From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:39202) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dRx5d-0006FL-QY for qemu-devel@nongnu.org; Mon, 03 Jul 2017 04:52:18 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dRx5d-0005wu-2S for qemu-devel@nongnu.org; Mon, 03 Jul 2017 04:52:17 -0400 Received: from mail-wm0-x232.google.com ([2a00:1450:400c:c09::232]:38349) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1dRx5c-0005u5-Rb for qemu-devel@nongnu.org; Mon, 03 Jul 2017 04:52:16 -0400 Received: by mail-wm0-x232.google.com with SMTP id f67so47078940wmh.1 for ; Mon, 03 Jul 2017 01:52:15 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: References: <1498728533-23160-1-git-send-email-frederic.konrad@adacore.com> <1498728533-23160-3-git-send-email-frederic.konrad@adacore.com> <7e7d94ce-a33c-6253-8dcf-475e93fe388f@adacore.com> From: Peter Maydell Date: Mon, 3 Jul 2017 09:51:53 +0100 Message-ID: Content-Type: text/plain; charset="UTF-8" Subject: Re: [Qemu-devel] [PATCH v1 2/3] arm: fix the armv7m reset state List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: KONRAD Frederic Cc: QEMU Developers , =?UTF-8?B?QWxleCBCZW5uw6ll?= , Paolo Bonzini , alistair@alistair23.me On 3 July 2017 at 08:31, KONRAD Frederic wrote: > On 06/30/2017 11:06 AM, Peter Maydell wrote: >> On 30 June 2017 at 09:24, KONRAD Frederic >> wrote: >>> If I choose (b) I won't be able to load it to SRAM and it is >>> basically the same result I'll need to move or modify the config. >> >> >> I don't understand this, though. Option (b) is probably painful >> to implement (I don't have a good idea of how to do it) but >> it ought to mean that the ELF files that work on the board >> also work for QEMU (regardless of how the board model >> implemented the aliased flash). >> > > Yes that's exactly what I want. > > Basically the 0x00000000 alias can point to the SRAM or the ROM > during the reset depending on some boot config. The ELF is > directly loaded in the ROM or in the SRAM and my patch allows to > fetch the two first words in the reset handler to make it work > for any boot config. Yes, but it only works if you implemented it that way round, and not for board implementations which put the real device at 0 and the alias at high memory. I'd like a fix which deals with all of this, not just with the particular arrangement your board implementation has. thanks -- PMM