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.

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

DAOS-17109 cart: clients URI lookup of rank0 can lead to crash #15891

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

DAOS-17109 cart: clients URI lookup of rank0 can lead to crash #15891

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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 src/cart/crt_group.c
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
/*
* (C) Copyright 2016-2024 Intel Corporation.
* (C) Copyright 2025 Hewlett Packard Enterprise Development LP
*
* SPDX-License-Identifier: BSD-2-Clause-Patent
*/
Expand Down Expand Up @@ -1090,6 +1091,8 @@ crt_grp_priv_create(struct crt_grp_priv **grp_priv_created,

grp_priv->gp_size = 0;
grp_priv->gp_refcount = 1;
grp_priv->gp_psr_rank = CRT_NO_RANK;

Copy link
Collaborator

Choose a reason for hiding this comment

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

how does setting it to no rank fix the issue ? the title says look up of rank 0, is rank 0 set to CRT_NO_RANK ?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

during URI_LOOKUP handling there is a following logic block:
if (base_addr == NULL && !crt_is_service()) { D_RWLOCK_RDLOCK(&grp_priv->gp_rwlock); if (tgt_ep->ep_rank == grp_priv->gp_psr_rank && dst_tag == 0) { D_STRNDUP(uri, grp_priv->gp_psr_uri, CRT_ADDR_STR_MAX_LEN); D_RWLOCK_UNLOCK(&grp_priv->gp_rwlock);

If rank being looked up (tgt_ep->ep_rank) matches a psr_rank then we can use psr's uri.
The problem is that we inialized before gp_psr_rank to 0, causing this block then to assume that psr rank has been set (to 0) and therefor gp_psr_uri is valid.

Setting gp_psr_rank to CRT_NO_RANK at init ensures 'gp_psr_rank' is treated properly in checks like above.

rc = D_RWLOCK_INIT(&grp_priv->gp_rwlock, NULL);
if (rc)
D_GOTO(out_swim_lock, rc);
Expand Down
Loading