bpf: Fast-Path approach for BPF program termination #5911
Closed
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
Author
|
Upstream branch: e12873e |
kernel-patches-daemon-bpf-rc
bot
force-pushed
the
bpf-next_base
branch
from
2d215b0 to
05282f4
Compare
Author
|
Upstream branch: 93a83d0 |
kernel-patches-daemon-bpf-rc
bot
force-pushed
the
series/999812=>bpf-next
branch
from
6b8279c to
0dea227
Compare
kernel-patches-daemon-bpf-rc
bot
force-pushed
the
bpf-next_base
branch
from
05282f4 to
e2bfc81
Compare
Introduced the definition of struct bpf_term_aux_states required to support fast-path termination of BPF programs. Added the memory allocation and free logic for newly added term_states feild in struct bpf_prog. Signed-off-by: Raj Sahu <[email protected]> Signed-off-by: Siddharth Chintamaneni <[email protected]>
Create callsites tables and store jit indexes of RET_NULL calls to poke them later with dummy functions. Additional to jit indexes, meta data about helpers/kfuncs/loops is stored. Later this could be extended to remaining potential long running iterator helpers/kfuncs. Signed-off-by: Raj Sahu <[email protected]> Signed-off-by: Siddharth Chintamaneni <[email protected]>
Update softlock detection logic to detect any stalls due to BPF programs. When softlockup is detected, bpf_die will be added to a workqueue on a CPU. With this implementation termination handler will only get triggered when CONFIG_SOFTLOCKUP_DETECTOR is enabled. Inside bpf_die, we perform the text_poke to stub helpers/kfuncs. The current implementation handles termination of long running bpf_loop iterators both inlining and non-inlining case. The limitation of this implementation is that the termination handler atleast need a single CPU to run. Signed-off-by: Raj Sahu <[email protected]> Signed-off-by: Siddharth Chintamaneni <[email protected]>
…sted bpf loops Adds tests checks for loops termination which are nested. 32/1 bpf_termination/bpf_termination:OK 32 bpf_termination:OK Summary: 1/1 PASSED, 0 SKIPPED, 0 FAILED Signed-off-by: Raj Sahu <[email protected]> Signed-off-by: Siddharth Chintamaneni <[email protected]>
Author
|
Upstream branch: 60ef541 |
kernel-patches-daemon-bpf-rc
bot
force-pushed
the
series/999812=>bpf-next
branch
from
0dea227 to
286cfff
Compare
Author
|
At least one diff in series https://patchwork.kernel.org/project/netdevbpf/list/?series=999812 expired. Closing PR. |
kernel-patches-daemon-bpf-rc
bot
deleted the
series/999812=>bpf-next
branch
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Pull request for series with
subject: bpf: Fast-Path approach for BPF program termination
version: 1
url: https://patchwork.kernel.org/project/netdevbpf/list/?series=999812