Skip to content

PML/SPML/UCX: added evaluation of mmap events - v4.0 #6486

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Mar 14, 2019
Merged

PML/SPML/UCX: added evaluation of mmap events - v4.0 #6486

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
14c271f
PML/SPML/UCX: added evaluation of mmap events
Mar 12, 2019
bed8141
COMMON/UCX: rewording of hooks suggestion
Mar 14, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions config/ompi_check_ucx.m4
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -112,6 +112,9 @@ AC_DEFUN([OMPI_CHECK_UCX],[
ucp_request_check_status, ucp_put_nb, ucp_get_nb],
[], [],
[#include <ucp/api/ucp.h>])
AC_CHECK_DECLS([ucm_test_events],
[], [],
[#include <ucm/api/ucm.h>])
AC_CHECK_DECLS([UCP_ATOMIC_POST_OP_AND,
UCP_ATOMIC_POST_OP_OR,
UCP_ATOMIC_POST_OP_XOR,
Expand Down
1 change: 1 addition & 0 deletions ompi/mca/pml/ucx/pml_ucx.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -422,6 +422,7 @@ int mca_pml_ucx_add_procs(struct ompi_proc_t **procs, size_t nprocs)
}
}

opal_common_ucx_mca_proc_added();
return OMPI_SUCCESS;
}

Expand Down
21 changes: 21 additions & 0 deletions opal/mca/common/ucx/common_ucx.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,27 @@ static void opal_common_ucx_mca_fence_complete_cb(int status, void *fenced)
*(int*)fenced = 1;
}

void opal_common_ucx_mca_proc_added(void)
{
#if HAVE_DECL_UCM_TEST_EVENTS
static int warned = 0;
static char *mem_hooks_suggestion = "Pls try adding --mca opal_common_ucx_opal_mem_hooks 1 "
"to mpirun/oshrun command line to resolve this issue.";
ucs_status_t status;

if (!warned) {
status = ucm_test_events(UCM_EVENT_VM_UNMAPPED);
if (status != UCS_OK) {
MCA_COMMON_UCX_WARN("UCX is unable to handle VM_UNMAP event. "
"This may cause performance degradation or data "
"corruption. %s",
opal_common_ucx.opal_mem_hooks ? "" : mem_hooks_suggestion);
warned = 1;
}
}
#endif
}

OPAL_DECLSPEC int opal_common_ucx_mca_pmix_fence(ucp_worker_h worker)
{
volatile int fenced = 0;
Expand Down
10 changes: 6 additions & 4 deletions opal/mca/common/ucx/common_ucx.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,10 +39,11 @@ BEGIN_C_DECLS
#define MCA_COMMON_UCX_QUOTE(_x) \
_MCA_COMMON_UCX_QUOTE(_x)

#define MCA_COMMON_UCX_ERROR(...) \
opal_output_verbose(0, opal_common_ucx.output, \
__FILE__ ":" MCA_COMMON_UCX_QUOTE(__LINE__) \
" Error: " __VA_ARGS__)
#define MCA_COMMON_UCX_ERROR(...) \
MCA_COMMON_UCX_VERBOSE(0, " Error: " __VA_ARGS__)

#define MCA_COMMON_UCX_WARN(...) \
MCA_COMMON_UCX_VERBOSE(0, " Warning: " __VA_ARGS__)

#define MCA_COMMON_UCX_VERBOSE(_level, ... ) \
if (((_level) <= MCA_COMMON_UCX_MAX_VERBOSE) && \
Expand Down Expand Up @@ -96,6 +97,7 @@ extern opal_common_ucx_module_t opal_common_ucx;

OPAL_DECLSPEC void opal_common_ucx_mca_register(void);
OPAL_DECLSPEC void opal_common_ucx_mca_deregister(void);
OPAL_DECLSPEC void opal_common_ucx_mca_proc_added(void);
OPAL_DECLSPEC void opal_common_ucx_empty_complete_cb(void *request, ucs_status_t status);
OPAL_DECLSPEC int opal_common_ucx_mca_pmix_fence(ucp_worker_h worker);
OPAL_DECLSPEC void opal_common_ucx_mca_var_register(const mca_base_component_t *component);
Expand Down
2 changes: 2 additions & 0 deletions oshmem/mca/spml/ucx/spml_ucx.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,6 +138,8 @@ int mca_spml_ucx_del_procs(ompi_proc_t** procs, size_t nprocs)

mca_spml_ucx_ctx_default.ucp_peers = NULL;

opal_common_ucx_mca_proc_added();

return ret;
}

Expand Down