reviewers
Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning, 1 inconclusive)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
CodeAnt AI is reviewing your PR. Thanks for using CodeAnt! 🎉We're free for open-source projects. if you're enjoying it, help us grow by sharing. Share on X · |
PR Reviewer Guide 🔍Here are some key observations to aid the review process:
|
codeant-ai
bot
added
the
size:XL
![qodo-code-review[bot]](https://avatars.githubusercontent.com/in/484649?s=60&v=4){kind=small}
There was a problem hiding this comment.
High-level Suggestion
The force and torque calculations in s_compute_ib_forces are incorrect because the pressure gradient and cell volume are computed using cell coordinates instead of grid spacing and cell dimensions. This invalidates the fluid-structure interaction results. [High-level, importance: 9]
Solution Walkthrough:
Before:
subroutine s_compute_ib_forces(pressure)
...
do i = 0, m
do j = 0, n
do k = 0, p
...
! Incorrect pressure gradient using cell-center coordinates in denominator
pressure_divergence(1) = (pressure(i+1,j,k) - pressure(i-1,j,k)) / (2._wp * x_cc(i))
pressure_divergence(2) = (pressure(i,j+1,k) - pressure(i,j-1,k)) / (2._wp * y_cc(j))
! Incorrect cell volume calculation using coordinates
cell_volume = x_cc(i) * y_cc(j)
if (num_dims == 3) then
pressure_divergence(3) = (pressure(i,j,k+1) - pressure(i,j,k-1)) / (2._wp * z_cc(k))
cell_volume = cell_volume * z_cc(k)
end if
forces(ib_idx, :) = forces(ib_idx, :) - (pressure_divergence * cell_volume)
...
end do
end do
end do
end subroutine
After:
subroutine s_compute_ib_forces(pressure)
...
! Grid spacings (dx, dy, dz) should be available or computed
! e.g., dx = x_cc(1) - x_cc(0) for a uniform grid
do i = 0, m
do j = 0, n
do k = 0, p
...
! Correct pressure gradient using grid spacing
pressure_divergence(1) = (pressure(i+1,j,k) - pressure(i-1,j,k)) / (2._wp * dx)
pressure_divergence(2) = (pressure(i,j+1,k) - pressure(i,j-1,k)) / (2._wp * dy)
! Correct cell volume calculation
cell_volume = dx * dy
if (num_dims == 3) then
pressure_divergence(3) = (pressure(i,j,k+1) - pressure(i,j,k-1)) / (2._wp * dz)
cell_volume = cell_volume * dz
end if
forces(ib_idx, :) = forces(ib_idx, :) - (pressure_divergence * cell_volume)
...
end do
end do
end do
end subroutine
src/simulation/m_ibm.fpp
Outdated
| pressure_divergence(1) = (pressure(i+1, j, k) - pressure(i-1, j, k)) / (2._wp * x_cc(i)) | ||
| pressure_divergence(2) = (pressure(i, j+1, k) - pressure(i, j-1, k)) / (2._wp * y_cc(j)) | ||
| cell_volume = x_cc(i) * y_cc(j) | ||
| if (num_dims == 3) then | ||
| pressure_divergence(3) = (pressure(i, j, k+1) - pressure(i, j, k-1)) / (2._wp * z_cc(k)) | ||
| cell_volume = cell_volume * z_cc(k) | ||
| else | ||
| pressure_divergence(3) = 0._wp | ||
| end if |
There was a problem hiding this comment.
✅ Suggestion: Correct the finite difference calculation for pressure_divergence and cell_volume. The denominators for the gradient should use grid spacing differences, and the volume should be calculated from grid spacings, not cell-center coordinates. [possible issue, importance: 9]
| pressure_divergence(1) = (pressure(i+1, j, k) - pressure(i-1, j, k)) / (2._wp * x_cc(i)) | |
| pressure_divergence(2) = (pressure(i, j+1, k) - pressure(i, j-1, k)) / (2._wp * y_cc(j)) | |
| cell_volume = x_cc(i) * y_cc(j) | |
| if (num_dims == 3) then | |
| pressure_divergence(3) = (pressure(i, j, k+1) - pressure(i, j, k-1)) / (2._wp * z_cc(k)) | |
| cell_volume = cell_volume * z_cc(k) | |
| else | |
| pressure_divergence(3) = 0._wp | |
| end if | |
| pressure_divergence(1) = (pressure(i+1, j, k) - pressure(i-1, j, k)) / (x_cc(i+1) - x_cc(i-1)) | |
| pressure_divergence(2) = (pressure(i, j+1, k) - pressure(i, j-1, k)) / (y_cc(j+1) - y_cc(j-1)) | |
| cell_volume = (x_cc(i) - x_cc(i-1)) * (y_cc(j) - y_cc(j-1)) | |
| if (num_dims == 3) then | |
| pressure_divergence(3) = (pressure(i, j, k+1) - pressure(i, j, k-1)) / (z_cc(k+1) - z_cc(k-1)) | |
| cell_volume = cell_volume * (z_cc(k) - z_cc(k-1)) | |
| else | |
| pressure_divergence(3) = 0._wp | |
| end if |
| if (p == 0) then | ||
| axis = [0, 1, 0] | ||
| else if (sqrt(sum(axis**2)) == 0) then | ||
| ! if the object is not actually rotating at this time, return a dummy value and exit | ||
| patch_ib(ib_marker)%moment = 1._wp | ||
| return | ||
| else | ||
| axis = axis / sqrt(sum(axis)) | ||
| end if |
There was a problem hiding this comment.
Suggestion: Correct the normalization of the axis vector by using the sum of squares of its components (sum(axis**2)) instead of the sum of the components (sum(axis)) to calculate its magnitude. [possible issue, importance: 9]
| if (p == 0) then | |
| axis = [0, 1, 0] | |
| else if (sqrt(sum(axis**2)) == 0) then | |
| ! if the object is not actually rotating at this time, return a dummy value and exit | |
| patch_ib(ib_marker)%moment = 1._wp | |
| return | |
| else | |
| axis = axis / sqrt(sum(axis)) | |
| end if | |
| if (p == 0) then | |
| axis = [0, 1, 0] | |
| else if (sqrt(sum(axis**2)) == 0) then | |
| ! if the object is not actually rotating at this time, return a dummy value and exit | |
| patch_ib(ib_marker)%moment = 1._wp | |
| return | |
| else | |
| axis = axis / sqrt(sum(axis**2)) | |
| end if |
src/simulation/m_time_steppers.fpp
Outdated
| ! update the angular velocity with the torque value | ||
| patch_ib(i)%angular_vel = (patch_ib(i)%angular_vel * patch_ib(i)%moment) + (rk_coef(s, 3)*dt*patch_ib(i)%torque/rk_coef(s, 4)) ! add the torque to the angular momentum | ||
| call s_compute_moment_of_inertia(i, patch_ib(i)%angular_vel) ! update the moment of inertia to be based on the direction of the angular momentum | ||
| patch_ib(i)%angular_vel = patch_ib(i)%angular_vel / patch_ib(i)%moment ! convert back to angular velocity with the new moment of inertia |
There was a problem hiding this comment.
Suggestion: Refactor the angular velocity update to be physically consistent. The moment of inertia should not be recomputed within the time step based on the updated angular momentum. [possible issue, importance: 8]
| ! update the angular velocity with the torque value | |
| patch_ib(i)%angular_vel = (patch_ib(i)%angular_vel * patch_ib(i)%moment) + (rk_coef(s, 3)*dt*patch_ib(i)%torque/rk_coef(s, 4)) ! add the torque to the angular momentum | |
| call s_compute_moment_of_inertia(i, patch_ib(i)%angular_vel) ! update the moment of inertia to be based on the direction of the angular momentum | |
| patch_ib(i)%angular_vel = patch_ib(i)%angular_vel / patch_ib(i)%moment ! convert back to angular velocity with the new moment of inertia | |
| ! update the angular velocity with the torque value | |
| patch_ib(i)%angular_vel = patch_ib(i)%angular_vel + (rk_coef(s, 3)*dt*patch_ib(i)%torque / patch_ib(i)%moment)/rk_coef(s, 4) | |
| ! The moment of inertia should be updated based on the new orientation of the body, which is not done here. | |
| ! For a rigid body, the moment of inertia tensor in the body-fixed frame is constant. | |
| ! What is needed is to transform it to the lab frame. | |
| ! The call to s_compute_moment_of_inertia here seems to be based on a misunderstanding. | |
| ! If the intent is to handle shape changes or axis of rotation changes, this needs a more robust implementation. | |
| ! For now, removing the re-computation within the time step seems safer. | |
| ! call s_compute_moment_of_inertia(i, patch_ib(i)%angular_vel) |
tests/127A967A/golden.txt
Outdated
| D/cons.2.00.000000.dat 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 | ||
| D/cons.2.00.000050.dat 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999998 0.04999999999985 0.04999999999852 0.04999999995641 0.0499999983967 0.04999997952971 0.04999978191689 0.04999796506702 0.04999025889306 0.04998169706106 0.04999611085642 0.04999938528897 0.05000003406488 0.05000000856657 0.05000000100545 0.05000000001462 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999998 0.04999999999991 0.04999999999923 0.04999999996488 0.04999999803882 0.04999997229677 0.04999967269375 0.04999665422966 0.04997168524811 0.04987327254808 0.04979311624675 0.04995397841543 0.04999594739398 0.05000077788507 0.0500001639964 0.05000002052222 0.05000000142172 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999997 0.04999999999991 0.04999999998695 0.04999999844259 0.04999997623937 0.04999968873807 0.0499965010374 0.04996504309478 0.04973820835164 0.04886555932451 0.04843778808391 0.04966225373664 0.05002701523979 0.05001309247728 0.0500022467731 0.05000026951288 0.05000001831563 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999997 0.04999999999993 0.04999999999827 0.04999999908656 0.04999998428982 0.04999977563247 0.04999723010338 0.04997053645753 0.04970923688668 0.04791285240187 0.04046420333642 0.03907204776879 0.04957555210315 0.05051144009196 0.05015540600603 0.05002287714675 0.05000246467859 0.05000017642509 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999999 0.04999999999998 0.04999999999169 0.04999999947116 0.04999999119532 0.04999986124473 0.04999804330651 0.04997620856266 0.04976347728702 0.04755588730479 0.03688418211879 0.01437060350092 0.02045694202298 0.05422678992115 0.05645452469339 0.05155752111215 0.05020233245261 0.05001961721525 0.05000145488255 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999994 0.04999999987414 0.04999999764279 0.049999962487 0.04999945230823 0.04999309608248 0.0499266930397 0.04936811276406 0.04440131390219 0.02711307233116 0.01470605861931 0.00087507308674 0.04613735708056 0.06908811263212 0.05544404130184 0.05072523836215 0.05007832371291 0.05000602983079 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000025 0.050000000056 0.05000000091518 0.05000001346286 0.05000017294338 0.05000189222029 0.05001739310026 0.05013022588402 0.05111953845441 0.05069538395107 0.00090001380328 0.00083929051022 0.00083936399812 0.05228187392241 0.05012122851237 0.04995499547067 0.0499919416536 0.04999903814445 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000009 0.0500000001691 0.05000000315234 0.05000004921188 0.05000070158163 0.05000864152319 0.05009005344984 0.05076081742011 0.05691156993021 0.07660369574895 0.05047641812066 0.0008425467421 0.02029819842763 0.03199499351858 0.04415345893543 0.04911822882368 0.04990431464006 0.04999245866319 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000001 0.05000000000005 0.05000000001208 0.05000000075044 0.05000001240977 0.05000018666652 0.05000249132331 0.05002826113878 0.05027356434779 0.05211426405864 0.05984506837099 0.05371160985389 0.02079466022408 0.01733300719257 0.03791279186699 0.04723145096985 0.04970555863263 0.04996803051822 0.04999760643662 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000001 0.05000000000005 0.05000000000011 0.0500000000101 0.05000000129503 0.05000002126439 0.0500002882534 0.05000333083001 0.05003513661765 0.05023318838838 0.05095275596936 0.04991870163798 0.03660642996376 0.03896622693327 0.04747990834842 0.04960725690713 0.04995782245821 0.0499959550632 0.0499997084973 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000001 0.05000000000006 0.05000000000015 0.05000000002912 0.05000000220327 0.05000003139675 0.05000038524923 0.05000377343971 0.05002264459563 0.05006307436213 0.04968943232059 0.04768905480554 0.04844277043544 0.04966184466254 0.04995016077492 0.04999470130797 0.04999951229451 0.04999996574246 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000003 0.05000000000014 0.05000000000127 0.05000000005654 0.05000000265932 0.05000003604149 0.05000031792185 0.05000162231062 0.05000073863141 0.04993019223311 0.04969990123289 0.04982000925175 0.04995949705861 0.04999482397968 0.04999946848261 0.04999995327899 0.04999999687019 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000004 0.05000000000028 0.05000000000267 0.05000000008561 0.05000000249016 0.05000002120161 0.05000008543336 0.04999942946791 0.04999252765673 0.04996559869146 0.04998179773187 0.04999605999322 0.04999954734343 0.04999995482017 0.04999999625434 0.04999999988223 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000006 0.05000000000055 0.05000000000221 0.0500000000624 0.05000000113051 0.05000000270023 0.04999989932472 0.04999925232776 0.0499967450211 0.04999842667202 0.04999967545278 0.04999996543917 0.04999999672522 0.04999999985538 0.04999999999801 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000006 0.05000000000045 0.0500000000012 0.04999999999997 0.04999999978849 0.04999999170712 0.04999993680487 0.04999973756309 0.04999988169797 0.0499999768142 0.04999999773868 0.04999999989704 0.0499999999987 0.04999999999946 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000004 0.05000000000017 0.04999999999968 0.04999999998615 0.04999999948328 0.04999999539019 0.04999998154085 0.04999999211345 0.04999999855465 0.0499999999601 0.04999999999938 0.04999999999966 0.04999999999992 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05000000000001 0.04999999999995 0.04999999999907 0.04999999999818 0.04999999980508 0.04999999888801 0.04999999960797 0.04999999999297 0.04999999999979 0.04999999999987 0.04999999999995 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999988 0.04999999999938 0.04999999999949 0.04999999999973 0.04999999999979 0.04999999999962 0.04999999999996 0.04999999999998 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999999 0.04999999999989 0.04999999999987 0.04999999999996 0.04999999999995 0.04999999999995 0.04999999999999 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.04999999999999 0.04999999999998 0.04999999999999 0.04999999999999 0.04999999999999 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 0.05 | ||
| D/cons.3.00.000000.dat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 | ||
| D/cons.3.00.000050.dat -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -5e-14 -3.9e-13 -1.67e-12 -3.238e-11 -1.602e-09 -1.685203e-08 -1.3879433e-07 -1.20253567e-06 -2.48090056e-06 4.8065816e-07 2.6775886e-06 6.2848626e-07 5.263255e-08 1.95458e-09 -5.4975e-10 -4.098e-11 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -3e-14 -2.6e-13 -1.37e-12 -5.46e-11 -2.17158e-09 -2.666603e-08 -2.6752521e-07 -2.18830974e-06 -1.752609875e-05 -3.211928825e-05 7.66511127e-06 3.647123626e-05 7.51967815e-06 4.9668817e-07 -1.402219e-08 -1.067843e-08 -9.1592e-10 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -2e-14 -1.4e-13 -7.6e-13 -5.087e-11 -2.22457e-09 -3.063205e-08 -3.5114471e-07 -3.39969244e-06 -2.750061072e-05 -0.00018648476564 -0.00027706630843 9.528364695e-05 0.0003330373763 6.687850733e-05 3.2919989e-06 -1.1214102e-06 -2.2039798e-07 -1.580863e-08 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -3.4e-13 -2.576e-11 -1.8256e-09 -2.796068e-08 -3.5602536e-07 -3.91305267e-06 -3.657325698e-05 -0.00029593131579 -0.00179746420834 -0.00240061641215 0.00159538828075 0.00252336445901 0.00038246941389 4.94108514e-06 -2.013154284e-05 -3.59999296e-06 -2.4640266e-07 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -4.5e-13 -6.37e-12 -1.30309e-09 -2.128751e-08 -3.015354e-07 -3.69717504e-06 -3.81679221e-05 -0.00030819093772 -0.00264159178646 -0.00918696605728 -0.00894502422236 0.00790304543832 0.01264497615098 0.00118139905033 -0.00068990604601 -0.00022691359074 -2.794972064e-05 -2.31274576e-06 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -7e-14 -3e-13 -4.2078e-10 -8.27975e-09 -1.2572284e-07 -1.67158865e-06 -1.902987943e-05 -0.00018185379269 -0.00137772693871 -0.01213731773214 -0.0400116682209 -0.05298121495587 -0.1008045121587 -0.03345851742991 -0.0121939798663 -0.00569391521925 -0.00093531569062 -0.00012955329572 -1.138967415e-05 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -9e-14 -6.51e-12 -1.53741e-09 -2.59862e-08 -3.863481e-07 -4.99058985e-06 -5.467985609e-05 -0.00049818810497 -0.0037464779014 -0.02621611747289 -0.07038919342009 -0.10367757133706 -0.09668251912213 -0.09669098460091 -0.0442989083387 -0.02338333186679 -0.00450880022162 -0.00052879220748 -4.894860962e-05 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -7e-14 -2.8e-13 -4.6039e-10 -8.7796e-09 -1.2999687e-07 -1.67723378e-06 -1.840823211e-05 -0.00016761865249 -0.00130118300606 -0.00833918749171 -0.01798631430591 -0.04032600790647 -0.09705762249572 -0.04967291675366 -0.03372943932347 -0.01048973693236 -0.00142846475316 -0.0001999938414 -1.738735182e-05 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -4.8e-13 -1.298e-11 -1.56446e-09 -2.48467e-08 -3.3932365e-07 -3.9442354e-06 -3.726442649e-05 -0.00029614138144 -0.00115164241236 0.00100102189618 0.01253987189686 0.00539062607243 -0.00901288704113 -0.00876673598871 -0.0026601678922 -0.00033376428405 -4.401176322e-05 -3.49810397e-06 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -8e-14 -4.6e-13 -3.392e-11 -2.01121e-09 -2.84468e-08 -3.2460469e-07 -3.08366754e-06 -2.646324632e-05 -5.620222336e-05 0.00042622892921 0.00311887118677 0.00194634041412 -0.00283639975938 -0.00209858305702 -0.00035444153341 -4.528772689e-05 -5.06617314e-06 -3.7266019e-07 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -3e-14 -1.7e-13 -8.1e-13 -7.562e-11 -2.69897e-09 -3.5334e-08 -4.0434199e-07 -2.3295561e-06 2.16260429e-06 7.769877415e-05 0.00044554496059 0.00013680966491 -0.00037923180177 -0.00023783432061 -3.549254234e-05 -4.57215344e-06 -4.9449213e-07 -3.457382e-08 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -4e-14 -2.9e-13 -2.1e-12 -5.735e-11 -2.09036e-09 -2.262854e-08 -1.1067959e-07 5.1652494e-07 9.72943745e-06 5.266275082e-05 1.053522087e-05 -4.549812061e-05 -2.423801213e-05 -3.11928838e-06 -3.9561964e-07 -4.097067e-08 -2.67848e-09 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -5.2e-13 -3.22e-12 -9.174e-11 -1.36949e-09 -1.14039e-09 5.778961e-08 9.2516025e-07 5.01145346e-06 9.1222982e-07 -4.56887504e-06 -2.06191627e-06 -2.4019976e-07 -2.995784e-08 -2.91654e-09 -9.48e-11 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -3e-13 -3.25e-12 -6.422e-11 1.0899e-10 5.92543e-09 7.997296e-08 4.0541908e-07 6.47887e-08 -3.8675841e-07 -1.507329e-07 -1.65617e-08 -2.00472e-09 -6.73e-11 -3.06e-12 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -2e-14 -1e-13 1.4e-13 1.6e-12 2.8909e-10 5.53166e-09 2.958209e-08 3.83231e-09 -2.821985e-08 -9.97116e-09 -1.01878e-09 -2.377e-11 -1.56e-12 -2.6e-13 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 1.1e-13 9.2e-13 1.43e-12 2.8934e-10 2.0374e-09 1.7924e-10 -1.91671e-09 -5.905e-10 -3.86e-12 -7e-14 -9e-14 -2e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 1e-13 2.9e-13 3.99e-12 8.751e-11 1.344e-11 -9.66e-11 -8.07e-12 -6e-14 -2e-14 -2e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 6e-14 1e-13 1.7e-13 4e-14 -2.9e-13 -7e-14 -3e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 1e-14 1e-14 -1e-14 -0.0 -0.0 -1e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 -0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 |
There was a problem hiding this comment.
✅ Suggestion: Replace all occurrences of negative zero (-0.0) with 0.0 in the test data file. This will improve test robustness by ensuring a canonical string representation for zero. [general, importance: 4]
| D/cons.3.00.000050.dat -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -5e-14 -3.9e-13 -1.67e-12 -3.238e-11 -1.602e-09 -1.685203e-08 -1.3879433e-07 -1.20253567e-06 -2.48090056e-06 4.8065816e-07 2.6775886e-06 6.2848626e-07 5.263255e-08 1.95458e-09 -5.4975e-10 -4.098e-11 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -3e-14 -2.6e-13 -1.37e-12 -5.46e-11 -2.17158e-09 -2.666603e-08 -2.6752521e-07 -2.18830974e-06 -1.752609875e-05 -3.211928825e-05 7.66511127e-06 3.647123626e-05 7.51967815e-06 4.9668817e-07 -1.402219e-08 -1.067843e-08 -9.1592e-10 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -2e-14 -1.4e-13 -7.6e-13 -5.087e-11 -2.22457e-09 -3.063205e-08 -3.5114471e-07 -3.39969244e-06 -2.750061072e-05 -0.00018648476564 -0.00027706630843 9.528364695e-05 0.0003330373763 6.687850733e-05 3.2919989e-06 -1.1214102e-06 -2.2039798e-07 -1.580863e-08 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -3.4e-13 -2.576e-11 -1.8256e-09 -2.796068e-08 -3.5602536e-07 -3.91305267e-06 -3.657325698e-05 -0.00029593131579 -0.00179746420834 -0.00240061641215 0.00159538828075 0.00252336445901 0.00038246941389 4.94108514e-06 -2.013154284e-05 -3.59999296e-06 -2.4640266e-07 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -4.5e-13 -6.37e-12 -1.30309e-09 -2.128751e-08 -3.015354e-07 -3.69717504e-06 -3.81679221e-05 -0.00030819093772 -0.00264159178646 -0.00918696605728 -0.00894502422236 0.00790304543832 0.01264497615098 0.00118139905033 -0.00068990604601 -0.00022691359074 -2.794972064e-05 -2.31274576e-06 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -7e-14 -3e-13 -4.2078e-10 -8.27975e-09 -1.2572284e-07 -1.67158865e-06 -1.902987943e-05 -0.00018185379269 -0.00137772693871 -0.01213731773214 -0.0400116682209 -0.05298121495587 -0.1008045121587 -0.03345851742991 -0.0121939798663 -0.00569391521925 -0.00093531569062 -0.00012955329572 -1.138967415e-05 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -9e-14 -6.51e-12 -1.53741e-09 -2.59862e-08 -3.863481e-07 -4.99058985e-06 -5.467985609e-05 -0.00049818810497 -0.0037464779014 -0.02621611747289 -0.07038919342009 -0.10367757133706 -0.09668251912213 -0.09669098460091 -0.0442989083387 -0.02338333186679 -0.00450880022162 -0.00052879220748 -4.894860962e-05 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -7e-14 -2.8e-13 -4.6039e-10 -8.7796e-09 -1.2999687e-07 -1.67723378e-06 -1.840823211e-05 -0.00016761865249 -0.00130118300606 -0.00833918749171 -0.01798631430591 -0.04032600790647 -0.09705762249572 -0.04967291675366 -0.03372943932347 -0.01048973693236 -0.00142846475316 -0.0001999938414 -1.738735182e-05 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -4.8e-13 -1.298e-11 -1.56446e-09 -2.48467e-08 -3.3932365e-07 -3.9442354e-06 -3.726442649e-05 -0.00029614138144 -0.00115164241236 0.00100102189618 0.01253987189686 0.00539062607243 -0.00901288704113 -0.00876673598871 -0.0026601678922 -0.00033376428405 -4.401176322e-05 -3.49810397e-06 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -8e-14 -4.6e-13 -3.392e-11 -2.01121e-09 -2.84468e-08 -3.2460469e-07 -3.08366754e-06 -2.646324632e-05 -5.620222336e-05 0.00042622892921 0.00311887118677 0.00194634041412 -0.00283639975938 -0.00209858305702 -0.00035444153341 -4.528772689e-05 -5.06617314e-06 -3.7266019e-07 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -3e-14 -1.7e-13 -8.1e-13 -7.562e-11 -2.69897e-09 -3.5334e-08 -4.0434199e-07 -2.3295561e-06 2.16260429e-06 7.769877415e-05 0.00044554496059 0.00013680966491 -0.00037923180177 -0.00023783432061 -3.549254234e-05 -4.57215344e-06 -4.9449213e-07 -3.457382e-08 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -4e-14 -2.9e-13 -2.1e-12 -5.735e-11 -2.09036e-09 -2.262854e-08 -1.1067959e-07 5.1652494e-07 9.72943745e-06 5.266275082e-05 1.053522087e-05 -4.549812061e-05 -2.423801213e-05 -3.11928838e-06 -3.9561964e-07 -4.097067e-08 -2.67848e-09 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -5.2e-13 -3.22e-12 -9.174e-11 -1.36949e-09 -1.14039e-09 5.778961e-08 9.2516025e-07 5.01145346e-06 9.1222982e-07 -4.56887504e-06 -2.06191627e-06 -2.4019976e-07 -2.995784e-08 -2.91654e-09 -9.48e-11 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 -6e-14 -3e-13 -3.25e-12 -6.422e-11 1.0899e-10 5.92543e-09 7.997296e-08 4.0541908e-07 6.47887e-08 -3.8675841e-07 -1.507329e-07 -1.65617e-08 -2.00472e-09 -6.73e-11 -3.06e-12 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -2e-14 -1e-13 1.4e-13 1.6e-12 2.8909e-10 5.53166e-09 2.958209e-08 3.83231e-09 -2.821985e-08 -9.97116e-09 -1.01878e-09 -2.377e-11 -1.56e-12 -2.6e-13 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -1e-14 1.1e-13 9.2e-13 1.43e-12 2.8934e-10 2.0374e-09 1.7924e-10 -1.91671e-09 -5.905e-10 -3.86e-12 -7e-14 -9e-14 -2e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 1e-13 2.9e-13 3.99e-12 8.751e-11 1.344e-11 -9.66e-11 -8.07e-12 -6e-14 -2e-14 -2e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 6e-14 1e-13 1.7e-13 4e-14 -2.9e-13 -7e-14 -3e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 1e-14 1e-14 -1e-14 -0.0 -0.0 -1e-14 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 -0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 0.0 0.0 0.0 0.0 0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 | |
| D/cons.3.00.000050.dat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -5e-14 -3.9e-13 -1.67e-12 -3.238e-11 -1.602e-09 -1.685203e-08 -1.3879433e-07 -1.20253567e-06 -2.48090056e-06 4.8065816e-07 2.6775886e-06 6.2848626e-07 5.263255e-08 1.95458e-09 -5.4975e-10 -4.098e-11 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -3e-14 -2.6e-13 -1.37e-12 -5.46e-11 -2.17158e-09 -2.666603e-08 -2.6752521e-07 -2.18830974e-06 -1.752609875e-05 -3.211928825e-05 7.66511127e-06 3.647123626e-05 7.51967815e-06 4.9668817e-07 -1.402219e-08 -1.067843e-08 -9.1592e-10 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -2e-14 -1.4e-13 -7.6e-13 -5.087e-11 -2.22457e-09 -3.063205e-08 -3.5114471e-07 -3.39969244e-06 -2.750061072e-05 -0.00018648476564 -0.00027706630843 9.528364695e-05 0.0003330373763 6.687850733e-05 3.2919989e-06 -1.1214102e-06 -2.2039798e-07 -1.580863e-08 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -6e-14 -3.4e-13 -2.576e-11 -1.8256e-09 -2.796068e-08 -3.5602536e-07 -3.91305267e-06 -3.657325698e-05 -0.00029593131579 -0.00179746420834 -0.00240061641215 0.00159538828075 0.00252336445901 0.00038246941389 4.94108514e-06 -2.013154284e-05 -3.59999296e-06 -2.4640266e-07 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -6e-14 -4.5e-13 -6.37e-12 -1.30309e-09 -2.128751e-08 -3.015354e-07 -3.69717504e-06 -3.81679221e-05 -0.00030819093772 -0.00264159178646 -0.00918696605728 -0.00894502422236 0.00790304543832 0.01264497615098 0.00118139905033 -0.00068990604601 -0.00022691359074 -2.794972064e-05 -2.31274576e-06 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -7e-14 -3e-13 -4.2078e-10 -8.27975e-09 -1.2572284e-07 -1.67158865e-06 -1.902987943e-05 -0.00018185379269 -0.00137772693871 -0.01213731773214 -0.0400116682209 -0.05298121495587 -0.1008045121587 -0.03345851742991 -0.0121939798663 -0.00569391521925 -0.00093531569062 -0.00012955329572 -1.138967415e-05 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -9e-14 -6.51e-12 -1.53741e-09 -2.59862e-08 -3.863481e-07 -4.99058985e-06 -5.467985609e-05 -0.00049818810497 -0.0037464779014 -0.02621611747289 -0.07038919342009 -0.10367757133706 -0.09668251912213 -0.09669098460091 -0.0442989083387 -0.02338333186679 -0.00450880022162 -0.00052879220748 -4.894860962e-05 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -7e-14 -2.8e-13 -4.6039e-10 -8.7796e-09 -1.2999687e-07 -1.67723378e-06 -1.840823211e-05 -0.00016761865249 -0.00130118300606 -0.00833918749171 -0.01798631430591 -0.04032600790647 -0.09705762249572 -0.04967291675366 -0.03372943932347 -0.01048973693236 -0.00142846475316 -0.0001999938414 -1.738735182e-05 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -6e-14 -4.8e-13 -1.298e-11 -1.56446e-09 -2.48467e-08 -3.3932365e-07 -3.9442354e-06 -3.726442649e-05 -0.00029614138144 -0.00115164241236 0.00100102189618 0.01253987189686 0.00539062607243 -0.00901288704113 -0.00876673598871 -0.0026601678922 -0.00033376428405 -4.401176322e-05 -3.49810397e-06 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -8e-14 -4.6e-13 -3.392e-11 -2.01121e-09 -2.84468e-08 -3.2460469e-07 -3.08366754e-06 -2.646324632e-05 -5.620222336e-05 0.00042622892921 0.00311887118677 0.00194634041412 -0.00283639975938 -0.00209858305702 -0.00035444153341 -4.528772689e-05 -5.06617314e-06 -3.7266019e-07 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -3e-14 -1.7e-13 -8.1e-13 -7.562e-11 -2.69897e-09 -3.5334e-08 -4.0434199e-07 -2.3295561e-06 2.16260429e-06 7.769877415e-05 0.00044554496059 0.00013680966491 -0.00037923180177 -0.00023783432061 -3.549254234e-05 -4.57215344e-06 -4.9449213e-07 -3.457382e-08 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -4e-14 -2.9e-13 -2.1e-12 -5.735e-11 -2.09036e-09 -2.262854e-08 -1.1067959e-07 5.1652494e-07 9.72943745e-06 5.266275082e-05 1.053522087e-05 -4.549812061e-05 -2.423801213e-05 -3.11928838e-06 -3.9561964e-07 -4.097067e-08 -2.67848e-09 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -6e-14 -5.2e-13 -3.22e-12 -9.174e-11 -1.36949e-09 -1.14039e-09 5.778961e-08 9.2516025e-07 5.01145346e-06 9.1222982e-07 -4.56887504e-06 -2.06191627e-06 -2.4019976e-07 -2.995784e-08 -2.91654e-09 -9.48e-11 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 -6e-14 -3e-13 -3.25e-12 -6.422e-11 1.0899e-10 5.92543e-09 7.997296e-08 4.0541908e-07 6.47887e-08 -3.8675841e-07 -1.507329e-07 -1.65617e-08 -2.00472e-09 -6.73e-11 -3.06e-12 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -2e-14 -1e-13 1.4e-13 1.6e-12 2.8909e-10 5.53166e-09 2.958209e-08 3.83231e-09 -2.821985e-08 -9.97116e-09 -1.01878e-09 -2.377e-11 -1.56e-12 -2.6e-13 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -1e-14 1.1e-13 9.2e-13 1.43e-12 2.8934e-10 2.0374e-09 1.7924e-10 -1.91671e-09 -5.905e-10 -3.86e-12 -7e-14 -9e-14 -2e-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1e-13 2.9e-13 3.99e-12 8.751e-11 1.344e-11 -9.66e-11 -8.07e-12 -6e-14 -2e-14 -2e-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6e-14 1e-13 1.7e-13 4e-14 -2.9e-13 -7e-14 -3e-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1e-14 1e-14 -1e-14 0.0 0.0 -1e-14 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -0.0 |
docs/documentation/case.md
Outdated
| | `model_threshold` | Real | Ray fraction inside the model patch above which the fraction is set to one.| | ||
| | `moving_ibm` | Integer | Sets the method used for IB movement. | | ||
| | `vel(i)` | Real | Initial velocity of the moving IB in the i-th direction. | | ||
| | `angular_vel(i)` | Real | Initial angular velcoity of the moving IB in the i-th direction. | |
There was a problem hiding this comment.
Suggestion: Corrects several typos and grammatical errors in the documentation for moving immersed boundaries to improve clarity. [general, importance: 3]
![codeant-ai[bot]](https://avatars.githubusercontent.com/in/646884?s=60&v=4){kind=small}
| @@ -0,0 +1,105 @@ | |||
| import json | |||
| import math | |||
There was a problem hiding this comment.
Suggestion: Unused import: math is imported but never used anywhere; keep dead imports out of the codebase to avoid confusion and maintain clean dependencies — remove the import. [possible bug]
Severity Level: Critical 🚨
| import math | |
| # removed unused import 'math' |
Why it matters? ⭐
The module imports math but never uses it anywhere in the current PR diff. Removing the import removes dead code and linter warnings and has no functional impact.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** examples/2D_mibm_cylinder_in_cross_flow/case.py
**Line:** 2:2
**Comment:**
*Possible Bug: Unused import: `math` is imported but never used anywhere; keep dead imports out of the codebase to avoid confusion and maintain clean dependencies — remove the import.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.| import json | ||
| import math | ||
|
|
||
| Mu = 1.84e-05 |
There was a problem hiding this comment.
Suggestion: Unused variable: Mu is assigned but never referenced; this is dead code and can be removed to avoid misleading readers about intended viscous computations. [possible bug]
Severity Level: Critical 🚨
| Mu = 1.84e-05 | |
| # removed unused variable 'Mu' |
Why it matters? ⭐
The variable Mu is defined but not referenced anywhere in the file after the PR. It's dead code and should be removed to avoid confusion; this is a harmless cleanup that doesn't change behavior.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** examples/2D_mibm_cylinder_in_cross_flow/case.py
**Line:** 4:4
**Comment:**
*Possible Bug: Unused variable: `Mu` is assigned but never referenced; this is dead code and can be removed to avoid misleading readers about intended viscous computations.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.| gam_a = 1.4 | ||
|
|
||
| # Configuring case dictionary | ||
| print( |
There was a problem hiding this comment.
Suggestion: Module-level side effect: the code prints the JSON at import time which causes unexpected output whenever this module is imported; wrap the printing in an if __name__ == "__main__": guard so the JSON is only printed when the script is executed directly. [logic error]
Severity Level: Minor
| print( | |
| if __name__ == "__main__": |
Why it matters? ⭐
The module currently performs a print at import time which is a side effect and can surprise callers who import this module. Wrapping the print in if name == "main": is the conventional, low-risk fix to ensure the JSON is only emitted when the script is executed directly.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** examples/2D_mibm_cylinder_in_cross_flow/case.py
**Line:** 8:8
**Comment:**
*Logic Error: Module-level side effect: the code prints the JSON at import time which causes unexpected output whenever this module is imported; wrap the printing in an `if __name__ == "__main__":` guard so the JSON is only printed when the script is executed directly.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.| !! Patch conditions for moving imersed boundaries | ||
| integer :: moving_ibm ! 0 for no moving, 1 for moving, 2 for moving on forced path | ||
| real(wp) :: mass, moment ! mass and moment of inertia of object used to compute forces in 2-way coupling | ||
| real(wp), dimension(1:3) :: force, torque ! vectors for the computed force and torque values applied to an IB |
There was a problem hiding this comment.
Suggestion: The new force and torque arrays use an explicit lower bound notation dimension(1:3) while many other 3-component vectors in this file use dimension(3) or L(3)/R(3) notation; this inconsistency can confuse callers and increase the risk of indexing mistakes. Use the canonical dimension(3) form for 3-element vectors to match the rest of the module and avoid subtle off-by-bound assumptions. [logic error]
Severity Level: Minor
| real(wp), dimension(1:3) :: force, torque ! vectors for the computed force and torque values applied to an IB | |
| real(wp), dimension(3) :: force, torque ! vectors for the computed force and torque values applied to an IB |
Why it matters? ⭐
Changing dimension(1:3) to dimension(3) makes the declaration consistent with other 3-component vectors in the module and avoids accidental reliance on a non-unit lower bound. It's a cosmetic/clarity improvement with no behavioral impact.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/common/m_derived_types.fpp
**Line:** 336:336
**Comment:**
*Logic Error: The new `force` and `torque` arrays use an explicit lower bound notation `dimension(1:3)` while many other 3-component vectors in this file use `dimension(3)` or `L(3)/R(3)` notation; this inconsistency can confuse callers and increase the risk of indexing mistakes. Use the canonical `dimension(3)` form for 3-element vectors to match the rest of the module and avoid subtle off-by-bound assumptions.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.| integer, intent(in) :: num_vectors, vector_length | ||
| real(wp), dimension(:, :), intent(in) :: var_loc | ||
| real(wp), dimension(:, :), intent(out) :: var_glb |
There was a problem hiding this comment.
Suggestion: Intent mismatch for in-place reduction: var_glb is declared with INTENT(OUT) but the subroutine uses MPI_IN_PLACE when the caller passes the same buffer for send and receive; INTENT(OUT) allows the compiler to undefine or reallocate the actual argument before the call which makes in-place MPI operations unsafe. Change var_glb to INTENT(INOUT) so the caller's data is preserved on entry for MPI_IN_PLACE use. [logic error]
Severity Level: Minor
| integer, intent(in) :: num_vectors, vector_length | |
| real(wp), dimension(:, :), intent(in) :: var_loc | |
| real(wp), dimension(:, :), intent(out) :: var_glb | |
| real(wp), dimension(:, :), intent(inout) :: var_glb |
Why it matters? ⭐
This is a valid correctness issue. INTENT(OUT) permits the compiler to treat the actual argument as undefined on entry (and even reallocate/erase it), which makes using MPI_IN_PLACE unsafe when the caller passes the same buffer for send and receive. Changing var_glb to INTENT(INOUT) accurately documents and guarantees the value is preserved on entry and is the correct contract for potential in-place MPI calls.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/common/m_mpi_common.fpp
**Line:** 479:481
**Comment:**
*Logic Error: Intent mismatch for in-place reduction: `var_glb` is declared with INTENT(OUT) but the subroutine uses MPI_IN_PLACE when the caller passes the same buffer for send and receive; INTENT(OUT) allows the compiler to undefine or reallocate the actual argument before the call which makes in-place MPI operations unsafe. Change `var_glb` to INTENT(INOUT) so the caller's data is preserved on entry for MPI_IN_PLACE use.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
src/simulation/m_ibm.fpp
Outdated
| q_prim_vf(E_idx)%sf(j, k, l) = pres_IP | ||
| else | ||
| ! TODO :: improve for two fluid | ||
| q_prim_vf(E_idx)%sf(j, k, l) = pres_IP / (1._wp - 2._wp*abs(levelset%sf(j,k,l,patch_id)*alpha_rho_IP(q)/pres_IP) * dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j,k,l,patch_id,:))) |
There was a problem hiding this comment.
Suggestion: Division-by-zero risk when computing modified pressure for moving immersed boundaries: the formula divides by pres_IP (and later by the computed denominator) with no guard, so if pres_IP == 0 (or the denominator becomes 0) this will raise a runtime error or produce NaNs; guard the division and handle the zero/near-zero case before dividing. [possible bug]
Severity Level: Critical 🚨
| q_prim_vf(E_idx)%sf(j, k, l) = pres_IP / (1._wp - 2._wp*abs(levelset%sf(j,k,l,patch_id)*alpha_rho_IP(q)/pres_IP) * dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j,k,l,patch_id,:))) | |
| if (pres_IP == 0._wp) then | |
| ! protect against division by zero: fall back to the interpolated pressure | |
| q_prim_vf(E_idx)%sf(j, k, l) = pres_IP | |
| else | |
| buf = 1._wp - 2._wp*abs(levelset%sf(j,k,l,patch_id)*alpha_rho_IP(q)/pres_IP) * dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j,k,l,patch_id,:)) | |
| if (abs(buf) < 1.e-30_wp) then | |
| ! denominator too small, avoid unstable division | |
| q_prim_vf(E_idx)%sf(j, k, l) = pres_IP | |
| else | |
| q_prim_vf(E_idx)%sf(j, k, l) = pres_IP / buf | |
| end if | |
| end if |
Why it matters? ⭐
The current expression divides by pres_IP and a denominator that can be zero or very small, which can produce NaNs or runtime errors. Guarding the division (check pres_IP and the computed denominator) is a legitimate safety fix. The proposed improved code prevents unstable divisions and falls back to a safe value.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/simulation/m_ibm.fpp
**Line:** 268:268
**Comment:**
*Possible Bug: Division-by-zero risk when computing modified pressure for moving immersed boundaries: the formula divides by `pres_IP` (and later by the computed denominator) with no guard, so if `pres_IP == 0` (or the denominator becomes 0) this will raise a runtime error or produce NaNs; guard the division and handle the zero/near-zero case before dividing.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
src/simulation/m_ibm.fpp
Outdated
| if (patch_ib(ib_marker)%geometry == 2) then ! circle | ||
| patch_ib(ib_marker)%moment = 0.5 * patch_ib(ib_marker)%mass * (patch_ib(ib_marker)%radius)**2 | ||
| elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle | ||
| patch_ib(ib_marker)%moment = patch_ib(i)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp |
There was a problem hiding this comment.
✅ Suggestion: Wrong variable used when computing analytic moment for rectangular geometry: the expression uses patch_ib(i)%mass but i is undefined here (should use the subroutine argument ib_marker); this will read a wrong/undefined mass value and produce incorrect result. [logic error]
Severity Level: Minor
| patch_ib(ib_marker)%moment = patch_ib(i)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp | |
| patch_ib(ib_marker)%moment = patch_ib(ib_marker)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp |
Why it matters? ⭐
The expression reads patch_ib(i)%mass but there is no loop variable i in this branch — the subroutine parameter is ib_marker. Using i is incorrect and will fetch the wrong struct/mass. Replacing with patch_ib(ib_marker)%mass fixes an obvious logic bug.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/simulation/m_ibm.fpp
**Line:** 1090:1090
**Comment:**
*Logic Error: Wrong variable used when computing analytic moment for rectangular geometry: the expression uses `patch_ib(i)%mass` but `i` is undefined here (should use the subroutine argument `ib_marker`); this will read a wrong/undefined mass value and produce incorrect result.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
src/simulation/m_ibm.fpp
Outdated
| $:GPU_PARALLEL_LOOP(private='[position,closest_point_along_axis,vector_to_axis,distance_to_axis]', copy='[moment,count]', copyin='[ib_marker,cell_volume,axis]', collapse=3) | ||
| do i = 0, m | ||
| do j = 0, j |
There was a problem hiding this comment.
✅ Suggestion: Loop bound typo in the moment-of-inertia grid scan: the inner loop is written do j = 0, j which uses the loop variable j as its own bound and is almost certainly incorrect (will either be a no-op or undefined behavior); it should use the grid size (e.g., n) instead. [logic error]
Severity Level: Minor
| do j = 0, j | |
| do j = 0, n |
Why it matters? ⭐
The inner loop bound "do j = 0, j" is clearly a typo and will not iterate over the intended grid (uses the loop variable as its own bound). Changing it to "0, n" (or appropriate grid size) fixes a logic bug that would otherwise make the loop a no-op or unpredictable.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/simulation/m_ibm.fpp
**Line:** 1105:1105
**Comment:**
*Logic Error: Loop bound typo in the moment-of-inertia grid scan: the inner loop is written `do j = 0, j` which uses the loop variable `j` as its own bound and is almost certainly incorrect (will either be a no-op or undefined behavior); it should use the grid size (e.g., `n`) instead.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.| if (ib) then | ||
| ! check if any IBMS are moving, and if so, update the markers, ghost points, levelsets, and levelset norms | ||
| if (moving_immersed_boundary_flag) then | ||
| call s_compute_ib_forces(q_prim_vf(E_idx)%sf(0:m, 0:n, 0:p)) ! compute the force and torque on the IB from the fluid |
There was a problem hiding this comment.
Suggestion: Unsafe use of q_prim_vf when computing IB forces: q_prim_vf is only allocated in initialization when .not. igr. Calling s_compute_ib_forces(q_prim_vf(E_idx)%sf(...)) without guarding for igr can dereference an unallocated array and crash when igr is true. Wrap the call in a guard so it's only executed when q_prim_vf is available. [possible bug]
Severity Level: Critical 🚨
| call s_compute_ib_forces(q_prim_vf(E_idx)%sf(0:m, 0:n, 0:p)) ! compute the force and torque on the IB from the fluid | |
| if (.not. igr) then | |
| call s_compute_ib_forces(q_prim_vf(E_idx)%sf(0:m, 0:n, 0:p)) ! compute the force and torque on the IB from the fluid | |
| end if |
Why it matters? ⭐
This is a valid runtime-safety concern: q_prim_vf is only allocated in initialization when (.not. igr).
The PR added an unconditional call that will dereference q_prim_vf when igr is true, which can crash.
Guarding the call is the right minimal fix — though note the IB block later still passes q_prim_vf into s_ibm_correct_state
unconditionally, so fixing just this line avoids one crash but doesn't fully resolve the broader inconsistency.
The suggestion is correct and actionable.
Prompt for AI Agent 🤖
This is a comment left during a code review.
**Path:** src/simulation/m_time_steppers.fpp
**Line:** 614:614
**Comment:**
*Possible Bug: Unsafe use of `q_prim_vf` when computing IB forces: `q_prim_vf` is only allocated in initialization when `.not. igr`. Calling `s_compute_ib_forces(q_prim_vf(E_idx)%sf(...))` without guarding for `igr` can dereference an unallocated array and crash when `igr` is true. Wrap the call in a guard so it's only executed when `q_prim_vf` is available.
Validate the correctness of the flagged issue. If correct, How can I resolve this? If you propose a fix, implement it and please make it concise.
|
CodeAnt AI finished reviewing your PR. |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 7
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
toolchain/bootstrap/modules.sh (1)
38-48: Fix interactive selection guard and keep HiPerGator option consistent with help/prompt
- Line 39:
if [ -v $u_c ]; thenis almost certainly wrong.-vexpects a variable name, not its value, and this will misbehave (or error) whenu_cis unset or holds a system code. If the goal is "prompt only when-c/--computerwas not supplied", use something like:if [ -z "${u_c+x}" ]; then # or simply: if [ -z "$u_c" ]; then ... fi
- Lines 11–14 vs 41–48: the new Florida/HiPerGator
(h)choice is in the interactive menu but not in theshow_help“Options” list. Please addFlorida (h)there so CLI help matches the interactive options.- Line 48: the colorized prompt string has a lot of adjacent quoted segments (
"$G""a$W"...), which triggers SC2027. Behavior is fine, but consider simplifying for readability and linters, e.g.:log_n "(${G}a${W}/${G}f${W}/${G}s${W}/${G}w${W}/${C}b${W}/${C}e${CR}/.../${OR}h${CR}): "
🧹 Nitpick comments (5)
src/common/m_derived_types.fpp (1)
335-336: Consider adding documentation comments for new fields.The new fields
mass,moment,force, andtorquelack documentation. Following the codebase style (e.g.,!< Description), consider adding brief descriptions:!! Patch conditions for moving imersed boundaries integer :: moving_ibm ! 0 for no moving, 1 for moving, 2 for moving on forced path -real(wp) :: mass, moment ! mass and moment of inertia of object used to compute forces in 2-way coupling -real(wp), dimension(1:3) :: force, torque ! vectors for the computed force and torque values applied to an IB +real(wp) :: mass !< Mass of IB object for two-way coupling +real(wp) :: moment !< Moment of inertia about rotation axis for two-way coupling +real(wp), dimension(1:3) :: force !< Computed pressure force vector on IB +real(wp), dimension(1:3) :: torque !< Computed torque vector on IB (in local coordinates)src/simulation/m_ibm.fpp (1)
199-212: Missing GPU parallelization for interior pressure loop.This triple-nested loop over the grid is not wrapped with GPU parallel directives, unlike similar loops in this file. For large grids, this could cause performance issues.
! set the Moving IBM interior Pressure Values do patch_id = 1, num_ibs if (patch_ib(patch_id)%moving_ibm == 2) then + $:GPU_PARALLEL_LOOP(collapse=3, private='[j,k,l]') do j = 0, m do k = 0, n do l = 0, p if (ib_markers%sf(j, k, l) == patch_id) then q_prim_vf(E_idx)%sf(j, k, l) = 1._wp end if end do end do end do + $:END_GPU_PARALLEL_LOOP() end if end doexamples/2D_mibm_cylinder_in_cross_flow/case.py (1)
1-103: Case setup matches intended 2D moving-cylinder demo; only minor cleanups
- The dictionary entries for the domain, single fluid, rectangular IC patch, and circular
patch_ib(1)(withmoving_ibm = 2, downwardvel(2), andmass) are consistent with a 2D cylinder-in-cross-flow example and with the new IBM mass field.Muand themathimport are currently unused; you can drop them or, if desired, computeRefromMuto keep the case self-consistent.- Please double-check that
moving_ibm = 2is the correct flag for the new two-way pressure-driven coupling mode, and that this value is documented alongside the other options indocs/documentation/case.md.src/common/m_mpi_common.fpp (1)
475-497: Add shape assertions and clarify alias handling ins_mpi_allreduce_vectors_sumThe MPI call pattern is fine, but two improvements would make this safer and more self-documenting:
- Add Fypp
ASSERTchecks so misuse is caught early, e.g.:@:ASSERT(size(var_loc, 1) == num_vectors .and. size(var_loc, 2) == vector_length, & 's_mpi_allreduce_vectors_sum: shape/length mismatch for var_loc') @:ASSERT(size(var_glb, 1) == num_vectors .and. size(var_glb, 2) == vector_length, & 's_mpi_allreduce_vectors_sum: shape/length mismatch for var_glb')This guards against passing slices or mismatched dimensions while still using packed
num_vectors*vector_lengthin the MPI call.
- The
loc(var_loc) == loc(var_glb)in-place detection relies on the non-standardlocintrinsic and on aliasing two distinct dummy arguments, which is technically non-conforming Fortran. If you don’t strictly need in-place semantics here, consider requiring distinct arrays and dropping the alias path; otherwise, document that this routine assumes compilers withlocand acceptscall s_mpi_allreduce_vectors_sum(A, A, ...)as a deliberate extension.docs/documentation/case.md (1)
378-381: Clean up table formatting (stray.column, hard tabs, and missing trailing pipes)Several of the modified tables now have small formatting issues that markdownlint is flagging:
- Simulation algorithm table header and first row (around lines 378–381) include an extra
.in the first column (| Parameter . | ...and| `bc_[x,y,z]%%beg[end]` . | ...).- The formatted-output and ensemble-averaged bubble model tables (lines ~578–615 and ~781–798), and the patch types table (lines ~1038–1059) still contain hard tabs and at least one line missing a trailing pipe, which triggers MD010/MD055/MD037.
All of this is cosmetic, but it’s easy to fix by:
- Removing the stray
.tokens from the first column,- Replacing tabs with spaces, and
- Ensuring each table row has a consistent set of leading and trailing
|characters.This will keep the docs rendering cleanly and quiet markdownlint.
Also applies to: 578-615, 781-798, 1038-1059
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (13)
docs/documentation/case.md(7 hunks)examples/2D_mibm_cylinder_in_cross_flow/case.py(1 hunks)src/common/m_derived_types.fpp(1 hunks)src/common/m_mpi_common.fpp(1 hunks)src/pre_process/m_global_parameters.fpp(1 hunks)src/simulation/m_ibm.fpp(4 hunks)src/simulation/m_mpi_proxy.fpp(1 hunks)src/simulation/m_time_steppers.fpp(2 hunks)tests/127A967A/golden-metadata.txt(1 hunks)tests/127A967A/golden.txt(1 hunks)toolchain/bootstrap/modules.sh(1 hunks)toolchain/mfc/run/case_dicts.py(2 hunks)toolchain/util.sh(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
**/*.{fpp,f90}: Use 2-space indentation; continuation lines align beneath &
Use lower-case keywords and intrinsics (do, end subroutine, etc.)
Name modules with m_ pattern (e.g., m_transport)
Name public subroutines with s_ pattern (e.g., s_compute_flux)
Name public functions with f_ pattern
Keep subroutine size ≤ 500 lines, helper subroutines ≤ 150 lines, functions ≤ 100 lines, files ≤ 1000 lines
Limit routine arguments to ≤ 6; use derived-type params struct if more are needed
Forbid goto statements (except in legacy code), COMMON blocks, and save globals
Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Call s_mpi_abort() for errors, never use stop or error stop
**/*.{fpp,f90}: Indent 2 spaces; continuation lines align under&
Use lower-case keywords and intrinsics (do,end subroutine, etc.)
Name modules withm_<feature>prefix (e.g.,m_transport)
Name public subroutines ass_<verb>_<noun>(e.g.,s_compute_flux) and functions asf_<verb>_<noun>
Keep private helpers in the module; avoid nested procedures
Enforce size limits: subroutine ≤ 500 lines, helper ≤ 150, function ≤ 100, module/file ≤ 1000
Limit subroutines to ≤ 6 arguments; otherwise pass a derived-type 'params' struct
Avoidgotostatements (except unavoidable legacy); avoid global state (COMMON,save)
Every variable must haveintent(in|out|inout)specification and appropriatedimension/allocatable/pointer
Uses_mpi_abort(<msg>)for error termination instead ofstop
Use!>style documentation for header comments; follow Doxygen Fortran format with!! @paramand!! @returntags
Useimplicit nonestatement in all modules
Useprivatedeclaration followed by explicitpublicexports in modules
Use derived types with pointers for encapsulation (e.g.,pointer, dimension(:,:,:) => null())
Usepureandelementalattributes for side-effect-free functions; combine them for array ...
Files:
src/simulation/m_mpi_proxy.fppsrc/simulation/m_time_steppers.fppsrc/pre_process/m_global_parameters.fppsrc/common/m_mpi_common.fppsrc/common/m_derived_types.fppsrc/simulation/m_ibm.fpp
src/simulation/**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
src/simulation/**/*.{fpp,f90}: Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Avoid stop/error stop inside GPU device code
Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
src/simulation/**/*.{fpp,f90}: Mark GPU-callable helpers with$:GPU_ROUTINE(function_name='...', parallelism='[seq]')immediately after declaration
Do not use OpenACC or OpenMP directives directly; use Fypp macros fromsrc/common/include/parallel_macros.fppinstead
Wrap tight loops with$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')macro; addcollapse=nfor safe nested loop merging
Declare loop-local variables withprivate='[...]'in GPU parallel loop macros
Allocate large arrays withmanagedor move them into a persistent$:GPU_ENTER_DATA(...)region at start-up
Do not placestoporerror stopinside device code
Files:
src/simulation/m_mpi_proxy.fppsrc/simulation/m_time_steppers.fppsrc/simulation/m_ibm.fpp
src/**/*.fpp
📄 CodeRabbit inference engine (.cursor/rules/mfc-agent-rules.mdc)
src/**/*.fpp: Use.fppfile extension for Fypp preprocessed files; CMake transpiles them to.f90
Start module files with Fypp include for macros:#:include 'macros.fpp'
Use the fyppASSERTmacro for validating conditions:@:ASSERT(predicate, message)
Use fypp macro@:ALLOCATE(var1, var2)for device-aware allocation instead of standard Fortranallocate
Use fypp macro@:DEALLOCATE(var1, var2)for device-aware deallocation instead of standard Fortrandeallocate
Files:
src/simulation/m_mpi_proxy.fppsrc/simulation/m_time_steppers.fppsrc/pre_process/m_global_parameters.fppsrc/common/m_mpi_common.fppsrc/common/m_derived_types.fppsrc/simulation/m_ibm.fpp
🧠 Learnings (13)
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Add and update focused tests for changes; ensure no regressions in golden test outputs without clear justification
Applied to files:
tests/127A967A/golden-metadata.txttests/127A967A/golden.txt
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Draft a step-by-step plan before making changes; build after each step using `./mfc.sh build -t pre_process simulation -j $(nproc)`
Applied to files:
tests/127A967A/golden-metadata.txt
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
Applied to files:
tests/127A967A/golden-metadata.txtsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Compile with Cray `ftn` or NVIDIA `nvfortran` for GPU offloading; also build CPU-only with GNU `gfortran` and Intel `ifx`/`ifort` for portability
Applied to files:
tests/127A967A/golden-metadata.txt
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Declare loop-local variables with `private='[...]'` in GPU parallel loop macros
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Use `private` declaration followed by explicit `public` exports in modules
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Do not use OpenACC or OpenMP directives directly; use Fypp macros from `src/common/include/parallel_macros.fpp` instead
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Use `pure` and `elemental` attributes for side-effect-free functions; combine them for array operations
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight loops with `$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')` macro; add `collapse=n` for safe nested loop merging
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Mark GPU-callable helpers with `$:GPU_ROUTINE(function_name='...', parallelism='[seq]')` immediately after declaration
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large arrays with `managed` or move them into a persistent `$:GPU_ENTER_DATA(...)` region at start-up
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Applied to files:
src/simulation/m_ibm.fpp
🧬 Code graph analysis (3)
toolchain/bootstrap/modules.sh (1)
toolchain/util.sh (2)
log(12-12)log_n(13-13)
examples/2D_mibm_cylinder_in_cross_flow/case.py (1)
toolchain/mfc/test/cases.py (5)
alter_ib(377-422)alter_2d(293-320)alter_mixlayer_perturb(685-716)alter_num_fluids(224-291)alter_grcbc(78-120)
src/simulation/m_ibm.fpp (1)
toolchain/mfc/test/cases.py (1)
alter_ib(377-422)
🪛 markdownlint-cli2 (0.18.1)
docs/documentation/case.md
385-385: Spaces inside emphasis markers
(MD037, no-space-in-emphasis)
386-386: Hard tabs
Column: 13
(MD010, no-hard-tabs)
387-387: Hard tabs
Column: 12
(MD010, no-hard-tabs)
394-394: Hard tabs
Column: 15
(MD010, no-hard-tabs)
395-395: Hard tabs
Column: 13
(MD010, no-hard-tabs)
396-396: Hard tabs
Column: 16
(MD010, no-hard-tabs)
397-397: Hard tabs
Column: 10
(MD010, no-hard-tabs)
408-408: Hard tabs
Column: 108
(MD010, no-hard-tabs)
409-409: Hard tabs
Column: 172
(MD010, no-hard-tabs)
410-410: Hard tabs
Column: 14
(MD010, no-hard-tabs)
411-411: Hard tabs
Column: 132
(MD010, no-hard-tabs)
411-411: Spaces inside emphasis markers
(MD037, no-space-in-emphasis)
412-412: Hard tabs
Column: 99
(MD010, no-hard-tabs)
413-413: Hard tabs
Column: 23
(MD010, no-hard-tabs)
580-580: Hard tabs
Column: 80
(MD010, no-hard-tabs)
581-581: Hard tabs
Column: 61
(MD010, no-hard-tabs)
582-582: Hard tabs
Column: 51
(MD010, no-hard-tabs)
585-585: Hard tabs
Column: 64
(MD010, no-hard-tabs)
586-586: Table pipe style
Expected: leading_and_trailing; Actual: leading_only; Missing trailing pipe
(MD055, table-pipe-style)
587-587: Hard tabs
Column: 78
(MD010, no-hard-tabs)
588-588: Hard tabs
Column: 85
(MD010, no-hard-tabs)
589-589: Hard tabs
Column: 85
(MD010, no-hard-tabs)
590-590: Hard tabs
Column: 75
(MD010, no-hard-tabs)
591-591: Hard tabs
Column: 71
(MD010, no-hard-tabs)
592-592: Hard tabs
Column: 91
(MD010, no-hard-tabs)
593-593: Hard tabs
Column: 91
(MD010, no-hard-tabs)
594-594: Hard tabs
Column: 82
(MD010, no-hard-tabs)
596-596: Hard tabs
Column: 89
(MD010, no-hard-tabs)
597-597: Hard tabs
Column: 74
(MD010, no-hard-tabs)
598-598: Hard tabs
Column: 86
(MD010, no-hard-tabs)
605-605: Hard tabs
Column: 97
(MD010, no-hard-tabs)
606-606: Hard tabs
Column: 55
(MD010, no-hard-tabs)
607-607: Hard tabs
Column: 33
(MD010, no-hard-tabs)
607-607: Hard tabs
Column: 62
(MD010, no-hard-tabs)
783-783: Hard tabs
Column: 30
(MD010, no-hard-tabs)
783-783: Hard tabs
Column: 66
(MD010, no-hard-tabs)
784-784: Hard tabs
Column: 30
(MD010, no-hard-tabs)
785-785: Hard tabs
Column: 13
(MD010, no-hard-tabs)
785-785: Hard tabs
Column: 28
(MD010, no-hard-tabs)
786-786: Hard tabs
Column: 11
(MD010, no-hard-tabs)
786-786: Hard tabs
Column: 24
(MD010, no-hard-tabs)
787-787: Hard tabs
Column: 30
(MD010, no-hard-tabs)
788-788: Hard tabs
Column: 8
(MD010, no-hard-tabs)
788-788: Hard tabs
Column: 28
(MD010, no-hard-tabs)
789-789: Hard tabs
Column: 16
(MD010, no-hard-tabs)
789-789: Hard tabs
Column: 28
(MD010, no-hard-tabs)
789-789: Hard tabs
Column: 31
(MD010, no-hard-tabs)
790-790: Hard tabs
Column: 8
(MD010, no-hard-tabs)
790-790: Hard tabs
Column: 25
(MD010, no-hard-tabs)
791-791: Hard tabs
Column: 9
(MD010, no-hard-tabs)
791-791: Hard tabs
Column: 24
(MD010, no-hard-tabs)
792-792: Hard tabs
Column: 12
(MD010, no-hard-tabs)
792-792: Hard tabs
Column: 26
(MD010, no-hard-tabs)
793-793: Hard tabs
Column: 10
(MD010, no-hard-tabs)
793-793: Hard tabs
Column: 30
(MD010, no-hard-tabs)
793-793: Hard tabs
Column: 32
(MD010, no-hard-tabs)
794-794: Hard tabs
Column: 15
(MD010, no-hard-tabs)
794-794: Hard tabs
Column: 29
(MD010, no-hard-tabs)
794-794: Hard tabs
Column: 31
(MD010, no-hard-tabs)
795-795: Hard tabs
Column: 10
(MD010, no-hard-tabs)
795-795: Hard tabs
Column: 28
(MD010, no-hard-tabs)
795-795: Hard tabs
Column: 31
(MD010, no-hard-tabs)
796-796: Hard tabs
Column: 10
(MD010, no-hard-tabs)
796-796: Hard tabs
Column: 28
(MD010, no-hard-tabs)
796-796: Hard tabs
Column: 31
(MD010, no-hard-tabs)
797-797: Hard tabs
Column: 10
(MD010, no-hard-tabs)
797-797: Hard tabs
Column: 28
(MD010, no-hard-tabs)
797-797: Hard tabs
Column: 31
(MD010, no-hard-tabs)
1040-1040: Hard tabs
Column: 23
(MD010, no-hard-tabs)
1041-1041: Hard tabs
Column: 17
(MD010, no-hard-tabs)
1042-1042: Hard tabs
Column: 20
(MD010, no-hard-tabs)
1043-1043: Hard tabs
Column: 21
(MD010, no-hard-tabs)
1044-1044: Hard tabs
Column: 18
(MD010, no-hard-tabs)
1045-1045: Hard tabs
Column: 14
(MD010, no-hard-tabs)
1046-1046: Hard tabs
Column: 21
(MD010, no-hard-tabs)
1047-1047: Hard tabs
Column: 17
(MD010, no-hard-tabs)
1048-1048: Hard tabs
Column: 17
(MD010, no-hard-tabs)
1049-1049: Hard tabs
Column: 19
(MD010, no-hard-tabs)
1050-1050: Hard tabs
Column: 22
(MD010, no-hard-tabs)
1051-1051: Hard tabs
Column: 20
(MD010, no-hard-tabs)
1052-1052: Hard tabs
Column: 21
(MD010, no-hard-tabs)
🪛 Shellcheck (0.11.0)
toolchain/bootstrap/modules.sh
[warning] 48-48: The surrounding quotes actually unquote this. Remove or escape them.
(SC2027)
[warning] 48-48: The surrounding quotes actually unquote this. Remove or escape them.
(SC2027)
toolchain/util.sh
[warning] 8-8: R appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 8-8: C appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 8-8: G appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 8-8: BR appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 9-9: Y appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 9-9: M appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 9-9: B appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 9-9: OR appears unused. Verify use (or export if used externally).
(SC2034)
[warning] 9-9: W appears unused. Verify use (or export if used externally).
(SC2034)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (9)
- GitHub Check: Code Cleanliness Check
- GitHub Check: Coverage Test on CodeCov
- GitHub Check: Github (ubuntu, mpi, no-debug, true)
- GitHub Check: Github (ubuntu, mpi, debug, true)
- GitHub Check: Github (ubuntu, no-mpi, single, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, debug, false)
- GitHub Check: cubic · AI code reviewer
- GitHub Check: Build & Publish
🔇 Additional comments (11)
src/simulation/m_mpi_proxy.fpp (1)
205-207: LGTM!The addition of
massto the MPI broadcast list is consistent with the new field added toib_patch_parametersinm_derived_types.fpp. The broadcast follows the existing pattern for scalarreal(wp)fields.toolchain/mfc/run/case_dicts.py (2)
118-118: LGTM!The addition of
massto thepatch_ibattribute mappings is correctly typed asParamType.REALand follows the existing pattern for other IB parameters.
367-367: LGTM!Symmetric addition of
massto the SIMULATION dictionary maintains consistency with PRE_PROCESS.src/simulation/m_time_steppers.fpp (2)
626-641: Approve the two-way coupling integration with a note on dependencies.The RK integration for velocity and angular velocity with force/torque coupling looks correct. However, this code depends on
s_compute_moment_of_inertiawhich has critical bugs (see m_ibm.fpp review). Once those are fixed, this integration should work correctly.Note: The angular momentum conservation approach (multiply by old moment, add torque impulse, recompute moment, divide by new moment) is physically sound for objects whose moment of inertia changes with rotation axis.
614-614: Verify force computation uses correct primitive variable.The call to
s_compute_ib_forcespassesq_prim_vf(E_idx)%sf(0:m, 0:n, 0:p). Confirm thatE_idxcorrectly indexes the pressure field in primitive variables (not energy as in conservative variables).src/simulation/m_ibm.fpp (2)
1133-1133: Invalid GPU update syntax for derived type field.The
$:GPU_UPDATEdirective withpatch_ib(ib_marker)%momentmay not work correctly. The entirepatch_ibarray should be updated instead.-$:GPU_UPDATE(device='[patch_ib(ib_marker)%moment]') +$:GPU_UPDATE(device='[patch_ib]')⛔ Skipped due to learnings
Learnt from: CR Repo: MFlowCode/MFC PR: 0 File: .cursor/rules/mfc-agent-rules.mdc:0-0 Timestamp: 2025-11-24T21:50:46.909Z Learning: Applies to src/simulation/**/*.{fpp,f90} : Mark GPU-callable helpers with `$:GPU_ROUTINE(function_name='...', parallelism='[seq]')` immediately after declarationLearnt from: CR Repo: MFlowCode/MFC PR: 0 File: .github/copilot-instructions.md:0-0 Timestamp: 2025-11-24T21:50:16.713Z Learning: Applies to src/simulation/**/*.{fpp,f90} : Avoid stop/error stop inside GPU device code
263-270: Redundant loop and unclear multi-fluid handling.The loop
do q = 1, num_fluidsiterates butqis never used in the loop body. The pressure is simply overwrittennum_fluidstimes with the same value. The TODO comment also suggests this needs attention.-! set the pressure -do q = 1, num_fluids - if (patch_ib(patch_id)%moving_ibm == 0) then - q_prim_vf(E_idx)%sf(j, k, l) = pres_IP - else - ! TODO :: improve for two fluid - q_prim_vf(E_idx)%sf(j, k, l) = pres_IP/(1._wp - 2._wp*abs(levelset%sf(j, k, l, patch_id)*alpha_rho_IP(q)/pres_IP)*dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j, k, l, patch_id, :))) - end if -end do +! set the pressure +if (patch_ib(patch_id)%moving_ibm == 0) then + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP +else + ! TODO :: improve for two fluid + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP/(1._wp - 2._wp*abs(levelset%sf(j, k, l, patch_id)*alpha_rho_IP(1)/pres_IP)*dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j, k, l, patch_id, :))) +end ifLikely an incorrect or invalid review comment.
toolchain/util.sh (1)
4-10: ORANGE/OR color additions look goodThe new
ORANGEescape andORalias integrate cleanly with the existing color scheme and are used by the updated module prompt without affecting existing behavior.tests/127A967A/golden-metadata.txt (1)
1-159: Golden metadata is consistent and helpful for reproducibilityThe metadata cleanly captures invocation, build configurations, and hardware details for test
127A967A, which aligns with the project’s guidance to document golden outputs for regressions tracking.tests/127A967A/golden.txt (1)
1-11: Data-only golden baseline looks fine; rely on tests to validateThis file is pure reference data (conservative fields and
ib_markers) for test127A967A. There’s nothing structurally concerning; just ensure CI re-runs this case and that changes to these values are intentional given the new IBM two-way coupling.src/pre_process/m_global_parameters.fpp (1)
534-565: Good: new IBMmassandmomentfields are now defaultedInitializing
patch_ib(i)%massandpatch_ib(i)%momenttodflt_realalongside the otherpatch_ibfields ins_assign_default_values_to_user_inputsavoids uninitialized state when these aren’t set in a case file and keeps the pre-process side consistent with the extendedib_patch_parameterstype.
| | Parameter | Type | Description | | ||
| | ---: | :----: | :--- | | ||
| | `geometry` | Integer | Geometry configuration of the patch.| | ||
| | `x[y,z]_centroid` | Real | Centroid of the applied geometry in the [x,y,z]-direction. | | ||
| | `length_x[y,z]` | Real | Length, if applicable, in the [x,y,z]-direction. | | ||
| | `radius` | Real | Radius, if applicable, of the applied geometry. | | ||
| | `theta` | Real | Angle of attach applied to airfoil IB patches | | ||
| | `c` | Real | NACA airfoil parameters (see below) | | ||
| | `t` | Real | NACA airfoil parameters (see below) | | ||
| | `m` | Real | NACA airfoil parameters (see below) | | ||
| | `p` | Real | NACA airfoil parameters (see below) | | ||
| | `slip` | Logical | Apply a slip boundary | | ||
| | `model_filepath` | String | Path to an STL or OBJ file (not all OBJs are supported). | | ||
| | `model_scale(i)` | Real | Model's (applied) scaling factor for component $i$. | | ||
| | `model_rotate(i)` | Real | Model's (applied) angle of rotation about axis $i$. | | ||
| | `model_translate(i)` | Real | Model's $i$-th component of (applied) translation. | | ||
| | `model_spc` | Integer | Number of samples per cell when discretizing the model into the grid. | | ||
| | `model_threshold` | Real | Ray fraction inside the model patch above which the fraction is set to one.| | ||
| | `geometry` | Integer | Geometry configuration of the patch.| | ||
| | `x[y,z]_centroid` | Real | Centroid of the applied geometry in the [x,y,z]-direction. | | ||
| | `length_x[y,z]` | Real | Length, if applicable, in the [x,y,z]-direction. | | ||
| | `radius` | Real | Radius, if applicable, of the applied geometry. | | ||
| | `theta` | Real | Angle of attach applied to airfoil IB patches | | ||
| | `c` | Real | NACA airfoil parameters (see below) | | ||
| | `t` | Real | NACA airfoil parameters (see below) | | ||
| | `m` | Real | NACA airfoil parameters (see below) | | ||
| | `p` | Real | NACA airfoil parameters (see below) | | ||
| | `slip` | Logical | Apply a slip boundary | | ||
| | `model_filepath` | String | Path to an STL or OBJ file (not all OBJs are supported). | | ||
| | `model_scale(i)` | Real | Model's (applied) scaling factor for component $i$. | | ||
| | `model_rotate(i)` | Real | Model's (applied) angle of rotation about axis $i$. | | ||
| | `model_translate(i)` | Real | Model's $i$-th component of (applied) translation. | | ||
| | `model_spc` | Integer | Number of samples per cell when discretizing the model into the grid. | | ||
| | `model_threshold` | Real | Ray fraction inside the model patch above which the fraction is set to one.| | ||
| | `moving_ibm` | Integer | Sets the method used for IB movement. | | ||
| | `vel(i)` | Real | Initial velocity of the moving IB in the i-th direction. | | ||
| | `angular_vel(i)` | Real | Initial angular velocity of the moving IB in the i-th direction. | |
There was a problem hiding this comment.
Document all moving_ibm modes and tighten the new IBM parameter descriptions
- The new
moving_ibm,vel(i), andangular_vel(i)rows are useful, but the text only explains modes 0 and 1. The code and new example use an additional value (e.g.,moving_ibm = 2) for two-way, pressure-driven motion. Please extend the description to cover all supported values (0 = fixed, 1 = prescribed/one-way, 2 = two-way coupling with fluid forces/torques) so users can correctly select the new behavior. - For clarity, you may also want to state explicitly that
vel(i)/angular_vel(i)are initial conditions, and that in mode 1 they remain constant, whereas in mode 2 they are updated by the time integrator based on the computed forces/torques. - Minor wording nits while you’re here: fix “applied forces to the fluid based upon it's own motion” → “applies forces to the fluid based upon its own motion”.
Also applies to: 337-341
| if (p == 0) then | ||
| axis = [0, 1, 0] | ||
| else if (sqrt(sum(axis**2)) == 0) then | ||
| ! if the object is not actually rotating at this time, return a dummy value and exit | ||
| patch_ib(ib_marker)%moment = 1._wp | ||
| return | ||
| else | ||
| axis = axis/sqrt(sum(axis)) | ||
| end if |
There was a problem hiding this comment.
Optional argument handling issue in 2D case.
When p == 0 (2D), the code assigns to axis without checking if it was provided. Since axis is declared optional, assigning to it when not present is undefined behavior. Additionally, the axis parameter should have intent(inout) if it's being modified.
-subroutine s_compute_moment_of_inertia(ib_marker, axis)
+subroutine s_compute_moment_of_inertia(ib_marker, axis_in)
- real(wp), dimension(3), optional :: axis !< the axis about which we compute the moment. Only required in 3D.
+ real(wp), dimension(3), intent(in), optional :: axis_in !< the axis about which we compute the moment. Only required in 3D.
integer, intent(in) :: ib_marker
real(wp) :: moment, distance_to_axis, cell_volume
- real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis
+ real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis, axis
integer :: i, j, k, count
if (p == 0) then
- axis = [0, 1, 0]
- else if (sqrt(sum(axis**2)) == 0) then
+ axis = [0._wp, 0._wp, 1._wp]
+ else if (.not. present(axis_in) .or. sqrt(sum(axis_in**2)) < 1.e-16_wp) then
! if the object is not actually rotating at this time, return a dummy value and exit
patch_ib(ib_marker)%moment = 1._wp
return
else
- axis = axis/sqrt(sum(axis))
+ axis = axis_in/sqrt(sum(axis_in**2))
end ifCommittable suggestion skipped: line range outside the PR's diff.
🤖 Prompt for AI Agents
In src/simulation/m_ibm.fpp around lines 1076 to 1084, the branch for p == 0
assigns to the optional argument axis (undefined if not present) and axis should
be declared intent(inout) if modified; fix by (a) guarding any write/read of
axis with present(axis) (e.g., only assign axis = [0,1,0] when present(axis) is
true), (b) avoid touching axis when not present (use a local dummy variable
instead), and (c) update the routine signature to declare axis as intent(inout)
if the caller is expected to receive the modified axis. Ensure other uses in
this block also check present(axis) before reading or normalizing it.
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
11 issues found across 13 files
Prompt for AI agents (all 11 issues)
Check if these issues are valid — if so, understand the root cause of each and fix them.
<file name="examples/2D_mibm_cylinder_in_cross_flow/case.py">
<violation number="1" location="examples/2D_mibm_cylinder_in_cross_flow/case.py:98">
P2: Incorrect comment: Comment says "z-axis rotation" but this line sets the `mass` parameter. This appears to be a copy-paste error.</violation>
</file>
<file name="docs/documentation/case.md">
<violation number="1" location="docs/documentation/case.md:312">
P2: Typo: 'velcoity' should be 'velocity'.</violation>
<violation number="2" location="docs/documentation/case.md:339">
P2: Typo: 'velcoity' should be 'velocity'.</violation>
<violation number="3" location="docs/documentation/case.md:341">
P2: Typo: 'velcoity' should be 'velocity'.</violation>
<violation number="4" location="docs/documentation/case.md:378">
P1: Stray period characters break markdown table formatting. The `. ` on a standalone line and appended to table headers will cause rendering issues.</violation>
<violation number="5" location="docs/documentation/case.md:781">
P1: Stray `. ` characters in table header row break markdown table formatting.</violation>
</file>
<file name="src/simulation/m_ibm.fpp">
<violation number="1" location="src/simulation/m_ibm.fpp:1025">
P1: Pressure gradient calculation uses cell center coordinate `x_cc(i)` instead of grid spacing. Central difference formula should divide by `2*dx` (grid spacing), not the coordinate value `x_cc(i)`.</violation>
<violation number="2" location="src/simulation/m_ibm.fpp:1054">
P2: Debug print statement should be removed from production code.</violation>
<violation number="3" location="src/simulation/m_ibm.fpp:1090">
P1: Wrong variable used: `patch_ib(i)%mass` should be `patch_ib(ib_marker)%mass`. Variable `i` is undefined in this context.</violation>
<violation number="4" location="src/simulation/m_ibm.fpp:1105">
P0: Loop variable `j` is used as its own upper bound. This should be `do j = 0, n` to iterate over the y-dimension grid.</violation>
<violation number="5" location="src/simulation/m_ibm.fpp:1121">
P1: Incorrect distance calculation: `sum(vector_to_axis)` sums components, not computing squared distance. Should be `sum(vector_to_axis**2)` per the comment.</violation>
</file>
Since this is your first cubic review, here's how it works:
- cubic automatically reviews your code and comments on bugs and improvements
- Teach cubic by replying to its comments. cubic learns from your replies and gets better over time
- Ask questions if you need clarification on any suggestion
Reply to cubic to teach it or ask questions. Re-run a review with @cubic-dev-ai review this PR
| "patch_ib(1)%angular_vel(1)": 0.0, # x-axis rotational velocity in radians per second | ||
| "patch_ib(1)%angular_vel(2)": 0.0, # y-axis rotation | ||
| "patch_ib(1)%angular_vel(3)": 0.0, # z-axis rotation | ||
| "patch_ib(1)%mass": 0.0001, # z-axis rotation |
There was a problem hiding this comment.
P2: Incorrect comment: Comment says "z-axis rotation" but this line sets the mass parameter. This appears to be a copy-paste error.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At examples/2D_mibm_cylinder_in_cross_flow/case.py, line 98:
<comment>Incorrect comment: Comment says "z-axis rotation" but this line sets the `mass` parameter. This appears to be a copy-paste error.</comment>
<file context>
@@ -0,0 +1,105 @@
+ "patch_ib(1)%angular_vel(1)": 0.0, # x-axis rotational velocity in radians per second
+ "patch_ib(1)%angular_vel(2)": 0.0, # y-axis rotation
+ "patch_ib(1)%angular_vel(3)": 0.0, # z-axis rotation
+ "patch_ib(1)%mass": 0.0001, # z-axis rotation
+ # Fluids Physical Parameters
+ "fluid_pp(1)%gamma": 1.0e00 / (gam_a - 1.0e00), # 2.50(Not 1.40)
</file context>
| "patch_ib(1)%mass": 0.0001, # z-axis rotation | |
| "patch_ib(1)%mass": 0.0001, # mass of the immersed boundary object |
docs/documentation/case.md
Outdated
|
|
||
| - `vel(i)` is the initial linear velocity of the IB in the x, y, z direction for i=1, 2, 3. When `moving_ibm` equals 1, this velcoity is constant. | ||
|
|
||
| - `angular_vel(i)` is the initial angular velocity of the IB about the x, y, z axes for i=1, 2, 3 in radians per second. When `moving_ibm` equals 1, this angular velcoity is constant. |
There was a problem hiding this comment.
P2: Typo: 'velcoity' should be 'velocity'.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At docs/documentation/case.md, line 341:
<comment>Typo: 'velcoity' should be 'velocity'.</comment>
<file context>
@@ -331,6 +334,12 @@ Additional details on this specification can be found in [The Naca Airfoil Serie
+
+- `vel(i)` is the initial linear velocity of the IB in the x, y, z direction for i=1, 2, 3. When `moving_ibm` equals 1, this velcoity is constant.
+
+- `angular_vel(i)` is the initial angular velocity of the IB about the x, y, z axes for i=1, 2, 3 in radians per second. When `moving_ibm` equals 1, this angular velcoity is constant.
+
### 5. Fluid Material’s
</file context>
| - `angular_vel(i)` is the initial angular velocity of the IB about the x, y, z axes for i=1, 2, 3 in radians per second. When `moving_ibm` equals 1, this angular velcoity is constant. | |
| - `angular_vel(i)` is the initial angular velocity of the IB about the x, y, z axes for i=1, 2, 3 in radians per second. When `moving_ibm` equals 1, this angular velocity is constant. |
docs/documentation/case.md
Outdated
| | `model_threshold` | Real | Ray fraction inside the model patch above which the fraction is set to one.| | ||
| | `moving_ibm` | Integer | Sets the method used for IB movement. | | ||
| | `vel(i)` | Real | Initial velocity of the moving IB in the i-th direction. | | ||
| | `angular_vel(i)` | Real | Initial angular velcoity of the moving IB in the i-th direction. | |
There was a problem hiding this comment.
P2: Typo: 'velcoity' should be 'velocity'.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At docs/documentation/case.md, line 312:
<comment>Typo: 'velcoity' should be 'velocity'.</comment>
<file context>
@@ -291,22 +291,25 @@ This is enabled by adding `'elliptic_smoothing': "T",` and `'elliptic_smoothing_
+| `model_threshold` | Real | Ray fraction inside the model patch above which the fraction is set to one.|
+| `moving_ibm` | Integer | Sets the method used for IB movement. |
+| `vel(i)` | Real | Initial velocity of the moving IB in the i-th direction. |
+| `angular_vel(i)` | Real | Initial angular velcoity of the moving IB in the i-th direction. |
These parameters should be prepended with `patch_ib(j)%` where $j$ is the patch index.
</file context>
| | `angular_vel(i)` | Real | Initial angular velcoity of the moving IB in the i-th direction. | | |
| | `angular_vel(i)` | Real | Initial angular velocity of the moving IB in the i-th direction. | |
docs/documentation/case.md
Outdated
|
|
||
| - `moving_ibm` sets the method by which movement will be applied to the immersed boundary. Using 0 will result in no movement. Using 1 will result 1-way coupling where the boundary moves at a constant rate and applied forces to the fluid based upon it's own motion. In 1-way coupling, the fluid does not apply forces back onto the IB. | ||
|
|
||
| - `vel(i)` is the initial linear velocity of the IB in the x, y, z direction for i=1, 2, 3. When `moving_ibm` equals 1, this velcoity is constant. |
There was a problem hiding this comment.
P2: Typo: 'velcoity' should be 'velocity'.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At docs/documentation/case.md, line 339:
<comment>Typo: 'velcoity' should be 'velocity'.</comment>
<file context>
@@ -331,6 +334,12 @@ Additional details on this specification can be found in [The Naca Airfoil Serie
+- `moving_ibm` sets the method by which movement will be applied to the immersed boundary. Using 0 will result in no movement. Using 1 will result 1-way coupling where the boundary moves at a constant rate and applied forces to the fluid based upon it's own motion. In 1-way coupling, the fluid does not apply forces back onto the IB.
+
+- `vel(i)` is the initial linear velocity of the IB in the x, y, z direction for i=1, 2, 3. When `moving_ibm` equals 1, this velcoity is constant.
+
+- `angular_vel(i)` is the initial angular velocity of the IB about the x, y, z axes for i=1, 2, 3 in radians per second. When `moving_ibm` equals 1, this angular velcoity is constant.
</file context>
| - `vel(i)` is the initial linear velocity of the IB in the x, y, z direction for i=1, 2, 3. When `moving_ibm` equals 1, this velcoity is constant. | |
| - `vel(i)` is the initial linear velocity of the IB in the x, y, z direction for i=1, 2, 3. When `moving_ibm` equals 1, this velocity is constant. |
| | Parameter | Type | Description | | ||
| | ---: | :----: | :--- | | ||
| | `bubbles_euler`| Logical | Ensemble-averaged bubble modeling | | ||
| | Parameter . | Type | Description | |
There was a problem hiding this comment.
P1: Stray . characters in table header row break markdown table formatting.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At docs/documentation/case.md, line 781:
<comment>Stray `. ` characters in table header row break markdown table formatting.</comment>
<file context>
@@ -769,23 +778,23 @@ Implementation of the parameters into the model follow [Ando (2010)](references.
-| Parameter | Type | Description |
-| ---: | :----: | :--- |
-| `bubbles_euler`| Logical | Ensemble-averaged bubble modeling |
+| Parameter . | Type | Description |
+| ---: . | :----: | :--- |
+| `bubbles_euler`. | Logical | Ensemble-averaged bubble modeling |
</file context>
| | Parameter . | Type | Description | | |
| | Parameter | Type | Description | |
src/simulation/m_ibm.fpp
Outdated
| patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB | ||
| end do | ||
| print *, patch_ib(1)%force(1:2) |
There was a problem hiding this comment.
P2: Debug print statement should be removed from production code.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1054:
<comment>Debug print statement should be removed from production code.</comment>
<file context>
@@ -969,6 +994,67 @@ contains
+ patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB
+ end do
+
+ print *, patch_ib(1)%force(1:2)
+
+ end subroutine s_compute_ib_forces
</file context>
✅ Addressed in 0c221ae
src/simulation/m_ibm.fpp
Outdated
| else | ||
| radial_vector = [x_cc(i), y_cc(j), 0._wp] - [patch_ib(ib_idx)%x_centroid, patch_ib(ib_idx)%y_centroid, 0._wp] ! get the vector pointing to the grid cell | ||
| end if | ||
| pressure_divergence(1) = (pressure(i+1, j, k) - pressure(i-1, j, k)) / (2._wp * x_cc(i)) |
There was a problem hiding this comment.
P1: Pressure gradient calculation uses cell center coordinate x_cc(i) instead of grid spacing. Central difference formula should divide by 2*dx (grid spacing), not the coordinate value x_cc(i).
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1025:
<comment>Pressure gradient calculation uses cell center coordinate `x_cc(i)` instead of grid spacing. Central difference formula should divide by `2*dx` (grid spacing), not the coordinate value `x_cc(i)`.</comment>
<file context>
@@ -969,6 +994,67 @@ contains
+ else
+ radial_vector = [x_cc(i), y_cc(j), 0._wp] - [patch_ib(ib_idx)%x_centroid, patch_ib(ib_idx)%y_centroid, 0._wp] ! get the vector pointing to the grid cell
+ end if
+ pressure_divergence(1) = (pressure(i+1, j, k) - pressure(i-1, j, k)) / (2._wp * x_cc(i))
+ pressure_divergence(2) = (pressure(i, j+1, k) - pressure(i, j-1, k)) / (2._wp * y_cc(j))
+ cell_volume = x_cc(i) * y_cc(j)
</file context>
✅ Addressed in cfe6714
src/simulation/m_ibm.fpp
Outdated
| ! project the position along the axis to find the closest distance to the rotation axis | ||
| closest_point_along_axis = axis*dot_product(axis, position) | ||
| vector_to_axis = position - closest_point_along_axis | ||
| distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared |
There was a problem hiding this comment.
P1: Incorrect distance calculation: sum(vector_to_axis) sums components, not computing squared distance. Should be sum(vector_to_axis**2) per the comment.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1121:
<comment>Incorrect distance calculation: `sum(vector_to_axis)` sums components, not computing squared distance. Should be `sum(vector_to_axis**2)` per the comment.</comment>
<file context>
@@ -978,6 +1064,78 @@ contains
+ ! project the position along the axis to find the closest distance to the rotation axis
+ closest_point_along_axis = axis*dot_product(axis, position)
+ vector_to_axis = position - closest_point_along_axis
+ distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared
+
+ ! compute the position component of the moment
</file context>
| distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared | |
| distance_to_axis = sum(vector_to_axis**2) ! saves the distance to the axis squared |
✅ Addressed in a216ed4
src/simulation/m_ibm.fpp
Outdated
| if (patch_ib(ib_marker)%geometry == 2) then ! circle | ||
| patch_ib(ib_marker)%moment = 0.5 * patch_ib(ib_marker)%mass * (patch_ib(ib_marker)%radius)**2 | ||
| elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle | ||
| patch_ib(ib_marker)%moment = patch_ib(i)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp |
There was a problem hiding this comment.
P1: Wrong variable used: patch_ib(i)%mass should be patch_ib(ib_marker)%mass. Variable i is undefined in this context.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1090:
<comment>Wrong variable used: `patch_ib(i)%mass` should be `patch_ib(ib_marker)%mass`. Variable `i` is undefined in this context.</comment>
<file context>
@@ -978,6 +1064,78 @@ contains
+ if (patch_ib(ib_marker)%geometry == 2) then ! circle
+ patch_ib(ib_marker)%moment = 0.5 * patch_ib(ib_marker)%mass * (patch_ib(ib_marker)%radius)**2
+ elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle
+ patch_ib(ib_marker)%moment = patch_ib(i)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp
+ elseif (patch_ib(ib_marker)%geometry == 8) then ! sphere
+ patch_ib(ib_marker)%moment = 0.4 * patch_ib(ib_marker)%mass * (patch_ib(ib_marker)%radius)**2
</file context>
| patch_ib(ib_marker)%moment = patch_ib(i)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp | |
| patch_ib(ib_marker)%moment = patch_ib(ib_marker)%mass * (patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2) / 6._wp |
✅ Addressed in a216ed4
src/simulation/m_ibm.fpp
Outdated
| $:GPU_PARALLEL_LOOP(private='[position,closest_point_along_axis,vector_to_axis,distance_to_axis]', copy='[moment,count]', copyin='[ib_marker,cell_volume,axis]', collapse=3) | ||
| do i = 0, m | ||
| do j = 0, j |
There was a problem hiding this comment.
P0: Loop variable j is used as its own upper bound. This should be do j = 0, n to iterate over the y-dimension grid.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1105:
<comment>Loop variable `j` is used as its own upper bound. This should be `do j = 0, n` to iterate over the y-dimension grid.</comment>
<file context>
@@ -978,6 +1064,78 @@ contains
+
+ $:GPU_PARALLEL_LOOP(private='[position,closest_point_along_axis,vector_to_axis,distance_to_axis]', copy='[moment,count]', copyin='[ib_marker,cell_volume,axis]', collapse=3)
+ do i = 0, m
+ do j = 0, j
+ do k = 0, p
+ if (ib_markers%sf(i, j, k) == ib_marker) then
</file context>
| do j = 0, j | |
| do j = 0, n |
✅ Addressed in a4967c9
There was a problem hiding this comment.
9 issues found across 13 files
Prompt for AI agents (all 9 issues)
Check if these issues are valid — if so, understand the root cause of each and fix them.
<file name="examples/2D_mibm_cylinder_in_cross_flow/case.py">
<violation number="1" location="examples/2D_mibm_cylinder_in_cross_flow/case.py:98">
P2: Incorrect comment: The comment `# z-axis rotation` is wrong for the `mass` parameter. This appears to be a copy-paste error from the angular velocity comments above.</violation>
</file>
<file name="docs/documentation/case.md">
<violation number="1" location="docs/documentation/case.md:379">
P2: Erroneous `. ` characters in table cells will corrupt markdown table rendering. Remove the `. ` from each affected cell.</violation>
<violation number="2" location="docs/documentation/case.md:781">
P2: Erroneous `. ` characters in table cells will corrupt markdown table rendering. Remove the `. ` from each affected cell.</violation>
</file>
<file name="src/common/m_mpi_common.fpp">
<violation number="1" location="src/common/m_mpi_common.fpp:495">
P2: Missing non-MPI fallback: when `MFC_MPI` is not defined, `var_glb` is never assigned, leaving it uninitialized. Similar subroutine `s_mpi_allreduce_integer_sum` includes an `#else` block to copy `var_loc` to `var_glb` for serial execution.</violation>
</file>
<file name="src/simulation/m_ibm.fpp">
<violation number="1" location="src/simulation/m_ibm.fpp:1025">
P1: Pressure gradient calculation uses `x_cc(i)` (cell center coordinate) instead of grid spacing `dx(i)` in denominator. For central difference: `(P[i+1]-P[i-1])/(2*dx)`. Similarly, `cell_volume` should use grid spacings `dx(i)*dy(j)`, not coordinates.</violation>
<violation number="2" location="src/simulation/m_ibm.fpp:1054">
P2: Debug print statement should be removed from production code.</violation>
<violation number="3" location="src/simulation/m_ibm.fpp:1090">
P0: Wrong index variable: `patch_ib(i)%mass` should be `patch_ib(ib_marker)%mass`. Variable `i` is undefined at this point in the code path - it's only used as a loop variable in the numerical approximation section below.</violation>
<violation number="4" location="src/simulation/m_ibm.fpp:1104">
P0: Loop bound error: `do j = 0, j` uses `j` as both loop variable and upper bound. This will cause incorrect iteration since `j` is uninitialized. Should be `do j = 0, n` to iterate over the y-dimension.</violation>
<violation number="5" location="src/simulation/m_ibm.fpp:1120">
P1: Incorrect distance calculation: `sum(vector_to_axis)` computes the sum of components, not the squared distance. For moment of inertia, this should be `sum(vector_to_axis**2)` to get the squared perpendicular distance from the rotation axis.</violation>
</file>
Reply to cubic to teach it or ask questions. Re-run a review with @cubic-dev-ai review this PR
| "patch_ib(1)%angular_vel(1)": 0.0, # x-axis rotational velocity in radians per second | ||
| "patch_ib(1)%angular_vel(2)": 0.0, # y-axis rotation | ||
| "patch_ib(1)%angular_vel(3)": 0.0, # z-axis rotation | ||
| "patch_ib(1)%mass": 0.0001, # z-axis rotation |
There was a problem hiding this comment.
P2: Incorrect comment: The comment # z-axis rotation is wrong for the mass parameter. This appears to be a copy-paste error from the angular velocity comments above.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At examples/2D_mibm_cylinder_in_cross_flow/case.py, line 98:
<comment>Incorrect comment: The comment `# z-axis rotation` is wrong for the `mass` parameter. This appears to be a copy-paste error from the angular velocity comments above.</comment>
<file context>
@@ -0,0 +1,105 @@
+ "patch_ib(1)%angular_vel(1)": 0.0, # x-axis rotational velocity in radians per second
+ "patch_ib(1)%angular_vel(2)": 0.0, # y-axis rotation
+ "patch_ib(1)%angular_vel(3)": 0.0, # z-axis rotation
+ "patch_ib(1)%mass": 0.0001, # z-axis rotation
+ # Fluids Physical Parameters
+ "fluid_pp(1)%gamma": 1.0e00 / (gam_a - 1.0e00), # 2.50(Not 1.40)
</file context>
| "patch_ib(1)%mass": 0.0001, # z-axis rotation | |
| "patch_ib(1)%mass": 0.0001, # mass of the immersed boundary |
| | Parameter | Type | Description | | ||
| | ---: | :----: | :--- | | ||
| | `bubbles_euler`| Logical | Ensemble-averaged bubble modeling | | ||
| | Parameter . | Type | Description | |
There was a problem hiding this comment.
P2: Erroneous . characters in table cells will corrupt markdown table rendering. Remove the . from each affected cell.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At docs/documentation/case.md, line 781:
<comment>Erroneous `. ` characters in table cells will corrupt markdown table rendering. Remove the `. ` from each affected cell.</comment>
<file context>
@@ -769,23 +778,23 @@ Implementation of the parameters into the model follow [Ando (2010)](references.
-| Parameter | Type | Description |
-| ---: | :----: | :--- |
-| `bubbles_euler`| Logical | Ensemble-averaged bubble modeling |
+| Parameter . | Type | Description |
+| ---: . | :----: | :--- |
+| `bubbles_euler`. | Logical | Ensemble-averaged bubble modeling |
</file context>
| | Parameter . | Type | Description | | |
| | Parameter | Type | Description | |
| | `igr_iter_solver` | Integer | Solution method for IGR elliptic solve [1] Jacobi [2] Gauss-Seidel | | ||
| | `num_igr_iters` | Integer | Number of iterations for for the IGR elliptic solve (default 2) | | ||
| . | ||
| | Parameter . | Type | Description | |
There was a problem hiding this comment.
P2: Erroneous . characters in table cells will corrupt markdown table rendering. Remove the . from each affected cell.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At docs/documentation/case.md, line 379:
<comment>Erroneous `. ` characters in table cells will corrupt markdown table rendering. Remove the `. ` from each affected cell.</comment>
<file context>
@@ -366,62 +375,62 @@ Details of implementation of viscosity in MFC can be found in [Coralic (2015)](r
-| `igr_iter_solver` | Integer | Solution method for IGR elliptic solve [1] Jacobi [2] Gauss-Seidel |
-| `num_igr_iters` | Integer | Number of iterations for for the IGR elliptic solve (default 2) |
+.
+| Parameter . | Type | Description |
+| ---: . | :----: | :--- |
+| `bc_[x,y,z]%%beg[end]` . | Integer | Beginning [ending] boundary condition in the $[x,y,z]$-direction (negative integer, see table [Boundary Conditions](#boundary-conditions)) |
</file context>
| | Parameter . | Type | Description | | |
| | Parameter | Type | Description | |
| mpi_p, MPI_SUM, MPI_COMM_WORLD, ierr) | ||
| end if | ||
| #endif |
There was a problem hiding this comment.
P2: Missing non-MPI fallback: when MFC_MPI is not defined, var_glb is never assigned, leaving it uninitialized. Similar subroutine s_mpi_allreduce_integer_sum includes an #else block to copy var_loc to var_glb for serial execution.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/common/m_mpi_common.fpp, line 495:
<comment>Missing non-MPI fallback: when `MFC_MPI` is not defined, `var_glb` is never assigned, leaving it uninitialized. Similar subroutine `s_mpi_allreduce_integer_sum` includes an `#else` block to copy `var_loc` to `var_glb` for serial execution.</comment>
<file context>
@@ -472,6 +472,30 @@ contains
+ mpi_p, MPI_SUM, MPI_COMM_WORLD, ierr)
+ end if
+
+#endif
+
+ end subroutine s_mpi_allreduce_vectors_sum
</file context>
| #endif | |
| #else | |
| var_glb = var_loc | |
| #endif |
✅ Addressed in 8531946
src/simulation/m_ibm.fpp
Outdated
| patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB | ||
| end do | ||
| print *, patch_ib(1)%force(1:2) |
There was a problem hiding this comment.
P2: Debug print statement should be removed from production code.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1054:
<comment>Debug print statement should be removed from production code.</comment>
<file context>
@@ -969,6 +994,67 @@ contains
+ patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB
+ end do
+
+ print *, patch_ib(1)%force(1:2)
+
+ end subroutine s_compute_ib_forces
</file context>
✅ Addressed in 0c221ae
src/simulation/m_ibm.fpp
Outdated
| else | ||
| radial_vector = [x_cc(i), y_cc(j), 0._wp] - [patch_ib(ib_idx)%x_centroid, patch_ib(ib_idx)%y_centroid, 0._wp] ! get the vector pointing to the grid cell | ||
| end if | ||
| pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(2._wp*x_cc(i)) |
There was a problem hiding this comment.
P1: Pressure gradient calculation uses x_cc(i) (cell center coordinate) instead of grid spacing dx(i) in denominator. For central difference: (P[i+1]-P[i-1])/(2*dx). Similarly, cell_volume should use grid spacings dx(i)*dy(j), not coordinates.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1025:
<comment>Pressure gradient calculation uses `x_cc(i)` (cell center coordinate) instead of grid spacing `dx(i)` in denominator. For central difference: `(P[i+1]-P[i-1])/(2*dx)`. Similarly, `cell_volume` should use grid spacings `dx(i)*dy(j)`, not coordinates.</comment>
<file context>
@@ -969,6 +994,67 @@ contains
+ else
+ radial_vector = [x_cc(i), y_cc(j), 0._wp] - [patch_ib(ib_idx)%x_centroid, patch_ib(ib_idx)%y_centroid, 0._wp] ! get the vector pointing to the grid cell
+ end if
+ pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(2._wp*x_cc(i))
+ pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(2._wp*y_cc(j))
+ cell_volume = x_cc(i)*y_cc(j)
</file context>
✅ Addressed in cfe6714
src/simulation/m_ibm.fpp
Outdated
| ! project the position along the axis to find the closest distance to the rotation axis | ||
| closest_point_along_axis = axis*dot_product(axis, position) | ||
| vector_to_axis = position - closest_point_along_axis | ||
| distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared |
There was a problem hiding this comment.
P1: Incorrect distance calculation: sum(vector_to_axis) computes the sum of components, not the squared distance. For moment of inertia, this should be sum(vector_to_axis**2) to get the squared perpendicular distance from the rotation axis.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1120:
<comment>Incorrect distance calculation: `sum(vector_to_axis)` computes the sum of components, not the squared distance. For moment of inertia, this should be `sum(vector_to_axis**2)` to get the squared perpendicular distance from the rotation axis.</comment>
<file context>
@@ -978,6 +1064,77 @@ contains
+ ! project the position along the axis to find the closest distance to the rotation axis
+ closest_point_along_axis = axis*dot_product(axis, position)
+ vector_to_axis = position - closest_point_along_axis
+ distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared
+
+ ! compute the position component of the moment
</file context>
| distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared | |
| distance_to_axis = sum(vector_to_axis**2) ! saves the distance to the axis squared |
✅ Addressed in a216ed4
src/simulation/m_ibm.fpp
Outdated
| if (patch_ib(ib_marker)%geometry == 2) then ! circle | ||
| patch_ib(ib_marker)%moment = 0.5*patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%radius)**2 | ||
| elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle | ||
| patch_ib(ib_marker)%moment = patch_ib(i)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp |
There was a problem hiding this comment.
P0: Wrong index variable: patch_ib(i)%mass should be patch_ib(ib_marker)%mass. Variable i is undefined at this point in the code path - it's only used as a loop variable in the numerical approximation section below.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1090:
<comment>Wrong index variable: `patch_ib(i)%mass` should be `patch_ib(ib_marker)%mass`. Variable `i` is undefined at this point in the code path - it's only used as a loop variable in the numerical approximation section below.</comment>
<file context>
@@ -978,6 +1064,77 @@ contains
+ if (patch_ib(ib_marker)%geometry == 2) then ! circle
+ patch_ib(ib_marker)%moment = 0.5*patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%radius)**2
+ elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle
+ patch_ib(ib_marker)%moment = patch_ib(i)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp
+ elseif (patch_ib(ib_marker)%geometry == 8) then ! sphere
+ patch_ib(ib_marker)%moment = 0.4*patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%radius)**2
</file context>
| patch_ib(ib_marker)%moment = patch_ib(i)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp | |
| patch_ib(ib_marker)%moment = patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp |
✅ Addressed in a216ed4
src/simulation/m_ibm.fpp
Outdated
| $:GPU_PARALLEL_LOOP(private='[position,closest_point_along_axis,vector_to_axis,distance_to_axis]', copy='[moment,count]', copyin='[ib_marker,cell_volume,axis]', collapse=3) | ||
| do i = 0, m | ||
| do j = 0, j |
There was a problem hiding this comment.
P0: Loop bound error: do j = 0, j uses j as both loop variable and upper bound. This will cause incorrect iteration since j is uninitialized. Should be do j = 0, n to iterate over the y-dimension.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1104:
<comment>Loop bound error: `do j = 0, j` uses `j` as both loop variable and upper bound. This will cause incorrect iteration since `j` is uninitialized. Should be `do j = 0, n` to iterate over the y-dimension.</comment>
<file context>
@@ -978,6 +1064,77 @@ contains
+
+ $:GPU_PARALLEL_LOOP(private='[position,closest_point_along_axis,vector_to_axis,distance_to_axis]', copy='[moment,count]', copyin='[ib_marker,cell_volume,axis]', collapse=3)
+ do i = 0, m
+ do j = 0, j
+ do k = 0, p
+ if (ib_markers%sf(i, j, k) == ib_marker) then
</file context>
| do j = 0, j | |
| do j = 0, n |
✅ Addressed in a4967c9
|
CodeAnt AI is running Incremental review Thanks for using CodeAnt! 🎉We're free for open-source projects. if you're enjoying it, help us grow by sharing. Share on X · |
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
src/simulation/m_ibm.fpp (1)
36-39: Missing public exports for new subroutines.The subroutines
s_compute_ib_forcesands_compute_moment_of_inertiaare called fromm_time_steppers.fppbut are not listed in the public exports. They will not be accessible outside this module.; public :: s_initialize_ibm_module, & s_ibm_setup, & s_ibm_correct_state, & - s_finalize_ibm_module + s_finalize_ibm_module, & + s_compute_ib_forces, & + s_compute_moment_of_inertia
♻️ Duplicate comments (7)
src/simulation/m_ibm.fpp (7)
1032-1034: 3D pressure gradient also uses cell center instead of grid spacing.-pressure_divergence(3) = (pressure(i, j, k + 1) - pressure(i, j, k - 1))/(2._wp*z_cc(k)) -cell_volume = cell_volume*z_cc(k) +pressure_divergence(3) = (pressure(i, j, k + 1) - pressure(i, j, k - 1))/(z_cc(k + 1) - z_cc(k - 1)) +cell_volume = cell_volume*dz(k)
1081-1089: Optional argument handling is unsafe and normalization is incorrect.Multiple issues flagged by previous reviewers:
- Assigning to
axiswhenp == 0is undefined behavior ifaxiswas not passed- Line 1088 uses
sqrt(sum(axis))which sums components, not squares+real(wp), dimension(3) :: local_axis + if (p == 0) then - axis = [0, 1, 0] -else if (sqrt(sum(axis**2)) == 0) then + local_axis = [0._wp, 0._wp, 1._wp] ! z-axis for 2D rotation +else if (.not. present(axis) .or. sqrt(sum(axis**2)) < 1.e-16_wp) then ! if the object is not actually rotating at this time, return a dummy value and exit patch_ib(ib_marker)%moment = 1._wp return else - axis = axis/sqrt(sum(axis)) + local_axis = axis/sqrt(sum(axis**2)) end if
1027-1029: Critical bug: Pressure gradient and cell volume use cell centers instead of grid spacing.This has been flagged by multiple reviewers. The central difference denominator should use grid spacing (
dx,dy,dz) or coordinate differences (x_cc(i+1) - x_cc(i-1)), not cell center coordinates.-pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(2._wp*x_cc(i)) -pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(2._wp*y_cc(j)) -cell_volume = x_cc(i)*y_cc(j) +pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(x_cc(i + 1) - x_cc(i - 1)) +pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(y_cc(j + 1) - y_cc(j - 1)) +cell_volume = dx(i)*dy(j)
1094-1095: Wrong variable: usespatch_ib(i)instead ofpatch_ib(ib_marker).Variable
iis undefined at this point—it's only declared as a loop variable later. This will read garbage or cause a compile error.elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle - patch_ib(ib_marker)%moment = patch_ib(i)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp + patch_ib(ib_marker)%moment = patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp
1109-1109: Loop bound error:do j = 0, jusesjas its own bound.This should be
do j = 0, nto iterate over the y-dimension grid.-do j = 0, j +do j = 0, n
1125-1125: Incorrect distance calculation: sums components instead of computing squared magnitude.The comment says "distance to the axis squared" but
sum(vector_to_axis)just adds the vector components. Should besum(vector_to_axis**2).-distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared +distance_to_axis = sum(vector_to_axis**2) ! saves the distance to the axis squared
1072-1075: Missingintentspecification foraxisparameter.Per coding guidelines, every argument must have explicit
intent. Theaxisparameter is declared asoptionalbut lacks an intent specification. Since it's being modified in the 2D case and normalized in 3D, it should beintent(inout).-real(wp), dimension(3), optional :: axis !< the axis about which we compute the moment. Only required in 3D. +real(wp), dimension(3), intent(inout), optional :: axis !< the axis about which we compute the moment. Only required in 3D.As per coding guidelines: "Every variable must have
intent(in|out|inout)specification."
🧹 Nitpick comments (1)
src/simulation/m_ibm.fpp (1)
199-212: Missing GPU parallelization for interior pressure loop.This nested loop modifies
q_prim_vf(E_idx)%sfwhich is GPU data, but lacks GPU parallelization directives. Other similar triple-nested loops in this file use$:GPU_PARALLEL_LOOP. Additionally, the hardcoded pressure value1._wplacks physical justification—consider making this a configurable parameter or documenting why this specific value is used.! set the Moving IBM interior Pressure Values do patch_id = 1, num_ibs if (patch_ib(patch_id)%moving_ibm == 2) then + $:GPU_PARALLEL_LOOP(collapse=3) do j = 0, m do k = 0, n do l = 0, p if (ib_markers%sf(j, k, l) == patch_id) then q_prim_vf(E_idx)%sf(j, k, l) = 1._wp end if end do end do end do + $:END_GPU_PARALLEL_LOOP() end if end do
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
src/simulation/m_ibm.fpp(4 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
**/*.{fpp,f90}: Use 2-space indentation; continuation lines align beneath &
Use lower-case keywords and intrinsics (do, end subroutine, etc.)
Name modules with m_ pattern (e.g., m_transport)
Name public subroutines with s_ pattern (e.g., s_compute_flux)
Name public functions with f_ pattern
Keep subroutine size ≤ 500 lines, helper subroutines ≤ 150 lines, functions ≤ 100 lines, files ≤ 1000 lines
Limit routine arguments to ≤ 6; use derived-type params struct if more are needed
Forbid goto statements (except in legacy code), COMMON blocks, and save globals
Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Call s_mpi_abort() for errors, never use stop or error stop
**/*.{fpp,f90}: Indent 2 spaces; continuation lines align under&
Use lower-case keywords and intrinsics (do,end subroutine, etc.)
Name modules withm_<feature>prefix (e.g.,m_transport)
Name public subroutines ass_<verb>_<noun>(e.g.,s_compute_flux) and functions asf_<verb>_<noun>
Keep private helpers in the module; avoid nested procedures
Enforce size limits: subroutine ≤ 500 lines, helper ≤ 150, function ≤ 100, module/file ≤ 1000
Limit subroutines to ≤ 6 arguments; otherwise pass a derived-type 'params' struct
Avoidgotostatements (except unavoidable legacy); avoid global state (COMMON,save)
Every variable must haveintent(in|out|inout)specification and appropriatedimension/allocatable/pointer
Uses_mpi_abort(<msg>)for error termination instead ofstop
Use!>style documentation for header comments; follow Doxygen Fortran format with!! @paramand!! @returntags
Useimplicit nonestatement in all modules
Useprivatedeclaration followed by explicitpublicexports in modules
Use derived types with pointers for encapsulation (e.g.,pointer, dimension(:,:,:) => null())
Usepureandelementalattributes for side-effect-free functions; combine them for array ...
Files:
src/simulation/m_ibm.fpp
src/simulation/**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
src/simulation/**/*.{fpp,f90}: Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Avoid stop/error stop inside GPU device code
Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
src/simulation/**/*.{fpp,f90}: Mark GPU-callable helpers with$:GPU_ROUTINE(function_name='...', parallelism='[seq]')immediately after declaration
Do not use OpenACC or OpenMP directives directly; use Fypp macros fromsrc/common/include/parallel_macros.fppinstead
Wrap tight loops with$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')macro; addcollapse=nfor safe nested loop merging
Declare loop-local variables withprivate='[...]'in GPU parallel loop macros
Allocate large arrays withmanagedor move them into a persistent$:GPU_ENTER_DATA(...)region at start-up
Do not placestoporerror stopinside device code
Files:
src/simulation/m_ibm.fpp
src/**/*.fpp
📄 CodeRabbit inference engine (.cursor/rules/mfc-agent-rules.mdc)
src/**/*.fpp: Use.fppfile extension for Fypp preprocessed files; CMake transpiles them to.f90
Start module files with Fypp include for macros:#:include 'macros.fpp'
Use the fyppASSERTmacro for validating conditions:@:ASSERT(predicate, message)
Use fypp macro@:ALLOCATE(var1, var2)for device-aware allocation instead of standard Fortranallocate
Use fypp macro@:DEALLOCATE(var1, var2)for device-aware deallocation instead of standard Fortrandeallocate
Files:
src/simulation/m_ibm.fpp
🧠 Learnings (11)
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight loops with `$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')` macro; add `collapse=n` for safe nested loop merging
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Avoid stop/error stop inside GPU device code
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Every variable must have `intent(in|out|inout)` specification and appropriate `dimension` / `allocatable` / `pointer`
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Declare loop-local variables with `private='[...]'` in GPU parallel loop macros
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Favor array operations over explicit loops when possible; use `collapse=N` directive to optimize nested loops
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Mark GPU-callable helpers with `$:GPU_ROUTINE(function_name='...', parallelism='[seq]')` immediately after declaration
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large arrays with `managed` or move them into a persistent `$:GPU_ENTER_DATA(...)` region at start-up
Applied to files:
src/simulation/m_ibm.fpp
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (17)
- GitHub Check: Coverage Test on CodeCov
- GitHub Check: Code Cleanliness Check
- GitHub Check: Github (macos, mpi, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, debug, false)
- GitHub Check: Github (macos, mpi, debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, false)
- GitHub Check: Self Hosted (cpu, none, gt)
- GitHub Check: Github (ubuntu, mpi, debug, true)
- GitHub Check: Github (ubuntu, no-mpi, single, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, true)
- GitHub Check: Self Hosted (gpu, omp, gt)
- GitHub Check: Self Hosted (gpu, omp, frontier)
- GitHub Check: Self Hosted (cpu, none, frontier)
- GitHub Check: Self Hosted (gpu, acc, frontier)
- GitHub Check: Self Hosted (gpu, acc, gt)
- GitHub Check: cubic · AI code reviewer
- GitHub Check: Build & Publish
🔇 Additional comments (1)
src/simulation/m_ibm.fpp (1)
1137-1138: Ensure derived type member is properly managed in GPU data region.The syntax
$:GPU_UPDATE(device='[patch_ib(ib_marker)%moment]')is valid OpenACC, but derived type members—especially if allocatable or pointers—require explicit data management for portability across nvfortran, gfortran, and Cray compilers. Verify thatpatch_ibis present in an$:GPU_ENTER_DATA(...)region at start-up, or explicitly managemomentwith appropriate data clauses. Ifmomentis an allocatable/pointer member, use separate enter/exit data for it rather than relying on implicit deep-copy behavior.
src/simulation/m_ibm.fpp
Outdated
| $:GPU_ATOMIC(atomic='update') | ||
| forces(ib_idx, :) = forces(ib_idx, :) - (pressure_divergence*cell_volume) | ||
| $:GPU_ATOMIC(atomic='update') | ||
| torques(ib_idx, :) = torques(ib_idx, :) + (cross_product(radial_vector, pressure_divergence)*cell_volume) |
There was a problem hiding this comment.
Atomic update on array slice may have undefined behavior.
GPU atomic operations on array slices forces(ib_idx, :) and torques(ib_idx, :) may not atomically update all elements together. Consider using separate atomic updates for each component.
! Update the force values atomically to prevent race conditions
-$:GPU_ATOMIC(atomic='update')
-forces(ib_idx, :) = forces(ib_idx, :) - (pressure_divergence*cell_volume)
-$:GPU_ATOMIC(atomic='update')
-torques(ib_idx, :) = torques(ib_idx, :) + (cross_product(radial_vector, pressure_divergence)*cell_volume)
+$:GPU_ATOMIC(atomic='update')
+forces(ib_idx, 1) = forces(ib_idx, 1) - pressure_divergence(1)*cell_volume
+$:GPU_ATOMIC(atomic='update')
+forces(ib_idx, 2) = forces(ib_idx, 2) - pressure_divergence(2)*cell_volume
+$:GPU_ATOMIC(atomic='update')
+forces(ib_idx, 3) = forces(ib_idx, 3) - pressure_divergence(3)*cell_volume
+! Similar expansion needed for torquesCommittable suggestion skipped: line range outside the PR's diff.
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (7)
src/simulation/m_ibm.fpp (7)
261-270: Loop repeatedly overwrites the same pressure value; only the last iteration persists.The loop over
q = 1, num_fluidsoverwritesq_prim_vf(E_idx)%sf(j, k, l)in each iteration, so only the value fromq == num_fluidsis retained. This is wasteful and likely a logic error for multi-fluid cases.Apply this diff to remove the redundant loop (using the first fluid's properties for now, per the TODO):
- ! set the pressure - do q = 1, num_fluids - if (patch_ib(patch_id)%moving_ibm == 0) then - q_prim_vf(E_idx)%sf(j, k, l) = pres_IP - else - ! TODO :: improve for two fluid - q_prim_vf(E_idx)%sf(j, k, l) = pres_IP/(1._wp - 2._wp*abs(levelset%sf(j, k, l, patch_id)*alpha_rho_IP(q)/pres_IP)*dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j, k, l, patch_id, :))) - end if - end do + ! set the pressure + if (patch_ib(patch_id)%moving_ibm == 0) then + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP + else + ! TODO :: improve for multi-fluid - currently uses first fluid's properties + if (abs(pres_IP) > 1.e-16_wp) then + buf = 1._wp - 2._wp*abs(levelset%sf(j, k, l, patch_id)*alpha_rho_IP(1)/pres_IP) * & + dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j, k, l, patch_id, :)) + if (abs(buf) > 1.e-16_wp) then + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP/buf + else + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP + end if + else + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP + end if + end ifBased on past review comments.
1026-1037: Critical: Pressure gradient uses cell-center coordinates instead of grid spacing.The finite-difference gradient divides by cell-center coordinates (
x_cc(i),y_cc(j),z_cc(k)) rather than grid spacing. The correct central-difference formula is(P[i+1] - P[i-1]) / (x_cc(i+1) - x_cc(i-1))or(P[i+1] - P[i-1]) / (2*dx(i)). Similarly,cell_volumeshould bedx(i)*dy(j)*dz(k), not the product of coordinates.Apply this diff to fix the gradient and volume calculations:
- ! use a finite difference to compute the 2D components of the gradient of the pressure and cell volume - pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(2._wp*x_cc(i)) - pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(2._wp*y_cc(j)) - cell_volume = x_cc(i)*y_cc(j) + ! use a finite difference to compute the 2D components of the gradient of the pressure and cell volume + pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(x_cc(i + 1) - x_cc(i - 1)) + pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(y_cc(j + 1) - y_cc(j - 1)) + cell_volume = dx(i)*dy(j) ! add the 3D component, if we are working in 3 dimensions if (num_dims == 3) then - pressure_divergence(3) = (pressure(i, j, k + 1) - pressure(i, j, k - 1))/(2._wp*z_cc(k)) - cell_volume = cell_volume*z_cc(k) + pressure_divergence(3) = (pressure(i, j, k + 1) - pressure(i, j, k - 1))/(z_cc(k + 1) - z_cc(k - 1)) + cell_volume = cell_volume*dz(k) else pressure_divergence(3) = 0._wp end ifBased on past review comments.
1040-1043: Atomic updates on array slices may not work correctly on all GPU compilers.The
$:GPU_ATOMICdirective is applied to array slicesforces(ib_idx, :)andtorques(ib_idx, :). Some GPU compilers may not atomically update all elements of a slice together, leading to race conditions. Update each component separately with individual atomic directives.Apply this diff:
! Update the force values atomically to prevent race conditions $:GPU_ATOMIC(atomic='update') - forces(ib_idx, :) = forces(ib_idx, :) - (pressure_divergence*cell_volume) + forces(ib_idx, 1) = forces(ib_idx, 1) - pressure_divergence(1)*cell_volume + $:GPU_ATOMIC(atomic='update') + forces(ib_idx, 2) = forces(ib_idx, 2) - pressure_divergence(2)*cell_volume + $:GPU_ATOMIC(atomic='update') + forces(ib_idx, 3) = forces(ib_idx, 3) - pressure_divergence(3)*cell_volume + + $:GPU_ATOMIC(atomic='update') + torques(ib_idx, 1) = torques(ib_idx, 1) + cross_product(radial_vector, pressure_divergence)(1)*cell_volume + $:GPU_ATOMIC(atomic='update') + torques(ib_idx, 2) = torques(ib_idx, 2) + cross_product(radial_vector, pressure_divergence)(2)*cell_volume $:GPU_ATOMIC(atomic='update') - torques(ib_idx, :) = torques(ib_idx, :) + (cross_product(radial_vector, pressure_divergence)*cell_volume) + torques(ib_idx, 3) = torques(ib_idx, 3) + cross_product(radial_vector, pressure_divergence)(3)*cell_volumeBased on past review comments.
1074-1089: Optional argumentaxislacksintentand is assigned without checkingpresent().The optional argument
axisis declared without anintentspecification and is assigned at line 1082 without checkingpresent(axis). Additionally, line 1088 incorrectly normalizesaxisbysqrt(sum(axis))instead ofsqrt(sum(axis**2))— this sums components rather than computing the magnitude.Apply this diff:
- subroutine s_compute_moment_of_inertia(ib_marker, axis) + subroutine s_compute_moment_of_inertia(ib_marker, axis_in) - real(wp), dimension(3), optional :: axis !< the axis about which we compute the moment. Only required in 3D. + real(wp), dimension(3), intent(in), optional :: axis_in !< the axis about which we compute the moment. Only required in 3D. integer, intent(in) :: ib_marker real(wp) :: moment, distance_to_axis, cell_volume - real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis + real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis, axis integer :: i, j, k, count if (p == 0) then - axis = [0, 1, 0] - else if (sqrt(sum(axis**2)) == 0) then + axis = [0._wp, 0._wp, 1._wp] + else if (.not. present(axis_in) .or. sqrt(sum(axis_in**2)) < 1.e-16_wp) then ! if the object is not actually rotating at this time, return a dummy value and exit patch_ib(ib_marker)%moment = 1._wp return else - axis = axis/sqrt(sum(axis)) + axis = axis_in/sqrt(sum(axis_in**2)) end ifBased on past review comments and coding guidelines requiring explicit
intent.
1095-1095: Critical: Wrong variableiused instead ofib_marker.Line 1095 references
patch_ib(i)%mass, butiis undefined in this branch (it's only a loop iterator declared later). This should bepatch_ib(ib_marker)%mass.Apply this diff:
elseif (patch_ib(ib_marker)%geometry == 3) then ! rectangle - patch_ib(ib_marker)%moment = patch_ib(i)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp + patch_ib(ib_marker)%moment = patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wpBased on past review comments.
1108-1110: Critical: Loop bound typo —jused as its own upper bound.Line 1109 is written as
do j = 0, j, which uses the loop variablejas its own bound. This will cause incorrect iteration (either a no-op or undefined behavior). It should bedo j = 0, nto iterate over the y-dimension.Apply this diff:
$:GPU_PARALLEL_LOOP(private='[position,closest_point_along_axis,vector_to_axis,distance_to_axis]', copy='[moment,count]', copyin='[ib_marker,cell_volume,axis]', collapse=3) do i = 0, m - do j = 0, j + do j = 0, n do k = 0, pBased on past review comments.
1125-1125: Critical: Distance calculation sums components instead of computing squared distance.Line 1125 computes
distance_to_axis = sum(vector_to_axis), which sums the vector components. For moment-of-inertia calculations, this should besum(vector_to_axis**2)(the squared distance from the rotation axis), as indicated by the comment.Apply this diff:
- distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared + distance_to_axis = sum(vector_to_axis**2) ! saves the distance to the axis squaredBased on past review comments.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
src/simulation/m_ibm.fpp(4 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
**/*.{fpp,f90}: Use 2-space indentation; continuation lines align beneath &
Use lower-case keywords and intrinsics (do, end subroutine, etc.)
Name modules with m_ pattern (e.g., m_transport)
Name public subroutines with s_ pattern (e.g., s_compute_flux)
Name public functions with f_ pattern
Keep subroutine size ≤ 500 lines, helper subroutines ≤ 150 lines, functions ≤ 100 lines, files ≤ 1000 lines
Limit routine arguments to ≤ 6; use derived-type params struct if more are needed
Forbid goto statements (except in legacy code), COMMON blocks, and save globals
Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Call s_mpi_abort() for errors, never use stop or error stop
**/*.{fpp,f90}: Indent 2 spaces; continuation lines align under&
Use lower-case keywords and intrinsics (do,end subroutine, etc.)
Name modules withm_<feature>prefix (e.g.,m_transport)
Name public subroutines ass_<verb>_<noun>(e.g.,s_compute_flux) and functions asf_<verb>_<noun>
Keep private helpers in the module; avoid nested procedures
Enforce size limits: subroutine ≤ 500 lines, helper ≤ 150, function ≤ 100, module/file ≤ 1000
Limit subroutines to ≤ 6 arguments; otherwise pass a derived-type 'params' struct
Avoidgotostatements (except unavoidable legacy); avoid global state (COMMON,save)
Every variable must haveintent(in|out|inout)specification and appropriatedimension/allocatable/pointer
Uses_mpi_abort(<msg>)for error termination instead ofstop
Use!>style documentation for header comments; follow Doxygen Fortran format with!! @paramand!! @returntags
Useimplicit nonestatement in all modules
Useprivatedeclaration followed by explicitpublicexports in modules
Use derived types with pointers for encapsulation (e.g.,pointer, dimension(:,:,:) => null())
Usepureandelementalattributes for side-effect-free functions; combine them for array ...
Files:
src/simulation/m_ibm.fpp
src/simulation/**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
src/simulation/**/*.{fpp,f90}: Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Avoid stop/error stop inside GPU device code
Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
src/simulation/**/*.{fpp,f90}: Mark GPU-callable helpers with$:GPU_ROUTINE(function_name='...', parallelism='[seq]')immediately after declaration
Do not use OpenACC or OpenMP directives directly; use Fypp macros fromsrc/common/include/parallel_macros.fppinstead
Wrap tight loops with$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')macro; addcollapse=nfor safe nested loop merging
Declare loop-local variables withprivate='[...]'in GPU parallel loop macros
Allocate large arrays withmanagedor move them into a persistent$:GPU_ENTER_DATA(...)region at start-up
Do not placestoporerror stopinside device code
Files:
src/simulation/m_ibm.fpp
src/**/*.fpp
📄 CodeRabbit inference engine (.cursor/rules/mfc-agent-rules.mdc)
src/**/*.fpp: Use.fppfile extension for Fypp preprocessed files; CMake transpiles them to.f90
Start module files with Fypp include for macros:#:include 'macros.fpp'
Use the fyppASSERTmacro for validating conditions:@:ASSERT(predicate, message)
Use fypp macro@:ALLOCATE(var1, var2)for device-aware allocation instead of standard Fortranallocate
Use fypp macro@:DEALLOCATE(var1, var2)for device-aware deallocation instead of standard Fortrandeallocate
Files:
src/simulation/m_ibm.fpp
🧠 Learnings (11)
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight loops with `$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')` macro; add `collapse=n` for safe nested loop merging
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Avoid stop/error stop inside GPU device code
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Every variable must have `intent(in|out|inout)` specification and appropriate `dimension` / `allocatable` / `pointer`
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Declare loop-local variables with `private='[...]'` in GPU parallel loop macros
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Favor array operations over explicit loops when possible; use `collapse=N` directive to optimize nested loops
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large arrays with `managed` or move them into a persistent `$:GPU_ENTER_DATA(...)` region at start-up
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Mark GPU-callable helpers with `$:GPU_ROUTINE(function_name='...', parallelism='[seq]')` immediately after declaration
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Applied to files:
src/simulation/m_ibm.fpp
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (16)
- GitHub Check: Coverage Test on CodeCov
- GitHub Check: cubic · AI code reviewer
- GitHub Check: Github (macos, mpi, no-debug, false)
- GitHub Check: Self Hosted (gpu, omp, gt)
- GitHub Check: Github (ubuntu, no-mpi, single, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, false)
- GitHub Check: Self Hosted (gpu, acc, gt)
- GitHub Check: Self Hosted (cpu, none, frontier)
- GitHub Check: Github (macos, mpi, debug, false)
- GitHub Check: Self Hosted (cpu, none, gt)
- GitHub Check: Self Hosted (gpu, acc, frontier)
- GitHub Check: Github (ubuntu, mpi, debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, true)
- GitHub Check: Github (ubuntu, mpi, debug, true)
- GitHub Check: Code Cleanliness Check
- GitHub Check: Build & Publish
|
CodeAnt AI is running Incremental review Thanks for using CodeAnt! 🎉We're free for open-source projects. if you're enjoying it, help us grow by sharing. Share on X · |
There was a problem hiding this comment.
Actionable comments posted: 0
♻️ Duplicate comments (7)
src/common/m_mpi_common.fpp (1)
477-481: Intent mismatch for in-place MPI reduction.The
var_glbparameter is declaredintent(out), but whenloc(var_loc) == loc(var_glb), the subroutine usesMPI_IN_PLACE(line 488). Withintent(out), the compiler may treat the actual argument as undefined on entry, making the in-place operation unsafe. Change tointent(inout).Apply this diff:
- real(wp), dimension(:, :), intent(out) :: var_glb + real(wp), dimension(:, :), intent(inout) :: var_glbBased on past review comments.
src/simulation/m_ibm.fpp (6)
261-270: Redundant loop overwrites pressure value.The loop over
q = 1, num_fluidsrepeatedly overwritesq_prim_vf(E_idx)%sf(j, k, l), retaining only the last fluid's value. Either remove the loop and compute once, or implement proper multi-fluid accumulation as indicated by the TODO comment.Based on past review comments.
1028-1030: Critical: Pressure gradient uses coordinates instead of grid spacing.The finite difference divides by cell-center coordinates (
x_cc(i),y_cc(j)) instead of grid spacing. Central difference requires:(P[i+1] - P[i-1]) / (x_cc(i+1) - x_cc(i-1)). Similarly,cell_volumeshould use grid spacingsdx(i)*dy(j), not coordinates.Apply this diff:
- pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(2._wp*x_cc(i)) - pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(2._wp*y_cc(j)) - cell_volume = x_cc(i)*y_cc(j) + pressure_divergence(1) = (pressure(i + 1, j, k) - pressure(i - 1, j, k))/(x_cc(i + 1) - x_cc(i - 1)) + pressure_divergence(2) = (pressure(i, j + 1, k) - pressure(i, j - 1, k))/(y_cc(j + 1) - y_cc(j - 1)) + cell_volume = dx(i)*dy(j)Based on past review comments.
1034-1035: Critical: 3D gradient and volume use coordinates instead of spacing.The 3D gradient calculation has the same issue: divides by
z_cc(k)(coordinate) instead of grid spacing. Volume should multiply bydz(k), notz_cc(k).Apply this diff:
- pressure_divergence(3) = (pressure(i, j, k + 1) - pressure(i, j, k - 1))/(2._wp*z_cc(k)) - cell_volume = cell_volume*z_cc(k) + pressure_divergence(3) = (pressure(i, j, k + 1) - pressure(i, j, k - 1))/(z_cc(k + 1) - z_cc(k - 1)) + cell_volume = cell_volume*dz(k)Based on past review comments.
1082-1090: Critical: Unsafe optional argument access and incorrect normalization.Two issues:
- Line 1083 assigns to optional
axiswithout checkingpresent(axis)(undefined behavior if not provided)- Line 1089 normalizes using
sqrt(sum(axis))instead ofsqrt(sum(axis**2))(incorrect magnitude calculation)Consider declaring a local
axisvariable and copying from the optional parameter when present:- real(wp), dimension(3), optional :: axis + real(wp), dimension(3), intent(in), optional :: axis integer, intent(in) :: ib_marker real(wp) :: moment, distance_to_axis, cell_volume - real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis + real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis, axis_local integer :: i, j, k, count if (p == 0) then - axis = [0, 1, 0] - else if (sqrt(sum(axis**2)) == 0) then + axis_local = [0._wp, 0._wp, 1._wp] + else if (.not. present(axis) .or. sqrt(sum(axis**2)) < 1.e-16_wp) then patch_ib(ib_marker)%moment = 1._wp return else - axis = axis/sqrt(sum(axis)) + axis_local = axis/sqrt(sum(axis**2)) end ifBased on past review comments.
1096-1096: Critical: Wrong variable index.Uses
patch_ib(i)%massbutiis undefined here (loop variable declared later). Should bepatch_ib(ib_marker)%mass.Apply this diff:
- patch_ib(ib_marker)%moment = patch_ib(i)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wp + patch_ib(ib_marker)%moment = patch_ib(ib_marker)%mass*(patch_ib(ib_marker)%length_x**2 + patch_ib(ib_marker)%length_y**2)/6._wpBased on past review comments.
1126-1126: Critical: Incorrect distance calculation.
sum(vector_to_axis)sums components (x + y + z), not the squared distance. For moment of inertia, this should besum(vector_to_axis**2)to compute r² = x² + y² + z².Apply this diff:
- distance_to_axis = sum(vector_to_axis) ! saves the distance to the axis squared + distance_to_axis = sum(vector_to_axis**2) ! saves the distance to the axis squaredBased on past review comments.
🧹 Nitpick comments (1)
src/simulation/m_ibm.fpp (1)
199-212: Consider GPU parallelization for nested loop.This 4-level nested loop lacks GPU parallelization. For
src/simulation/**/*.fppfiles, tight loops should use the$:GPU_PARALLEL_LOOPmacro for performance.Consider applying this pattern:
+ $:GPU_PARALLEL_LOOP(private='[patch_id,j,k,l]', collapse=3) do patch_id = 1, num_ibs if (patch_ib(patch_id)%moving_ibm == 2) then do j = 0, m do k = 0, n do l = 0, p if (ib_markers%sf(j, k, l) == patch_id) then q_prim_vf(E_idx)%sf(j, k, l) = 1._wp end if end do end do end do end if end do + $:END_GPU_PARALLEL_LOOP()As per coding guidelines for GPU code in simulation modules.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
src/common/m_mpi_common.fpp(1 hunks)src/simulation/m_ibm.fpp(4 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
**/*.{fpp,f90}: Use 2-space indentation; continuation lines align beneath &
Use lower-case keywords and intrinsics (do, end subroutine, etc.)
Name modules with m_ pattern (e.g., m_transport)
Name public subroutines with s_ pattern (e.g., s_compute_flux)
Name public functions with f_ pattern
Keep subroutine size ≤ 500 lines, helper subroutines ≤ 150 lines, functions ≤ 100 lines, files ≤ 1000 lines
Limit routine arguments to ≤ 6; use derived-type params struct if more are needed
Forbid goto statements (except in legacy code), COMMON blocks, and save globals
Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Call s_mpi_abort() for errors, never use stop or error stop
**/*.{fpp,f90}: Indent 2 spaces; continuation lines align under&
Use lower-case keywords and intrinsics (do,end subroutine, etc.)
Name modules withm_<feature>prefix (e.g.,m_transport)
Name public subroutines ass_<verb>_<noun>(e.g.,s_compute_flux) and functions asf_<verb>_<noun>
Keep private helpers in the module; avoid nested procedures
Enforce size limits: subroutine ≤ 500 lines, helper ≤ 150, function ≤ 100, module/file ≤ 1000
Limit subroutines to ≤ 6 arguments; otherwise pass a derived-type 'params' struct
Avoidgotostatements (except unavoidable legacy); avoid global state (COMMON,save)
Every variable must haveintent(in|out|inout)specification and appropriatedimension/allocatable/pointer
Uses_mpi_abort(<msg>)for error termination instead ofstop
Use!>style documentation for header comments; follow Doxygen Fortran format with!! @paramand!! @returntags
Useimplicit nonestatement in all modules
Useprivatedeclaration followed by explicitpublicexports in modules
Use derived types with pointers for encapsulation (e.g.,pointer, dimension(:,:,:) => null())
Usepureandelementalattributes for side-effect-free functions; combine them for array ...
Files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
src/simulation/**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
src/simulation/**/*.{fpp,f90}: Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Avoid stop/error stop inside GPU device code
Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
src/simulation/**/*.{fpp,f90}: Mark GPU-callable helpers with$:GPU_ROUTINE(function_name='...', parallelism='[seq]')immediately after declaration
Do not use OpenACC or OpenMP directives directly; use Fypp macros fromsrc/common/include/parallel_macros.fppinstead
Wrap tight loops with$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')macro; addcollapse=nfor safe nested loop merging
Declare loop-local variables withprivate='[...]'in GPU parallel loop macros
Allocate large arrays withmanagedor move them into a persistent$:GPU_ENTER_DATA(...)region at start-up
Do not placestoporerror stopinside device code
Files:
src/simulation/m_ibm.fpp
src/**/*.fpp
📄 CodeRabbit inference engine (.cursor/rules/mfc-agent-rules.mdc)
src/**/*.fpp: Use.fppfile extension for Fypp preprocessed files; CMake transpiles them to.f90
Start module files with Fypp include for macros:#:include 'macros.fpp'
Use the fyppASSERTmacro for validating conditions:@:ASSERT(predicate, message)
Use fypp macro@:ALLOCATE(var1, var2)for device-aware allocation instead of standard Fortranallocate
Use fypp macro@:DEALLOCATE(var1, var2)for device-aware deallocation instead of standard Fortrandeallocate
Files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
🧠 Learnings (17)
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight loops with `$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')` macro; add `collapse=n` for safe nested loop merging
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Avoid stop/error stop inside GPU device code
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Every variable must have `intent(in|out|inout)` specification and appropriate `dimension` / `allocatable` / `pointer`
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Declare loop-local variables with `private='[...]'` in GPU parallel loop macros
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Favor array operations over explicit loops when possible; use `collapse=N` directive to optimize nested loops
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Avoid `goto` statements (except unavoidable legacy); avoid global state (`COMMON`, `save`)
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Each variable should have one clear purpose; do not use the same variable for multiple purposes
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Do not use OpenACC or OpenMP directives directly; use Fypp macros from `src/common/include/parallel_macros.fpp` instead
Applied to files:
src/simulation/m_ibm.fppsrc/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large arrays with `managed` or move them into a persistent `$:GPU_ENTER_DATA(...)` region at start-up
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Mark GPU-callable helpers with `$:GPU_ROUTINE(function_name='...', parallelism='[seq]')` immediately after declaration
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Use `private` declaration followed by explicit `public` exports in modules
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Use `s_mpi_abort(<msg>)` for error termination instead of `stop`
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Call s_mpi_abort(<msg>) for errors, never use stop or error stop
Applied to files:
src/common/m_mpi_common.fpp
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (14)
- GitHub Check: Code Cleanliness Check
- GitHub Check: Coverage Test on CodeCov
- GitHub Check: Self Hosted (gpu, acc, gt)
- GitHub Check: Github (ubuntu, no-mpi, single, no-debug, false)
- GitHub Check: Github (macos, mpi, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, debug, false)
- GitHub Check: Github (ubuntu, mpi, debug, true)
- GitHub Check: Self Hosted (cpu, none, gt)
- GitHub Check: Github (ubuntu, mpi, no-debug, false)
- GitHub Check: Github (macos, mpi, debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, true)
- GitHub Check: Self Hosted (gpu, omp, gt)
- GitHub Check: cubic · AI code reviewer
- GitHub Check: Build & Publish
🔇 Additional comments (1)
src/common/m_mpi_common.fpp (1)
495-496: LGTM! Non-MPI fallback correctly implemented.The serial execution path correctly assigns
var_loctovar_glb, addressing the concern raised in past review comments about missing non-MPI fallback.
There was a problem hiding this comment.
Actionable comments posted: 1
♻️ Duplicate comments (4)
tests/127A967A/golden.txt (1)
6-6: Normalize negative zero to positive zero for test robustness.Line 6 contains many
-0.0values mixed with0.0values and small finite magnitudes. For canonical floating-point representation and test stability, replace all occurrences of-0.0with0.0. This aligns with the prior review suggestion for similar test data and ensures the test isn't sensitive to sign-of-zero variations that can arise from different numerical implementations.-D/cons.3.00.000050.dat -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -5e-14 -3.9e-13 -1.67e-12 -3.223e-11 -1.59892e-09 -1.681843e-08 -1.3856516e-07 -1.19948516e-06 -2.46667983e-06 4.7305218e-07 2.66766224e-06 6.2837466e-07 5.273083e-08 1.96152e-09 -5.4952e-10 -4.108e-11 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 -0.0 ... +D/cons.3.00.000050.dat 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 -5e-14 -3.9e-13 -1.67e-12 -3.223e-11 -1.59892e-09 -1.681843e-08 -1.3856516e-07 -1.19948516e-06 -2.46667983e-06 4.7305218e-07 2.66766224e-06 6.2837466e-07 5.273083e-08 1.96152e-09 -5.4952e-10 -4.108e-11 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ...Generate a verification script to apply this normalization:
#!/bin/bash # Replace all `-0.0` with `0.0` in the golden test file sed -i 's/ -0\.0 / 0.0 /g' tests/127A967A/golden.txt # Also handle cases at line start/end sed -i 's/^-0\.0 /0.0 /' tests/127A967A/golden.txt sed -i 's/ -0\.0$/ 0.0/' tests/127A967A/golden.txtsrc/common/m_mpi_common.fpp (1)
479-481: Intent mismatch for in-place MPI reduction.
var_glbis declaredintent(out), but whenloc(var_loc) == loc(var_glb)the code usesMPI_IN_PLACE, which requires the buffer content to be valid on entry. Withintent(out), the compiler may leave the actual argument undefined before the call, making in-place reduction unsafe.- real(wp), dimension(:, :), intent(out) :: var_glb + real(wp), dimension(:, :), intent(inout) :: var_glbsrc/simulation/m_ibm.fpp (2)
259-269: Division-by-zero risks remain unguarded.The expression on line 267 divides by
pres_IPandpatch_ib(patch_id)%mass, and the denominator(1._wp - ...)can also approach zero. Previous reviews flagged this. Guard these divisions:! set the pressure if (patch_ib(patch_id)%moving_ibm == 0) then q_prim_vf(E_idx)%sf(j, k, l) = pres_IP else - q_prim_vf(E_idx)%sf(j, k, l) = 0._wp - $:GPU_LOOP(parallelism='[seq]') - do q = 1, num_fluids - q_prim_vf(E_idx)%sf(j, k, l) = q_prim_vf(E_idx)%sf(j, k, l) + pres_IP/(1._wp - 2._wp*abs(levelset%sf(j, k, l, patch_id)*alpha_rho_IP(q)/pres_IP)*dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j, k, l, patch_id, :))) - end do + if (abs(pres_IP) < 1.e-16_wp .or. abs(patch_ib(patch_id)%mass) < 1.e-16_wp) then + q_prim_vf(E_idx)%sf(j, k, l) = pres_IP + else + q_prim_vf(E_idx)%sf(j, k, l) = 0._wp + $:GPU_LOOP(parallelism='[seq]') + do q = 1, num_fluids + buf = 1._wp - 2._wp*abs(levelset%sf(j, k, l, patch_id)*alpha_rho_IP(q)/pres_IP)* & + dot_product(patch_ib(patch_id)%force/patch_ib(patch_id)%mass, levelset_norm%sf(j, k, l, patch_id, :)) + if (abs(buf) > 1.e-16_wp) then + q_prim_vf(E_idx)%sf(j, k, l) = q_prim_vf(E_idx)%sf(j, k, l) + pres_IP/buf + else + q_prim_vf(E_idx)%sf(j, k, l) = q_prim_vf(E_idx)%sf(j, k, l) + pres_IP + end if + end do + end if end if
1077-1094: Optional argument handling and normalization error.Two issues in this section:
Line 1079:
axisis declaredoptionalbut lacksintentspecification. Per coding guidelines, every argument must have explicit intent.Line 1093: Normalization uses
sqrt(sum(axis))which sums components, not their squares. Should besqrt(sum(axis**2))to compute the magnitude.-subroutine s_compute_moment_of_inertia(ib_marker, axis) +subroutine s_compute_moment_of_inertia(ib_marker, axis_in) - real(wp), dimension(3), optional :: axis !< the axis about which we compute the moment. Only required in 3D. + real(wp), dimension(3), intent(in), optional :: axis_in !< the axis about which we compute the moment. Only required in 3D. integer, intent(in) :: ib_marker real(wp) :: moment, distance_to_axis, cell_volume - real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis + real(wp), dimension(3) :: position, closest_point_along_axis, vector_to_axis, axis integer :: i, j, k, count if (p == 0) then - axis = [0, 1, 0] - else if (sqrt(sum(axis**2)) == 0) then + axis = [0._wp, 0._wp, 1._wp] + else if (.not. present(axis_in) .or. sqrt(sum(axis_in**2)) < 1.e-16_wp) then ! if the object is not actually rotating at this time, return a dummy value and exit patch_ib(ib_marker)%moment = 1._wp return else - axis = axis/sqrt(sum(axis)) + axis = axis_in/sqrt(sum(axis_in**2)) end if
🧹 Nitpick comments (1)
src/simulation/m_ibm.fpp (1)
1042-1046: Atomic update on array slices may not be element-wise atomic.GPU atomic operations on array slices like
forces(ib_idx, :)may not atomically update all three components together. Consider expanding to separate atomic updates per component to ensure correctness:! Update the force values atomically to prevent race conditions -$:GPU_ATOMIC(atomic='update') -forces(ib_idx, :) = forces(ib_idx, :) - (pressure_divergence*cell_volume) -$:GPU_ATOMIC(atomic='update') -torques(ib_idx, :) = torques(ib_idx, :) - (cross_product(radial_vector, pressure_divergence)*cell_volume) +$:GPU_ATOMIC(atomic='update') +forces(ib_idx, 1) = forces(ib_idx, 1) - pressure_divergence(1)*cell_volume +$:GPU_ATOMIC(atomic='update') +forces(ib_idx, 2) = forces(ib_idx, 2) - pressure_divergence(2)*cell_volume +$:GPU_ATOMIC(atomic='update') +forces(ib_idx, 3) = forces(ib_idx, 3) - pressure_divergence(3)*cell_volume +torque_contrib = cross_product(radial_vector, pressure_divergence)*cell_volume +$:GPU_ATOMIC(atomic='update') +torques(ib_idx, 1) = torques(ib_idx, 1) - torque_contrib(1) +$:GPU_ATOMIC(atomic='update') +torques(ib_idx, 2) = torques(ib_idx, 2) - torque_contrib(2) +$:GPU_ATOMIC(atomic='update') +torques(ib_idx, 3) = torques(ib_idx, 3) - torque_contrib(3)
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (5)
examples/2D_mibm_cylinder_in_cross_flow/case.py(1 hunks)src/common/m_mpi_common.fpp(1 hunks)src/simulation/m_ibm.fpp(4 hunks)tests/127A967A/golden-metadata.txt(1 hunks)tests/127A967A/golden.txt(1 hunks)
🚧 Files skipped from review as they are similar to previous changes (2)
- examples/2D_mibm_cylinder_in_cross_flow/case.py
- tests/127A967A/golden-metadata.txt
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
**/*.{fpp,f90}: Use 2-space indentation; continuation lines align beneath &
Use lower-case keywords and intrinsics (do, end subroutine, etc.)
Name modules with m_ pattern (e.g., m_transport)
Name public subroutines with s_ pattern (e.g., s_compute_flux)
Name public functions with f_ pattern
Keep subroutine size ≤ 500 lines, helper subroutines ≤ 150 lines, functions ≤ 100 lines, files ≤ 1000 lines
Limit routine arguments to ≤ 6; use derived-type params struct if more are needed
Forbid goto statements (except in legacy code), COMMON blocks, and save globals
Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Call s_mpi_abort() for errors, never use stop or error stop
**/*.{fpp,f90}: Indent 2 spaces; continuation lines align under&
Use lower-case keywords and intrinsics (do,end subroutine, etc.)
Name modules withm_<feature>prefix (e.g.,m_transport)
Name public subroutines ass_<verb>_<noun>(e.g.,s_compute_flux) and functions asf_<verb>_<noun>
Keep private helpers in the module; avoid nested procedures
Enforce size limits: subroutine ≤ 500 lines, helper ≤ 150, function ≤ 100, module/file ≤ 1000
Limit subroutines to ≤ 6 arguments; otherwise pass a derived-type 'params' struct
Avoidgotostatements (except unavoidable legacy); avoid global state (COMMON,save)
Every variable must haveintent(in|out|inout)specification and appropriatedimension/allocatable/pointer
Uses_mpi_abort(<msg>)for error termination instead ofstop
Use!>style documentation for header comments; follow Doxygen Fortran format with!! @paramand!! @returntags
Useimplicit nonestatement in all modules
Useprivatedeclaration followed by explicitpublicexports in modules
Use derived types with pointers for encapsulation (e.g.,pointer, dimension(:,:,:) => null())
Usepureandelementalattributes for side-effect-free functions; combine them for array ...
Files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
src/**/*.fpp
📄 CodeRabbit inference engine (.cursor/rules/mfc-agent-rules.mdc)
src/**/*.fpp: Use.fppfile extension for Fypp preprocessed files; CMake transpiles them to.f90
Start module files with Fypp include for macros:#:include 'macros.fpp'
Use the fyppASSERTmacro for validating conditions:@:ASSERT(predicate, message)
Use fypp macro@:ALLOCATE(var1, var2)for device-aware allocation instead of standard Fortranallocate
Use fypp macro@:DEALLOCATE(var1, var2)for device-aware deallocation instead of standard Fortrandeallocate
Files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
src/simulation/**/*.{fpp,f90}
📄 CodeRabbit inference engine (.github/copilot-instructions.md)
src/simulation/**/*.{fpp,f90}: Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Avoid stop/error stop inside GPU device code
Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
src/simulation/**/*.{fpp,f90}: Mark GPU-callable helpers with$:GPU_ROUTINE(function_name='...', parallelism='[seq]')immediately after declaration
Do not use OpenACC or OpenMP directives directly; use Fypp macros fromsrc/common/include/parallel_macros.fppinstead
Wrap tight loops with$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')macro; addcollapse=nfor safe nested loop merging
Declare loop-local variables withprivate='[...]'in GPU parallel loop macros
Allocate large arrays withmanagedor move them into a persistent$:GPU_ENTER_DATA(...)region at start-up
Do not placestoporerror stopinside device code
Files:
src/simulation/m_ibm.fpp
🧠 Learnings (22)
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Add and update focused tests for changes; ensure no regressions in golden test outputs without clear justification
Applied to files:
tests/127A967A/golden.txt
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight GPU loops with !$acc parallel loop gang vector default(present) reduction(...); add collapse(n) when safe; declare loop-local variables with private(...)
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Use `private` declaration followed by explicit `public` exports in modules
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Declare loop-local variables with `private='[...]'` in GPU parallel loop macros
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Wrap tight loops with `$:GPU_PARALLEL_FOR(private='[...]', copy='[...]')` macro; add `collapse=n` for safe nested loop merging
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Do not use OpenACC or OpenMP directives directly; use Fypp macros from `src/common/include/parallel_macros.fpp` instead
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Keep private helpers in the module; avoid nested procedures
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Every variable must have `intent(in|out|inout)` specification and appropriate `dimension` / `allocatable` / `pointer`
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Every argument must have explicit intent; use dimension/allocatable/pointer as appropriate
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Each variable should have one clear purpose; do not use the same variable for multiple purposes
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Use `s_mpi_abort(<msg>)` for error termination instead of `stop`
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Avoid `goto` statements (except unavoidable legacy); avoid global state (`COMMON`, `save`)
Applied to files:
src/common/m_mpi_common.fppsrc/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/**/*.fpp : Use fypp macro `@:ALLOCATE(var1, var2)` for device-aware allocation instead of standard Fortran `allocate`
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Call s_mpi_abort(<msg>) for errors, never use stop or error stop
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to **/*.{fpp,f90} : Forbid goto statements (except in legacy code), COMMON blocks, and save globals
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/**/*.fpp : Use fypp macro `@:DEALLOCATE(var1, var2)` for device-aware deallocation instead of standard Fortran `deallocate`
Applied to files:
src/common/m_mpi_common.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Avoid stop/error stop inside GPU device code
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to **/*.{fpp,f90} : Favor array operations over explicit loops when possible; use `collapse=N` directive to optimize nested loops
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large arrays with `managed` or move them into a persistent `$:GPU_ENTER_DATA(...)` region at start-up
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Ensure GPU code compiles with Cray ftn, NVIDIA nvfortran, GNU gfortran, and Intel ifx/ifort compilers
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:46.909Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .cursor/rules/mfc-agent-rules.mdc:0-0
Timestamp: 2025-11-24T21:50:46.909Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Mark GPU-callable helpers with `$:GPU_ROUTINE(function_name='...', parallelism='[seq]')` immediately after declaration
Applied to files:
src/simulation/m_ibm.fpp
📚 Learning: 2025-11-24T21:50:16.713Z
Learnt from: CR
Repo: MFlowCode/MFC PR: 0
File: .github/copilot-instructions.md:0-0
Timestamp: 2025-11-24T21:50:16.713Z
Learning: Applies to src/simulation/**/*.{fpp,f90} : Allocate large GPU arrays with managed memory or move them into persistent !$acc enter data regions at start-up
Applied to files:
src/simulation/m_ibm.fpp
🧬 Code graph analysis (1)
src/simulation/m_ibm.fpp (1)
toolchain/mfc/test/cases.py (1)
alter_ib(377-422)
🪛 GitHub Actions: Pretty
src/simulation/m_ibm.fpp
[error] 198-210: MFC formatting changed code. The formatting step exited with code 1 (diffs present). Run the formatting tool again or commit the changes. Command: './mfc.sh format -j $(nproc)'
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (14)
- GitHub Check: Coverage Test on CodeCov
- GitHub Check: Github (ubuntu, mpi, no-debug, true)
- GitHub Check: Github (macos, mpi, debug, false)
- GitHub Check: Github (ubuntu, no-mpi, single, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, debug, false)
- GitHub Check: Github (ubuntu, mpi, no-debug, false)
- GitHub Check: Github (ubuntu, mpi, debug, true)
- GitHub Check: Github (macos, mpi, no-debug, false)
- GitHub Check: Self Hosted (gpu, acc, gt)
- GitHub Check: Self Hosted (gpu, omp, gt)
- GitHub Check: Self Hosted (cpu, none, gt)
- GitHub Check: Code Cleanliness Check
- GitHub Check: cubic · AI code reviewer
- GitHub Check: Build & Publish
🔇 Additional comments (3)
src/common/m_mpi_common.fpp (1)
475-499: LGTM on the overall implementation.The new
s_mpi_allreduce_vectors_sumsubroutine correctly handles both in-place and separate buffer MPI reductions, with proper non-MPI fallback. The logic mirrors existing patterns in the module.src/simulation/m_ibm.fpp (2)
1096-1144: LGTM on moment of inertia calculations.The analytic formulas for circle (I = ½mr²), rectangle, and sphere (I = ⅖mr²) are correct. The numerical approximation branch properly accumulates
r²contributions viadot_product(vector_to_axis, vector_to_axis). Previous issues with loop bounds and variable names have been addressed.
199-210: Unable to verify due to repository access failure. The review comment raises two specific concerns that require codebase examination:
Formatting issue: Claims CI flagged formatting changes needed. Verification requires inspecting the actual file and checking
git diffoutput.Logic scope mismatch: Suggests the comment "Moving IBM interior Pressure Values" is misleading because the condition
ib_markers%sf(j, k, l) /= 0applies to all IB interiors, not just moving ones. Verification requires understanding whether this code is within a moving IBM-specific subroutine context.Without repository access, both claims cannot be confirmed or refuted. Manual verification by examining the file context around lines 199-210 and checking the CI logs is recommended.
| patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB | ||
| end do | ||
| print *, forces(1, 1:2) |
There was a problem hiding this comment.
Remove debug print statement.
This debug output will produce excessive noise in production runs. Remove or guard with a debug flag.
- print *, forces(1, 1:2)📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| print *, forces(1, 1:2) |
🤖 Prompt for AI Agents
In src/simulation/m_ibm.fpp around line 1064, there is a debug print statement
"print *, forces(1, 1:2)" that will produce excessive runtime output; remove
this line or wrap it in a conditional that checks a debug/logging flag (e.g., if
(debug) then ... end if) so it only prints when debugging is enabled, and ensure
any added flag is initialized/controlled by existing logging or
compilation/debug configuration.
There was a problem hiding this comment.
2 issues found across 5 files (changes from recent commits).
Prompt for AI agents (all 2 issues)
Check if these issues are valid — if so, understand the root cause of each and fix them.
<file name="src/simulation/m_ibm.fpp">
<violation number="1" location="src/simulation/m_ibm.fpp:1064">
P2: Debug print statement should be removed before merging. This will pollute output logs and may cause performance issues in parallel execution.</violation>
</file>
<file name="examples/2D_mibm_cylinder_in_cross_flow/case.py">
<violation number="1" location="examples/2D_mibm_cylinder_in_cross_flow/case.py:98">
P2: Misleading comment: The comment says `# z-axis rotation` but this parameter sets the mass of the immersed boundary, not rotation. This appears to be a copy-paste error from the previous line.</violation>
</file>
Reply to cubic to teach it or ask questions. Re-run a review with @cubic-dev-ai review this PR
| patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB | ||
| end do | ||
| print *, forces(1, 1:2) |
There was a problem hiding this comment.
P2: Debug print statement should be removed before merging. This will pollute output logs and may cause performance issues in parallel execution.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At src/simulation/m_ibm.fpp, line 1064:
<comment>Debug print statement should be removed before merging. This will pollute output logs and may cause performance issues in parallel execution.</comment>
<file context>
@@ -1059,6 +1061,8 @@ contains
patch_ib(i)%torque(:) = matmul(patch_ib(i)%rotation_matrix_inverse, torques(i, :)) ! torques must be computed in the local coordinates of the IB
end do
+ print *, forces(1, 1:2)
+
end subroutine s_compute_ib_forces
</file context>
| "patch_ib(1)%angular_vel(1)": 0.0, # x-axis rotational velocity in radians per second | ||
| "patch_ib(1)%angular_vel(2)": 0.0, # y-axis rotation | ||
| "patch_ib(1)%angular_vel(3)": 0.0, # z-axis rotation | ||
| "patch_ib(1)%mass": 1.0e-6, # z-axis rotation |
There was a problem hiding this comment.
P2: Misleading comment: The comment says # z-axis rotation but this parameter sets the mass of the immersed boundary, not rotation. This appears to be a copy-paste error from the previous line.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At examples/2D_mibm_cylinder_in_cross_flow/case.py, line 98:
<comment>Misleading comment: The comment says `# z-axis rotation` but this parameter sets the mass of the immersed boundary, not rotation. This appears to be a copy-paste error from the previous line.</comment>
<file context>
@@ -95,7 +95,7 @@
"patch_ib(1)%angular_vel(2)": 0.0, # y-axis rotation
"patch_ib(1)%angular_vel(3)": 0.0, # z-axis rotation
- "patch_ib(1)%mass": 0.0001, # z-axis rotation
+ "patch_ib(1)%mass": 1.0e-6, # z-axis rotation
# Fluids Physical Parameters
"fluid_pp(1)%gamma": 1.0e00 / (gam_a - 1.0e00), # 2.50(Not 1.40)
</file context>
| "patch_ib(1)%mass": 1.0e-6, # z-axis rotation | |
| "patch_ib(1)%mass": 1.0e-6, # mass of the immersed boundary |
1 participant
User description
User description
Description
This PR has the contents for an initial 2-way fluid-structure interaction addition to the code, only considering drag due to to pressure. There will be another PR following that adds viscosity in which we include the contributions due to the viscous stress tensor. This PR adds a new method for computing the interior pressure to an IB which reduces to the same method as before when the object is stationary, a method for numerically or analytically setting the moment of inertia, and a way to compute the forces and toques applied to an object from the pressure gradient.
There
Type of change
Please delete options that are not relevant.
Scope
If you cannot check the above box, please split your PR into multiple PRs that each have a common goal.
How Has This Been Tested?
I have run
Test Configuration:
Checklist
docs/)examples/that demonstrate my new feature performing as expected.They run to completion and demonstrate "interesting physics"
./mfc.sh formatbefore committing my codeIf your code changes any code source files (anything in
src/simulation)To make sure the code is performing as expected on GPU devices, I have:
nvtxranges so that they can be identified in profiles./mfc.sh run XXXX --gpu -t simulation --nsys, and have attached the output file (.nsys-rep) and plain text results to this PR./mfc.sh run XXXX --gpu -t simulation --rsys --hip-trace, and have attached the output file and plain text results to this PR.PR Type
Enhancement
Description
Implements two-way fluid-structure interaction via volume integration method
Adds force and torque computation on immersed boundaries from pressure gradients
Introduces moment of inertia calculation with analytical and numerical methods
Extends time-stepping integration to apply computed forces to moving objects
Adds example case demonstrating cylinder in cross-flow with two-way coupling
Diagram Walkthrough
File Walkthrough
5 files
Add force computation and moment of inertia methodsIntegrate force/torque into time-stepping schemeAdd MPI reduction for vector arraysAdd mass, moment, force, torque fields to IB typeInitialize mass and moment of inertia fields3 files
Broadcast IB mass parameter across MPI ranksAdd mass parameter to IB configuration parsingAdd HiPerGator cluster option to bootstrap2 files
New example case for cylinder in cross-flowDocument moving IB parameters and two-way coupling1 files
Add orange color support to shell utilities1 files
Add test metadata for two-way coupling case1 files
CodeAnt-AI Description
Enable inviscid two-way coupling: fluid pressure drives immersed-boundary translation and rotation
What Changed
Impact
✅ Two-way coupled immersed boundaries✅ Fluid pressure drives IB motion✅ Automatic moment-of-inertia estimation for moving objects💡 Usage Guide
Checking Your Pull Request
Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.
Talking to CodeAnt AI
Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:
This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.
Example
Preserve Org Learnings with CodeAnt
You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:
This helps CodeAnt AI learn and adapt to your team's coding style and standards.
Example
Retrigger review
Ask CodeAnt AI to review the PR again, by typing:
Check Your Repository Health
To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.
Summary by CodeRabbit
New Features
Documentation
Tests
Infrastructure
✏️ Tip: You can customize this high-level summary in your review settings.