1
1
openmpi/opal/mca/btl/uct/btl_uct_am.h
Nathan Hjelm e6f84e79de btl/uct: fix deadlock in connection code
This commit fixes a deadlock that can occur when using a TL that
supports the connect to endpoint model. The deadlock was occurring
while processing an incoming connection requests. This was done from
an active-message callback. For some unknown reason (at this time)
this callback was sometimes hanging. To avoid the issue the connection
active-message is saved for later processing.

At the same time I cleaned up the connection code to eliminate
duplicate messages when possible.

This commit also fixes some bugs in the active-message send path:

 - Correctly set all fragment fields in prepare_src.

 - Fix bug when using buffered-send. We were not reading the return
   code correctly (which is in bytes). This resulted in a message
   getting sent multiple times.

 - Don't try to progress sends from the btl_send function when in an
   active-message callback. It could lead to deep recursion and an
   eventual crash if we get a trace like
   send->progress->am_complete->ob1_callback->send->am_complete...

Closes #5820
Closes #5821

Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
(cherry picked from commit 707d35deeb)
Signed-off-by: Nathan Hjelm <hjelmn@me.com>
2018-10-16 19:16:11 -06:00

39 строки
1.7 KiB
C

/* -*- Mode: C; c-basic-offset:4 ; indent-tabs-mode:nil -*- */
/*
* Copyright (c) 2018 Los Alamos National Security, LLC. All rights
* reserved.
* $COPYRIGHT$
*
* Additional copyrights may follow
*
* $HEADER$
*/
#if !defined(MCA_BTL_UCT_AM_H)
#define MCA_BTL_UCT_AM_H
#include "btl_uct_frag.h"
struct mca_btl_base_descriptor_t *mca_btl_uct_prepare_src (mca_btl_base_module_t *btl,
mca_btl_base_endpoint_t *endpoint,
opal_convertor_t *convertor,
uint8_t order, size_t reserve,
size_t *size, uint32_t flags);
int mca_btl_uct_sendi (mca_btl_base_module_t *btl, mca_btl_base_endpoint_t *endpoint, opal_convertor_t *convertor,
void *header, size_t header_size, size_t payload_size, uint8_t order, uint32_t flags,
mca_btl_base_tag_t tag, mca_btl_base_descriptor_t **descriptor);
int mca_btl_uct_send (mca_btl_base_module_t *btl, mca_btl_base_endpoint_t *endpoint, mca_btl_base_descriptor_t *descriptor,
mca_btl_base_tag_t tag);
int mca_btl_uct_send_frag (mca_btl_uct_module_t *uct_btl, mca_btl_uct_base_frag_t *frag, bool append);
mca_btl_base_descriptor_t *mca_btl_uct_alloc (mca_btl_base_module_t *btl, mca_btl_base_endpoint_t *endpoint,
uint8_t order, size_t size, uint32_t flags);
int mca_btl_uct_free (mca_btl_base_module_t *btl, mca_btl_base_descriptor_t *des);
#endif /* !defined(MCA_BTL_UCT_AM_H) */