linux-kernel - Re: [PATCH -v8] x86: Disable x2apic if nox2apic is specified or intr-remap can not be enabled

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 20 Dec 2011 18:24:21 -0800
From:	Yinghai Lu <yinghai@...nel.org>
To:	Suresh Siddha <suresh.b.siddha@...el.com>
Cc:	Ingo Molnar <mingo@...e.hu>, Thomas Gleixner <tglx@...utronix.de>,
	"H. Peter Anvin" <hpa@...or.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH -v8] x86: Disable x2apic if nox2apic is specified or
 intr-remap can not be enabled

On Tue, Dec 20, 2011 at 5:49 PM, Yinghai Lu <yinghai@...nel.org> wrote:
> On Mon, Dec 19, 2011 at 10:51 PM, Suresh Siddha
> <suresh.b.siddha@...el.com> wrote:
>> While reviewing the code, I ended up doing some of this cleanup of your
>> patch. I have appended the modified version. Can you please incorporate
>> the smpboot checks before the AP bringup (apic-id should be less than
>> 255 in !x2apic_mode etc) and split the patch into couple of patches
>> (apic_probe etc) along with other needed cleanups.
>
> We could handle AP's apic id > 255 panic problem later if needed in
> another patch.
>
> Actually on system with apic id > 255, intr-remapping aka DMAR must be
> working, otherwise those big box can not be sold.
>
> And in this case, nox2apic must be passed to disable x2apic, because
> dmar is correct in both case kexec or first boot.
>
> when nox2apic is passed,  those entries will be skipped early. and BSP
> x2apic get disabled early.
> and code path is more like bios enable xapic only.
>
> Can we let this patch get into tip and have test cycle at first?
> at least we should not let it break any working set up.

put your updated version as v9.

like to let ingo or hpa or thomas to decide v8 or v9 to be used.

        Changes:
                 1. clear x2apic cpu features
                 2. disable x2apic later instead of check_x2apic()

Thanks

Yinghai

View attachment "nox2apic_v9.patch" of type "text/x-patch" (10358 bytes)