From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yann E. MORIN Date: Mon, 10 Jul 2017 23:57:48 +0200 Subject: [Buildroot] [PATCH 2/4 v3] package/kvm-unit-tests: introduce _ARCH_SUPPORTS In-Reply-To: <20170710234741.554c7a44@windsurf.lan> References: <996d1d52c27da1a7725ed425ac895c3a7b5b7301.1499592591.git.yann.morin.1998@free.fr> <20170710180949.764a0fdb@windsurf.lan> <20170710162056.GB3014@scaer> <20170710234741.554c7a44@windsurf.lan> Message-ID: <20170710215748.GD22898@scaer> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Thomas, All, On 2017-07-10 23:47 +0200, Thomas Petazzoni spake thusly: > On Mon, 10 Jul 2017 18:20:56 +0200, Yann E. MORIN wrote: > > > What about the x86_64 dependency on a host compiler >= 4.5 ? > > In the manual, it is said that one should not add a comment about a > > dependency on the host toolchain... > > The manual says one thing, reality says another: > > package/collectd/Config.in:comment "write_prometheus needs a toolchain w/ C++, host gcc >= 4.5" > package/cppcms/Config.in:comment "icu support needs a toolchain w/ gcc >= 4.8, host gcc >= 4.8" > package/icu/Config.in:comment "icu needs a toolchain w/ C++, wchar, threads, gcc >= 4.8, host gcc >= 4.8" > package/kodi/Config.in:comment "kodi needs python w/ .py modules, a uClibc or glibc toolchain w/ C++, locale, threads, wchar, dynamic library, gcc >= 4.8, host gcc >= 4.6" > package/mfgtools/Config.in.host:comment "mfgtools needs host gcc >= 4.8" > package/midori/Config.in:comment "midori needs libgtk3 and a glibc toolchain w/ C++, gcc >= 4.9, host gcc >= 4.8" > package/php/Config.ext:comment "intl support needs a toolchain w/ C++, wchar, threads, gcc >= 4.8, host gcc >= 4.8" > package/protobuf-c/Config.in:comment "protobuf-c needs a toolchain w/ C++, threads, host gcc >= 4.5" > package/python-mwscrape2slob/Config.in:comment "python-mwscrape2slob needs a toolchain w/ C++, gcc >= 4.8, host gcc >= 4.8" > package/python-pyicu/Config.in:comment "python-pyicu needs a toolchain w/ C++, gcc >= 4.8, host gcc >= 4.8" > package/python-slob/Config.in:comment "python-slob needs a toolchain w/ C++, gcc >= 4.8, host gcc >= 4.8" > package/qt5/qt5base/Config.in:comment "icu support needs a toolchain w/ gcc >= 4.8, host gcc >= 4.8" > package/qt5/qt5webkit/Config.in:comment "qt5webkit needs a toolchain w/ dynamic library, gcc >= 4.8, host gcc >= 4.8" > package/riemann-c-client/Config.in:comment "riemann-c-client needs a toolchain w/ C++, threads, host gcc >= 4.5" > package/webkitgtk/Config.in:comment "webkitgtk needs libgtk3 and a glibc toolchain w/ C++, gcc >= 4.9, host gcc >= 4.8" > > I'm not sure why we would not add Config.in comment for dependencies on > the host toolchain. What's the reasoning for *not* adding such > dependencies ? Because there is not much a user can do about the toolchain on host host machine? As much as it is "possible" to change the version for the cross-gcc, the native compiler is often not replaceable. This is most probably the case on enterprise-grade servers, where users are not root (and are stuck with a precambrian host compiler)... I wrote the comment, checked the manual for the template, then removed the comment. Oh well, here it is again, then: comment "kvm-unit-tests needs a host gcc >= 4.5" depends on BR2_x86_64 depends on !BR2_HOST_GCC_AT_LEAST_4_5 Hehe! ;-) Regards, Yann E. MORIN. > Best regards, > > Thomas > -- > Thomas Petazzoni, CTO, Free Electrons > Embedded Linux and Kernel engineering > http://free-electrons.com -- .-----------------.--------------------.------------------.--------------------. | 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. | '------------------------------^-------^------------------^--------------------'