Message-ID: <1312964587.26263.168.camel@zakaz.uk.xensource.com>
Date:	Wed, 10 Aug 2011 09:23:07 +0100
From:	Ian Campbell <Ian.Campbell@...citrix.com>
To:	"stefano.stabellini@...citrix.com" <stefano.stabellini@...citrix.com>
CC:	"konrad.wilk@...cle.com" <konrad.wilk@...cle.com>,
	"xen-devel@...ts.xensource.com" <xen-devel@...ts.xensource.com>,
	"Jeremy Fitzhardinge" <Jeremy.Fitzhardinge@...rix.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] xen: modify kernel mappings corresponding to granted
 pages

On Tue, 2011-07-26 at 17:55 +0100, stefano.stabellini@...citrix.com
wrote:
> From: Stefano Stabellini <stefano.stabellini@...citrix.com>
> 
> If we want to use granted pages for AIO, changing the mappings of a user
> vma and the corresponding p2m is not enough, we also need to update the
> kernel mappings accordingly.
> On x86_64 it is easy, we can issue another HYPERVISOR_grant_table_op
> right away in m2p_add_override. We can remove the mappings using another
> HYPERVISOR_grant_table_op in m2p_remove_override.
> On x86_32 it is more difficult because the pages are highmem pages

Aren't kernel mappings of highmem pages always in the fixmap? Therefore
can we maybe catch this at that layer in the pvops infrastructure? That
would avoid the impact on all set_pte calls and (perhaps) provide more
context regarding the mapping. For example that might mean we could keep
a bitmap (or other data structure as appropriate) mapping the fixmap IDX
(which is a smallish integer, far more manageable than {P,M}FN) into
granted-ness in order to speed up the check for this case. Avoiding the
m2p_override lookup in the common case seems like something which would
be worth putting some effort into.

> and
> therefore we need to catch the set_pte that tries to map a granted page
> and issue an HYPERVISOR_grant_table_op instead.

> Same thing for unmapping them: we need to catch the pte clear or the
> set_pte that try to unmap a granted page and issue an
> HYPERVISOR_grant_table_op.
> 
> Signed-off-by: Stefano Stabellini <stefano.stabellini@...citrix.com>
> ---
>  arch/x86/include/asm/xen/page.h |    5 ++-
>  arch/x86/xen/mmu.c              |   68 +++++++++++++++++++++++++++++++++++++++
>  arch/x86/xen/p2m.c              |   47 ++++++++++++++++++++------
>  drivers/xen/gntdev.c            |   27 +++++++++++++++-
>  drivers/xen/grant-table.c       |    4 +-
>  include/xen/grant_table.h       |    1 +
>  6 files changed, 137 insertions(+), 15 deletions(-)
> 
> diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h
> index 64a619d..ec7bbfb 100644
> --- a/arch/x86/include/asm/xen/page.h
> +++ b/arch/x86/include/asm/xen/page.h
> @@ -12,6 +12,7 @@
>  #include <asm/pgtable.h>
>  
>  #include <xen/interface/xen.h>
> +#include <xen/grant_table.h>
>  #include <xen/features.h>
>  
>  /* Xen machine address */
> @@ -31,8 +32,10 @@ typedef struct xpaddr {
>  #define INVALID_P2M_ENTRY	(~0UL)
>  #define FOREIGN_FRAME_BIT	(1UL<<(BITS_PER_LONG-1))
>  #define IDENTITY_FRAME_BIT	(1UL<<(BITS_PER_LONG-2))
> +#define GRANT_FRAME_BIT	(1UL<<(BITS_PER_LONG-3))
>  #define FOREIGN_FRAME(m)	((m) | FOREIGN_FRAME_BIT)
>  #define IDENTITY_FRAME(m)	((m) | IDENTITY_FRAME_BIT)
> +#define GRANT_FRAME(m)	((m) | GRANT_FRAME_BIT)

GRANT_FRAME_BIT, despite being declared like
{FOREIGN,IDENTITY}_FRAME_BIT is not actually used in the upper bits of
an MFN but rather is used as a flag in page->private.

If it's not to be used in an MFN in the style of the others then there
is no need for it to follow that pattern. The declaration of GRANT_FRAME
is unused and confusing since the bit isn't even used that way. If
page->private is to be used as a bitfield then it needs to be clear
which bits are used in that context vs. the MFN high bits, i.e. it
should be declared somewhere else with an appropriate comment.

More importantly though are we sure that page->private is always
available for use by this code? The private field is for the use of the
owner of the page. In the case where the page is a granted page then we
are (ultimately) the owner, but in the case where the page isn't a
granted page at all the owner could be another subsystem altogether and
there could be false positives if that bit happens to be set.

>  
>  /* Maximum amount of memory we can handle in a domain in pages */
>  #define MAX_DOMAIN_PAGES						\
> @@ -48,7 +51,7 @@ extern unsigned long set_phys_range_identity(unsigned long pfn_s,
>  					     unsigned long pfn_e);
>  
>  extern int m2p_add_override(unsigned long mfn, struct page *page,
> -			    bool clear_pte);
> +			    struct gnttab_map_grant_ref *kmap_op);
>  extern int m2p_remove_override(struct page *page, bool clear_pte);
>  extern struct page *m2p_find_override(unsigned long mfn);
>  extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn);
> diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
> index 4e37a7c0..13f20d8 100644
> --- a/arch/x86/xen/mmu.c
> +++ b/arch/x86/xen/mmu.c
> @@ -282,8 +282,70 @@ static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval)
>  	return true;
>  }
>  
> +#ifdef CONFIG_HIGHMEM
> +static int xen_unmap_granted_page(pte_t *ptep)
> +{
> +	unsigned long mfn;
> +	struct page *page;
> +
> +	if (pte_flags(*ptep) & _PAGE_USER)
> +		return 1;
> +	mfn = (ptep->pte & PTE_PFN_MASK) >> PAGE_SHIFT;
> +	page = m2p_find_override(mfn);
> +	if (page != NULL && (page->private & GRANT_FRAME_BIT)) {
> +		int ret;
> +		struct gnttab_unmap_grant_ref kunmap_op;
> +		struct gnttab_map_grant_ref *kmap_op =
> +			(struct gnttab_map_grant_ref *) page->index;
> +		kunmap_op.host_addr = kmap_op->host_addr;
> +		kunmap_op.handle = kmap_op->handle;
> +		kunmap_op.dev_bus_addr = 0;
> +		ret = HYPERVISOR_grant_table_op(
> +				GNTTABOP_unmap_grant_ref, &kunmap_op, 1);
> +		WARN(ret, "m2p_remove_override: pfn %lx mfn %lx, failed to "
> +			"modify kernel mappings", page_to_pfn(page), mfn);
> +		return ret;
> +	}
> +	return 1;
> +}
> +#endif
> +
>  static void xen_set_pte(pte_t *ptep, pte_t pteval)
>  {
> +#ifdef CONFIG_HIGHMEM
> +	if (!(pte_flags(pteval) & _PAGE_USER)) {
> +		int ret;
> +		struct page *page;
> +		unsigned long mfn = (pteval.pte & PTE_PFN_MASK) >> PAGE_SHIFT;
> +		page = m2p_find_override(mfn);
> +		/*
> +		 * if this is a granted page we need to use a grant table
> +		 * hypercall to map it instead
> +		 */
> +		if (page != NULL && (page->private & GRANT_FRAME_BIT)) {
> +			struct gnttab_map_grant_ref *kmap_op =
> +				(struct gnttab_map_grant_ref *) page->index;
> +			unsigned long old_mfn = kmap_op->dev_bus_addr;
> +			kmap_op->host_addr =
> +				arbitrary_virt_to_machine(ptep).maddr;
> +			kmap_op->dev_bus_addr = 0;
> +			ret = HYPERVISOR_grant_table_op(
> +					GNTTABOP_map_grant_ref, kmap_op, 1);
> +			WARN(ret, "xen_set_pte: pfn %lx mfn %lx, failed to "
> +				"modify kernel mappings",
> +				page_to_pfn(page), mfn);
> +			kmap_op->dev_bus_addr = old_mfn;
> +			return;
> +		}
> +		/*
> +		 * even if it is not a granted page, the old page we are about
> +		 * to overwrite could be a granted page and in that case we need
> +		 * to unmap it using a grant table hypercall
> +		 */
> +		xen_unmap_granted_page(ptep);
> +	}
> +#endif
> +
>  	if (!xen_batched_set_pte(ptep, pteval))
>  		native_set_pte(ptep, pteval);
>  }
> @@ -548,6 +610,12 @@ static void xen_set_pte_atomic(pte_t *ptep, pte_t pte)
>  
>  static void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
>  {
> +	/*
> +	 * check if this is a granted page and unmap it using a grant table
> +	 * hypercall in that case
> +	 */
> +	if (!xen_unmap_granted_page(ptep))
> +		return;
>  	if (!xen_batched_set_pte(ptep, native_make_pte(0)))
>  		native_pte_clear(mm, addr, ptep);
>  }
> diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
> index 58efeb9..1c4d2b5 100644
> --- a/arch/x86/xen/p2m.c
> +++ b/arch/x86/xen/p2m.c
> @@ -161,6 +161,7 @@
>  #include <asm/xen/page.h>
>  #include <asm/xen/hypercall.h>
>  #include <asm/xen/hypervisor.h>
> +#include <xen/grant_table.h>
>  
>  #include "xen-ops.h"
>  
> @@ -676,7 +677,8 @@ static unsigned long mfn_hash(unsigned long mfn)
>  }
>  
>  /* Add an MFN override for a particular page */
> -int m2p_add_override(unsigned long mfn, struct page *page, bool clear_pte)
> +int m2p_add_override(unsigned long mfn, struct page *page,
> +		struct gnttab_map_grant_ref *kmap_op)
>  {
>  	unsigned long flags;
>  	unsigned long pfn;
> @@ -699,9 +701,18 @@ int m2p_add_override(unsigned long mfn, struct page *page, bool clear_pte)
>  	if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn))))
>  		return -ENOMEM;
>  
> -	if (clear_pte && !PageHighMem(page))
> -		/* Just zap old mapping for now */
> -		pte_clear(&init_mm, address, ptep);
> +	if (kmap_op != NULL) {
> +		if (!PageHighMem(page)) {
> +			int ret = HYPERVISOR_grant_table_op(
> +					GNTTABOP_map_grant_ref, kmap_op, 1);
> +			WARN(ret, "m2p_add_override: pfn %lx mfn %lx, "
> +				"failed to modify kernel mappings", pfn, mfn);
> +		}
> +		page->private |= GRANT_FRAME_BIT;
> +		/* let's use dev_bus_addr to record the old mfn instead */
> +		kmap_op->dev_bus_addr = page->index;
> +		page->index = (unsigned long) kmap_op;
> +	}
>  	spin_lock_irqsave(&m2p_override_lock, flags);
>  	list_add(&page->lru,  &m2p_overrides[mfn_hash(mfn)]);
>  	spin_unlock_irqrestore(&m2p_override_lock, flags);
> @@ -735,13 +746,27 @@ int m2p_remove_override(struct page *page, bool clear_pte)
>  	spin_lock_irqsave(&m2p_override_lock, flags);
>  	list_del(&page->lru);
>  	spin_unlock_irqrestore(&m2p_override_lock, flags);
> -	set_phys_to_machine(pfn, page->index);
>  
> -	if (clear_pte && !PageHighMem(page))
> -		set_pte_at(&init_mm, address, ptep,
> -				pfn_pte(pfn, PAGE_KERNEL));
> -		/* No tlb flush necessary because the caller already
> -		 * left the pte unmapped. */
> +	if (clear_pte) {
> +		struct gnttab_map_grant_ref *map_op =
> +			(struct gnttab_map_grant_ref *) page->index;
> +		set_phys_to_machine(pfn, map_op->dev_bus_addr);
> +		if (!PageHighMem(page)) {
> +			int ret;
> +			struct gnttab_unmap_grant_ref unmap_op;
> +			unmap_op.host_addr = map_op->host_addr;
> +			unmap_op.handle = map_op->handle;
> +			unmap_op.dev_bus_addr = 0;
> +			ret = HYPERVISOR_grant_table_op(
> +					GNTTABOP_unmap_grant_ref, &unmap_op, 1);
> +			WARN(ret, "m2p_remove_override: pfn %lx mfn %lx, "
> +				"failed to modify kernel mappings", pfn, mfn);
> +			set_pte_at(&init_mm, address, ptep,
> +					pfn_pte(pfn, PAGE_KERNEL));
> +			__flush_tlb_single(address);
> +		}
> +	} else
> +		set_phys_to_machine(pfn, page->index);
>  
>  	return 0;
>  }
> @@ -758,7 +783,7 @@ struct page *m2p_find_override(unsigned long mfn)
>  	spin_lock_irqsave(&m2p_override_lock, flags);
>  
>  	list_for_each_entry(p, bucket, lru) {
> -		if (p->private == mfn) {
> +		if ((p->private & (~GRANT_FRAME_BIT)) == mfn) {
>  			ret = p;
>  			break;
>  		}
> diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
> index f914b26..ca41772 100644
> --- a/drivers/xen/gntdev.c
> +++ b/drivers/xen/gntdev.c
> @@ -83,6 +83,7 @@ struct grant_map {
>  	struct ioctl_gntdev_grant_ref *grants;
>  	struct gnttab_map_grant_ref   *map_ops;
>  	struct gnttab_unmap_grant_ref *unmap_ops;
> +	struct gnttab_map_grant_ref   *kmap_ops;
>  	struct page **pages;
>  };
>  
> @@ -116,10 +117,12 @@ static struct grant_map *gntdev_alloc_map(struct gntdev_priv *priv, int count)
>  	add->grants    = kzalloc(sizeof(add->grants[0])    * count, GFP_KERNEL);
>  	add->map_ops   = kzalloc(sizeof(add->map_ops[0])   * count, GFP_KERNEL);
>  	add->unmap_ops = kzalloc(sizeof(add->unmap_ops[0]) * count, GFP_KERNEL);
> +	add->kmap_ops  = kzalloc(sizeof(add->kmap_ops[0])  * count, GFP_KERNEL);
>  	add->pages     = kzalloc(sizeof(add->pages[0])     * count, GFP_KERNEL);
>  	if (NULL == add->grants    ||
>  	    NULL == add->map_ops   ||
>  	    NULL == add->unmap_ops ||
> +	    NULL == add->kmap_ops  ||
>  	    NULL == add->pages)
>  		goto err;
>  
> @@ -129,6 +132,7 @@ static struct grant_map *gntdev_alloc_map(struct gntdev_priv *priv, int count)
>  	for (i = 0; i < count; i++) {
>  		add->map_ops[i].handle = -1;
>  		add->unmap_ops[i].handle = -1;
> +		add->kmap_ops[i].handle = -1;
>  	}
>  
>  	add->index = 0;
> @@ -142,6 +146,7 @@ err:
>  	kfree(add->grants);
>  	kfree(add->map_ops);
>  	kfree(add->unmap_ops);
> +	kfree(add->kmap_ops);
>  	kfree(add);
>  	return NULL;
>  }
> @@ -243,10 +248,30 @@ static int map_grant_pages(struct grant_map *map)
>  			gnttab_set_unmap_op(&map->unmap_ops[i], addr,
>  				map->flags, -1 /* handle */);
>  		}
> +	} else {
> +		for (i = 0; i < map->count; i++) {
> +			unsigned level;
> +			unsigned long address = (unsigned long)
> +				pfn_to_kaddr(page_to_pfn(map->pages[i]));
> +			pte_t *ptep;
> +			u64 pte_maddr = 0;
> +			if (!PageHighMem(map->pages[i])) {
> +				ptep = lookup_address(address, &level);
> +				pte_maddr =
> +					arbitrary_virt_to_machine(ptep).maddr;
> +			}
> +			gnttab_set_map_op(&map->kmap_ops[i], pte_maddr,
> +				map->flags |
> +				GNTMAP_host_map |
> +				GNTMAP_contains_pte,
> +				map->grants[i].ref,
> +				map->grants[i].domid);
> +		}
>  	}
>  
>  	pr_debug("map %d+%d\n", map->index, map->count);
> -	err = gnttab_map_refs(map->map_ops, map->pages, map->count);
> +	err = gnttab_map_refs(map->map_ops, use_ptemod ? map->kmap_ops : NULL,
> +			map->pages, map->count);
>  	if (err)
>  		return err;
>  
> diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
> index fd725cd..1e8669a 100644
> --- a/drivers/xen/grant-table.c
> +++ b/drivers/xen/grant-table.c
> @@ -448,6 +448,7 @@ unsigned int gnttab_max_grant_frames(void)
>  EXPORT_SYMBOL_GPL(gnttab_max_grant_frames);
>  
>  int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
> +			struct gnttab_map_grant_ref *kmap_ops,
>  		    struct page **pages, unsigned int count)
>  {
>  	int i, ret;
> @@ -488,8 +489,7 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
>  			 */
>  			return -EOPNOTSUPP;
>  		}
> -		ret = m2p_add_override(mfn, pages[i],
> -				       map_ops[i].flags & GNTMAP_contains_pte);
> +		ret = m2p_add_override(mfn, pages[i], &kmap_ops[i]);
>  		if (ret)
>  			return ret;
>  	}
> diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h
> index b1fab6b..6b99bfb 100644
> --- a/include/xen/grant_table.h
> +++ b/include/xen/grant_table.h
> @@ -156,6 +156,7 @@ unsigned int gnttab_max_grant_frames(void);
>  #define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr))
>  
>  int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
> +			struct gnttab_map_grant_ref *kmap_ops,
>  		    struct page **pages, unsigned int count);
>  int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
>  		      struct page **pages, unsigned int count);


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists