rpm: disable static build support
PATCH 1/1 rpm: disable static build support
The RPM package requires dynamic linking support (dlfcn.h) for its capabilities related to plugins (and possibly more). The following commit adds a dependency to !BR2_STATIC_LIBS.
Indirectly fixes (see [1]):
http://autobuild.buildroot.net/results/139/1394a6e43529a3969572912dcbdd5606770beeeb/
http://autobuild.buildroot.net/results/f4f/f4f8d21e70fcd79668847407586db2bf3e56feb5/
http://autobuild.buildroot.net/results/80e/80e40419d657ee16b4d38f08d894822e532ab63d/
http://autobuild.buildroot.net/results/f36/f368e1e7310a3cf8266fd351b8b76496868af7e7/
[1]: https://github.com/rpm-software-management/rpm/pull/118
--- package/rpm/Config.in | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
package/rpm/Config.in +3 -2
@@ -1,10 +1,11 @@ -comment "rpm needs a toolchain w/ threads" - depends on !BR2_TOOLCHAIN_HAS_THREADS +comment "rpm needs a toolchain w/ dynamic library and threads" + depends on !BR2_TOOLCHAIN_HAS_THREADS || BR2_STATIC_LIBS depends on BR2_PACKAGE_BUSYBOX_SHOW_OTHERS depends on BR2_USE_MMU config BR2_PACKAGE_RPM bool "rpm" + depends on !BR2_STATIC_LIBS # dlfcn.h depends on BR2_PACKAGE_BUSYBOX_SHOW_OTHERS depends on BR2_TOOLCHAIN_HAS_THREADS # beecrypt depends on BR2_USE_MMU # fork()
Signature · 1 line
2.8.4.windows.1
PATCH 1/1 rpm: disable static build support
On 2017-01-06 19:30 -0500, James Knight spake thusly:
Quoted 10 lines
The RPM package requires dynamic linking support (
dlfcn.h) for its capabilities related to plugins (and possibly more). The following commit adds a dependency to !BR2_STATIC_LIBS.Indirectly fixes (see [1]):
http://autobuild.buildroot.net/results/139/1394a6e43529a3969572912dcbdd5606770beeeb/
http://autobuild.buildroot.net/results/f4f/f4f8d21e70fcd79668847407586db2bf3e56feb5/
http://autobuild.buildroot.net/results/80e/80e40419d657ee16b4d38f08d894822e532ab63d/
http://autobuild.buildroot.net/results/f36/f368e1e7310a3cf8266fd351b8b76496868af7e7/
Although I agree that a depednency on !static is correct because of the use of dlfcn.h et al., the autobuild failures you report are, as you said, unrelated: rpm fails to build because it does not link with -lz.
So I don't think they should be in the commit log. You may put them in a post-commit-log (after the --- line), though.
Quoted 1 line
[1]: https://github.com/rpm-software-management/rpm/pull/118
And I tend to agree with there reasoning. Besides, this is not enough because popt then fails to link because it misses symbols from libintl. And so on...
Quoted 1 line
Signed-off-by James Knight <[email protected]>
Otherwise, I'm OK with that.
Regards,
Yann E. MORIN.
Quoted 29 lines
--- package/rpm/Config.in | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)package/rpm/Config.in +3
-2@@ -1,10 +1,11 @@ -comment "rpm needs a toolchain w/ threads" - depends on !BR2_TOOLCHAIN_HAS_THREADS +comment "rpm needs a toolchain w/ dynamic library and threads" + depends on !BR2_TOOLCHAIN_HAS_THREADS || BR2_STATIC_LIBS depends on BR2_PACKAGE_BUSYBOX_SHOW_OTHERS depends on BR2_USE_MMU config BR2_PACKAGE_RPM bool "rpm" + depends on !BR2_STATIC_LIBS # dlfcn.h depends on BR2_PACKAGE_BUSYBOX_SHOW_OTHERS depends on BR2_TOOLCHAIN_HAS_THREADS # beecrypt depends on BR2_USE_MMU # fork()Signature · 6 lines
2.8.4.windows.1 _______________________________________________ buildroot mailing list buildroot at busybox.net http://lists.busybox.net/mailman/listinfo/buildroot
Signature · 6 lines
.-----------------.--------------------.------------------.--------------------. | Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: | | +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ | | +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no | | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. | '------------------------------^-------^------------------^--------------------'
PATCH 1/1 rpm: disable static build support
Yann,
Quoted 1 line · Yann E. MORIN
Although I agree that a depednency on !static ... I don't think they should be in the commit log.
Ok; I'll cleanup the commit message and resend.
Thanks for the comments.