From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Tue, 22 Nov 2016 09:35:15 +0100 Subject: [Buildroot] [PATCH 2/2] toolchain: simplify the conditions for BR2_TOOLCHAIN_ARM_HAS_SYNC_8 In-Reply-To: <20161121234340.30463-2-arnout@mind.be> References: <20161121234340.30463-1-arnout@mind.be> <20161121234340.30463-2-arnout@mind.be> Message-ID: <20161122093515.6db0df30@free-electrons.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net Hello, On Tue, 22 Nov 2016 00:43:40 +0100, Arnout Vandecappelle (Essensium/Mind) wrote: > config BR2_TOOLCHAIN_ARM_HAS_SYNC_8 > bool > default y > depends on BR2_arm || BR2_armeb > - depends on BR2_TOOLCHAIN_GCC_AT_LEAST_4_7 > - depends on BR2_TOOLCHAIN_USES_GLIBC || BR2_ARM_CPU_ARMV7A > + depends on BR2_TOOLCHAIN_GCC_AT_LEAST_4_8 Can we make this instead: depends on (BR2_TOOLCHAIN_GCC_AT_LEAST_4_7 && BR2_TOOLCHAIN_USES_GLIBC) || BR2_TOOLCHAIN_GCC_AT_LEAST_4_8 This way, glibc toolchains (which were never affected by the _write issue) still expose the fact that they support the __sync_8 thing. Otherwise, your change introduces a regression: a 4.7 glibc external toolchain was prior to your patch exposing the fact that it supports 8-byte sync, but no longer after your patch. Also, what about the __atomic 8 bytes? As we discussed yesterday, it's in fact affected by the same issue. How do we deal with that? Do we introduce per-size BR2_TOOLCHAIN_HAS_ATOMIC_ options? Best regards, Thomas -- Thomas Petazzoni, CTO, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com