Messages in this thread | | | Subject | Re: [RFC PATCH] x86/apic: Fix BUG due to multiple allocation of legacy vectors. | From | imran.f.khan@oracle ... | Date | Thu, 27 May 2021 15:13:26 +1000 |
| |
Hi Thomas,
On 26/5/21 12:51 am, Thomas Gleixner wrote: > Imran, > > On Mon, May 24 2021 at 13:29, imran f. khan wrote: [...] > >> Even my current setup that was crashing with in-house 5.4 and 5.8 >> kernels, boots fine if I boot it with noapic option removed from kernel >> boot parameters. > > May I ask the obvious question why the changelog of your patch did not > mention that this happens only with 'noapic' on the kernel command line? > I was not aware of it earlier. After your first feedback I debugged the code further with ML tag and found out about the noapic dependency. [...] > > Correct. So you almost decoded the underlying problem. > [...] > So the right thing to do is to ensure that the legacy interrupts are > marked as system vectors in case that the IO/APIC is disabled via > config, kernel command line or lack of enumeration. See below. > Thanks for clarifying it. I have tested your change in my local setup as well and it works well.
Thanks, Imran
| |