Skip to content
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.

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

prov/rxm: rxm_mc introduction #7

Draft
wants to merge 77 commits into
base: main
Choose a base branch
from
Draft

Conversation

grom72
Copy link
Collaborator

@grom72 grom72 commented Dec 13, 2022

rxm_mc added to hide implementation details of collective operations' multicast address

Signed-off-by: Tomasz Gromadzki [email protected]


This change is Reviewable

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 3 of 4 files at r1, all commit messages.
Reviewable status: 3 of 4 files reviewed, 9 unresolved discussions (waiting on @grom72)


-- commits line 5 at r1:
... details of multicast address of collective operations.

Code quote:

  rxm_mc added to hide implementation details of collective operations
  multicast address

prov/rxm/src/rxm_ep.c line 1059 at r1 (raw file):

	struct rxm_mc *rxm_mc;
	struct util_coll_mc *coll_mc;
        ofi_ep_lock_acquire(&rxm_ep->util_ep);

You can unify the alignment here and use one tab instead of spaces in this line.

Code quote:

        ofi_ep_lock_acquire(&rxm_ep->util_ep);

prov/rxm/src/rxm_ep.c line 1061 at r1 (raw file):

        ofi_ep_lock_acquire(&rxm_ep->util_ep);
	(*req) = rxm_get_coll_buf(rxm_ep);
        ofi_ep_lock_release(&rxm_ep->util_ep);

You can unify the alignment here and use one tab instead of spaces in this line.

Code quote:

        ofi_ep_lock_release(&rxm_ep->util_ep);

prov/rxm/src/rxm_ep.c line 1070 at r1 (raw file):

	(*req)->app_context = context;

	rxm_mc = (struct rxm_mc*) ((uintptr_t) *coll_addr) ;

redundant space

Code quote:

) ;

prov/rxm/src/rxm_ep.c line 1078 at r1 (raw file):

		coll_mc = container_of(rxm_mc->util_coll_mc_fid,
			struct util_coll_mc, mc_fid);
		*coll_addr = fi_mc_addr(&coll_mc->mc_fid);			

3 redundant tabs at the end of line.

Code quote:

;		·

prov/rxm/src/rxm_ep.c line 1081 at r1 (raw file):

		(*coll_ep) = rxm_ep->util_coll_ep;
	}
	else {
	} else {

Code quote:

	}
	else {

prov/rxm/src/rxm_ep.c line 1081 at r1 (raw file):

		(*coll_ep) = rxm_ep->util_coll_ep;
	}
	else {

Isn't this still needed here?

} else if (rxm_ep->offload_coll_mask & BIT(coll_op)) {

Can be set

(*coll_ep) = rxm_ep->offload_coll_ep;

without the above condition being true?

Code quote:

else {

prov/rxm/src/rxm_ep.c line 1087 at r1 (raw file):

		coll_mc = container_of(rxm_mc->offload_coll_mc_fid,
			struct util_coll_mc, mc_fid);
		*coll_addr = fi_mc_addr(&coll_mc->mc_fid);	

1 redundant tab at the end of line.

Code quote:

);·

prov/rxm/src/rxm_ep.c line 1222 at r1 (raw file):

	ssize_t ret;

redundant empty line added

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r1.
Reviewable status: all files reviewed, 18 unresolved discussions (waiting on @grom72)


prov/rxm/src/rxm_eq.c line 73 at r1 (raw file):

	const struct fi_eq_entry *in_entry = buf;
	struct rxm_mc *mc;
	if (event != FI_JOIN_COMPLETE) {

Missing one empty line in between.

Code quote:

	struct rxm_mc *mc;
	if (event != FI_JOIN_COMPLETE) {

prov/rxm/src/rxm_eq.c line 75 at r1 (raw file):

	if (event != FI_JOIN_COMPLETE) {
		return ofi_eq_write(eq_fid, event, buf, len, flags);
	}	

1 redundant tab at the end of line.

Code quote:


prov/rxm/src/rxm_eq.c line 82 at r1 (raw file):

		if (mc->util_coll_join_completed == -1) {
			mc->util_coll_join_completed = 1;
			fi_close (&mc->mc_fid.fid);

1 redundant space in between

Code quote:

fi_close (&mc

prov/rxm/src/rxm_eq.c line 91 at r1 (raw file):

		mc->offload_coll_join_completed = 1;
	}
	if (mc->util_coll_join_completed == 1 && 

Missing one empty line in between.

Code quote:

	}
	if (mc->util_coll_join_completed == 1 &&

prov/rxm/src/rxm_eq.c line 91 at r1 (raw file):

		mc->offload_coll_join_completed = 1;
	}
	if (mc->util_coll_join_completed == 1 && 

1 redundant space at the end of line.

Code quote:

&&·

prov/rxm/src/rxm_eq.c line 92 at r1 (raw file):

	}
	if (mc->util_coll_join_completed == 1 && 
		(mc->offload_coll_join_completed == 1 || 

1 redundant space at the end of line.

Code quote:

||·

prov/rxm/src/rxm_eq.c line 92 at r1 (raw file):

	}
	if (mc->util_coll_join_completed == 1 && 
		(mc->offload_coll_join_completed == 1 || 

4 spaces instead of the 2nd tab

Code quote:

		(mc->offload_coll_join_completed == 1 ||
		!mc->offload_coll_mc_fid)) {

prov/rxm/src/rxm_eq.c line 99 at r1 (raw file):

		return ofi_eq_write(eq_fid, event, &entry, len, flags);
	}
	return len;

Missing one empty line in between.

Code quote:

	}
	return len;

prov/rxm/src/rxm_eq.c line 101 at r1 (raw file):

	return len;
};
static struct fi_ops_eq rxm_eq_ops = {

Missing one empty line in between.

Code quote:

};
static struct fi_ops_eq rxm_eq_ops = {

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 19 unresolved discussions (waiting on @grom72)


prov/rxm/src/rxm_eq.c line 87 at r1 (raw file):

			mc->util_coll_join_completed = 1;
		}
	} if (in_entry->fid == &mc->offload_coll_mc_fid->fid) {

??? new line or else if ?

Code quote:

} if (in

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r1, 2 of 3 files at r2, all commit messages.
Reviewable status: 4 of 5 files reviewed, 19 unresolved discussions (waiting on @ldorau)


-- commits line 5 at r1:

Previously, ldorau (Lukasz Dorau) wrote…

... details of multicast address of collective operations.

Done.


prov/rxm/src/rxm_ep.c line 1059 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

You can unify the alignment here and use one tab instead of spaces in this line.

Done.


prov/rxm/src/rxm_ep.c line 1061 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

You can unify the alignment here and use one tab instead of spaces in this line.

Done.


prov/rxm/src/rxm_ep.c line 1070 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

redundant space

Done.


prov/rxm/src/rxm_ep.c line 1078 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

3 redundant tabs at the end of line.

Done.


prov/rxm/src/rxm_ep.c line 1081 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

Isn't this still needed here?

} else if (rxm_ep->offload_coll_mask & BIT(coll_op)) {

Can be set

(*coll_ep) = rxm_ep->offload_coll_ep;

without the above condition being true?

It is covered above by !(rxm_ep->offload_coll_mask & BIT(coll_op))


prov/rxm/src/rxm_ep.c line 1081 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…
	} else {

Done.


prov/rxm/src/rxm_ep.c line 1087 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

1 redundant tab at the end of line.

Done.

Code quote:

rxm_ep_broadcast

prov/rxm/src/rxm_ep.c line 1222 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

redundant empty line added

Done.


prov/rxm/src/rxm_eq.c line 73 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

Missing one empty line in between.

Done.


prov/rxm/src/rxm_eq.c line 75 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

1 redundant tab at the end of line.

Done.


prov/rxm/src/rxm_eq.c line 82 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

1 redundant space in between

Done.


prov/rxm/src/rxm_eq.c line 87 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

??? new line or else if ?

Done.


prov/rxm/src/rxm_eq.c line 91 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

Missing one empty line in between.

Done.


prov/rxm/src/rxm_eq.c line 91 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

1 redundant space at the end of line.

Done.


prov/rxm/src/rxm_eq.c line 92 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

1 redundant space at the end of line.

Done.


prov/rxm/src/rxm_eq.c line 92 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

4 spaces instead of the 2nd tab

Done.


prov/rxm/src/rxm_eq.c line 99 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

Missing one empty line in between.

Done.


prov/rxm/src/rxm_eq.c line 101 at r1 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

Missing one empty line in between.

Done.

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 3 files at r2.
Reviewable status: all files reviewed, 19 unresolved discussions (waiting on @ldorau)

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 3 of 3 files at r2, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @grom72)


prov/rxm/src/rxm_eq.c line 92 at r2 (raw file):

		mc->offload_coll_join_completed = 1;
	} else {
		assert(0); /* we do not expect any other fid*/

fid */ - add a space at the end

Code quote:

fid*/

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r3.
Reviewable status: 2 of 5 files reviewed, 1 unresolved discussion (waiting on @ldorau)


prov/rxm/src/rxm_eq.c line 92 at r2 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

fid */ - add a space at the end

Done.

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r1, 3 of 4 files at r3, all commit messages.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @ldorau)

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r3.
Reviewable status: :shipit: complete! all files reviewed, all discussions resolved (waiting on @grom72)

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 4 files at r3, all commit messages.
Reviewable status: all files reviewed, 7 unresolved discussions (waiting on @grom72)


-- commits line 18 at r3:
prov

Code quote:

ptov

-- commits line 21 at r3:
completion

Code quote:

compeltion

-- commits line 22 at r3:
the same number of references ?

Code quote:

the same references

-- commits line 28 at r3:
with the FI_PEER flag

Code quote:

with FI_PEER flag

-- commits line 28 at r3:
restoreS

Code quote:

restore

-- commits line 28 at r3:
uses it ... ?

Code quote:

uses

-- commits line 29 at r3:
reporting the join ...

Code quote:

reporting join

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 8 unresolved discussions (waiting on @grom72)


-- commits line 18 at r3:
??? I do not understand the title ...

Code quote:

pass submit

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 2 of 4 files at r3.
Reviewable status: all files reviewed, 10 unresolved discussions (waiting on @grom72)


prov/coll/src/coll_coll.c line 932 at r3 (raw file):

		return -FI_ENOSYS;

	if ((flags & FI_PEER)) {

(flags & FI_PEER)

Code quote:

((flags & FI_PEER))

prov/coll/src/coll_coll.c line 954 at r3 (raw file):

		return -FI_ENOMEM;

	if ((flags & FI_PEER))

.

Code quote:

((flags & FI_PEER))

@grom72 grom72 force-pushed the prov-rxm-mc branch 4 times, most recently from cf5b84d to eba75ba Compare December 14, 2022 09:36
Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed (commit messages unreviewed), 10 unresolved discussions (waiting on @ldorau)


-- commits line 18 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

prov

Done.


-- commits line 18 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

??? I do not understand the title ...

Done.


-- commits line 21 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

completion

Done.


-- commits line 22 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

the same number of references ?

Done.


-- commits line 28 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

with the FI_PEER flag

Done.


-- commits line 28 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

restoreS

Done.


-- commits line 28 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

uses it ... ?

Done.


-- commits line 29 at r3:

Previously, ldorau (Lukasz Dorau) wrote…

reporting the join ...

Done.

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed all commit messages.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @grom72)


-- commits line 18 at r4:
provides

Code quote:

provids

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 1 files at r5, all commit messages.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @ldorau)


-- commits line 18 at r4:

Previously, ldorau (Lukasz Dorau) wrote…

provides

Done.


prov/coll/src/coll_coll.c line 932 at r3 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

(flags & FI_PEER)

Done.


prov/coll/src/coll_coll.c line 954 at r3 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

.

Done.

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed 1 of 1 files at r5, all commit messages.
Reviewable status: :shipit: complete! all files reviewed, all discussions resolved (waiting on @grom72)

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @grom72)


prov/rxm/src/rxm_ep.c line 468 at r5 (raw file):

			&rxm_mc->offload_coll_mc_fid, &peer_context);
		if (ret) {
			/*mark util_coll_mc to be removed as soon as

/* mark - add one space

Code quote:

/*mark

prov/rxm/src/rxm_eq.c line 102 at r5 (raw file):

	if (mc->util_coll_join_completed == 1 &&
	    (mc->offload_coll_join_completed == 1 ||
	     !mc->offload_coll_mc_fid)) {

There is one space too many.
image.png

Code quote:

	     !mc->offload_coll_mc_fid)) {

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

:lgtm:

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @grom72)

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @ldorau)


prov/rxm/src/rxm_eq.c line 102 at r5 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

There is one space too many.
image.png

It is intentionally added to mark the bracket that opens the part of the condition

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @grom72)


prov/rxm/src/rxm_eq.c line 102 at r5 (raw file):

Previously, grom72 (Tomasz Gromadzki) wrote…

It is intentionally added to mark the bracket that opens the part of the condition

OK, for me 4 spaces in both lines looks better:
image.png

Copy link
Collaborator Author

@grom72 grom72 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @ldorau)


prov/rxm/src/rxm_eq.c line 102 at r5 (raw file):

Previously, ldorau (Lukasz Dorau) wrote…

OK, for me 4 spaces in both lines looks better:
image.png

What do you think about this version:

	if (mc->util_coll_join_completed == 1
	    && (mc->offload_coll_join_completed == 1
	        || !mc->offload_coll_mc_fid)) {

Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @grom72)


prov/rxm/src/rxm_eq.c line 102 at r5 (raw file):

Previously, grom72 (Tomasz Gromadzki) wrote…

What do you think about this version:

	if (mc->util_coll_join_completed == 1
	    && (mc->offload_coll_join_completed == 1
	        || !mc->offload_coll_mc_fid)) {

It looks good. It can be. (I am afraid our cstyle wouldn't accept it ;-))

wzamazon and others added 22 commits January 20, 2023 11:10
FI_SOURCE adds overhead so if it is not needed by the application, we should not
return it.

Rxm already removes it, but this patch moves the omission to the util code so other
providers taking the util path will have the same optimization.

Signed-off-by: Alexia Ingerson <[email protected]>
Clarifies usage for incoming peer srx changes

Signed-off-by: Alexia Ingerson <[email protected]>
Because both client and server utilize all Neuron cores, they cannot request
-D neuron on the same node. Thus skip those tests.

Signed-off-by: Wenduo Wang <[email protected]>
RDM endpoint now uses rxr_ep->base_ep.av

Signed-off-by: Sai Sunku <[email protected]>
This commit removes the rdm_ep member of rxr_ep. The RDM endpoint no
longer uses the DGRAM endpoint as a lower endpoint

Signed-off-by: Sai Sunku <[email protected]>
SHMEM summary is missing an '_' character in the log file and a token reset line
Adding this back will properly display SHMEM as summarized by the summarizer.

Signed-off-by: Zach Dworkin <[email protected]>
Add back DSA testing since Node is back online

Signed-off-by: Zach Dworkin <[email protected]>
Declare missing hmem ops function pointers for Neuron

Signed-off-by: Wenduo Wang <[email protected]>
Add missing function pointers for Neuron hmem ops

Signed-off-by: Wenduo Wang <[email protected]>
include,src/hmem: Define missing neuron hmem ops function pointers
fabtests/pytest/common.py: skip neuron to neuron tests on single node
The function efa_base_ep_construct() did not handle the
case when fi_dupinfo() failed to allocate info object.

The handling of calloc() failure is not right either.

This patch addressed both issues.

Signed-off-by: Wei Zhang <[email protected]>
prov/efa: fix the error handling of efa_base_ep_construct()
rxr_pkt_trigger_handshake() initializes a tx_entry to send
a RTW packet. It did not initialize the ep field of the tx_entry,
this patch fixed the issue.

Signed-off-by: Wei Zhang <[email protected]>
Replace redundant synapseai hmem ops with ofi hmem functions

Signed-off-by: Wenduo Wang <[email protected]>
Cosmetic changes

Signed-off-by: Wenduo Wang <[email protected]>
…rigger_handshake

prov/efa: set tx_entry->ep in rxr_pkt_trigger_handshake()
prov/efa: fix comments in rxr_op_entry.c
Running fabtests with ASAN has revealed 2 memory leaks with RDM.
When a new endpoint (active or passive) is created, the fi_info
object is duplicated and can be deleted by the caller.

Signed-off-by: Sylvain Didelot <[email protected]>
rxm_mc added to hide implementation details of multicast address of collective
operations.

Signed-off-by: Tomasz Gromadzki <[email protected]>
Collective provider (i.e. SHARP) requires to execute collective
operation on the supporting collective provider while running join() operation.
Such operation is executed on peer_ep with peer_mc as address of the operation.

Signed-off-by: Tomasz Gromadzki <[email protected]>
util_coll:fi_join() called with the FI_PEER flag restores peer_mc_context.mc_fid and
uses it as the actual context of fi_join() operation. This includes also reporting
the join operation completion with mc_fid as event's context.

Signed-off-by: Tomasz Gromadzki <[email protected]>
mc_ref counts all mc created for rxm_ep. rxm_ep cannot be closed
until all mc's are closed properly.

Signed-off-by: Tomasz Gromadzki <[email protected]>
Copy link
Member

@ldorau ldorau left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Reviewed all commit messages.
Reviewable status: 14 of 129 files reviewed, all discussions resolved (waiting on @grom72)

grom72 pushed a commit that referenced this pull request Mar 24, 2023
If a posted receive matches with a saved receive, we may need to
increment the rx counter.  Set the rx counter increment callback
to match that of the posted receive.  This fixes an assert in
xnet_cntr_inc() accessing a NULL cntr_inc function pointer.

Program received signal SIGABRT, Aborted.
0x0000155552d4d37f in raise () from /lib64/libc.so.6
#0  0x0000155552d4d37f in raise () from /lib64/libc.so.6
#1  0x0000155552d37db5 in abort () from /lib64/libc.so.6
#2  0x0000155552d37c89 in __assert_fail_base.cold.0 () from /lib64/libc.so.6
#3  0x0000155552d45a76 in __assert_fail () from /lib64/libc.so.6
#4  0x00001555522967f9 in xnet_cntr_inc (ep=0x6e4c70, xfer_entry=0x6f7a30) at prov/tcp/src/xnet_cq.c:347
#5  0x0000155552296836 in xnet_report_cntr_success (ep=0x6e4c70, cq=0x6ca930, xfer_entry=0x6f7a30) at prov/tcp/src/xnet_cq.c:354
#6  0x000015555229970d in xnet_complete_saved (saved_entry=0x6f7a30) at prov/tcp/src/xnet_progress.c:153
#7  0x0000155552299961 in xnet_recv_saved (saved_entry=0x6f7a30, rx_entry=0x6f7840) at prov/tcp/src/xnet_progress.c:188
#8  0x00001555522946f8 in xnet_srx_tag (srx=0x6dd1c0, recv_entry=0x6f7840) at prov/tcp/src/xnet_srx.c:445
#9  0x0000155552294bb1 in xnet_srx_trecv (ep_fid=0x6dd1c0, buf=0x6990c4, len=4, desc=0x0, src_addr=0, tag=21474836494, ignore=3458764513820540928, context=0x7ffffffeb180) at prov/tcp/src/xnet_srx.c:558
ofiwg#10 0x000015555228f60e in fi_trecv (ep=0x6dd1c0, buf=0x6990c4, len=4, desc=0x0, src_addr=0, tag=21474836494, ignore=3458764513820540928, context=0x7ffffffeb180) at ./include/rdma/fi_tagged.h:91
ofiwg#11 0x00001555522900a7 in xnet_rdm_trecv (ep_fid=0x6d9fe0, buf=0x6990c4, len=4, desc=0x0, src_addr=0, tag=21474836494, ignore=3458764513820540928, context=0x7ffffffeb180) at prov/tcp/src/xnet_rdm.c:212

Signed-off-by: Sean Hefty <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.