[OE-core] [oe-core master v2] elfutils: Fix size passed to snprintf for invalid sh_name case

Luo Zhenhua-B19537 B19537 at freescale.com
Wed Mar 13 02:40:57 UTC 2013


V3 is sent out to include a new patch for this build issue, please ignore this one. 


Best Regards,

Zhenhua

> -----Original Message-----
> From: Luo Zhenhua-B19537
> Sent: Tuesday, March 12, 2013 11:59 AM
> To: openembedded-core at lists.openembedded.org
> Cc: Liu Ting-B28495; Guo Chunrong-B40290; Luo Zhenhua-B19537
> Subject: [oe-core master v2] elfutils: Fix size passed to snprintf for
> invalid sh_name case
> 
> Following build error will appear on Fedora18.
> nm.c: In function 'show_symbols_sysv':
> nm.c:756:27: error: argument to 'sizeof' in 'snprintf' call is the same
> expression as the destination; did you mean to provide an explicit length?
> [-Werror=sizeof-pointer-memaccess]
>     snprintf (name, sizeof name, "[invalid sh_name %#" PRIx32 "]",
>                            ^
> 
> Signed-off-by: Zhenhua Luo <zhenhua.luo at freescale.com>
> ---
>  ...ssed-to-snprintf-for-invalid-sh_name-case.patch |   27
> ++++++++++++++++++++
>  meta/recipes-devtools/elfutils/elfutils_0.148.bb   |    3 +-
>  2 files changed, 29 insertions(+), 1 deletions(-)  create mode 100644
> meta/recipes-devtools/elfutils/elfutils/nm-Fix-size-passed-to-snprintf-
> for-invalid-sh_name-case.patch
> 
> diff --git a/meta/recipes-devtools/elfutils/elfutils/nm-Fix-size-passed-
> to-snprintf-for-invalid-sh_name-case.patch b/meta/recipes-
> devtools/elfutils/elfutils/nm-Fix-size-passed-to-snprintf-for-invalid-
> sh_name-case.patch
> new file mode 100644
> index 0000000..2b5dad3
> --- /dev/null
> +++ b/meta/recipes-devtools/elfutils/elfutils/nm-Fix-size-passed-to-snpr
> +++ intf-for-invalid-sh_name-case.patch
> @@ -0,0 +1,27 @@
> +Upstream-Status: Backport
> +
> +nm: Fix size passed to snprintf for invalid sh_name case.
> +native build failed as following on Fedora18:
> +nm.c: In function 'show_symbols_sysv':
> +nm.c:756:27: error: argument to 'sizeof' in 'snprintf' call is the same
> expression as the destination; did you mean to provide an explicit length?
> [-Werror=sizeof-pointer-memaccess]
> +    snprintf (name, sizeof name, "[invalid sh_name %#" PRIx32 "]",
> +                           ^
> +
> +The original commit is
> +http://git.fedorahosted.org/cgit/elfutils.git/commit/src/nm.c?id=57bd66
> +cabf6e6b9ecf622cdbf350804897a8df58
> +
> +Signed-off-by: Zhenhua Luo <zhenhua.luo at freescale.com>
> +
> +--- elfutils-0.148/src/nm.c.org	2013-03-11 22:36:11.000000000 -0500
> ++++ elfutils-0.148/src/nm.c	2013-03-11 22:46:09.000000000 -0500
> +@@ -752,8 +752,9 @@
> + 			       gelf_getshdr (scn, &shdr_mem)->sh_name);
> +       if (unlikely (name == NULL))
> + 	{
> +-	  name = alloca (sizeof "[invalid sh_name 0x12345678]");
> +-	  snprintf (name, sizeof name, "[invalid sh_name %#" PRIx32 "]",
> ++	  const size_t bufsz = sizeof "[invalid sh_name 0x12345678]";
> ++	  name = alloca (bufsz);
> ++	  snprintf (name, bufsz, "[invalid sh_name %#" PRIx32 "]",
> + 		    gelf_getshdr (scn, &shdr_mem)->sh_name);
> + 	}
> +       scnnames[elf_ndxscn (scn)] = name;
> diff --git a/meta/recipes-devtools/elfutils/elfutils_0.148.bb
> b/meta/recipes-devtools/elfutils/elfutils_0.148.bb
> index 45931b0..bc9e9c6 100644
> --- a/meta/recipes-devtools/elfutils/elfutils_0.148.bb
> +++ b/meta/recipes-devtools/elfutils/elfutils_0.148.bb
> @@ -6,7 +6,7 @@ LIC_FILES_CHKSUM =
> "file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3\
> 
> file://EXCEPTION;md5=570adcb0c1218ab57f2249c67d0ce417"
>  DEPENDS = "libtool bzip2 zlib virtual/libintl"
> 
> -PR = "r9"
> +PR = "r10"
> 
>  SRC_URI = "https://fedorahosted.org/releases/e/l/elfutils/elfutils-
> ${PV}.tar.bz2"
> 
> @@ -31,6 +31,7 @@ SRC_URI += "\
>  	file://mempcpy.patch \
>  	file://fix_for_gcc-4.7.patch \
>  	file://dso-link-change.patch \
> +	file://nm-Fix-size-passed-to-snprintf-for-invalid-sh_name-
> case.patch \
>  "
>  # Only apply when building uclibc based target recipe
> SRC_URI_append_libc-uclibc = " file://uclibc-support.patch"
> --
> 1.7.5.4






More information about the Openembedded-core mailing list