linux-kernel - Re: [PATCH v9 15/29] x86/insn-eval: Add utility functions to get segment descriptor base address and limit

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:   Wed, 11 Oct 2017 18:24:03 -0700
From:   Ricardo Neri <ricardo.neri-calderon@...ux.intel.com>
To:     Borislav Petkov <bp@...e.de>
Cc:     Ingo Molnar <mingo@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        "H. Peter Anvin" <hpa@...or.com>,
        Andy Lutomirski <luto@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Brian Gerst <brgerst@...il.com>,
        Chris Metcalf <cmetcalf@...lanox.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Masami Hiramatsu <mhiramat@...nel.org>,
        Huang Rui <ray.huang@....com>, Jiri Slaby <jslaby@...e.cz>,
        Jonathan Corbet <corbet@....net>,
        "Michael S. Tsirkin" <mst@...hat.com>,
        Paul Gortmaker <paul.gortmaker@...driver.com>,
        Vlastimil Babka <vbabka@...e.cz>,
        Chen Yucong <slaoub@...il.com>,
        "Ravi V. Shankar" <ravi.v.shankar@...el.com>,
        Shuah Khan <shuah@...nel.org>, linux-kernel@...r.kernel.org,
        x86@...nel.org, Adam Buchbinder <adam.buchbinder@...il.com>,
        Colin Ian King <colin.king@...onical.com>,
        Lorenzo Stoakes <lstoakes@...il.com>,
        Qiaowei Ren <qiaowei.ren@...el.com>,
        Arnaldo Carvalho de Melo <acme@...hat.com>,
        Adrian Hunter <adrian.hunter@...el.com>,
        Kees Cook <keescook@...omium.org>,
        Thomas Garnier <thgarnie@...gle.com>,
        Dmitry Vyukov <dvyukov@...gle.com>
Subject: Re: [PATCH v9 15/29] x86/insn-eval: Add utility functions to get
 segment descriptor base address and limit

On Wed, 2017-10-11 at 22:16 +0200, Borislav Petkov wrote:
> On Wed, Oct 11, 2017 at 12:57:01PM -0700, Ricardo Neri wrote:
> > 
> > This is meant to be an error case. In long mode,
> > only INAT_SEG_REG_IGNORE/FS/GS
> > are valid. All other indices are invalid.
> > 
> > Perhaps we could return -EINVAL instead?
> So, my question is, when are you ever going to have that case? What
> constellation of events would ever hit this else branch for long mode?
> Because it looks impossible to me. What I can imagine only is something
> like this:
> 
>                 else if (seg_reg != INAT_SEG_REG_IGNORE)
> 			WARN_ONCE(1, "This should never happen!\n");
> 
> assertion.

To clarify, I think you mean seg_reg_idx.

Yes, it would be impossible to hit this else branch provided that callers don't
attempt to use an invalid seg_reg_idx while in long mode. Probably this is not
critical as this is a static function and as such we control who can call it and
make sure seg_reg_idx is always valid (i.e., INAT_SEG_REG_IGNORE/FS/GS in long
mode).



> But you don't really need that - you can simply ignore seg_reg in that
> case:
> 
>         if (user_64bit_mode(regs)) {
>                 /*
>                  * Only FS or GS will have a base address, the rest of
>                  * the segments' bases are forced to 0.
>                  */
>                 unsigned long base;
> 
>                 if (seg_reg == INAT_SEG_REG_FS)
>                         rdmsrl(MSR_FS_BASE, base);
>                 else if (seg_reg == INAT_SEG_REG_GS)
>                         /*
>                          * swapgs was called at the kernel entry point. Thus,
>                          * MSR_KERNEL_GS_BASE will have the user-space GS
> base.
>                          */
>                         rdmsrl(MSR_KERNEL_GS_BASE, base);
>                 else
>                         base = 0;
> 
>                 return base;
>         }
> 
> Or am I missing something?

My intention is to let the caller know about the invalid seg_reg_idx instead of
silently correcting the caller's input by ignoring seg_reg_idx.

On the other hand, in long mode, hardware ignore all segment registers except FS
and GS.

Hence, I guess I can remove the check in question.

Thanks and BR,
Ricardo