libbpf-tools: Fix llx printf modifiers in futexctn #4836
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
The perf buffer in exit snoop doesn't maintain 8 byte alignment for start_time and exit_time in struct event. When building with "-fsanitize=alignment -fsanitize-undefined-trap-on-error" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in biosnoop doesn't maintain 8 byte alignment for delta, qdelta, ts and sector in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in filelife doesn't maintain 8 byte alignment for delta_ns in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in fsslower doesn't maintain 8 byte alignment for delta_us, end_ns, offset and size in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in gethostlatency doesn't maintain 8 byte alignment for time in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in opensnoop doesn't maintain 8 byte alignment for ts and callers in struct event. When building with " -fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in runqslower doesn't maintain 8 byte alignment for delta_us in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in statsnoop doesn't maintain 8 byte alignment for ts_ns in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in tcpconnect doesn't maintain 8 byte alignment for ts_us in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in tcplife doesn't maintain 8 byte alignment for saddr, daddr, ts_us, span_us, rx_b and tx_b in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in tcpstates doesn't maintain 8 byte alignment for saddr, daddr, skaddr, ts_us and delta_us in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in tcptracer doesn't maintain 8 byte alignment for saddr_v4, daddr_v4 and ts_us in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
The perf buffer in drsnoop doesn't maintain 8 byte alignment for delta_ns, nr_reclaimed and nr_free_pages in struct event. When building with "-fsanitize=alignment -fsanitize-trap=undefined" failures happen in handle_event. Fix these by copying the event from the perf buffer before accessing. This is similar to a fix in exitsnoop where different ways to handle misaligned pointers were discussed: iovisor#4760 Signed-off-by: Ian Rogers <[email protected]>
Fix the following warning:
```
$ make futexctn
...
futexctn.c: In function ‘print_stack’:
futexctn.c:236:59: warning: format ‘%llx’ expects argument of type ‘long long unsigned int’, but argument 3 has type ‘uint64_t’ {aka ‘long unsigned int’} [-Wformat=]
236 | printf(" #%-2d 0x%016llx [unknown]\n", idx++, ip[i]);
| ~~~~~~^ ~~~~~
| | |
| long long unsigned int uint64_t {aka long unsigned int}
| %016lx
futexctn.c:249:51: warning: format ‘%llx’ expects argument of type ‘long long unsigned int’, but argument 3 has type ‘uint64_t’ {aka ‘long unsigned int’} [-Wformat=]
249 | printf(" #%-2d 0x%016llx", idx++, ip[i]);
| ~~~~~~^ ~~~~~
| | |
| | uint64_t {aka long unsigned int}
| long long unsigned int
| %016lx
...
```
by using inttypes.h.
Signed-off-by: Ian Rogers <[email protected]>
captain5050
requested review from
drzaeus77,
goldshtn,
yonghong-song,
4ast,
brendangregg and
davemarchevsky
as code owners
|
In libbpf-tools/javagc.c, we already have So '%zx' should work in your case. To preserve the code consistency, could you use '%zx' style instead? |
|
BTW, this patch has conflict with latest futexctn.c. You need to rebase your patch on the top of latest master 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.
Fix llx printf modifiers by using inttypes.h's PRIx64.