Verify format specifiers in bpf_trace_printk in rewriter #1046
Conversation
This file contains 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
pchaigno
force-pushed
the
check-fmt-printk
branch
from
6882f8c
to
bd9e50c
Compare
drzaeus77
approved these changes
Mar 13, 2017
| @@ -523,6 +524,54 @@ bool BTypeVisitor::VisitCallExpr(CallExpr *Call) { | |||
| return true; | |||
| } | |||
|
|
|||
| bool BTypeVisitor::checkFormatSpecifiers(string fmt, SourceLocation loc) { | |||
There was a problem hiding this comment.
Why not const string& to avoid copy?
| } | ||
| if (fmt[i] != '%') | ||
| continue; | ||
| if (nb_specifiers >= 3) { |
There was a problem hiding this comment.
Seems like printing literal "%%" isn't supported here or in the kernel. Is that the right interpretation?
There was a problem hiding this comment.
Yes. @4ast Is that on purpose?
|
Buildbot, you're not listening! |
pchaigno
force-pushed
the
check-fmt-printk
branch
from
bd9e50c
to
533c25e
Compare
|
Build failed legitimately on fedora it seems. Failed in py_test_clang with: |
|
The three errors are expected (see the Ubuntu builds), but we get a segfault afterwards: I'll have to start a Fedora to check it out. |
|
@pchaigno Have you had any luck reproducing the build issue? |
|
@goldshtn Yes, I'm able to reproduce. It doesn't look related to the changes I made though, since I'm able to reproduce by copying the new tests on the master branch. Weirdly, if I replace: bpf_trace_printk("%0.2f\\n", 1.0);with bpf_trace_printk("%0.2f\\n", 1);the segfault disappears... I'm trying to get more information, but my day job doesn't leave me much free time. I'd appreciate any help :-) |
|
Yup, I can confirm that just an empty script that uses /cc @drzaeus77 |
|
Hmm, maybe something with how LLVM deals with floats (unsupported in bpf), that only happens on Fedora due to the LLVM version that is used. I would suggest fixing up this PR to use |
Verifies format specifiers while rewriting calls to bpf_trace_printk and prints a warning the printk will be rejected by the kernel at runtime. For tests, redirects stderr at the file descriptor level in order to catch warnings from the C library.
pchaigno
force-pushed
the
check-fmt-printk
branch
from
533c25e
to
9341942
Compare
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.
Verifies format specifiers while rewriting calls to bpf_trace_printk
and prints a warning the printk will be rejected by the kernel at
runtime.
For tests, redirects stderr at the file descriptor level in order to
catch warnings from the C library.
Fixes #124.
/cc @drzaeus77
I noticed the kernel doesn't check that the number of format
specifiers equals the number of arguments. Is that normal?
(
CAP_SYS_ADMINis required to callbpf_trace_printk, but it stillfeels weird.)