On Mon, Jul 19, 2021 at 11:20:33AM +0800, Yanan Wang wrote:
> In the SMP configuration, we should either specify a topology
> parameter with a reasonable value (equal to or greater than 1)
> or just leave it omitted and QEMU will calculate its value.
>
> Configurations which explicitly specify the topology parameters
> as zero like "sockets=0" are meaningless, so disallow them.
>
> Suggested-by: Andrew Jones <drjo...@redhat.com>
> Signed-off-by: Yanan Wang <wangyana...@huawei.com>
> ---
> hw/core/machine.c | 31 +++++++++++++++++++++++--------
> hw/i386/pc.c | 29 +++++++++++++++++++++--------
> qapi/machine.json | 4 ++--
> 3 files changed, 46 insertions(+), 18 deletions(-)
>
> diff --git a/hw/core/machine.c b/hw/core/machine.c
> index 775add0795..d73daa10f4 100644
> --- a/hw/core/machine.c
> +++ b/hw/core/machine.c
> @@ -745,11 +745,25 @@ static void smp_parse(MachineState *ms,
> SMPConfiguration *config, Error **errp)
> {
> unsigned cpus = config->has_cpus ? config->cpus : 0;
> unsigned sockets = config->has_sockets ? config->sockets : 0;
> + unsigned dies = config->has_dies ? config->dies : 1;
It looks odd to set dies=1 by default at initially, when everything
else is set to 0. I realize you're just copying existing pc_smp_parse
code in this respect, but I feel like could benefit from a separate
initialization with a comment to explain why we're hardcoding it
to 1....
> unsigned cores = config->has_cores ? config->cores : 0;
> unsigned threads = config->has_threads ? config->threads : 0;
> + unsigned maxcpus = config->has_maxcpus ? config->maxcpus : 0;
> +
> + if ((config->has_cpus && config->cpus == 0) ||
> + (config->has_sockets && config->sockets == 0) ||
> + (config->has_dies && config->dies == 0) ||
> + (config->has_cores && config->cores == 0) ||
> + (config->has_threads && config->threads == 0) ||
> + (config->has_maxcpus && config->maxcpus == 0)) {
> + error_setg(errp, "parameters must be equal to or greater than one"
> + "if provided");
> + return;
> + }
>
> - if (config->has_dies && config->dies != 0 && config->dies != 1) {
> + if (dies > 1) {
> error_setg(errp, "dies not supported by this machine's CPU
> topology");
> + return;
> }
.... eg how about here adding
/* Never try to assign multiple dies when defaulting omitted topology */
if (dies == 0) {
dies = 1;
}
> diff --git a/qapi/machine.json b/qapi/machine.json
> index c3210ee1fb..c11b2e6f73 100644
> --- a/qapi/machine.json
> +++ b/qapi/machine.json
> @@ -1288,8 +1288,8 @@
> ##
> # @SMPConfiguration:
> #
> -# Schema for CPU topology configuration. "0" or a missing value lets
> -# QEMU figure out a suitable value based on the ones that are provided.
> +# Schema for CPU topology configuration. A missing value lets QEMU
> +# figure out a suitable value based on the ones that are provided.
Hmm, so we had actually documented that '0' had the same semantics
as omitting a parameter. This was done in:
commit 1e63fe685804dfadddd643bf3860b1a59702d4bf
Author: Paolo Bonzini <pbonz...@redhat.com>
Date: Thu Jun 17 17:53:06 2021 +0200
machine: pass QAPI struct to mc->smp_parse
which hasn't been released yet.
This was possible, but never documented, with the traditiaonl -smp
impl before it was qapi-ified. I think that historical behaviour
was simply a side effect of the QemuOpts impl rather than an
intentional design, hence not documented.
At the very least I think need to get rid of this bit of docs about
"0" before this release, otherwise we'll have stronger need to
consider a real deprecation process.
Regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
