[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <3f98f2be-47e1-ecf1-ff3c-0b35ae93437a@amazon.de>
Date: Thu, 7 Jun 2018 17:48:23 +0200
From: Norbert Manthey <nmanthey@...zon.de>
To: Paolo Bonzini <pbonzini@...hat.com>
CC: Radim Krčmář <rkrcmar@...hat.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>, <x86@...nel.org>,
<kvm@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [less-CONFIG_NET 7/7] kvm: drop CONFIG_NET
Dear Paolo,
after you asked, I run a few more "make randconfig; make" rounds, and
found a problem with one of the commits (when using BPF_SYSCALL, NET and
SECCOMP_FILTER). I'll fix the issues and all other that might pop up,
and rerun the randconfig loop many more times to get a better coverage
and higher confidence. I'll update the series afterwards.
Best,
Norbert
On 06/07/2018 01:46 PM, Norbert Manthey wrote:
> On 06/07/2018 01:20 PM, Paolo Bonzini wrote:
>> On 06/06/2018 15:53, Norbert Manthey wrote:
>>> In some scenarios, we do not want to use CONFIG_NET even though we want
>>> to use CONFIG_KVM. To get this working, and to not pull in more
>>> dependencies, we need to disable CONFIG_TASKSTATS as well.
>>>
>>> Signed-off-by: Norbert Manthey <nmanthey@...zon.de>
>>> ---
>>> arch/x86/kvm/Kconfig | 5 ++---
>>> 1 file changed, 2 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
>>> index 92fd433..f209817 100644
>>> --- a/arch/x86/kvm/Kconfig
>>> +++ b/arch/x86/kvm/Kconfig
>>> @@ -22,8 +22,8 @@ config KVM
>>> tristate "Kernel-based Virtual Machine (KVM) support"
>>> depends on HAVE_KVM
>>> depends on HIGH_RES_TIMERS
>>> - # for TASKSTATS/TASK_DELAY_ACCT:
>>> - depends on NET && MULTIUSER
>>> + # TASK_DELAY_ACCT:
>>> + depends on MULTIUSER
>>> depends on X86_LOCAL_APIC
>>> select PREEMPT_NOTIFIERS
>>> select MMU_NOTIFIER
>>> @@ -37,7 +37,6 @@ config KVM
>>> select KVM_ASYNC_PF
>>> select USER_RETURN_NOTIFIER
>>> select KVM_MMIO
>>> - select TASKSTATS
>>> select TASK_DELAY_ACCT
>>> select PERF_EVENTS
>>> select HAVE_KVM_MSI
>>>
>> Assuming it compiles, it looks good. :)
>>
>> Paolo
>>
> It compiles, with the default configuration, as well as with the subset
> that I currently use. Since I have an actually small selection of config
> options, I wonder whether there is some best practices to make sure this
> change does not corrupt other configurations that rely on KVM selecting
> TASKSTATS.
>
> Norbert
Amazon Development Center Germany GmbH
Berlin - Dresden - Aachen
main office: Krausenstr. 38, 10117 Berlin
Geschaeftsfuehrer: Dr. Ralf Herbrich, Christian Schlaeger
Ust-ID: DE289237879
Eingetragen am Amtsgericht Charlottenburg HRB 149173 B
Powered by blists - more mailing lists