I compiled gentoo-sources-2.6.16-r1 and rt2500 with GCC 4.1 from portage. The resultant rt2500 kernel module is unusable - dmesg contains this line each time the system attempts to insert the module: "rt2500: falsely claims to have parameter ifname." rt2500 worked fine with 2.6.16 under GCC 3.4.6. emerge --info Portage 2.1_pre7-r5 (default-linux/x86/2006.0, gcc-4.1.0, glibc-2.4-r1, 2.6.14-suspend2-r10 i686) ================================================================= System uname: 2.6.14-suspend2-r10 i686 Intel(R) Celeron(R) CPU 3.06GHz Gentoo Base System version 1.12.0_pre17 distcc 2.18.3 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled] ccache version 2.4 [enabled] dev-lang/python: 2.3.5, 2.4.2-r1 sys-apps/sandbox: 1.2.17 sys-devel/autoconf: 2.13, 2.59-r7 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2 sys-devel/binutils: 2.16.1-r2 sys-devel/libtool: 1.5.22 virtual/os-headers: 2.6.11-r3 ACCEPT_KEYWORDS="x86 ~x86" AUTOCLEAN="yes" CBUILD="i686-pc-linux-gnu" CFLAGS="-Os -march=prescott -pipe -fomit-frame-pointer" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/kde/3/share/config /usr/share/X11/xkb /usr/share/config /var/qmail/control" CONFIG_PROTECT_MASK="/etc/eselect/compiler /etc/gconf /etc/revdep-rebuild /etc/splash /etc/terminfo /etc/texmf/web2c /etc/env.d" CXXFLAGS="-Os -march=prescott -pipe -fomit-frame-pointer" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig ccache confcache distlocks gpg metadata-transfer sandbox sfperms strict usersandbox" GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo" LDFLAGS="-Wl,-O1,--enable-new-dtags,--sort-common" LINGUAS="en_us" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.namerica.gentoo.org/gentoo-portage" USE="x86 X a52 aac acpi alsa animation apache2 artworkextra asf avi bash-completion beagle berkdb bitmap-fonts bluetooth bzip2 cairo cdda cdparanoia cdr chm cli crypt ctype cups daap dba dbus dga divx4linux dlloader dpms dri dts dv dvd dvdr dvdread emboss encode exif expat fam fame fastbuild fat fbsplash ffmpeg fftw firefox flac foomaticdb force-cgi-redirect fortran freetype ftp gaim gd gdbm geoip gif gimpprint glibc-omitfp glitz glut gnome gnutls gphoto2 gpm gstreamer gstreamer10 gtk gtk2 hal hardened imagemagick imlib innodb ipv6 isdnlog java joystick jpeg jpeg2k junit lcms ldap libfame libg++ libnotify libusb libwww lirc live lm_sensors lzo mad memlimit mikmod mjpeg mmap mmx mmxext mng mono motif mozilla mozsvg mp3 mpeg mpeg2 mplayer mysql mythtv nautilus ncurses network nls nptl nptlonly ntfs nvidia offensive ogg oggvorbis openexr opengl pam pango pcre pdf pdflib perl pic pie png pnp posix ppd ppds pppd python quicktime rar readline real rtc samba scanner screensaver sdl session sharedmem shout simplexml soap sockets sox speex spell spl sse sse2 ssl suspend2 svg tcpd theora threads tiff tokenizer transcode truetype truetype-fonts type-fonts type1-fonts udev unicode usb v4l v4l2 vcd videos vorbis webservices wifi win32codecs wmf wv xine xinerama xml xml2 xpm xprint xrandr xscreensaver xsl xv xvid xvmc zeroconf zlib elibc_glibc input_devices_evdev input_devices_joystick input_devices_keyboard input_devices_mouse kernel_linux linguas_en_us userland_GNU video_cards_nv video_cards_nvidia" Unset: ASFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL
What about if you use -O2 instead of -Os? Or get rid of the LDFLAGS?
(In reply to comment #1) > What about if you use -O2 instead of -Os? Or get rid of the LDFLAGS? > I tried again, commenting out LDFLAGS and CFLAGS entirely in make.conf (to use the Gentoo defaults) and the same error persists. Anything else I can try?
Created attachment 84478 [details] ebuild fixing module parameters The only difference from the original ebuild are two sed lines replacing MODULE_PARM macro call with module_param, which fixes the "falsely claims to have parameter ifname" message.
Hello, actually as far as I could test the fix is quite simple: it's enough to change two lines in the source so that the module uses module_param macro instead of the old MODULE_PARM. That enables you to successfully modprobe this module. *However*, 1.1.0_beta3 still won't work with kernel 2.6.16, you need a cvs version. See http://forums.gentoo.org/viewtopic-p-3250022.html for more information. Kind regards, David Ciecierski PS: I'll try to attach a corrected ebuild; sorry if this isn't how it's supposed to be done - never used a bugzilla before.
(In reply to comment #4) > Hello, > actually as far as I could test the fix is quite simple: it's enough to change > two lines in the source so that the module uses module_param macro instead of > the old MODULE_PARM. That enables you to successfully modprobe this module. > *However*, 1.1.0_beta3 still won't work with kernel 2.6.16, you need a cvs > version. See http://forums.gentoo.org/viewtopic-p-3250022.html for more > information. > > Kind regards, > David Ciecierski > > PS: I'll try to attach a corrected ebuild; sorry if this isn't how it's > supposed to be done - never used a bugzilla before. > That patch works for me! On GCC 4.1 with gentoo-sources-2.6.16-r1, the module works perfectly. I do not need to use a CVS version either... nor do I need to edit any of the additional files like that forums post indicates.
> That patch works for me! On GCC 4.1 with gentoo-sources-2.6.16-r1, the module > works perfectly. I do not need to use a CVS version either... nor do I need to > edit any of the additional files like that forums post indicates. That's excellent, for some reason this is not enough for me (kernel config / hw difference?). Would be useful to hear from others if cvs is rally needed or is it a marginally common problem. Kind regards, David Ciecierski
If that fixes it, sounds good to me :)
thanks, fixed
*** Bug 130163 has been marked as a duplicate of this bug. ***
