Skip to contentSkip to conversation

perf FTBFS due to -Werror with gcc 14.1.1

DISCUSSION Milian Wolff 3 participants 4 messages Jun 4, 2024

lore
Share & tools
lore
#0 lore raw

Hey all,

I cannot build the acme perf/core branch on Arch anymore with gcc 14.1.1 and Linux 6.9.3 due to multiple -Werror issues. Is this known and if so is anyone working on it? Otherwise we can try to get this fixed ourselves and upstream the patches then:

```
  CC      builtin-record.o
  CC      builtin-top.o
  CC      util/metricgroup.o
  CC      util/hist.o
  CC      util/synthetic-events.o
  CC      util/bpf-loader.o
tests/bpf.c: In function ‘epoll_pwait_loop’:
tests/bpf.c:36:17: error: argument 2 null where non-null expected [-
Werror=nonnull]
   36 |                 epoll_pwait(-(i + 1), NULL, 0, 0, NULL);
      |                 ^~~~~~~~~~~
In file included from tests/bpf.c:5:
/usr/include/sys/epoll.h:134:12: note: in a call to function ‘epoll_pwait’ 
declared ‘nonnull’
  134 | extern int epoll_pwait (int __epfd, struct epoll_event *__events,
      |            ^~~~~~~~~~~
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
tests/bpf.o] Error 1
make[3]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:140: 
tests] Error 2
make[3]: *** Waiting for unfinished jobs....
util/metricgroup.c: In function ‘setup_metric_events’:
util/metricgroup.c:289:39: error: ‘calloc’ sizes specified with ‘sizeof’ in 
the earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
  289 |         metric_events = calloc(sizeof(void *), ids_size + 1);
      |                                       ^~~~
util/metricgroup.c:289:39: note: earlier argument should specify number of 
elements, later size of each element
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
util/metricgroup.o] Error 1
make[4]: *** Waiting for unfinished jobs....
builtin-top.c: In function ‘prompt_integer’:
builtin-top.c:360:21: error: allocation of insufficient size ‘0’ for type 
‘char’ with size ‘1’ [-Werror=alloc-size]
  360 |         char *buf = malloc(0), *p;
      |                     ^~~~~~
cc1: all warnings being treated as errors
make[3]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
builtin-top.o] Error 1
util/bpf-loader.c: In function ‘config_bpf_program’:
util/bpf-loader.c:551:29: error: ‘calloc’ sizes specified with ‘sizeof’ in the 
earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
  551 |         priv = calloc(sizeof(*priv), 1);
      |                             ^
util/bpf-loader.c:551:29: note: earlier argument should specify number of 
elements, later size of each element
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
util/bpf-loader.o] Error 1
util/synthetic-events.c: In function ‘perf_event__synthesize_threads’:
util/synthetic-events.c:1042:44: error: ‘calloc’ sizes specified with ‘sizeof’ 
in the earlier argument and not in the later argument [-Werror=calloc-
transposed-args]
 1042 |         synthesize_threads = calloc(sizeof(pthread_t), thread_nr);
      |                                            ^~~~~~~~~
util/synthetic-events.c:1042:44: note: earlier argument should specify number 
of elements, later size of each element
util/synthetic-events.c:1046:29: error: ‘calloc’ sizes specified with ‘sizeof’ 
in the earlier argument and not in the later argument [-Werror=calloc-
transposed-args]
 1046 |         args = calloc(sizeof(*args), thread_nr);
      |                             ^
util/synthetic-events.c:1046:29: note: earlier argument should specify number 
of elements, later size of each element
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
util/synthetic-events.o] Error 1
builtin-record.c: In function ‘cmd_record’:
builtin-record.c:4091:62: error: ‘calloc’ sizes specified with ‘sizeof’ in the 
earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
 4091 |                 rec->switch_output.filenames = calloc(sizeof(char *),
      |                                                              ^~~~
builtin-record.c:4091:62: note: earlier argument should specify number of 
elements, later size of each element
cc1: all warnings being treated as errors
make[3]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
builtin-record.o] Error 1
util/hist.c: In function ‘hist_entry__init’:
util/hist.c:494:49: error: ‘calloc’ sizes specified with ‘sizeof’ in the 
earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
  494 |                 he->res_samples = calloc(sizeof(struct res_sample),
      |                                                 ^~~~~~
```

Thanks

Signature · 4 lines
Milian Wolff | [email protected] | Senior Software Engineer
KDAB (Deutschland) GmbH, a KDAB Group company
Tel: +49-30-521325470
KDAB - The Qt, C++ and OpenGL Experts
#1 lore raw
Quoted 3 lines · Milian Wolff

Hey all,

I cannot build the acme perf/core branch on Arch anymore with gcc 14.1.1 and

perf/core isn't being updated for quite a while, we nowadays update just the perf-tools (current merge window) and perf-tools-next (next merge window/devel).

Also I still keep my tree uptodate when I'm updating those branches, but nowadays the trees/branches to be used are:

https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools.git perf-tools

https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git perf-tools-next

As Namhyung Kim is a perf tools co-maintainer and keeps one of these trees+branch while I keep the other.

Right now I'm taking care of perf-tools while Namhyung is managing perf-tools.next.

Quoted 3 lines

Linux 6.9.3 due to multiple -Werror issues. Is this known and if so is anyone working on it? Otherwise we can try to get this fixed ourselves and upstream the patches then:

Can you please check if the problems are present in the above trees+branches?

- Arnaldo

Quoted 99 lines
```
  CC      builtin-record.o
  CC      builtin-top.o
  CC      util/metricgroup.o
  CC      util/hist.o
  CC      util/synthetic-events.o
  CC      util/bpf-loader.o
tests/bpf.c: In function ‘epoll_pwait_loop’:
tests/bpf.c:36:17: error: argument 2 null where non-null expected [-
Werror=nonnull]
   36 |                 epoll_pwait(-(i + 1), NULL, 0, 0, NULL);
      |                 ^~~~~~~~~~~
In file included from tests/bpf.c:5:
/usr/include/sys/epoll.h:134:12: note: in a call to function ‘epoll_pwait’ 
declared ‘nonnull’
  134 | extern int epoll_pwait (int __epfd, struct epoll_event *__events,
      |            ^~~~~~~~~~~
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
tests/bpf.o] Error 1
make[3]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:140: 
tests] Error 2
make[3]: *** Waiting for unfinished jobs....
util/metricgroup.c: In function ‘setup_metric_events’:
util/metricgroup.c:289:39: error: ‘calloc’ sizes specified with ‘sizeof’ in 
the earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
  289 |         metric_events = calloc(sizeof(void *), ids_size + 1);
      |                                       ^~~~
util/metricgroup.c:289:39: note: earlier argument should specify number of 
elements, later size of each element
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
util/metricgroup.o] Error 1
make[4]: *** Waiting for unfinished jobs....
builtin-top.c: In function ‘prompt_integer’:
builtin-top.c:360:21: error: allocation of insufficient size ‘0’ for type 
‘char’ with size ‘1’ [-Werror=alloc-size]
  360 |         char *buf = malloc(0), *p;
      |                     ^~~~~~
cc1: all warnings being treated as errors
make[3]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
builtin-top.o] Error 1
util/bpf-loader.c: In function ‘config_bpf_program’:
util/bpf-loader.c:551:29: error: ‘calloc’ sizes specified with ‘sizeof’ in the 
earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
  551 |         priv = calloc(sizeof(*priv), 1);
      |                             ^
util/bpf-loader.c:551:29: note: earlier argument should specify number of 
elements, later size of each element
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
util/bpf-loader.o] Error 1
util/synthetic-events.c: In function ‘perf_event__synthesize_threads’:
util/synthetic-events.c:1042:44: error: ‘calloc’ sizes specified with ‘sizeof’ 
in the earlier argument and not in the later argument [-Werror=calloc-
transposed-args]
 1042 |         synthesize_threads = calloc(sizeof(pthread_t), thread_nr);
      |                                            ^~~~~~~~~
util/synthetic-events.c:1042:44: note: earlier argument should specify number 
of elements, later size of each element
util/synthetic-events.c:1046:29: error: ‘calloc’ sizes specified with ‘sizeof’ 
in the earlier argument and not in the later argument [-Werror=calloc-
transposed-args]
 1046 |         args = calloc(sizeof(*args), thread_nr);
      |                             ^
util/synthetic-events.c:1046:29: note: earlier argument should specify number 
of elements, later size of each element
cc1: all warnings being treated as errors
make[4]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
util/synthetic-events.o] Error 1
builtin-record.c: In function ‘cmd_record’:
builtin-record.c:4091:62: error: ‘calloc’ sizes specified with ‘sizeof’ in the 
earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
 4091 |                 rec->switch_output.filenames = calloc(sizeof(char *),
      |                                                              ^~~~
builtin-record.c:4091:62: note: earlier argument should specify number of 
elements, later size of each element
cc1: all warnings being treated as errors
make[3]: *** [/home/milian/projects/src/linux/tools/build/Makefile.build:97: 
builtin-record.o] Error 1
util/hist.c: In function ‘hist_entry__init’:
util/hist.c:494:49: error: ‘calloc’ sizes specified with ‘sizeof’ in the 
earlier argument and not in the later argument [-Werror=calloc-transposed-
args]
  494 |                 he->res_samples = calloc(sizeof(struct res_sample),
      |                                                 ^~~~~~
```

Thanks

Signature · 4 lines
Milian Wolff | [email protected] | Senior Software Engineer
KDAB (Deutschland) GmbH, a KDAB Group company
Tel: +49-30-521325470
KDAB - The Qt, C++ and OpenGL Experts
#2 lore raw
Quoted 15 lines · On Dienstag, 4. Juni 2024 15:45:52 MESZ Arnaldo Carvalho de Melo wrote:
Quoted 4 lines · Milian Wolff

Hey all,

I cannot build the acme perf/core branch on Arch anymore with gcc 14.1.1 and

perf/core isn't being updated for quite a while, we nowadays update just the perf-tools (current merge window) and perf-tools-next (next merge window/devel).

Also I still keep my tree uptodate when I'm updating those branches, but nowadays the trees/branches to be used are:

https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools.git
perf-tools

Quoted 8 lines

https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git
perf-tools-next

As Namhyung Kim is a perf tools co-maintainer and keeps one of these trees+branch while I keep the other.

Right now I'm taking care of perf-tools while Namhyung is managing perf-tools.next.

Ah indeed that's it, thank you!

Which one of the two would you recommend for interested parties that want to keep track of perf tools development, i.e. what was perf/core?

Thanks

Signature · 4 lines
Milian Wolff | [email protected] | Senior Software Engineer
KDAB (Deutschland) GmbH, a KDAB Group company
Tel: +49-30-521325470
KDAB - The Qt, C++ and OpenGL Experts
#3 lore raw

Hello,

Quoted 22 lines · Milian Wolff
Quoted 21 lines · On Dienstag, 4. Juni 2024 15:45:52 MESZ Arnaldo Carvalho de Melo wrote:
Quoted 4 lines · Milian Wolff

Hey all,

I cannot build the acme perf/core branch on Arch anymore with gcc 14.1.1 and

perf/core isn't being updated for quite a while, we nowadays update just the perf-tools (current merge window) and perf-tools-next (next merge window/devel).

Also I still keep my tree uptodate when I'm updating those branches, but nowadays the trees/branches to be used are:

https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools.git
perf-tools

https://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git
perf-tools-next

As Namhyung Kim is a perf tools co-maintainer and keeps one of these trees+branch while I keep the other.

Quoted 7 lines
Quoted 2 lines

Right now I'm taking care of perf-tools while Namhyung is managing perf-tools.next.

Ah indeed that's it, thank you!

Which one of the two would you recommend for interested parties that want to keep track of perf tools development, i.e. what was perf/core?

The perf-tools-next is for new development (previously perf/core) and the perf-tools is for fixes in the current cycles (previously perf/urgent).

Thanks,
Namhyung