linux-kernel - Re: [PATCH v5 12/22] x86/virt/tdx: Convert all memory regions in memblock to TDX memory

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]

Message-ID: <b4c36488909b474af51b9b7c9d7857d9d6b43fb1.camel@intel.com>
Date:   Thu, 04 Aug 2022 10:14:54 +1200
From:   Kai Huang <kai.huang@...el.com>
To:     Dave Hansen <dave.hansen@...el.com>, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org
Cc:     seanjc@...gle.com, pbonzini@...hat.com, len.brown@...el.com,
        tony.luck@...el.com, rafael.j.wysocki@...el.com,
        reinette.chatre@...el.com, dan.j.williams@...el.com,
        peterz@...radead.org, ak@...ux.intel.com,
        kirill.shutemov@...ux.intel.com,
        sathyanarayanan.kuppuswamy@...ux.intel.com,
        isaku.yamahata@...el.com
Subject: Re: [PATCH v5 12/22] x86/virt/tdx: Convert all memory regions in
 memblock to TDX memory

On Wed, 2022-08-03 at 07:22 -0700, Dave Hansen wrote:
> On 8/2/22 18:30, Kai Huang wrote:
> > On Fri, 2022-07-08 at 11:34 +1200, Kai Huang wrote:
> > > > Why not just entirely remove the lower 1MB from the memblock structure
> > > > on TDX systems?  Do something equivalent to adding this on the kernel
> > > > command line:
> > > > 
> > > >  	memmap=1M$0x0
> > > I will explore this option.  Thanks!
> > Hi Dave,
> > 
> > After investigating and testing, we cannot simply remove first 1MB from e820
> > table which is similar to what 'memmap=1M$0x0' does, as the kernel needs low
> > memory as trampoline to bring up all APs.
> 
> OK, so don't remove it, but reserve it so that the trampoline code can
> use it.

It's already reserved in the existing reserve_real_mode().  What we need is to
*remove* the first 1MB from memblock.memory, so that the
for_each_mem_pfn_range() will just not get any memory below 1MB.  Otherwise we
need to explicitly skip the first 1MB in TDX code like what I did  in this
series.

-- 
Thanks,
-Kai