From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33896) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1emIyt-0008OB-T1 for qemu-devel@nongnu.org; Thu, 15 Feb 2018 07:49:45 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1emIyp-0003K9-TB for qemu-devel@nongnu.org; Thu, 15 Feb 2018 07:49:43 -0500 Sender: =?UTF-8?Q?Philippe_Mathieu=2DDaud=C3=A9?= References: <20180208055039.24666-1-penberg@iki.fi> <20180208055039.24666-4-penberg@iki.fi> From: =?UTF-8?Q?Philippe_Mathieu-Daud=c3=a9?= Message-ID: <9c52a265-6659-5c9c-27c7-eced8528fb63@amsat.org> Date: Thu, 15 Feb 2018 09:49:36 -0300 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [Qemu-arm] [PATCH v1 3/3] raspi: Add "raspi3" machine type List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell , Pekka Enberg Cc: qemu-arm , QEMU Developers On 02/15/2018 09:39 AM, Peter Maydell wrote: > On 8 February 2018 at 05:50, Pekka Enberg wrote: >> This patch adds a "raspi3" machine type, which can now be selected as >> the machine to run on by users via the "-M" command line option to QEMU. >> >> The machine type does *not* ignore memory transaction failures so we >> likely need to add some dummy devices later when people run something >> more complicated than what I'm using for testing. >> >> Signed-off-by: Pekka Enberg >> --- >> hw/arm/raspi.c | 21 +++++++++++++++++++++ >> 1 file changed, 21 insertions(+) >> >> diff --git a/hw/arm/raspi.c b/hw/arm/raspi.c >> index 66fe10e376..048ff23a51 100644 >> --- a/hw/arm/raspi.c >> +++ b/hw/arm/raspi.c >> @@ -187,3 +187,24 @@ static void raspi2_machine_init(MachineClass *mc) >> mc->ignore_memory_transaction_failures = true; >> }; >> DEFINE_MACHINE("raspi2", raspi2_machine_init) >> + >> +static void raspi3_init(MachineState *machine) >> +{ >> + raspi_init(machine, 3); >> +} >> + >> +static void raspi3_machine_init(MachineClass *mc) >> +{ >> + mc->desc = "Raspberry Pi 3"; >> + mc->init = raspi3_init; >> + mc->block_default_type = IF_SD; >> + mc->no_parallel = 1; >> + mc->no_floppy = 1; >> + mc->no_cdrom = 1; >> + mc->default_cpu_type = ARM_CPU_TYPE_NAME("cortex-a53"); >> + mc->max_cpus = BCM2836_NCPUS; >> + mc->min_cpus = BCM2836_NCPUS; >> + mc->default_cpus = BCM2836_NCPUS; >> + mc->default_ram_size = 1024 * 1024 * 1024; >> +} >> +DEFINE_MACHINE("raspi3", raspi3_machine_init) > > Hi. This patch breaks "make check", because it adds the raspi3 > to the arm-softmmu (32-bit guest CPUs only) build, where the > cortex-a53 CPU doesn't exist: > > e104462:xenial:qemu$ ./build/x86/arm-softmmu/qemu-system-arm -M raspi3 > ** > ERROR:/home/petmay01/linaro/qemu-from-laptop/qemu/qom/object.c:372:object_initialize_with_type: > assertion failed: (type != NULL) > Aborted (core dumped) > > The usual way we avoid this is that 64-bit only boards are > in their own source file, which is only compiled if the right > CONFIG_FOO is set by default-configs/aarch64-softmmu.mak. > In this case splitting the 64-bit board into its own source > file would be weird and awkward, so the simple thing is to > guard the raspi3 bits with #ifdef TARGET_AARCH64. Reviewed-by: Philippe Mathieu-Daudé > > (You might think we could define a CONFIG_RASPI3 in > aarch64-softmmu.mak and #ifdef on it, but for some reason > we don't expose those CONFIG_* to C code, possibly just because > we've never needed to in the past...) > > Since this was the only code change needed, I'm just going to make > it and apply the patchset to target-arm.next, rather than ask > you to do a respin. (There was also a stray space-at-end-of-line > in patch 2 which checkpatch grumbles about; I'll fix that up too.) > > thanks > -- PMM >