linux-kernel - Re: [PATCH 1/5 V5] Add a function(ioremap_encrypted) for kdump when AMD sme 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 for Android: free password hash cracker in your pocket

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

Date:   Tue, 3 Jul 2018 19:25:21 +0800
From:   lijiang <lijiang@...hat.com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     linux-kernel@...r.kernel.org, mingo@...hat.com, tglx@...utronix.de,
        hpa@...or.com, ebiederm@...ssion.com, joro@...tes.org,
        thomas.lendacky@....com, dyoung@...hat.com,
        kexec@...ts.infradead.org, iommu@...ts.linux-foundation.org,
        bhe@...hat.com
Subject: Re: [PATCH 1/5 V5] Add a function(ioremap_encrypted) for kdump when
 AMD sme enabled

在 2018年07月03日 17:39, Borislav Petkov 写道:
> On Tue, Jul 03, 2018 at 10:17:19AM +0800, lijiang wrote:
>> for example, the elfcorehdr. In fact, the elfcorehdr and notes
> 
> You mean this?
> 
>  ssize_t __weak elfcorehdr_read_notes(char *buf, size_t count, u64 *ppos)
>  {
> -       return read_from_oldmem(buf, count, ppos, 0);
> +       return read_from_oldmem(buf, count, ppos, 0, sme_active());
> 
> That looks encrypted to me.
> 
The elf notes is an old memory, it is encrypted.

But the elf header is a crash kernel reserved memory, which is unencrypted.
 ssize_t __weak elfcorehdr_read(char *buf, size_t count, u64 *ppos)
 {
-	return read_from_oldmem(buf, count, ppos, 0);
+	return read_from_oldmem(buf, count, ppos, 0, false);
 }

They call the same function(read_from_oldmem->ioremap_cache), so it is hard to
properly remap the memory if we don't use the parameter to distinguish. 

Regards,
Lianbo
>> call the same function(read_from_oldmem->ioremap_cache), in this case,
>> it is very difficult to properly remap the memory if the caller don't
>> care whether the memory is encrypted.
> 
> So beef up the logic in __ioremap_caller() to figure out based on the
> address whether to access the memory encrypted or not. You can find out
> the elfcorehdr address in the capture kernel.
>