linux/arch/x86/xen/mmu.c

839 lines
20 KiB
C
Raw Normal View History

xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
/*
* Xen mmu operations
*
* This file contains the various mmu fetch and update operations.
* The most important job they must perform is the mapping between the
* domain's pfn and the overall machine mfns.
*
* Xen allows guests to directly update the pagetable, in a controlled
* fashion. In other words, the guest modifies the same pagetable
* that the CPU actually uses, which eliminates the overhead of having
* a separate shadow pagetable.
*
* In order to allow this, it falls on the guest domain to map its
* notion of a "physical" pfn - which is just a domain-local linear
* address - into a real "machine address" which the CPU's MMU can
* use.
*
* A pgd_t/pmd_t/pte_t will typically contain an mfn, and so can be
* inserted directly into the pagetable. When creating a new
* pte/pmd/pgd, it converts the passed pfn into an mfn. Conversely,
* when reading the content back with __(pgd|pmd|pte)_val, it converts
* the mfn back into a pfn.
*
* The other constraint is that all pages which make up a pagetable
* must be mapped read-only in the guest. This prevents uncontrolled
* guest updates to the pagetable. Xen strictly enforces this, and
* will disallow any pagetable update which will end up mapping a
* pagetable page RW, and will disallow using any writable page as a
* pagetable.
*
* Naively, when loading %cr3 with the base of a new pagetable, Xen
* would need to validate the whole pagetable before going on.
* Naturally, this is quite slow. The solution is to "pin" a
* pagetable, which enforces all the constraints on the pagetable even
* when it is not actively in use. This menas that Xen can be assured
* that it is still valid when you do load it into %cr3, and doesn't
* need to revalidate it.
*
* Jeremy Fitzhardinge <jeremy@xensource.com>, XenSource Inc, 2007
*/
#include <linux/sched.h>
#include <linux/highmem.h>
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
#include <linux/bug.h>
#include <asm/pgtable.h>
#include <asm/tlbflush.h>
#include <asm/mmu_context.h>
#include <asm/paravirt.h>
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
#include <asm/xen/hypercall.h>
#include <asm/xen/hypervisor.h>
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
#include <xen/page.h>
#include <xen/interface/xen.h>
#include "multicalls.h"
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
#include "mmu.h"
#define P2M_ENTRIES_PER_PAGE (PAGE_SIZE / sizeof(unsigned long))
#define TOP_ENTRIES (MAX_DOMAIN_PAGES / P2M_ENTRIES_PER_PAGE)
/* Placeholder for holes in the address space */
static unsigned long p2m_missing[P2M_ENTRIES_PER_PAGE]
__attribute__((section(".data.page_aligned"))) =
{ [ 0 ... P2M_ENTRIES_PER_PAGE-1 ] = ~0UL };
/* Array of pointers to pages containing p2m entries */
static unsigned long *p2m_top[TOP_ENTRIES]
__attribute__((section(".data.page_aligned"))) =
{ [ 0 ... TOP_ENTRIES - 1] = &p2m_missing[0] };
/* Arrays of p2m arrays expressed in mfns used for save/restore */
static unsigned long p2m_top_mfn[TOP_ENTRIES]
__attribute__((section(".bss.page_aligned")));
xen: fix early bootup crash on native hardware -tip tree auto-testing found the following early bootup hang: --------------> get_memcfg_from_srat: assigning address to rsdp RSD PTR v0 [Nvidia] BUG: Int 14: CR2 ffd00040 EDI 8092fbfe ESI ffd00040 EBP 80b0aee8 ESP 80b0aed0 EBX 000f76f0 EDX 0000000e ECX 00000003 EAX ffd00040 err 00000000 EIP 802c055a CS 00000060 flg 00010006 Stack: ffd00040 80bc78d0 80b0af6c 80b1dbfe 8093d8ba 00000008 80b42810 80b4ddb4 80b42842 00000000 80b0af1c 801079c8 808e724e 00000000 80b42871 802c0531 00000100 00000000 0003fff0 80b0af40 80129999 00040100 00040100 00000000 Pid: 0, comm: swapper Not tainted 2.6.26-rc4-sched-devel.git #570 [<802c055a>] ? strncmp+0x11/0x25 [<80b1dbfe>] ? get_memcfg_from_srat+0xb4/0x568 [<801079c8>] ? mcount_call+0x5/0x9 [<802c0531>] ? strcmp+0xa/0x22 [<80129999>] ? printk+0x38/0x3a [<80129999>] ? printk+0x38/0x3a [<8011b122>] ? memory_present+0x66/0x6f [<80b216b4>] ? setup_memory+0x13/0x40c [<80b16b47>] ? propagate_e820_map+0x80/0x97 [<80b1622a>] ? setup_arch+0x248/0x477 [<80129999>] ? printk+0x38/0x3a [<80b11759>] ? start_kernel+0x6e/0x2eb [<80b110fc>] ? i386_start_kernel+0xeb/0xf2 ======================= <------ with this config: http://redhat.com/~mingo/misc/config-Wed_May_28_01_33_33_CEST_2008.bad The thing is, the crash makes little sense at first sight. We crash on a benign-looking printk. The code around it got changed in -tip but checking those topic branches individually did not reproduce the bug. Bisection led to this commit: | d5edbc1f75420935b1ec7e65df10c8f81cea82de is first bad commit | commit d5edbc1f75420935b1ec7e65df10c8f81cea82de | Author: Jeremy Fitzhardinge <jeremy@goop.org> | Date: Mon May 26 23:31:22 2008 +0100 | | xen: add p2m mfn_list_list Which is somewhat surprising, as on native hardware Xen client side should have little to no side-effects. After some head scratching, it turns out the following happened: randconfig enabled the following Xen options: CONFIG_XEN=y CONFIG_XEN_MAX_DOMAIN_MEMORY=8 # CONFIG_XEN_BLKDEV_FRONTEND is not set # CONFIG_XEN_NETDEV_FRONTEND is not set CONFIG_HVC_XEN=y # CONFIG_XEN_BALLOON is not set which activated this piece of code in arch/x86/xen/mmu.c: > @@ -69,6 +69,13 @@ > __attribute__((section(".data.page_aligned"))) = > { [ 0 ... TOP_ENTRIES - 1] = &p2m_missing[0] }; > > +/* Arrays of p2m arrays expressed in mfns used for save/restore */ > +static unsigned long p2m_top_mfn[TOP_ENTRIES] > + __attribute__((section(".bss.page_aligned"))); > + > +static unsigned long p2m_top_mfn_list[TOP_ENTRIES / P2M_ENTRIES_PER_PAGE] > + __attribute__((section(".bss.page_aligned"))); The problem is, you must only put variables into .bss.page_aligned that have a _size_ that is _exactly_ page aligned. In this case the size of p2m_top_mfn_list is not page aligned: 80b8d000 b p2m_top_mfn 80b8f000 b p2m_top_mfn_list 80b8f008 b softirq_stack 80b97008 b hardirq_stack 80b9f008 b bm_pte So all subsequent variables get unaligned which, depending on luck, breaks the kernel in various funny ways. In this case what killed the kernel first was the misaligned bootmap pte page, resulting in that creative crash above. Anyway, this was a fun bug to track down :-) I think the moral is that .bss.page_aligned is a dangerous construct in its current form, and the symptoms of breakage are very non-trivial, so i think we need build-time checks to make sure all symbols in .bss.page_aligned are truly page aligned. The Xen fix below gets the kernel booting again. Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-05-28 12:24:38 +00:00
static unsigned long p2m_top_mfn_list[
PAGE_ALIGN(TOP_ENTRIES / P2M_ENTRIES_PER_PAGE)]
__attribute__((section(".bss.page_aligned")));
static inline unsigned p2m_top_index(unsigned long pfn)
{
BUG_ON(pfn >= MAX_DOMAIN_PAGES);
return pfn / P2M_ENTRIES_PER_PAGE;
}
static inline unsigned p2m_index(unsigned long pfn)
{
return pfn % P2M_ENTRIES_PER_PAGE;
}
/* Build the parallel p2m_top_mfn structures */
void xen_setup_mfn_list_list(void)
{
unsigned pfn, idx;
for(pfn = 0; pfn < MAX_DOMAIN_PAGES; pfn += P2M_ENTRIES_PER_PAGE) {
unsigned topidx = p2m_top_index(pfn);
p2m_top_mfn[topidx] = virt_to_mfn(p2m_top[topidx]);
}
for(idx = 0; idx < ARRAY_SIZE(p2m_top_mfn_list); idx++) {
unsigned topidx = idx * P2M_ENTRIES_PER_PAGE;
p2m_top_mfn_list[idx] = virt_to_mfn(&p2m_top_mfn[topidx]);
}
BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info);
HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
virt_to_mfn(p2m_top_mfn_list);
HYPERVISOR_shared_info->arch.max_pfn = xen_start_info->nr_pages;
}
/* Set up p2m_top to point to the domain-builder provided p2m pages */
void __init xen_build_dynamic_phys_to_machine(void)
{
unsigned long *mfn_list = (unsigned long *)xen_start_info->mfn_list;
unsigned long max_pfn = min(MAX_DOMAIN_PAGES, xen_start_info->nr_pages);
unsigned pfn;
for(pfn = 0; pfn < max_pfn; pfn += P2M_ENTRIES_PER_PAGE) {
unsigned topidx = p2m_top_index(pfn);
p2m_top[topidx] = &mfn_list[pfn];
}
}
unsigned long get_phys_to_machine(unsigned long pfn)
{
unsigned topidx, idx;
if (unlikely(pfn >= MAX_DOMAIN_PAGES))
return INVALID_P2M_ENTRY;
topidx = p2m_top_index(pfn);
idx = p2m_index(pfn);
return p2m_top[topidx][idx];
}
EXPORT_SYMBOL_GPL(get_phys_to_machine);
static void alloc_p2m(unsigned long **pp, unsigned long *mfnp)
{
unsigned long *p;
unsigned i;
p = (void *)__get_free_page(GFP_KERNEL | __GFP_NOFAIL);
BUG_ON(p == NULL);
for(i = 0; i < P2M_ENTRIES_PER_PAGE; i++)
p[i] = INVALID_P2M_ENTRY;
if (cmpxchg(pp, p2m_missing, p) != p2m_missing)
free_page((unsigned long)p);
else
*mfnp = virt_to_mfn(p);
}
void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
{
unsigned topidx, idx;
if (unlikely(xen_feature(XENFEAT_auto_translated_physmap))) {
BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
return;
}
if (unlikely(pfn >= MAX_DOMAIN_PAGES)) {
BUG_ON(mfn != INVALID_P2M_ENTRY);
return;
}
topidx = p2m_top_index(pfn);
if (p2m_top[topidx] == p2m_missing) {
/* no need to allocate a page to store an invalid entry */
if (mfn == INVALID_P2M_ENTRY)
return;
alloc_p2m(&p2m_top[topidx], &p2m_top_mfn[topidx]);
}
idx = p2m_index(pfn);
p2m_top[topidx][idx] = mfn;
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
xmaddr_t arbitrary_virt_to_machine(unsigned long address)
{
unsigned int level;
pte_t *pte = lookup_address(address, &level);
unsigned offset = address & ~PAGE_MASK;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
BUG_ON(pte == NULL);
return XMADDR((pte_mfn(*pte) << PAGE_SHIFT) + offset);
}
void make_lowmem_page_readonly(void *vaddr)
{
pte_t *pte, ptev;
unsigned long address = (unsigned long)vaddr;
unsigned int level;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
pte = lookup_address(address, &level);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
BUG_ON(pte == NULL);
ptev = pte_wrprotect(*pte);
if (HYPERVISOR_update_va_mapping(address, ptev, 0))
BUG();
}
void make_lowmem_page_readwrite(void *vaddr)
{
pte_t *pte, ptev;
unsigned long address = (unsigned long)vaddr;
unsigned int level;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
pte = lookup_address(address, &level);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
BUG_ON(pte == NULL);
ptev = pte_mkwrite(*pte);
if (HYPERVISOR_update_va_mapping(address, ptev, 0))
BUG();
}
static bool page_pinned(void *ptr)
{
struct page *page = virt_to_page(ptr);
return PagePinned(page);
}
static void extend_mmu_update(const struct mmu_update *update)
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
{
struct multicall_space mcs;
struct mmu_update *u;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
mcs = xen_mc_extend_args(__HYPERVISOR_mmu_update, sizeof(*u));
if (mcs.mc != NULL)
mcs.mc->args[1]++;
else {
mcs = __xen_mc_entry(sizeof(*u));
MULTI_mmu_update(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
}
u = mcs.args;
*u = *update;
}
void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
{
struct mmu_update u;
preempt_disable();
xen_mc_batch();
u.ptr = virt_to_machine(ptr).maddr;
u.val = pmd_val_ma(val);
extend_mmu_update(&u);
xen_mc_issue(PARAVIRT_LAZY_MMU);
preempt_enable();
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
void xen_set_pmd(pmd_t *ptr, pmd_t val)
{
/* If page is not pinned, we can just update the entry
directly */
if (!page_pinned(ptr)) {
*ptr = val;
return;
}
xen_set_pmd_hyper(ptr, val);
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
/*
* Associate a virtual page frame with a given physical page frame
* and protection flags for that frame.
*/
void set_pte_mfn(unsigned long vaddr, unsigned long mfn, pgprot_t flags)
{
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
pgd = swapper_pg_dir + pgd_index(vaddr);
if (pgd_none(*pgd)) {
BUG();
return;
}
pud = pud_offset(pgd, vaddr);
if (pud_none(*pud)) {
BUG();
return;
}
pmd = pmd_offset(pud, vaddr);
if (pmd_none(*pmd)) {
BUG();
return;
}
pte = pte_offset_kernel(pmd, vaddr);
/* <mfn,flags> stored as-is, to permit clearing entries */
xen_set_pte(pte, mfn_pte(mfn, flags));
/*
* It's enough to flush this one mapping.
* (PGE mappings get flushed as well)
*/
__flush_tlb_one(vaddr);
}
void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
pte_t *ptep, pte_t pteval)
{
/* updates to init_mm may be done without lock */
if (mm == &init_mm)
preempt_disable();
if (mm == current->mm || mm == &init_mm) {
if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) {
struct multicall_space mcs;
mcs = xen_mc_entry(0);
MULTI_update_va_mapping(mcs.mc, addr, pteval, 0);
xen_mc_issue(PARAVIRT_LAZY_MMU);
goto out;
} else
if (HYPERVISOR_update_va_mapping(addr, pteval, 0) == 0)
goto out;
}
xen_set_pte(ptep, pteval);
out:
if (mm == &init_mm)
preempt_enable();
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
pte_t xen_ptep_modify_prot_start(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
{
/* Just return the pte as-is. We preserve the bits on commit */
return *ptep;
}
void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
pte_t *ptep, pte_t pte)
{
struct mmu_update u;
xen_mc_batch();
u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD;
u.val = pte_val_ma(pte);
extend_mmu_update(&u);
xen_mc_issue(PARAVIRT_LAZY_MMU);
}
/* Assume pteval_t is equivalent to all the other *val_t types. */
static pteval_t pte_mfn_to_pfn(pteval_t val)
{
if (val & _PAGE_PRESENT) {
unsigned long mfn = (val & PTE_MASK) >> PAGE_SHIFT;
pteval_t flags = val & ~PTE_MASK;
val = ((pteval_t)mfn_to_pfn(mfn) << PAGE_SHIFT) | flags;
}
return val;
}
static pteval_t pte_pfn_to_mfn(pteval_t val)
{
if (val & _PAGE_PRESENT) {
unsigned long pfn = (val & PTE_MASK) >> PAGE_SHIFT;
pteval_t flags = val & ~PTE_MASK;
val = ((pteval_t)pfn_to_mfn(pfn) << PAGE_SHIFT) | flags;
}
return val;
}
pteval_t xen_pte_val(pte_t pte)
{
return pte_mfn_to_pfn(pte.pte);
}
pgdval_t xen_pgd_val(pgd_t pgd)
{
return pte_mfn_to_pfn(pgd.pgd);
}
pte_t xen_make_pte(pteval_t pte)
{
pte = pte_pfn_to_mfn(pte);
return native_make_pte(pte);
}
pgd_t xen_make_pgd(pgdval_t pgd)
{
pgd = pte_pfn_to_mfn(pgd);
return native_make_pgd(pgd);
}
pmdval_t xen_pmd_val(pmd_t pmd)
{
return pte_mfn_to_pfn(pmd.pmd);
}
void xen_set_pud_hyper(pud_t *ptr, pud_t val)
{
struct mmu_update u;
preempt_disable();
xen_mc_batch();
u.ptr = virt_to_machine(ptr).maddr;
u.val = pud_val_ma(val);
extend_mmu_update(&u);
xen_mc_issue(PARAVIRT_LAZY_MMU);
preempt_enable();
}
void xen_set_pud(pud_t *ptr, pud_t val)
{
/* If page is not pinned, we can just update the entry
directly */
if (!page_pinned(ptr)) {
*ptr = val;
return;
}
xen_set_pud_hyper(ptr, val);
}
void xen_set_pte(pte_t *ptep, pte_t pte)
{
ptep->pte_high = pte.pte_high;
smp_wmb();
ptep->pte_low = pte.pte_low;
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
void xen_set_pte_atomic(pte_t *ptep, pte_t pte)
{
set_64bit((u64 *)ptep, pte_val_ma(pte));
}
void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
{
ptep->pte_low = 0;
smp_wmb(); /* make sure low gets written first */
ptep->pte_high = 0;
}
void xen_pmd_clear(pmd_t *pmdp)
{
set_pmd(pmdp, __pmd(0));
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
pmd_t xen_make_pmd(pmdval_t pmd)
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
{
pmd = pte_pfn_to_mfn(pmd);
return native_make_pmd(pmd);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
/*
(Yet another) pagetable walker. This one is intended for pinning a
pagetable. This means that it walks a pagetable and calls the
callback function on each page it finds making up the page table,
at every level. It walks the entire pagetable, but it only bothers
pinning pte pages which are below pte_limit. In the normal case
this will be TASK_SIZE, but at boot we need to pin up to
FIXADDR_TOP. But the important bit is that we don't pin beyond
there, because then we start getting into Xen's ptes.
*/
static int pgd_walk(pgd_t *pgd_base, int (*func)(struct page *, enum pt_level),
unsigned long limit)
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
{
pgd_t *pgd = pgd_base;
int flush = 0;
unsigned long addr = 0;
unsigned long pgd_next;
BUG_ON(limit > FIXADDR_TOP);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
if (xen_feature(XENFEAT_auto_translated_physmap))
return 0;
for (; addr != FIXADDR_TOP; pgd++, addr = pgd_next) {
pud_t *pud;
unsigned long pud_limit, pud_next;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
pgd_next = pud_limit = pgd_addr_end(addr, FIXADDR_TOP);
if (!pgd_val(*pgd))
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
continue;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
pud = pud_offset(pgd, 0);
if (PTRS_PER_PUD > 1) /* not folded */
flush |= (*func)(virt_to_page(pud), PT_PUD);
for (; addr != pud_limit; pud++, addr = pud_next) {
pmd_t *pmd;
unsigned long pmd_limit;
pud_next = pud_addr_end(addr, pud_limit);
if (pud_next < limit)
pmd_limit = pud_next;
else
pmd_limit = limit;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
if (pud_none(*pud))
continue;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
pmd = pmd_offset(pud, 0);
if (PTRS_PER_PMD > 1) /* not folded */
flush |= (*func)(virt_to_page(pmd), PT_PMD);
for (; addr != pmd_limit; pmd++) {
addr += (PAGE_SIZE * PTRS_PER_PTE);
if ((pmd_limit-1) < (addr-1)) {
addr = pmd_limit;
break;
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
if (pmd_none(*pmd))
continue;
flush |= (*func)(pmd_page(*pmd), PT_PTE);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
}
}
flush |= (*func)(virt_to_page(pgd_base), PT_PGD);
return flush;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
static spinlock_t *lock_pte(struct page *page)
{
spinlock_t *ptl = NULL;
#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
ptl = __pte_lockptr(page);
spin_lock(ptl);
#endif
return ptl;
}
static void do_unlock(void *v)
{
spinlock_t *ptl = v;
spin_unlock(ptl);
}
static void xen_do_pin(unsigned level, unsigned long pfn)
{
struct mmuext_op *op;
struct multicall_space mcs;
mcs = __xen_mc_entry(sizeof(*op));
op = mcs.args;
op->cmd = level;
op->arg1.mfn = pfn_to_mfn(pfn);
MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
}
static int pin_page(struct page *page, enum pt_level level)
{
unsigned pgfl = TestSetPagePinned(page);
int flush;
if (pgfl)
flush = 0; /* already pinned */
else if (PageHighMem(page))
/* kmaps need flushing if we found an unpinned
highpage */
flush = 1;
else {
void *pt = lowmem_page_address(page);
unsigned long pfn = page_to_pfn(page);
struct multicall_space mcs = __xen_mc_entry(0);
spinlock_t *ptl;
flush = 0;
ptl = NULL;
if (level == PT_PTE)
ptl = lock_pte(page);
MULTI_update_va_mapping(mcs.mc, (unsigned long)pt,
pfn_pte(pfn, PAGE_KERNEL_RO),
level == PT_PGD ? UVMF_TLB_FLUSH : 0);
if (level == PT_PTE)
xen_do_pin(MMUEXT_PIN_L1_TABLE, pfn);
if (ptl) {
/* Queue a deferred unlock for when this batch
is completed. */
xen_mc_callback(do_unlock, ptl);
}
}
return flush;
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
/* This is called just after a mm has been created, but it has not
been used yet. We need to make sure that its pagetable is all
read-only, and can be pinned. */
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
void xen_pgd_pin(pgd_t *pgd)
{
xen_mc_batch();
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
if (pgd_walk(pgd, pin_page, TASK_SIZE)) {
/* re-enable interrupts for kmap_flush_unused */
xen_mc_issue(0);
kmap_flush_unused();
xen_mc_batch();
}
xen_do_pin(MMUEXT_PIN_L3_TABLE, PFN_DOWN(__pa(pgd)));
xen_mc_issue(0);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
/*
* On save, we need to pin all pagetables to make sure they get their
* mfns turned into pfns. Search the list for any unpinned pgds and pin
* them (unpinned pgds are not currently in use, probably because the
* process is under construction or destruction).
*/
void xen_mm_pin_all(void)
{
unsigned long flags;
struct page *page;
spin_lock_irqsave(&pgd_lock, flags);
list_for_each_entry(page, &pgd_list, lru) {
if (!PagePinned(page)) {
xen_pgd_pin((pgd_t *)page_address(page));
SetPageSavePinned(page);
}
}
spin_unlock_irqrestore(&pgd_lock, flags);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
/*
* The init_mm pagetable is really pinned as soon as its created, but
* that's before we have page structures to store the bits. So do all
* the book-keeping now.
*/
static __init int mark_pinned(struct page *page, enum pt_level level)
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
{
SetPagePinned(page);
return 0;
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
void __init xen_mark_init_mm_pinned(void)
{
pgd_walk(init_mm.pgd, mark_pinned, FIXADDR_TOP);
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
static int unpin_page(struct page *page, enum pt_level level)
{
unsigned pgfl = TestClearPagePinned(page);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
if (pgfl && !PageHighMem(page)) {
void *pt = lowmem_page_address(page);
unsigned long pfn = page_to_pfn(page);
spinlock_t *ptl = NULL;
struct multicall_space mcs;
if (level == PT_PTE) {
ptl = lock_pte(page);
xen_do_pin(MMUEXT_UNPIN_TABLE, pfn);
}
mcs = __xen_mc_entry(0);
MULTI_update_va_mapping(mcs.mc, (unsigned long)pt,
pfn_pte(pfn, PAGE_KERNEL),
level == PT_PGD ? UVMF_TLB_FLUSH : 0);
if (ptl) {
/* unlock when batch completed */
xen_mc_callback(do_unlock, ptl);
}
}
return 0; /* never need to flush on unpin */
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
/* Release a pagetables pages back as normal RW */
static void xen_pgd_unpin(pgd_t *pgd)
{
xen_mc_batch();
xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
pgd_walk(pgd, unpin_page, TASK_SIZE);
xen_mc_issue(0);
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
/*
* On resume, undo any pinning done at save, so that the rest of the
* kernel doesn't see any unexpected pinned pagetables.
*/
void xen_mm_unpin_all(void)
{
unsigned long flags;
struct page *page;
spin_lock_irqsave(&pgd_lock, flags);
list_for_each_entry(page, &pgd_list, lru) {
if (PageSavePinned(page)) {
BUG_ON(!PagePinned(page));
printk("unpinning pinned %p\n", page_address(page));
xen_pgd_unpin((pgd_t *)page_address(page));
ClearPageSavePinned(page);
}
}
spin_unlock_irqrestore(&pgd_lock, flags);
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
void xen_activate_mm(struct mm_struct *prev, struct mm_struct *next)
{
spin_lock(&next->page_table_lock);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
xen_pgd_pin(next->pgd);
spin_unlock(&next->page_table_lock);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
void xen_dup_mmap(struct mm_struct *oldmm, struct mm_struct *mm)
{
spin_lock(&mm->page_table_lock);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
xen_pgd_pin(mm->pgd);
spin_unlock(&mm->page_table_lock);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
#ifdef CONFIG_SMP
/* Another cpu may still have their %cr3 pointing at the pagetable, so
we need to repoint it somewhere else before we can unpin it. */
static void drop_other_mm_ref(void *info)
{
struct mm_struct *mm = info;
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
if (__get_cpu_var(cpu_tlbstate).active_mm == mm)
leave_mm(smp_processor_id());
/* If this cpu still has a stale cr3 reference, then make sure
it has been flushed. */
if (x86_read_percpu(xen_current_cr3) == __pa(mm->pgd)) {
load_cr3(swapper_pg_dir);
arch_flush_lazy_cpu_mode();
}
}
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
static void drop_mm_ref(struct mm_struct *mm)
{
cpumask_t mask;
unsigned cpu;
if (current->active_mm == mm) {
if (current->mm == mm)
load_cr3(swapper_pg_dir);
else
leave_mm(smp_processor_id());
arch_flush_lazy_cpu_mode();
}
/* Get the "official" set of cpus referring to our pagetable. */
mask = mm->cpu_vm_mask;
/* It's possible that a vcpu may have a stale reference to our
cr3, because its in lazy mode, and it hasn't yet flushed
its set of pending hypercalls yet. In this case, we can
look at its actual current cr3 value, and force it to flush
if needed. */
for_each_online_cpu(cpu) {
if (per_cpu(xen_current_cr3, cpu) == __pa(mm->pgd))
cpu_set(cpu, mask);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}
if (!cpus_empty(mask))
smp_call_function_mask(mask, drop_other_mm_ref, mm, 1);
}
#else
static void drop_mm_ref(struct mm_struct *mm)
{
if (current->active_mm == mm)
load_cr3(swapper_pg_dir);
}
#endif
/*
* While a process runs, Xen pins its pagetables, which means that the
* hypervisor forces it to be read-only, and it controls all updates
* to it. This means that all pagetable updates have to go via the
* hypervisor, which is moderately expensive.
*
* Since we're pulling the pagetable down, we switch to use init_mm,
* unpin old process pagetable and mark it all read-write, which
* allows further operations on it to be simple memory accesses.
*
* The only subtle point is that another CPU may be still using the
* pagetable because of lazy tlb flushing. This means we need need to
* switch all CPUs off this pagetable before we can unpin it.
*/
void xen_exit_mmap(struct mm_struct *mm)
{
get_cpu(); /* make sure we don't move around */
drop_mm_ref(mm);
put_cpu();
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
spin_lock(&mm->page_table_lock);
/* pgd may not be pinned in the error exit path of execve */
if (page_pinned(mm->pgd))
xen_pgd_unpin(mm->pgd);
spin_unlock(&mm->page_table_lock);
xen: virtual mmu Xen pagetable handling, including the machinery to implement direct pagetables. Xen presents the real CPU's pagetables directly to guests, with no added shadowing or other layer of abstraction. Naturally this means the hypervisor must maintain close control over what the guest can put into the pagetable. When the guest modifies the pte/pmd/pgd, it must convert its domain-specific notion of a "physical" pfn into a global machine frame number (mfn) before inserting the entry into the pagetable. Xen will check to make sure the domain is allowed to create a mapping of the given mfn. Xen also requires that all mappings the guest has of its own active pagetable are read-only. This is relatively easy to implement in Linux because all pagetables share the same pte pages for kernel mappings, so updating the pte in one pagetable will implicitly update the mapping in all pagetables. Normally a pagetable becomes active when you point to it with cr3 (or the Xen equivalent), but when you do so, Xen must check the whole pagetable for correctness, which is clearly a performance problem. Xen solves this with pinning which keeps a pagetable effectively active even if its currently unused, which means that all the normal update rules are enforced. This means that it need not revalidate the pagetable when loading cr3. This patch has a first-cut implementation of pinning, but it is more fully implemented in a later patch. Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Chris Wright <chrisw@sous-sol.org>
2007-07-18 01:37:04 +00:00
}