From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23153 invoked by alias); 10 Nov 2014 10:12:22 -0000 Mailing-List: contact libffi-discuss-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libffi-discuss-owner@sourceware.org Received: (qmail 23140 invoked by uid 89); 10 Nov 2014 10:12:21 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-2.3 required=5.0 tests=AWL,BAYES_00,RP_MATCHES_RCVD,SPF_PASS autolearn=ham version=3.3.2 X-HELO: cam-smtp0.cambridge.arm.com Received: from fw-tnat.cambridge.arm.com (HELO cam-smtp0.cambridge.arm.com) (217.140.96.21) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES256-SHA encrypted) ESMTPS; Mon, 10 Nov 2014 10:12:20 +0000 Received: from arm.com (e106375-lin.cambridge.arm.com [10.1.203.34]) by cam-smtp0.cambridge.arm.com (8.13.8/8.13.8) with ESMTP id sAAACEk6032440; Mon, 10 Nov 2014 10:12:15 GMT Date: Mon, 10 Nov 2014 10:12:00 -0000 From: James Greenhalgh To: Richard Henderson Cc: "libffi-discuss@sourceware.org" Subject: Re: [PATCH 00/16] Go closures for aarch64 Message-ID: <20141110101213.GA8221@arm.com> References: <1414522393-19169-1-git-send-email-rth@twiddle.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1414522393-19169-1-git-send-email-rth@twiddle.net> User-Agent: Mutt/1.5.21 (2010-09-15) X-IsSubscribed: yes X-SW-Source: 2014/txt/msg00208.txt.bz2 On Tue, Oct 28, 2014 at 06:52:57PM +0000, Richard Henderson wrote: > This patch set fixes a compilation error since the iOS merge, > tidies up the port significantly, and finally adds support for > complex and Go closures. Hi Richard, Possibly an irrelevant comment for this patch series, but while rewriting the world, did you consider Jakub's comments in this thread on gcc-patches regarding .note.GNU-stack notes? https://gcc.gnu.org/ml/gcc-patches/2014-09/msg00820.html "I've noticed that on 4.8 branch libgo recently (in the last few months) started being linked with GNU_STACK 0x000000 0x00000000 0x00000000 0x00000 0x00000 RWE 0x10 i.e. requiring executable stack on powerpc-linux (32-bit). The problem is that we link into libffi linux64.o and linux64_closure.o unconditionally, both for 32-bit and 64-bit compilations, just for 32-bit ones all the assembly is ifdefed out, so they have just empty sections. The .note.GNU-stack section isn't emitted in that case either, which means that the linker conservatively treats those as possibly needing executable stack. The following patch should fix that, ok for trunk/4.9/4.8? BTW, I wonder if e.g. libffi/src/arm/trampoline.S or libffi/src/aarch64/sysv.S shouldn't have those notes too (note, both of those were added after 2008 when most of the *.S files were marked that way)." If it doesn't belong in this series, I'll propose a patch adding it once your patches have gone in. Thanks, James > Richard Henderson (16): > aarch64: Fix non-apple compilation > aarch64: Improve is_hfa > aarch64: Always distinguish LONGDOUBLE > aarch64: Simplify AARCH64_STACK_ALIGN > aarch64: Reduce the size of register_context > aarch64: Use correct return registers > aarch64: Treat void return as not passed in registers > aarch64: Tidy up abi manipulation > aarch64: Merge prep_args with ffi_call > aarch64: Move return value handling into ffi_call_SYSV > aarch64: Move return value handling into ffi_closure_SYSV > aarch64: Unify scalar fp and hfa handling > aarch64: Remove aarch64_flags > aarch64: Add support for complex types > aarch64: Move x8 out of call_context > aarch64: Add support for Go closures > > src/aarch64/ffi.c | 1477 ++++++++++++++++------------------------ > src/aarch64/ffitarget.h | 14 +- > src/aarch64/internal.h | 67 ++ > src/aarch64/sysv.S | 589 +++++++++------- > testsuite/libffi.call/call.exp | 10 +- > 5 files changed, 1008 insertions(+), 1149 deletions(-) > create mode 100644 src/aarch64/internal.h > > -- > 1.9.3 > >