linux-kernel - Re: [PATCH v4 11/39] x86/mm: Update pte_modify for _PAGE

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, 27 Dec 2022 23:31:37 +0000
From:   "Edgecombe, Rick P" <rick.p.edgecombe@...el.com>
To:     "bp@...en8.de" <bp@...en8.de>
CC:     "akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
        "tglx@...utronix.de" <tglx@...utronix.de>,
        "linux-arch@...r.kernel.org" <linux-arch@...r.kernel.org>,
        "kcc@...gle.com" <kcc@...gle.com>,
        "Lutomirski, Andy" <luto@...nel.org>,
        "nadav.amit@...il.com" <nadav.amit@...il.com>,
        "kirill.shutemov@...ux.intel.com" <kirill.shutemov@...ux.intel.com>,
        "Schimpe, Christina" <christina.schimpe@...el.com>,
        "peterz@...radead.org" <peterz@...radead.org>,
        "corbet@....net" <corbet@....net>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "jannh@...gle.com" <jannh@...gle.com>,
        "dethoma@...rosoft.com" <dethoma@...rosoft.com>,
        "x86@...nel.org" <x86@...nel.org>, "pavel@....cz" <pavel@....cz>,
        "rdunlap@...radead.org" <rdunlap@...radead.org>,
        "linux-api@...r.kernel.org" <linux-api@...r.kernel.org>,
        "john.allen@....com" <john.allen@....com>,
        "arnd@...db.de" <arnd@...db.de>,
        "jamorris@...ux.microsoft.com" <jamorris@...ux.microsoft.com>,
        "rppt@...nel.org" <rppt@...nel.org>,
        "bsingharora@...il.com" <bsingharora@...il.com>,
        "mike.kravetz@...cle.com" <mike.kravetz@...cle.com>,
        "oleg@...hat.com" <oleg@...hat.com>,
        "fweimer@...hat.com" <fweimer@...hat.com>,
        "keescook@...omium.org" <keescook@...omium.org>,
        "gorcunov@...il.com" <gorcunov@...il.com>,
        "Yu, Yu-cheng" <yu-cheng.yu@...el.com>,
        "andrew.cooper3@...rix.com" <andrew.cooper3@...rix.com>,
        "hpa@...or.com" <hpa@...or.com>,
        "mingo@...hat.com" <mingo@...hat.com>,
        "hjl.tools@...il.com" <hjl.tools@...il.com>,
        "linux-mm@...ck.org" <linux-mm@...ck.org>,
        "Syromiatnikov, Eugene" <esyr@...hat.com>,
        "Yang, Weijiang" <weijiang.yang@...el.com>,
        "linux-doc@...r.kernel.org" <linux-doc@...r.kernel.org>,
        "dave.hansen@...ux.intel.com" <dave.hansen@...ux.intel.com>,
        "Eranian, Stephane" <eranian@...gle.com>
Subject: Re: [PATCH v4 11/39] x86/mm: Update pte_modify for _PAGE_COW

On Tue, 2022-12-27 at 12:42 +0100, Borislav Petkov wrote:
> On Fri, Dec 02, 2022 at 04:35:38PM -0800, Rick Edgecombe wrote:
> >  static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
> >  {
> > +       pteval_t _page_chg_mask_no_dirty = _PAGE_CHG_MASK &
> > ~_PAGE_DIRTY;
> >         pteval_t val = pte_val(pte), oldval = val;
> > +       pte_t pte_result;
> >  
> >         /*
> >          * Chop off the NX bit (if present), and add the NX portion
> > of
> >          * the newprot (if present):
> >          */
> > -       val &= _PAGE_CHG_MASK;
> > -       val |= check_pgprot(newprot) & ~_PAGE_CHG_MASK;
> > +       val &= _page_chg_mask_no_dirty;
> > +       val |= check_pgprot(newprot) & ~_page_chg_mask_no_dirty;
> >         val = flip_protnone_guard(oldval, val, PTE_PFN_MASK);
> > -       return __pte(val);
> > +
> > +       pte_result = __pte(val);
> > +
> > +       /*
> > +        * Dirty bit is not preserved above so it can be done
> 
> Just for my own understanding: are you saying here that
> flip_protnone_guard() might end up setting _PAGE_DIRTY in val...
> 
> > +        * in a special way for the shadow stack case, where it
> > +        * needs to set _PAGE_COW. pte_mkcow() will do this in
> > +        * the case of shadow stack.
> > +        */
> > +       if (pte_dirty(pte_result))
> > +               pte_result = pte_mkcow(pte_result);
> 
> ... and in that case we need to turn it into a _PAGE_COW setting?
> 
The comment is referring to the dirty bits possibly coming from
newprot, but looking at it now I think the code was broken trying to
fix the recent soft dirty test breakage. Now it might lose pre-existing
dirty bits in the pte unessarily... I think. I'm temporarily without
access to my test equipment so will have to get back to you on this.
Thanks for flagging that something looks off.