Date: Tue, 11 Jan 2011 13:49:40 +0200 From: Kostik Belousov <kostikbel@gmail.com> To: Dimitry Andric <dim@freebsd.org> Cc: svn-src-projects@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r217260 - projects/binutils-2.17/contrib/binutils/ld/emultempl Message-ID: <20110111114940.GG2518@deviant.kiev.zoral.com.ua> In-Reply-To: <201101111048.p0BAmowJ017937@svn.freebsd.org> References: <201101111048.p0BAmowJ017937@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--+sHJum3is6Tsg7/J Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Jan 11, 2011 at 10:48:50AM +0000, Dimitry Andric wrote: > Author: dim > Date: Tue Jan 11 10:48:50 2011 > New Revision: 217260 > URL: http://svn.freebsd.org/changeset/base/217260 >=20 > Log: > Put in a temporary kludge to force ld to put .note sections just after > .interp sections. This seems to have been broken by binutils commit > 7f55fa5745fbe04b91e1fbd668d102f3166e03d8, but the specific reasons are > still unclear. > =20 > For now, this fix is needed to make sure .note.ABI-tag sections always > end up in the first page of the resulting executable. >=20 > Modified: > projects/binutils-2.17/contrib/binutils/ld/emultempl/elf32.em >=20 > Modified: projects/binutils-2.17/contrib/binutils/ld/emultempl/elf32.em > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D > --- projects/binutils-2.17/contrib/binutils/ld/emultempl/elf32.em Tue Jan= 11 06:07:36 2011 (r217259) > +++ projects/binutils-2.17/contrib/binutils/ld/emultempl/elf32.em Tue Jan= 11 10:48:50 2011 (r217260) > @@ -1540,8 +1540,7 @@ gld${EMULATION_NAME}_place_orphan (asect > if ((s->flags & SEC_ALLOC) =3D=3D 0) > ; > else if ((s->flags & SEC_LOAD) !=3D 0 > - && ((iself && sh_type =3D=3D SHT_NOTE) > - || (!iself && CONST_STRNEQ (secname, ".note")))) > + && CONST_STRNEQ (secname, ".note")) I am not sure, but it looks likely that (iself && sh_type =3D=3D SHT_NOTE) part of condition better be kept. > place =3D &hold[orphan_interp]; > else if ((s->flags & (SEC_LOAD | SEC_HAS_CONTENTS)) =3D=3D 0) > place =3D &hold[orphan_bss]; --+sHJum3is6Tsg7/J Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (FreeBSD) iEYEARECAAYFAk0sQ9QACgkQC3+MBN1Mb4jR8ACg4c8WuAO40cULeyqrT0SUQ7SO 0qIAoPOH8YheoZwY3UBrEXVeGp4vAfBH =XlrW -----END PGP SIGNATURE----- --+sHJum3is6Tsg7/J--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110111114940.GG2518>