Messages in this thread | | | Subject | Re: [PATCH v15 08/26] x86/mm: Introduce _PAGE_COW | From | "Yu, Yu-cheng" <> | Date | Tue, 8 Dec 2020 10:25:15 -0800 |
| |
On 12/8/2020 9:50 AM, Borislav Petkov wrote: > On Tue, Nov 10, 2020 at 08:21:53AM -0800, Yu-cheng Yu wrote: >> There is essentially no room left in the x86 hardware PTEs on some OSes >> (not Linux). That left the hardware architects looking for a way to >> represent a new memory type (shadow stack) within the existing bits. >> They chose to repurpose a lightly-used state: Write=0,Dirty=1. > > It is not clear to me what the definition and semantics of that bit is. > > +#define _PAGE_BIT_COW _PAGE_BIT_SOFTW5 /* copy-on-write */ > > Is it set by hw or by sw and hw uses it to know it is a shadow stack > page, and so on. > > I think you should lead with its definition.
Ok.
...
>> Write=0,Dirty=1 PTEs. In places where we do create them, we need to find >> an alternative way to represent them _without_ using the same hardware bit >> combination. Thus, enter _PAGE_COW. This results in the following: >> >> (a) A modified, copy-on-write (COW) page: (R/O + _PAGE_COW) >> (b) A R/O page that has been COW'ed: (R/O + _PAGE_COW) > > Both are "R/O + _PAGE_COW". Where's the difference? The dirty bit?
The PTEs are the same for both (a) and (b), but come from different routes.
>> The user page is in a R/O VMA, and get_user_pages() needs a writable >> copy. The page fault handler creates a copy of the page and sets >> the new copy's PTE as R/O and _PAGE_COW. >> (c) A shadow stack PTE: (R/O + _PAGE_DIRTY_HW) > > So W=0, D=1 ?
Yes.
>> (d) A shared shadow stack PTE: (R/O + _PAGE_COW) >> When a shadow stack page is being shared among processes (this happens >> at fork()), its PTE is cleared of _PAGE_DIRTY_HW, so the next shadow >> stack access causes a fault, and the page is duplicated and >> _PAGE_DIRTY_HW is set again. This is the COW equivalent for shadow >> stack pages, even though it's copy-on-access rather than copy-on-write. >> (e) A page where the processor observed a Write=1 PTE, started a write, set >> Dirty=1, but then observed a Write=0 PTE. > > How does that happen? Something changed the PTE's W bit to 0 in-between?
Yes.
...
>> diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h >> index b23697658b28..c88c7ccf0318 100644 >> --- a/arch/x86/include/asm/pgtable.h >> +++ b/arch/x86/include/asm/pgtable.h >> @@ -121,9 +121,9 @@ extern pmdval_t early_pmd_flags; >> * The following only work if pte_present() is true. >> * Undefined behaviour if not.. >> */ >> -static inline int pte_dirty(pte_t pte) >> +static inline bool pte_dirty(pte_t pte) >> { >> - return pte_flags(pte) & _PAGE_DIRTY_HW; >> + return pte_flags(pte) & _PAGE_DIRTY_BITS; > > Why? > > Does _PAGE_COW mean dirty too?
Yes. Basically [read-only & dirty] is created by software. Now the software uses a different bit.
>> @@ -343,6 +349,17 @@ static inline pte_t pte_mkold(pte_t pte) >> >> static inline pte_t pte_wrprotect(pte_t pte) >> { >> + /* >> + * Blindly clearing _PAGE_RW might accidentally create >> + * a shadow stack PTE (RW=0,Dirty=1). Move the hardware >> + * dirty value to the software bit. >> + */ >> + if (cpu_feature_enabled(X86_FEATURE_SHSTK)) { >> + pte.pte |= (pte.pte & _PAGE_DIRTY_HW) >> >> + _PAGE_BIT_DIRTY_HW << _PAGE_BIT_COW; > > Let that line stick out. And that shifting is not grokkable at a quick > glance, at least not to me. Simplify?
Ok.
>> static inline pmd_t pmd_wrprotect(pmd_t pmd) >> { >> + /* >> + * Blindly clearing _PAGE_RW might accidentally create >> + * a shadow stack PMD (RW=0,Dirty=1). Move the hardware >> + * dirty value to the software bit. > > This whole carefully sidestepping the possiblity of creating a shadow > stack pXd is kinda sucky... > >> diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h >> index 7462a574fc93..5f764d8d9bae 100644 >> --- a/arch/x86/include/asm/pgtable_types.h >> +++ b/arch/x86/include/asm/pgtable_types.h >> @@ -23,7 +23,8 @@ >> #define _PAGE_BIT_SOFTW2 10 /* " */ >> #define _PAGE_BIT_SOFTW3 11 /* " */ >> #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ >> -#define _PAGE_BIT_SOFTW4 58 /* available for programmer */ >> +#define _PAGE_BIT_SOFTW4 57 /* available for programmer */ >> +#define _PAGE_BIT_SOFTW5 58 /* available for programmer */ >> #define _PAGE_BIT_PKEY_BIT0 59 /* Protection Keys, bit 1/4 */ >> #define _PAGE_BIT_PKEY_BIT1 60 /* Protection Keys, bit 2/4 */ >> #define _PAGE_BIT_PKEY_BIT2 61 /* Protection Keys, bit 3/4 */ >> @@ -36,6 +37,16 @@ >> #define _PAGE_BIT_SOFT_DIRTY _PAGE_BIT_SOFTW3 /* software dirty tracking */ >> #define _PAGE_BIT_DEVMAP _PAGE_BIT_SOFTW4 >> >> +/* >> + * This bit indicates a copy-on-write page, and is different from >> + * _PAGE_BIT_SOFT_DIRTY, which tracks which pages a task writes to. >> + */ >> +#ifdef CONFIG_X86_64 > > CONFIG_X86_64 ? Do all x86 machines out there support CET? > > If anything, CONFIG_X86_CET...
Ok.
-- Yu-cheng
| |