November 2023 - petsc-dev - mailman3.cels.anl.gov

Re: [petsc-dev] DMPlex redistribution
by Adrian Croucher 30 Nov '23

30 Nov '23

hi Matt, I last tested this with PETSc v3.19 and that was still broken, so unless there was any progress on it in 3.20 I suspect that's still the case. So for our code we are still basically stuck with using PETSc 3.15 at the moment. Would be nice to be able to get that up to date again. Cheers, Adrian On 1/12/23 9:00 am, Matthew Knepley wrote: > On Tue, Mar 28, 2023 at 6:12 PM Adrian Croucher > <a.croucher(a)auckland.ac.nz> wrote: > > > Yes, I think so. The partitioning groups cells, and does not > expect cells > > in the overlap. We could generalize that, which is probably > > the right thing to do when redistributing meshes with overlap. I > will look > > at why it is complaining. Maybe I can just put a guard instead > > of an error in there. > > Yes, and it might be useful to be able to disable that check for > cells > in the overlap, for cases like this. > > > Okay, I lost this in my Inbox. Is redistribution of overlapping meshes > still broken? If so, we should make a test so that I can fix it. > > Thanks! > > Matt > > - Adrian > > -- > Dr Adrian Croucher > Senior Research Fellow > Department of Engineering Science > University of Auckland, New Zealand > email: a.croucher(a)auckland.ac.nz > tel: +64 (0)9 923 4611 > > > > -- > What most experimenters take for granted before they begin their > experiments is infinitely more interesting than any results to which > their experiments lead. > -- Norbert Wiener > > https://www.cse.buffalo.edu/~knepley/ > <http://www.cse.buffalo.edu/~knepley/> -- Dr Adrian Croucher Senior Research Fellow Department of Engineering Science University of Auckland, New Zealand email:[email protected] tel: +64 (0)9 923 4611

1 0

Re: [petsc-dev] DMPlex redistribution
by Matthew Knepley 30 Nov '23

30 Nov '23

On Tue, Mar 28, 2023 at 6:12 PM Adrian Croucher <a.croucher(a)auckland.ac.nz> wrote: > > Yes, I think so. The partitioning groups cells, and does not expect cells > > in the overlap. We could generalize that, which is probably > > the right thing to do when redistributing meshes with overlap. I will > look > > at why it is complaining. Maybe I can just put a guard instead > > of an error in there. > > Yes, and it might be useful to be able to disable that check for cells > in the overlap, for cases like this. > Okay, I lost this in my Inbox. Is redistribution of overlapping meshes still broken? If so, we should make a test so that I can fix it. Thanks! Matt > - Adrian > > -- > Dr Adrian Croucher > Senior Research Fellow > Department of Engineering Science > University of Auckland, New Zealand > email: a.croucher(a)auckland.ac.nz > tel: +64 (0)9 923 4611 > > -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>

1 0

Re: [petsc-dev] PetscHMapI passing
by Matthew Knepley 28 Nov '23

28 Nov '23

On Tue, Nov 28, 2023 at 5:23 PM Brandon Denton <bldenton(a)buffalo.edu> wrote: > Good Afternoon, > > Ok. I think I've developed a solution that meets everyone's interest. > Please review the below and let me know if you have any questions or > comments. > > Quick background. The objective of this function is to provide user > friendly access to geometric data associated with a DMPlex with a geometry > CAD model attached. This function will replace the originally conceived > PetscObjectQuery() method. > Yes, that is a good start. We might eventually want an object, but this is great to experiment with. I like it. Thanks, Matt > *Proposed Function Signature:* > DMPlexGetGeomCntrlPntAndWeightData(DM dm, PetscHMapI *cpHashTable, > PetscInt *cpCoordDataLength, PetscScalar **cpCoordData, PetscInt > *maxNumEquiv, Mat *cpEquiv, PetscHMapI *wHashTable, PetscInt *wDataLength, > PetscScalar **wData) > > The only input it the DM. All others are outputs. > > *Proposed Call* > PetscScalar *cpCoordData, *wData; > PetscInt cpCoordDataLength, wDataLength, maxNumEquiv; > PetscHMapI cpHashTable, wHashTable; > Mat cpEquiv; > > PetscCall(DMPlexGetGeomCntrlPntAndWeightData(dmNozzle, &cpHashTable, > &cpCoordDataLength, &cpCoordData, &maxNumEquiv, &cpEquiv, &wHashTable, > &wDataLength, &wData)); > > *Proposed Function Added to plexegads.c* > DMPlexGetGeomCntrlPntAndWeightData(DM dm, PetscHMapI *cpHashTable, > PetscInt *cpCoordDataLength, PetscScalar **cpCoordData, PetscInt > *maxNumEquiv, Mat *cpEquiv, PetscHMapI *wHashTable, PetscInt *wDataLength, > PetscScalar **wData) > { > PetscFunctionBeginHot; > #ifdef PETSC_HAVE_EGADS > PetscContainer modelObj, cpHashTableObj, wHashTableObj, cpCoordDataObj, > wDataObj, cpCoordDataLengthObj, wDataLengthObj, cpEquivObj, maxNumRelateObj; > PetscInt *cpCoordDataLengthPtr, *wDataLengthPtr, *maxNumEquivPtr; > PetscHMapI cpHashTableTemp, wHashTableTemp; > PetscScalar *cpCoordDataTemp, *wDataTemp; > Mat cpEquivTemp; > > /* Determine which type of EGADS model is attached to the DM */ > PetscCall(PetscObjectQuery((PetscObject) dm, "EGADS Model", (PetscObject > *) &modelObj)); > if (!modelObj) { > PetscCall(PetscObjectQuery((PetscObject) dm, "EGADSlite Model", > (PetscObject *) &modelObj)); > } > > if (!modelObj) {PetscFunctionReturn(PETSC_SUCCESS);} > > // Look to see if DM has Container for Geometry Control Point Data > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Hash Table", > (PetscObject *)&cpHashTableObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Coordinates", > (PetscObject *)&cpCoordDataObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Coordinate > Data Length", (PetscObject *)&cpCoordDataLengthObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Weights Hash > Table", (PetscObject *)&wHashTableObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Weight Data", > (PetscObject *)&wDataObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Weight Data > Length", (PetscObject *)&wDataLengthObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Equivalancy > Matrix", (PetscObject *)&cpEquivObj)); > PetscCall(PetscObjectQuery((PetscObject)dm, "Maximum Number Control > Point Equivalency", (PetscObject *)&maxNumRelateObj)); > > // Get attached EGADS model Control Point and Weights Hash Tables and > Data Arrays (pointer) > PetscCall(PetscContainerGetPointer(cpHashTableObj, (void > **)&cpHashTableTemp)); > PetscCall(PetscContainerGetPointer(cpCoordDataObj, (void > **)&cpCoordDataTemp)); > PetscCall(PetscContainerGetPointer(cpCoordDataLengthObj, (void > **)&cpCoordDataLengthPtr)); > PetscCall(PetscContainerGetPointer(wHashTableObj, (void > **)&wHashTableTemp)); > PetscCall(PetscContainerGetPointer(wDataObj, (void **)&wDataTemp)); > PetscCall(PetscContainerGetPointer(wDataLengthObj, (void > **)&wDataLengthPtr)); > PetscCall(PetscContainerGetPointer(cpEquivObj, (void **)&cpEquivTemp)); > PetscCall(PetscContainerGetPointer(maxNumRelateObj, (void > **)&maxNumEquivPtr)); > > *cpCoordDataLength = *cpCoordDataLengthPtr; > *wDataLength = *wDataLengthPtr; > *maxNumEquiv = *maxNumEquivPtr; > *cpEquiv = cpEquivTemp; > *cpHashTable = cpHashTableTemp; > *wHashTable = wHashTableTemp; > *cpCoordData = cpCoordDataTemp; > *wData = wDataTemp; > #endif > > PetscFunctionReturn(PETSC_SUCCESS); > } > > ------------------------------ > *From:* Matthew Knepley <knepley(a)gmail.com> > *Sent:* Tuesday, November 28, 2023 1:26 PM > *To:* Barry Smith <bsmith(a)petsc.dev> > *Cc:* Brandon Denton <bldenton(a)buffalo.edu>; petsc-dev(a)mcs.anl.gov < > petsc-dev(a)mcs.anl.gov> > *Subject:* Re: [petsc-dev] PetscHMapI passing > > On Mon, Nov 27, 2023 at 1:00 PM Barry Smith <bsmith(a)petsc.dev> wrote: > > > What errors do you get? > > I am trying to develop a function where the user can specify an empty > PetscHMapI variable and the function will retrieve the information attached > to the DMPlex. However, when I try something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) > > > I don't understand the design of specifying and empty one. Why not have > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI > *hashTable) > > and return the one you have inside your container? > > > This sounds like the right design. Does that work? > > Thanks, > > Matt > > > On Nov 27, 2023, at 9:11 AM, Brandon Denton via petsc-dev < > petsc-dev(a)mcs.anl.gov> wrote: > > Good Morning, > > I am trying to develop a function in PETSc that allows for the > retrieval of a PetscHMapI contained in a PetscContainer attached to a > DMPlex. The creation of this Hash Table and its attachment to the DMPlex is > automated as part of the CAD/Discretization integration I've been > developing. Currently, the only way to get this Hash Table is through the > PetscObjectQuery() functionality. In an effort to make it easier for users > to access and use this information, I am trying to develop a function where > the user can specify an empty PetscHMapI variable and the function will > retrieve the information attached to the DMPlex. However, when I try > something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) > > I get a ton of compilation errors associated with the PetscHMapI > designation. Is there a way around this? > > Thank you. > Brandon > > > > > -- > What most experimenters take for granted before they begin their > experiments is infinitely more interesting than any results to which their > experiments lead. > -- Norbert Wiener > > https://www.cse.buffalo.edu/~knepley/ > <http://www.cse.buffalo.edu/~knepley/> > -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>

1 0

Re: [petsc-dev] PetscHMapI passing
by Brandon Denton 28 Nov '23

28 Nov '23

Good Afternoon, Ok. I think I've developed a solution that meets everyone's interest. Please review the below and let me know if you have any questions or comments. Quick background. The objective of this function is to provide user friendly access to geometric data associated with a DMPlex with a geometry CAD model attached. This function will replace the originally conceived PetscObjectQuery() method. Proposed Function Signature: DMPlexGetGeomCntrlPntAndWeightData(DM dm, PetscHMapI *cpHashTable, PetscInt *cpCoordDataLength, PetscScalar **cpCoordData, PetscInt *maxNumEquiv, Mat *cpEquiv, PetscHMapI *wHashTable, PetscInt *wDataLength, PetscScalar **wData) The only input it the DM. All others are outputs. Proposed Call PetscScalar *cpCoordData, *wData; PetscInt cpCoordDataLength, wDataLength, maxNumEquiv; PetscHMapI cpHashTable, wHashTable; Mat cpEquiv; PetscCall(DMPlexGetGeomCntrlPntAndWeightData(dmNozzle, &cpHashTable, &cpCoordDataLength, &cpCoordData, &maxNumEquiv, &cpEquiv, &wHashTable, &wDataLength, &wData)); Proposed Function Added to plexegads.c DMPlexGetGeomCntrlPntAndWeightData(DM dm, PetscHMapI *cpHashTable, PetscInt *cpCoordDataLength, PetscScalar **cpCoordData, PetscInt *maxNumEquiv, Mat *cpEquiv, PetscHMapI *wHashTable, PetscInt *wDataLength, PetscScalar **wData) { PetscFunctionBeginHot; #ifdef PETSC_HAVE_EGADS PetscContainer modelObj, cpHashTableObj, wHashTableObj, cpCoordDataObj, wDataObj, cpCoordDataLengthObj, wDataLengthObj, cpEquivObj, maxNumRelateObj; PetscInt *cpCoordDataLengthPtr, *wDataLengthPtr, *maxNumEquivPtr; PetscHMapI cpHashTableTemp, wHashTableTemp; PetscScalar *cpCoordDataTemp, *wDataTemp; Mat cpEquivTemp; /* Determine which type of EGADS model is attached to the DM */ PetscCall(PetscObjectQuery((PetscObject) dm, "EGADS Model", (PetscObject *) &modelObj)); if (!modelObj) { PetscCall(PetscObjectQuery((PetscObject) dm, "EGADSlite Model", (PetscObject *) &modelObj)); } if (!modelObj) {PetscFunctionReturn(PETSC_SUCCESS);} // Look to see if DM has Container for Geometry Control Point Data PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Hash Table", (PetscObject *)&cpHashTableObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Coordinates", (PetscObject *)&cpCoordDataObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Coordinate Data Length", (PetscObject *)&cpCoordDataLengthObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Weights Hash Table", (PetscObject *)&wHashTableObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Weight Data", (PetscObject *)&wDataObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Weight Data Length", (PetscObject *)&wDataLengthObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Control Point Equivalancy Matrix", (PetscObject *)&cpEquivObj)); PetscCall(PetscObjectQuery((PetscObject)dm, "Maximum Number Control Point Equivalency", (PetscObject *)&maxNumRelateObj)); // Get attached EGADS model Control Point and Weights Hash Tables and Data Arrays (pointer) PetscCall(PetscContainerGetPointer(cpHashTableObj, (void **)&cpHashTableTemp)); PetscCall(PetscContainerGetPointer(cpCoordDataObj, (void **)&cpCoordDataTemp)); PetscCall(PetscContainerGetPointer(cpCoordDataLengthObj, (void **)&cpCoordDataLengthPtr)); PetscCall(PetscContainerGetPointer(wHashTableObj, (void **)&wHashTableTemp)); PetscCall(PetscContainerGetPointer(wDataObj, (void **)&wDataTemp)); PetscCall(PetscContainerGetPointer(wDataLengthObj, (void **)&wDataLengthPtr)); PetscCall(PetscContainerGetPointer(cpEquivObj, (void **)&cpEquivTemp)); PetscCall(PetscContainerGetPointer(maxNumRelateObj, (void **)&maxNumEquivPtr)); *cpCoordDataLength = *cpCoordDataLengthPtr; *wDataLength = *wDataLengthPtr; *maxNumEquiv = *maxNumEquivPtr; *cpEquiv = cpEquivTemp; *cpHashTable = cpHashTableTemp; *wHashTable = wHashTableTemp; *cpCoordData = cpCoordDataTemp; *wData = wDataTemp; #endif PetscFunctionReturn(PETSC_SUCCESS); } ________________________________ From: Matthew Knepley <knepley(a)gmail.com> Sent: Tuesday, November 28, 2023 1:26 PM To: Barry Smith <bsmith(a)petsc.dev> Cc: Brandon Denton <bldenton(a)buffalo.edu>; petsc-dev(a)mcs.anl.gov <petsc-dev(a)mcs.anl.gov> Subject: Re: [petsc-dev] PetscHMapI passing On Mon, Nov 27, 2023 at 1:00 PM Barry Smith <bsmith(a)petsc.dev<mailto:[email protected]>> wrote: What errors do you get? I am trying to develop a function where the user can specify an empty PetscHMapI variable and the function will retrieve the information attached to the DMPlex. However, when I try something like ... PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) I don't understand the design of specifying and empty one. Why not have PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI *hashTable) and return the one you have inside your container? This sounds like the right design. Does that work? Thanks, Matt On Nov 27, 2023, at 9:11 AM, Brandon Denton via petsc-dev <petsc-dev(a)mcs.anl.gov<mailto:[email protected]>> wrote: Good Morning, I am trying to develop a function in PETSc that allows for the retrieval of a PetscHMapI contained in a PetscContainer attached to a DMPlex. The creation of this Hash Table and its attachment to the DMPlex is automated as part of the CAD/Discretization integration I've been developing. Currently, the only way to get this Hash Table is through the PetscObjectQuery() functionality. In an effort to make it easier for users to access and use this information, I am trying to develop a function where the user can specify an empty PetscHMapI variable and the function will retrieve the information attached to the DMPlex. However, when I try something like ... PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) I get a ton of compilation errors associated with the PetscHMapI designation. Is there a way around this? Thank you. Brandon -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener https://www.cse.buffalo.edu/~knepley/<http://www.cse.buffalo.edu/~knepley/>

1 0

Re: [petsc-dev] PetscHMapI passing
by Matthew Knepley 28 Nov '23

28 Nov '23

On Mon, Nov 27, 2023 at 1:00 PM Barry Smith <bsmith(a)petsc.dev> wrote: > > What errors do you get? > > I am trying to develop a function where the user can specify an empty > PetscHMapI variable and the function will retrieve the information attached > to the DMPlex. However, when I try something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) > > > I don't understand the design of specifying and empty one. Why not have > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI > *hashTable) > > and return the one you have inside your container? > This sounds like the right design. Does that work? Thanks, Matt > On Nov 27, 2023, at 9:11 AM, Brandon Denton via petsc-dev < > petsc-dev(a)mcs.anl.gov> wrote: > > Good Morning, > > I am trying to develop a function in PETSc that allows for the > retrieval of a PetscHMapI contained in a PetscContainer attached to a > DMPlex. The creation of this Hash Table and its attachment to the DMPlex is > automated as part of the CAD/Discretization integration I've been > developing. Currently, the only way to get this Hash Table is through the > PetscObjectQuery() functionality. In an effort to make it easier for users > to access and use this information, I am trying to develop a function where > the user can specify an empty PetscHMapI variable and the function will > retrieve the information attached to the DMPlex. However, when I try > something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) > > I get a ton of compilation errors associated with the PetscHMapI > designation. Is there a way around this? > > Thank you. > Brandon > > > -- What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead. -- Norbert Wiener https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>

1 0

Re: [petsc-dev] PetscHMapI passing
by Barry Smith 27 Nov '23

27 Nov '23

What errors do you get? > I am trying to develop a function where the user can specify an empty PetscHMapI variable and the function will retrieve the information attached to the DMPlex. However, when I try something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) I don't understand the design of specifying and empty one. Why not have PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI *hashTable) and return the one you have inside your container? > On Nov 27, 2023, at 9:11 AM, Brandon Denton via petsc-dev <petsc-dev(a)mcs.anl.gov> wrote: > > Good Morning, > > I am trying to develop a function in PETSc that allows for the retrieval of a PetscHMapI contained in a PetscContainer attached to a DMPlex. The creation of this Hash Table and its attachment to the DMPlex is automated as part of the CAD/Discretization integration I've been developing. Currently, the only way to get this Hash Table is through the PetscObjectQuery() functionality. In an effort to make it easier for users to access and use this information, I am trying to develop a function where the user can specify an empty PetscHMapI variable and the function will retrieve the information attached to the DMPlex. However, when I try something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) > > I get a ton of compilation errors associated with the PetscHMapI designation. Is there a way around this? > > Thank you. > Brandon

1 0

Re: [petsc-dev] PetscHMapI passing
by Mark Adams 27 Nov '23

27 Nov '23

To add a method to PETSc there are several places that you need to add headers and registration stuff. I find a method that looks like my method and clone that by searching (git grep) for that (DMPlexGet something) method and cloning that code everywhere that you find that method. Mark On Mon, Nov 27, 2023 at 6:35 AM Brandon Denton via petsc-dev < petsc-dev(a)mcs.anl.gov> wrote: > Good Morning, > > I am trying to develop a function in PETSc that allows for the > retrieval of a PetscHMapI contained in a PetscContainer attached to a > DMPlex. The creation of this Hash Table and its attachment to the DMPlex is > automated as part of the CAD/Discretization integration I've been > developing. Currently, the only way to get this Hash Table is through the > PetscObjectQuery() functionality. In an effort to make it easier for users > to access and use this information, I am trying to develop a function where > the user can specify an empty PetscHMapI variable and the function will > retrieve the information attached to the DMPlex. However, when I try > something like ... > > PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) > > I get a ton of compilation errors associated with the PetscHMapI > designation. Is there a way around this? > > Thank you. > Brandon > > >

1 0

PetscHMapI passing
by Brandon Denton 27 Nov '23

27 Nov '23

Good Morning, I am trying to develop a function in PETSc that allows for the retrieval of a PetscHMapI contained in a PetscContainer attached to a DMPlex. The creation of this Hash Table and its attachment to the DMPlex is automated as part of the CAD/Discretization integration I've been developing. Currently, the only way to get this Hash Table is through the PetscObjectQuery() functionality. In an effort to make it easier for users to access and use this information, I am trying to develop a function where the user can specify an empty PetscHMapI variable and the function will retrieve the information attached to the DMPlex. However, when I try something like ... PETSC_EXTERN PetscErrorCode DMPlexGetGeomInfo(DM dm, PetscHMapI hashTable) I get a ton of compilation errors associated with the PetscHMapI designation. Is there a way around this? Thank you. Brandon

1 0

Release Announcement for Intel® SHMEM v1.0.0
by Rahman, Md 21 Nov '23

21 Nov '23

Hello OpenSHMEM community members, The Intel OpenSHMEM team is excited to announce the first open-source release of Intel® SHMEM v1.0.0. Intel SHMEM is a C++ software library that enables applications to use OpenSHMEM communication APIs with device kernels implemented in SYCL. It supports Intel HPC/AI-focused GPUs, beginning with the Intel® Data Center GPU Max Series. Intel SHMEM implements a Partitioned Global Address Space (PGAS) programming model and includes a subset of host-initiated operations as in the current OpenSHMEM standard. It also includes new device-initiated operations callable directly from GPU kernels. Below is a summarized list of features that are available in Intel SHMEM v1.0.0. * A complete specification<https://oneapi-src.github.io/ishmem/intro.html> detailing the programming model, supported API, example programs, build and run instructions, etc. * Device and host API support for OpenSHMEM 1.5 compliant point-to-point Remote Memory Access (RMA), Atomic Memory Operations, Signaling, Memory Ordering, and Synchronization Operations. * Device and host API support for OpenSHMEM collective operations. * Device API support for SYCL work-group and sub-group level extensions of Remote Memory Access, Signaling, Collective, Memory Ordering, and Synchronization Operations. * Support of C++ template function routines replacing the C11 Generic selection routines from the OpenSHMEM specification. * GPU RDMA (Remote Direct Memory Access) support when configured with Sandia OpenSHMEM with suitable Libfabric providers for high-performance networking services. * Choice of device memory (default) or Unified Shared Memory for the SHMEM Symmetric Heap. * A comprehensive set of unit tests for functionality of core operations. * A suite of performance benchmarks covering device-initiated operation performance for a subset of the operations. * An implementation of a 2D stencil kernel utilizing ISHMEM RMA operations. * Examples to illustrate different use cases of ISHMEM functionalities. * A launcher script to launch ISHMEM applications on the available SYCL devices in the system with the correct mapping. The full source code of Intel SHMEM v1.0.0 can be accessed through the GitHub repo: https://github.com/oneapi-src/ishmem. Please follow the instructions in README.md<https://github.com/oneapi-src/ishmem/blob/main/README.md> for how to build, install, and use Intel SHMEM in your application. Please refer to the examples given here<https://oneapi-src.github.io/ishmem/writing_programs.html> for how to write programs utilizing the Intel SHMEM APIs. Please also look at example codes<https://github.com/oneapi-src/ishmem/tree/main/examples> which demonstrate the usage of various APIs. If you would like to contribute to the project or raise any issues, please follow the guidelines provided in CONTRIBUTING.md<https://github.com/oneapi-src/ishmem/blob/main/CONTRIBUTING.md>. For any other questions, please send an email to ishmem-discuss(a)intel.com<mailto:[email protected]>. Enjoy! The Intel OpenSHMEM Team

1 0

Re: [petsc-dev] Petsc compilation issues with Xcode 15 on macOS Sonoma
by Satish Balay 13 Nov '23

13 Nov '23

Sounds like the issue is with using findPetsc - after petsc is installed [and petsc install went fine?] I think the current recommendation is to use pkg-config file interface via cmake -instead of findPetsc And the following works for me.. ./configure COPTFLAGS="-O3 -g" CXXOPTFLAGS="-O3 -g" FOPTFLAGS="-O3 -g" \ --download-openmpi --with-64-bit-indices --download-metis --download-parmetis \ --download-suitesparse --download-superlu_dist --download-hypre --download-scalapack --download-mumps \ --with-debugging=0 --with-clean A couple of notes: - we recommend not using 'sudo' when building PETSc [and build/use petsc from user account. one can also install in a system-wide location using a user account] - not use --with-cxx-dialect - and let configure determine it. - petsc does not use boost [its there for trilinos - but that part is broken] Satish On Mon, 13 Nov 2023, Barry Smith wrote: > > Please send configure.log (best to petsc-maint(a)mcs.anl.gov) > > > > On Nov 13, 2023, at 11:41 AM, Abhinav Singh <abhinavrajendra(a)gmail.com> wrote: > > > > Dear Petsc devs, > > > > I have been having issues lately when compiling petsc on MacOS with various libraries. This is probably due to the updated linker in MacOS and unfortunately the newer operating systems do not allow using older toolchains. > > > > on arm64, only 3.20 and up can be configured > > > > > > My configure command usually looks like this: > > 'sudo ./configure COPTFLAGS=-O3 -g CXXOPTFLAGS=-O3 -g FOPTFLAGS=-O3 -g --with-cxx-dialect=C++11 --with-mpi-dir=/opt/openfpm/dep_clang/MPI --with-64-bit-indices --with-parmetis-dir=/opt/openfpm/dep_clang/PARMETIS --with-metis-dir=/opt/openfpm/dep_clang/METIS --with-boost=yes --with-boost-dir=/opt/openfpm/dep_clang/BOOST --with-suitesparse-dir=/opt/openfpm/dep_clang/SUITESPARSE --download-superlu_dist --download-hypre --prefix=/opt/openfpm/dep_clang/PETSC --download-scalapack --download-mumps --with-debugging=0 --with-clean > > ' > > > > There are two main issues: > > > > 1) Duplicate 'LC_Paths' when compiling code with apple clang and gfortran. If I enable download_scalapack, the the cmake findPetsc fails with the following error on both x86 and arm64: > > """ > > kind: "try_run-v1" > > backtrace: > > - "/opt/homebrew/Cellar/cmake/3.27.7/share/cmake/Modules/Internal/CheckSourceRuns.cmake:93 (try_run)" > > - "/opt/homebrew/Cellar/cmake/3.27.7/share/cmake/Modules/CheckCSourceRuns.cmake:52 (cmake_check_source_runs)" > > - "cmake_modules/FindPackageMultipass.cmake:97 (check_c_source_runs)" > > - "cmake_modules/FindPETSc.cmake:284 (multipass_source_runs)" > > - "cmake_modules/FindPETSc.cmake:318 (petsc_test_runs)" > > - "CMakeLists.txt:69 (find_package)" > > checks: > > - "Performing Test MULTIPASS_TEST_4_petsc_works_all" > > directories: > > source: "/Users/absingh/openfpm_pdata/build/CMakeFiles/CMakeScratch/TryCompile-LJcmJB" > > binary: "/Users/absingh/openfpm_pdata/build/CMakeFiles/CMakeScratch/TryCompile-LJcmJB" > > cmakeVariables: > > CMAKE_C_FLAGS: "" > > CMAKE_EXE_LINKER_FLAGS: "" > > CMAKE_MODULE_PATH: "/Users/absingh/openfpm_pdata/cmake_modules/" > > CMAKE_OSX_ARCHITECTURES: "" > > CMAKE_OSX_DEPLOYMENT_TARGET: "" > > CMAKE_OSX_SYSROOT: "/Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk" > > buildResult: > > variable: "MULTIPASS_TEST_4_petsc_works_all_COMPILED" > > cached: true > > stdout: | > > Change Dir: '/Users/absingh/openfpm_pdata/build/CMakeFiles/CMakeScratch/TryCompile-LJcmJB' > > > > Run Build Command(s): /opt/homebrew/Cellar/cmake/3.27.7/bin/cmake -E env VERBOSE=1 /usr/bin/make -f Makefile cmTC_e0165/fast > > /Library/Developer/CommandLineTools/usr/bin/make -f CMakeFiles/cmTC_e0165.dir/build.make CMakeFiles/cmTC_e0165.dir/build > > Building C object CMakeFiles/cmTC_e0165.dir/src.c.o > > /Library/Developer/CommandLineTools/usr/bin/cc -DMULTIPASS_TEST_4_petsc_works_all -I/opt/openfpm/dep_clang/PETSC/include -I/opt/openfpm/dep_clang/SUITESPARSE/include -I/opt/openfpm/dep_clang/PARMETIS/include -I/opt/openfpm/dep_clang/METIS/include -I/opt/openfpm/dep_clang/BOOST/include -I/opt/openfpm/dep_clang/MPI/include -arch arm64 -isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk -MD -MT CMakeFiles/cmTC_e0165.dir/src.c.o -MF CMakeFiles/cmTC_e0165.dir/src.c.o.d -o CMakeFiles/cmTC_e0165.dir/src.c.o -c /Users/absingh/openfpm_pdata/build/CMakeFiles/CMakeScratch/TryCompile-LJcmJB/src.c > > Linking C executable cmTC_e0165 > > /opt/homebrew/Cellar/cmake/3.27.7/bin/cmake -E cmake_link_script CMakeFiles/cmTC_e0165.dir/link.txt --verbose=1 > > /Library/Developer/CommandLineTools/usr/bin/cc -arch arm64 -isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk -Wl,-search_paths_first -Wl,-headerpad_max_install_names CMakeFiles/cmTC_e0165.dir/src.c.o -o cmTC_e0165 -Wl,-rpath,/opt/openfpm/dep_clang/PETSC/lib -Wl,-rpath,/opt/openfpm/dep_clang/SUITESPARSE/lib /opt/openfpm/dep_clang/PETSC/lib/libpetsc.dylib /opt/openfpm/dep_clang/PETSC/lib/libHYPRE.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libspqr.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libumfpack.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libklu.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libcholmod.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libbtf.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libccolamd.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libcolamd.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libcamd.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libamd.dylib /opt/openfpm/dep_clang/SUITESPARSE/lib/libsuitesparseconfig.dylib /opt/ openfpm/ dep_clang/PETSC/lib/libdmumps.a /opt/openfpm/dep_clang/PETSC/lib/libmumps_common.a /opt/openfpm/dep_clang/PETSC/lib/libpord.a /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/lib/libpthread.tbd /opt/openfpm/dep_clang/PETSC/lib/libscalapack.dylib /opt/openfpm/dep_clang/PETSC/lib/libsuperlu_dist.dylib /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/lib/liblapack.tbd /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/lib/libblas.tbd /opt/openfpm/dep_clang/PARMETIS/lib/libparmetis.a /opt/openfpm/dep_clang/METIS/lib/libmetis.dylib /opt/openfpm/dep_clang/MPI/lib/libmpi_usempif08.dylib /opt/openfpm/dep_clang/MPI/lib/libmpi_usempi_ignore_tkr.a /opt/openfpm/dep_clang/MPI/lib/libmpi_mpifh.dylib /opt/openfpm/dep_clang/MPI/lib/libmpi.dylib /opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/libgfortran.dylib /opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/gcc/aarch64-apple-darwin23/13/libemutls_w.a /opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/libquadmath.d ylib /Li brary/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/lib/libc++.tbd /Library/Developer/CommandLineTools/SDKs/MacOSX14.0.sdk/usr/lib/libgcc_s.1.tbd > > ld: warning: duplicate LC_RPATH are deprecated ('/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/gcc/aarch64-apple-darwin23/13') > > ld: warning: duplicate LC_RPATH are deprecated ('/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/gcc') > > ld: warning: duplicate LC_RPATH are deprecated ('/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current') > > ld: warning: duplicate LC_RPATH are deprecated ('/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/gcc/aarch64-apple-darwin23/13') > > ld: warning: duplicate LC_RPATH are deprecated ('/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/gcc') > > ld: warning: duplicate LC_RPATH are deprecated ('/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current') > > ld: duplicate LC_RPATH '/opt/homebrew/Cellar/gcc/13.2.0/lib/gcc/current/gcc/aarch64-apple-darwin23/13' in '/opt/openfpm/dep_clang/PETSC/lib/libscalapack.2.2.1.dylib' > > clang: error: linker command failed with exit code 1 (use -v to see invocation) > > make[1]: *** [cmTC_e0165] Error 1 > > make: *** [cmTC_e0165/fast] Error 2 > > > > exitCode: 2 > > runResult: > > variable: "MULTIPASS_TEST_4_petsc_works_all_EXITCODE" > > cached: true > > """ > > > > Second, on arm64, I am unable to use the option "with-metis-dir=/path/to/metis". The installation works fine but when I run make check, I get the error that 'libmetis.dylib' is not found. This behavior continues even when I export 'DYLD_LIBRARY_PATH' to have the libmetis. CMake findPetsc also fails with the same error. > > > > Any advice would be great. > > > > Thank you! > > Abhinav > > > > > > > > > > > > > > > > > > >

1 0