Is it OK to stabilize =sys-apps/flashrom-0.9.6.1 ? If so, please CC all arches which have stable keywords for older versions of this package and add STABLEREQ keyword to the bug.
Please stabilize =sys-apps/flashrom-0.9.6.1.
amd64 stable
x86 stable
on arm: the ebuild says to enable CONFIG_NIC3COM, so I did: arm_zc linux # make allmodconfig arm_zc linux # grep NIC3COM .config | wc 0 0 0 we need to mask that use?
That's not a kernel config option, but a flashrom one. You can set it on the make command line when building flashrom.
(In reply to comment #5) > That's not a kernel config option, but a flashrom one. You can set it on the > make command line when building flashrom. ok probably there was a misunderstanding: I get: >>> Failed to emerge sys-apps/flashrom-0.9.6, Log file: >>> '/var/tmp/portage/sys-apps/flashrom-0.9.6/temp/build.log' * Package: sys-apps/flashrom-0.9.6 * Repository: gentoo * Maintainer: idl0r@gentoo.org peter@stuge.se * USE: arm bitbang_spi buspirate_spi dediprog drkaiser dummy elibc_glibc gfxnvidia internal kernel_linux nic3com nicintel nicintel_spi ogp_spi pony_spi satasii serprog userland_GNU wiki * FEATURES: sandbox userpriv usersandbox The following features are unavailable on your machine: CONFIG_NIC3COM=yes make: *** [featuresavailable] Error 1
(In reply to comment #6) > * USE: arm bitbang_spi buspirate_spi dediprog drkaiser dummy > elibc_glibc gfxnvidia internal kernel_linux nic3com nicintel nicintel_spi > ogp_spi pony_spi satasii serprog userland_GNU wiki > * FEATURES: sandbox userpriv usersandbox > The following features are unavailable on your machine: CONFIG_NIC3COM=yes That's probably an upstream bug. Albeit unlikely there is no reason why an arm system couldn't have a PCI bus with a 3Com network card on it. For now I guess the ebuild has to be made to mimic whatever rules the Makefile imposes, as to what options can be enabled on what systems. Quick fix is to disallow USE="gfxnvidia nic3com nicintel nicintel_spi satasii" on the arch. Better check the Makefile to see what systems it makes such assumptions on.
arm stable. Last arch, closing
(In reply to Peter Stuge from comment #7) the problem isn't that it's PCI, it's that the driver is written using x86 I/O port insns: https://en.wikibooks.org/wiki/X86_Assembly/Other_Instructions#I.2FO_Instructions and those indeed are not portable :)
(In reply to SpanKY from comment #9) > the problem isn't that it's PCI, it's that the driver is written using x86 > I/O port insns Yeah, that's an upstream bug alright.