From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:37875) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VGTUs-0001iu-2s for qemu-devel@nongnu.org; Mon, 02 Sep 2013 08:44:50 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VGTUn-0006CA-P2 for qemu-devel@nongnu.org; Mon, 02 Sep 2013 08:44:46 -0400 Received: from mail-lb0-f181.google.com ([209.85.217.181]:56704) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VGTUn-0006C3-HW for qemu-devel@nongnu.org; Mon, 02 Sep 2013 08:44:41 -0400 Received: by mail-lb0-f181.google.com with SMTP id u14so3839840lbd.26 for ; Mon, 02 Sep 2013 05:44:40 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <1372700120-8896-14-git-send-email-peter.maydell@linaro.org> References: <1372700120-8896-1-git-send-email-peter.maydell@linaro.org> <1372700120-8896-14-git-send-email-peter.maydell@linaro.org> From: Peter Maydell Date: Mon, 2 Sep 2013 13:44:20 +0100 Message-ID: Content-Type: text/plain; charset=UTF-8 Subject: Re: [Qemu-devel] [PATCH v5 13/21] linux-user: Add signal handling for AArch64 List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: QEMU Developers Cc: Andreas Schwab , Alexander Graf , "Mian M. Hamayun" , Patch Tracking On 1 July 2013 18:35, Peter Maydell wrote: > From: Andreas Schwab > > This patch adds signal handling for AArch64. The code is based on the > respective source in the Linux kernel. Couple of minor things I noticed, and will fix for an upcoming v6: > +typedef struct target_sigaltstack { > + abi_ulong ss_sp; > + abi_long ss_flags; The kernel defines this field as 'int', not 'long', so so should we (would only cause a problem on a big endian host I think, given the struct layout means ss_size ends up at the same offset either way.) > + abi_ulong ss_size; > +} target_stack_t; > + for (i = 0; i < 32 * 2; i++) { > + __put_user(env->vfp.regs[i], &aux->fpsimd.vregs[i]); > + } This is wrong for the (currently hypothetical) case of bigendian (the other __put_user() are fine because that macro does an endianness swap if needed, but this is actually an array of 32 int128_t. -- PMM