<div dir="ltr">Thanks.</div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jan 21, 2020 at 11:03 AM Jeremy Theler <<a href="mailto:jeremy@seamplex.com">jeremy@seamplex.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear Sam<br>
<br>
Probably you are already aware of the following paragraph, but just in<br>
case. Quote from <br>
<a href="https://www.gnu.org/prep/standards/standards.html#Memory-Usage" rel="noreferrer" target="_blank">https://www.gnu.org/prep/standards/standards.html#Memory-Usage</a><br>
<br>
<br>
Memory analysis tools such as valgrind can be useful, but don’t<br>
complicate a program merely to avoid their false alarms. For example,<br>
if memory is used until just before a process exits, don’t free it<br>
simply to silence such a tool. <br>
<br>
<br>
Regards<br>
--<br>
jeremy theler<br>
<a href="http://www.seamplex.com" rel="noreferrer" target="_blank">www.seamplex.com</a><br>
<br>
<br>
<br>
On Tue, 2020-01-21 at 08:49 -0800, Sam Guo wrote:<br>
> I use PETSc from my application. Sounds you are saying I just treat<br>
> ierr!=0 as an system error and no need to call Destroy functions.<br>
> <br>
> On Tuesday, January 21, 2020, Smith, Barry F. <<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>><br>
> wrote:<br>
> > <br>
> > > On Jan 20, 2020, at 6:32 PM, Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a>><br>
> > wrote:<br>
> > > <br>
> > > Hi Barry,<br>
> > > I understand ierr != 0 means something catastrophic. I just<br>
> > want to release all memory before I exit PETSc.<br>
> > <br>
> > In general not possible. If you run with the debug version and<br>
> > -malloc_debug it is possible but because of the unknown error it<br>
> > could be that the releasing of the memory causes a real crash.<br>
> > <br>
> > Is your main concern when you use PETSc for a large problem and<br>
> > it errors because it is "out of memory"?<br>
> > <br>
> > Barry<br>
> > <br>
> > <br>
> > > <br>
> > > Thanks,<br>
> > > Sam<br>
> > > <br>
> > > On Mon, Jan 20, 2020 at 4:06 PM Smith, Barry F. <<br>
> > <a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>> wrote:<br>
> > > <br>
> > > Sam,<br>
> > > <br>
> > > I am not sure what your goal is but PETSc error return codes<br>
> > are error return codes not exceptions. They mean that something<br>
> > catastrophic happened and there is no recovery. <br>
> > > <br>
> > > Note that PETSc solvers do not return nonzero error codes on<br>
> > failure to converge etc. You call, for example,<br>
> > KPSGetConvergedReason() after a KSP solve to see if it has failed,<br>
> > this is not a catastrophic failure. If a MatCreate() or any other<br>
> > call returns a nonzero ierr the game is up, you cannot continue<br>
> > running PETSc.<br>
> > > <br>
> > > Barry<br>
> > > <br>
> > > <br>
> > > > On Jan 20, 2020, at 5:41 PM, Matthew Knepley <<a href="mailto:knepley@gmail.com" target="_blank">knepley@gmail.com</a><br>
> > > wrote:<br>
> > > > <br>
> > > > Not if you initialize the pointers to zero: Mat A = NULL.<br>
> > > > <br>
> > > > Matt<br>
> > > > <br>
> > > > On Mon, Jan 20, 2020 at 6:31 PM Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a>><br>
> > wrote:<br>
> > > > I mean MatDestroy.<br>
> > > > <br>
> > > > On Mon, Jan 20, 2020 at 3:28 PM Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a>><br>
> > wrote:<br>
> > > > Does it hurt to call Destroy function without calling<br>
> > CreateFunction? For example <br>
> > > > Mat A, B;<br>
> > > > PetscErrorCode ierr1, ierr2;<br>
> > > > ierr1 = MatCreate(PETSC_COMM_WORLD,&A);<br>
> > > > if(ierr1 == 0)<br>
> > > > {<br>
> > > > ierr2 = MatCreate(PETSC_COMM_WORLD<br>
> > > > ,&B);<br>
> > > > <br>
> > > > }<br>
> > > > if(ierr1 !=0 || ierr2 != 0)<br>
> > > > {<br>
> > > > Destroy(&A);<br>
> > > > Destroy(&B); // if ierr1 !=0, MatCreat is not called on B.<br>
> > Does it hurt to call Destroy B here?<br>
> > > > }<br>
> > > > <br>
> > > > <br>
> > > > <br>
> > > > On Mon, Jan 20, 2020 at 11:11 AM Dave May <<br>
> > <a href="mailto:dave.mayhem23@gmail.com" target="_blank">dave.mayhem23@gmail.com</a>> wrote:<br>
> > > > <br>
> > > > <br>
> > > > On Mon 20. Jan 2020 at 19:47, Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a>><br>
> > wrote:<br>
> > > > Can I assume if there is MatCreat or VecCreate, I should clean<br>
> > up the memory myself?<br>
> > > > <br>
> > > > Yes. You will need to call the matching Destroy function.<br>
> > > > <br>
> > > > <br>
> > > > <br>
> > > > On Mon, Jan 20, 2020 at 10:45 AM Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a><br>
> > > wrote:<br>
> > > > I only include the first few lines of SLEPc example. What about<br>
> > following<br>
> > > > ierr = MatCreate(PETSC_COMM_WORLD,&A);CHKERRQ(ierr); <br>
> > > > ierr =<br>
> > MatSetSizes(A,PETSC_DECIDE,PETSC_DECIDE,n,n);CHKERRQ(ierr); <br>
> > > > Is there any memory lost?<br>
> > > > <br>
> > > > On Mon, Jan 20, 2020 at 10:41 AM Dave May <<br>
> > <a href="mailto:dave.mayhem23@gmail.com" target="_blank">dave.mayhem23@gmail.com</a>> wrote:<br>
> > > > <br>
> > > > <br>
> > > > On Mon 20. Jan 2020 at 19:39, Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a>><br>
> > wrote:<br>
> > > > I don't have a specific case yet. Currently every call of PETSc<br>
> > is checked. If ierr is not zero, print the error and return. For<br>
> > example,<br>
> > > > Mat A; /* problem matrix */ <br>
> > > > EPS eps; /* eigenproblem solver context */ <br>
> > > > EPSType type; <br>
> > > > PetscReal error,tol,re,im; <br>
> > > > PetscScalar kr,ki; Vec xr,xi; 25 <br>
> > > > PetscInt n=30,i,Istart,Iend,nev,maxit,its,nconv; <br>
> > > > PetscErrorCode ierr; <br>
> > > > ierr =<br>
> > SlepcInitialize(&argc,&argv,(char*)0,help);CHKERRQ(ierr); <br>
> > > > ierr = PetscOptionsGetInt(NULL,NULL,"-<br>
> > n",&n,NULL);CHKERRQ(ierr);<br>
> > > > ierr = PetscPrintf(PETSC_COMM_WORLD,"\n1-D Laplacian<br>
> > Eigenproblem, n=%D\n\n",n);CHKERRQ(ierr); <br>
> > > > <br>
> > > > I am wondering if the memory is lost by calling CHKERRQ.<br>
> > > > <br>
> > > > No.<br>
> > > > <br>
> > > > <br>
> > > > <br>
> > > > On Mon, Jan 20, 2020 at 10:14 AM Dave May <<br>
> > <a href="mailto:dave.mayhem23@gmail.com" target="_blank">dave.mayhem23@gmail.com</a>> wrote:<br>
> > > > <br>
> > > > <br>
> > > > On Mon 20. Jan 2020 at 19:11, Sam Guo <<a href="mailto:sam.guo@cd-adapco.com" target="_blank">sam.guo@cd-adapco.com</a>><br>
> > wrote:<br>
> > > > Dear PETSc dev team,<br>
> > > > If PETSc function returns an error, what's the correct way<br>
> > to clean PETSc?<br>
> > > > <br>
> > > > The answer depends on the error message reported. Send the<br>
> > complete error message and a better answer can be provided.<br>
> > > > <br>
> > > > Particularly how to clean up the memory?<br>
> > > > <br>
> > > > Totally depends on the objects which aren’t being freed. You<br>
> > need to provide more information<br>
> > > > <br>
> > > > Thanks<br>
> > > > Dave<br>
> > > > <br>
> > > > <br>
> > > > Thanks,<br>
> > > > Sam<br>
> > > > <br>
> > > > <br>
> > > > -- <br>
> > > > What most experimenters take for granted before they begin<br>
> > their experiments is infinitely more interesting than any results<br>
> > to which their experiments lead.<br>
> > > > -- Norbert Wiener<br>
> > > > <br>
> > > > <a href="https://www.cse.buffalo.edu/~knepley/" rel="noreferrer" target="_blank">https://www.cse.buffalo.edu/~knepley/</a><br>
> > > <br>
> > <br>
> > <br>
<br>
</blockquote></div>