Bluetooth: Host: Fix L2CAP EINPROGRESS refs #76835
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
1867fcb to
9f2dc36
Compare
zephyrbot
added
area: Bluetooth Host
jhedberg
reviewed
Aug 9, 2024
jori-nordic
reviewed
Aug 9, 2024
Contributor
jori-nordic
left a comment
jori-nordic
left a comment
There was a problem hiding this comment.
Please also clean-up and submit the test that found the bug.
Member
|
GitHub doesn't allow me to do inline review of commit messages, so I'll do it here:
The second function reference there should be |
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
2da64d7 to
9f9052e
Compare
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
9f9052e to
6f00334
Compare
Contributor
Author
|
Added test and fixed the commented-on issues. |
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
6f00334 to
ae2630d
Compare
jori-nordic
suggested changes
Aug 12, 2024
tests/bsim/bluetooth/host/l2cap/einprogress/tester/CMakeLists.txt
Outdated
Show resolved
Hide resolved
Thalley
requested changes
Aug 13, 2024
tests/bsim/bluetooth/host/l2cap/einprogress/tester/src/tester.c
Outdated
Show resolved
Hide resolved
tests/bsim/bluetooth/host/l2cap/einprogress/tester/src/tester.c
Outdated
Show resolved
Hide resolved
Contributor
Author
|
I'll delete the tester code, and rewrite the test. Basing it on a different test was a mistake. |
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
938ea3d to
ae43589
Compare
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
ae43589 to
201b1c1
Compare
jori-nordic
previously approved these changes
Sep 13, 2024
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
2 times, most recently
from
dae83b6 to
3ebe4e8
Compare
Fix discrepancy in reference management between calls to `bt_l2cap_chan_ops.recv` when the application returns `-EINPROGRESS`. There are two call sites, `l2cap_chan_le_recv_sdu` and `l2cap_chan_le_recv`, that were inconsistent. `l2cap_chan_le_recv_sdu` moves the reference, and this patch updates `l2cap_chan_le_recv` to do the same. This behavior is also now documented. This bug has existed since the introduction of this feature in 3151d26. Signed-off-by: Aleksander Wasaznik <[email protected]>
For ease of development, we should log the event as an error. Signed-off-by: Aleksander Wasaznik <[email protected]>
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
3ebe4e8 to
83cebe8
Compare
Thalley
previously approved these changes
Oct 16, 2024
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
2 times, most recently
from
8229fd8 to
f4b8925
Compare
Thalley
previously approved these changes
Oct 21, 2024
theob-pro
previously approved these changes
Oct 22, 2024
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
2 times, most recently
from
e0b7903 to
58da9bd
Compare
theob-pro
previously approved these changes
Oct 25, 2024
jhedberg
reviewed
Oct 25, 2024
Thalley
reviewed
Oct 25, 2024
subsys/bluetooth/host/hci_core.c
Outdated
Contributor
There was a problem hiding this comment.
I'm not a fan of public testing headers. If this is for internal testing, then I think such headers are better placed in a non-public place. Placing a testing.h makes more sense if it's a testing API that is designed to be used for users (maybe that is the intention?).
We actually recently moved the Mesh testing.h header file from the public include directory to internal.
This is needed for a test to catch a double-free. Signed-off-by: Aleksander Wasaznik <[email protected]>
This is shorthand for random static addresses. It's similar to `bt_addr_le_from_str`, but is a macro that results in an object literal, making it more versatile and less verbose. This macro only gives access to the first 255 random static addresses, but this ought to be enough addresses for testing. Signed-off-by: Aleksander Wasaznik <[email protected]>
The test implementation is based on a copy of the HFC multilink test. The test verifies that the stack respects the reference counting of SDU buffers when the L2CAP -EINPROGRESS feature is used. Signed-off-by: Aleksander Wasaznik <[email protected]>
alwa-nordic
force-pushed
the
fix-ref-l2cap-einprogress
branch
from
58da9bd to
e8b9c00
Compare
jhedberg
approved these changes
Oct 25, 2024
Thalley
approved these changes
Oct 26, 2024
Contributor
Thalley
left a comment
Thalley
left a comment
There was a problem hiding this comment.
As mentioned, I will not block this PR based on the testing API, but I do have my reservations about it.
This was referenced Dec 10, 2024
Merged
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix discrepancy in reference management between calls to
bt_l2cap_chan_ops.recvwhen the application returns-EINPROGRESS.There are two call sites,
l2cap_chan_le_recv_sduandl2cap_chan_le_recv, that were inconsistent.l2cap_chan_le_recv_sdumoves the reference, and this patch updatesl2cap_chan_le_recv_sduto do the same.This behavior is also now documented.
This bug has existed since the introduction of this feature in 3151d26.
Signed-off-by: Aleksander Wasaznik [email protected]