[Buildroot] [PATCH 2/2] package/oprofile: fux musl build

[Arnout Vandecappelle <arnout@mind.be>]

On 31-01-16 11:14, Bernd Kuhls wrote:
> The build error has not yet been found by the autobuilders:
> 
> operf.cpp: In function ?int __delete_old_previous_sample_data(const char*, const stat*, int, FTW*)?:
> operf.cpp:862:10: error: ?FTW_STOP? was not declared in this scope
>    return FTW_STOP;
>           ^
> operf.cpp:864:10: error: ?FTW_CONTINUE? was not declared in this scope
>    return FTW_CONTINUE;
>           ^
> operf.cpp: In function ?void convert_sample_data()?:
> operf.cpp:899:41: error: ?FTW_ACTIONRETVAL? was not declared in this scope
>                  int flags = FTW_DEPTH | FTW_ACTIONRETVAL;
> 
> Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
> ---
>  package/oprofile/0001-musl.patch | 53 ++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 53 insertions(+)
>  create mode 100644 package/oprofile/0001-musl.patch
> 
> diff --git a/package/oprofile/0001-musl.patch b/package/oprofile/0001-musl.patch
> new file mode 100644
> index 0000000..95ae200
> --- /dev/null
> +++ b/package/oprofile/0001-musl.patch
> @@ -0,0 +1,53 @@
> +Fix musl build
> +
> +Downloaded from:
> +https://github.com/openwrt-mirror/openwrt/blob/f22d5e25660106a48727c7aa5d1a73e4171a7987/package/devel/oprofile/patches/100-musl.patch
> +after I found a hint for the patch here:
> +http://patchwork.openembedded.org/patch/112675/
> +
> +Because openwrt removed the oprofile package from their repo last week
> +https://github.com/openwrt-mirror/openwrt/commit/aaf46a8524e138e1673a398e8d2dd9357405b313#diff-fe14456f94abf436d997e2c01c10f3bd
> +I decided to put this patch into the buildroot repo instead of adding
> +_PATCH to oprofile.mk

 Commit message could be a lot better:

Remove non-Posix use of FTW_ACTIONRETVAL

The musl implementation of nftw doesn't support the glibc extension
FTW_ACTIONRETVAL. Since none of the features of FTW_ACTIONRETVAL are
used here, just use the normal nftw return value.


> +
> +Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
> +
> +--- a/pe_profiling/operf.cpp
> ++++ b/pe_profiling/operf.cpp
> +@@ -857,11 +857,14 @@ static int __delete_old_previous_sample_
> +                                 int tflag  __attribute__((unused)),
> +                                 struct FTW *ftwbuf __attribute__((unused)))
> + {
> ++	int err;
> ++
> + 	if (remove(fpath)) {
> ++		err = errno;
> + 		perror("sample data removal error");
> +-		return FTW_STOP;
> ++		return err;
> + 	} else {
> +-		return FTW_CONTINUE;
> ++		return 0;
> + 	}
> + }
> + 
> +@@ -896,7 +899,7 @@ static void convert_sample_data(void)
> + 		return;
> + 
> + 	if (!operf_options::append) {
> +-                int flags = FTW_DEPTH | FTW_ACTIONRETVAL;
> ++                int flags = FTW_DEPTH;
> + 		errno = 0;
> + 		if (nftw(previous_sampledir.c_str(), __delete_old_previous_sample_data, 32, flags) !=0 &&
> + 				errno != ENOENT) {
> +--- a/libop/op_events.c
> ++++ b/libop/op_events.c
> +@@ -83,7 +83,7 @@ static int parse_hex(char const * str)
> + static u64 parse_long_hex(char const * str)
> + {
> + 	u64 value;
> +-	if (sscanf(str, "%Lx", &value) != 1)
> ++	if (sscanf(str, "0x%llx", &value) != 1)

 Why is this change needed? musl supports the standard %Lx specifier, no?

 Regards,
 Arnout

> + 		parse_error("expected long hexadecimal value");
> + 
> + 	fflush(stderr);
> 


-- 
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF