qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: seeteena <s1seetee@linux.vnet.ibm.com>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: agraf@suse.de, qemu-ppc@nongnu.org, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH v1] spapr.c: Update qemu's maxcpus for pseries machine.
Date: Mon, 4 Dec 2017 12:03:03 +0530	[thread overview]
Message-ID: <1854d711-d19d-343e-91d0-832f44d9c905@linux.vnet.ibm.com> (raw)
In-Reply-To: <20171201123713.GA2130@umbus.fritz.box>

Thanks David for the explanation.

I thought the changes been into upstream code for pseries to limit

maxcpus to 240 instead of 1024.



On 12/01/2017 06:07 PM, David Gibson wrote:
> On Fri, Dec 01, 2017 at 04:54:09PM +0530, Seeteena Thoufeek wrote:
>> Need to adjust the max cpus supported number from error message since
>> it was conflicting with KVM's.
>>
>> Steps to Reproduce:
>> 1.boot up with
>> "-smp 64,maxcpus=102464,cores=8,threads=1,sockets=8"
>>
>> qemu-kvm: Number of SMP CPUs requested (102464) exceeds max CPUs
>> supported by machine 'pseries-rhel7.4.0alt' (1024)
>>
>> 2. On KVM machine it shows
>>
>> boot up with
>> "-m 6G,maxmem=300G,slots=256 -smp 64,maxcpus=1024,cores=8,threads=1
>> ,sockets=128"
>>
>> Number of hotpluggable cpus requested (1024) exceeds the maximum cpus
>>   supported by KVM (240)
>>
>> It seemed that 1024 was useless since KVM only support 240 so far.
>> Hence,we need to adjust it to an reasonable value 240.
>>
>> Signed-off-by: Seeteena Thoufeek <s1seetee@linux.vnet.ibm.com>
> This has been written without adequate thought and investigation.
>
> First, in upstream code there's nothing wrong with having different
> limits in qemu and kernel - the can both vary depending on various
> factors.  Secondly, the 240 limit doesn't come from the kernel
> upstream - only in RHEL.
>
> Third, it doesn't even come from the kernel in RHEL - the downstream
> qemu lies about the kernel limit.
>
> I'm in the process of fixing this correctly, I expect to have
> something on Monday.
>
>> ---
>>   hw/ppc/spapr.c | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c
>> index 9efddea..c753254 100644
>> --- a/hw/ppc/spapr.c
>> +++ b/hw/ppc/spapr.c
>> @@ -3691,6 +3691,7 @@ static const TypeInfo spapr_machine_info = {
>>           if (latest) {                                                \
>>               mc->alias = "pseries";                                   \
>>               mc->is_default = 1;                                      \
>> +            mc->max_cpus = 240;                                      \
>>           }                                                            \
>>       }                                                                \
>>       static void spapr_machine_##suffix##_instance_init(Object *obj)  \

  reply	other threads:[~2017-12-04  6:33 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-01 11:24 [Qemu-devel] [PATCH v1] spapr.c: Update qemu's maxcpus for pseries machine Seeteena Thoufeek
2017-12-01 12:37 ` David Gibson
2017-12-04  6:33   ` seeteena [this message]
2017-12-01 17:20 ` [Qemu-devel] [Qemu-ppc] " Greg Kurz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1854d711-d19d-343e-91d0-832f44d9c905@linux.vnet.ibm.com \
    --to=s1seetee@linux.vnet.ibm.com \
    --cc=agraf@suse.de \
    --cc=david@gibson.dropbear.id.au \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).