/var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-linux-2.11.1/drivers/dahdi/dahdi-base.c: In function ‘dahdi_ioctl_iomux’: /var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-linux-2.11.1/drivers/dahdi/dahdi-base.c:5954:7: error: implicit declaration of function ‘signal_pending’ [-Werror=implicit-function-declaration] if (signal_pending(current)) { ^~~~~~~~~~~~~~ cc1: some warnings being treated as errors ------------------------------------------------------------------- This is an unstable amd64 chroot image at a tinderbox (==build bot) name: 13.0-systemd-abi32+64_20170506-112642 ------------------------------------------------------------------- gcc-config -l: [1] x86_64-pc-linux-gnu-6.3.0 * Available Python interpreters, in order of preference: [1] python3.4 [2] python2.7 (fallback) Available Ruby profiles: [1] ruby21 (with Rubygems) [2] ruby22 (with Rubygems) *
Created attachment 471998 [details] emerge-info.txt
Created attachment 472000 [details] emerge-history.txt
Created attachment 472002 [details] environment
Created attachment 472004 [details] etc.portage.tbz2
Created attachment 472006 [details] net-misc:dahdi-2.11.1:20170508-131930.log
Created attachment 472008 [details] temp.tbz2
the patches in this bug report helped fix the issue: https://issues.asterisk.org/jira/browse/DAHLIN-354
Created attachment 489714 [details, diff] Another patch to avoid building OPVXA1200 - as it fails with a compilation error Another patch to avoid building OPVXA1200 - as it fails with a compilation error
Hi, The don't build patch is inappropriate. We should rather fix the includes. Further problems building against 4.13.10 kernel: /var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-linux-2.11.1/drivers/dahdi/dahdi-sysfs.c:273:2: error: unknown field 'dev_attrs' specified in initializer .dev_attrs = span_dev_attrs, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=879043 and https://issues.asterisk.org/jira/browse/DAHLIN-356 I can confirm that with BOTH patches at the link Joe M supplied in comment 7, as well as the -v3 patch at the 356 DAHDI bug report dahdi also fails for me on opvxa1200. This is due to touch_softlockup_watchdog declaration having moved from linux/sched.h to linux/nmi.h - I'm not sure which kernel version exactly yet. This leaves: dahd i-linux-2.11.1/drivers/dahdi/xpp/xbus-pcm.c In file included from ./include/linux/kobject.h:21:0, from ./include/linux/module.h:17, from /var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-linux-2.11.1/drivers/dahdi/xpp/xbus-sysfs.c:24: /var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-linux-2.11.1/drivers/dahdi/xpp/xbus-sysfs.c:801:23: error: 'hwid_show' undeclared here (not in a function) static DEVICE_ATTR_RO(hwid); ^ ./include/linux/sysfs.h:116:10: note: in definition of macro '__ATTR_RO' .show = _name##_show, \ ^ /var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-linux-2.11.1/drivers/dahdi/xpp/xbus-sysfs.c:801:8 note: in expansion of macro 'DEVICE_ATTR_RO' static DEVICE_ATTR_RO(hwid); I can't find any reference to a hwid attribute other than this, which is introduced by dev_attrs_groups-v3.patch - any objections to simply mangling the patch? Sorry for only picking up on this report now - been using 4.0.9 kernel until very recently due to being the least unstable version in 4.X branch. I'll attach individual patches here in a bit. Just need to find which kernel version the nmi.h changed.
Created attachment 502798 [details, diff] signal pending patch
Created attachment 502804 [details, diff] atomic read patch
Created attachment 502806 [details, diff] dev attrs patch
Created attachment 502848 [details, diff] opxval200 compile fix # ebuild /usr/portage/net-misc/dahdi/dahdi-2.11.1.ebuild clean unpack ... # cd /var/tmp/portage/net-misc/dahdi-2.11.1/work/dahdi-patchset/ # cp ~/0001-signal_pending-is-now-in-linux-sched-signal.h-includ.patch 05-linux-sched-signal.diff # cp ~/0002-atomic_read-refcount_read.patch 06-linux-refcount.diff # cp ~/dev_attrs_groups-v3.patch 07-dev-attrs.diff # mv 98-non-digium-hardware-and-oslec.diff 97-non-digium-hardware-and-oslec.diff # mv 99-irqf-disabled.diff 98-irqf-disabled.diff # cp ~/opvxa1200-nmi.diff 99-opvxa1200-nmi.diff # ebuild /usr/portage/net-misc/dahdi/dahdi-2.11.1.ebuild compile ... >>> Source compiled. DISCLAIMER: I don't have all the hardware to test all of the various modules. It does compile and it does merge with the additional four patches. Three of the four patches comes from the Digium bug tracker (with a minor modification). The fourth is a trivial patch to accomodate commit d151b27d3f86589308cd8c891fdfd2db5f8e80d6 in kernel 4.11 as submitted by Ingo Molnar and acked by Linus. The commit message hints that linux/nmi.h may be renamed again at some point in the future, so when anyone picks that up, please contact me and I'll adjust the patch accordingly.
*** Bug 638380 has been marked as a duplicate of this bug. ***
Created attachment 535246 [details, diff] xpp-command-timer-core.diff
Created attachment 535248 [details, diff] 09-init-timer.diff Additional patches required for dahdi 2.11.1 to compile against 4.15+.
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=544f094826762efc335dcfcf8392e07b0883894d commit 544f094826762efc335dcfcf8392e07b0883894d Author: Tony Vroon <chainsaw@gentoo.org> AuthorDate: 2018-06-08 12:35:55 +0000 Commit: Tony Vroon <chainsaw@gentoo.org> CommitDate: 2018-06-08 12:36:16 +0000 net-misc/dahdi: Address build failures on newer kernels Reported by Toralf Förster; patches scavenged & rebased by Jaco Kroon. Closes: https://bugs.gentoo.org/617876 Package-Manager: Portage-2.3.40, Repoman-2.3.9 net-misc/dahdi/Manifest | 1 + net-misc/dahdi/dahdi-2.11.1-r1.ebuild | 80 +++++++++++++++++++++++++++++++++++ 2 files changed, 81 insertions(+)
