Date: Tue, 7 Oct 2014 16:08:50 -0400 From: Ed Maste <emaste@freebsd.org> To: Andrew Turner <andrew@fubar.geek.nz> Cc: "freebsd-arm@freebsd.org" <freebsd-arm@freebsd.org> Subject: Re: Why are arm libs branded as SYSV? Message-ID: <CAPyFy2CSAp=TYCzMTy1yz0fqrJzfRjq7%2B=qPobSoeJkUPrk0SA@mail.gmail.com> In-Reply-To: <20141007163440.02a402f8@bender.lan> References: <op.xnc7szg9kndu52@ronaldradial.radialsg.local> <20141007163440.02a402f8@bender.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
On 7 October 2014 11:34, Andrew Turner <andrew@fubar.geek.nz> wrote: >> On my ARM Sheevaplug: >> # file /usr/local/lib/libpcre.so.3 >> /usr/local/lib/libpcre.so.3: ELF 32-bit LSB shared object, ARM, >> EABI5 version 1 (SYSV), dynamically linked, stripped >> ... > > Because the EI_OSABI field, where this value comes from, is documented > to be zero in the ARM AAELF spec. It's somewhat confusing for file to report SYSV, since both ELFOSABI_SYSV and ELFOSABI_NONE are 0. It could perhaps output "SYSV/NONE", although I'm not sure that would be less confusing. I looked for references to EI_OSABI in the source tree. Aside from ldd the only tool I found that might need a change is gcore, which sets the field to ELFOSABI_FREEBSD in generated core files. I'm not sure that this will actually cause trouble in practice though.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAPyFy2CSAp=TYCzMTy1yz0fqrJzfRjq7%2B=qPobSoeJkUPrk0SA>