by Jan Kara

[permalink] [raw]

Subject: Re: [PATCH v6 09/13] filesystem-dax: Introduce dax_lock_mapping_entry()

On Fri 13-07-18 21:50:16, Dan Williams wrote:
> In preparation for implementing support for memory poison (media error)
> handling via dax mappings, implement a lock_page() equivalent. Poison
> error handling requires rmap and needs guarantees that the page->mapping
> association is maintained / valid (inode not freed) for the duration of
> the lookup.
>
> In the device-dax case it is sufficient to simply hold a dev_pagemap
> reference. In the filesystem-dax case we need to use the entry lock.
>
> Export the entry lock via dax_lock_mapping_entry() that uses
> rcu_read_lock() to protect against the inode being freed, and
> revalidates the page->mapping association under xa_lock().
>
> Cc: Christoph Hellwig <[email protected]>
> Cc: Matthew Wilcox <[email protected]>
> Cc: Ross Zwisler <[email protected]>
> Cc: Jan Kara <[email protected]>
> Signed-off-by: Dan Williams <[email protected]>

Just got back from vacation... This patch looks good to me. You can add:

Reviewed-by: Jan Kara <[email protected]>

Honza

> ---
> fs/dax.c | 109 ++++++++++++++++++++++++++++++++++++++++++++++++---
> include/linux/dax.h | 13 ++++++
> 2 files changed, 116 insertions(+), 6 deletions(-)
>
> diff --git a/fs/dax.c b/fs/dax.c
> index 4de11ed463ce..57ec272038da 100644
> --- a/fs/dax.c
> +++ b/fs/dax.c
> @@ -226,8 +226,8 @@ static inline void *unlock_slot(struct address_space *mapping, void **slot)
> *
> * Must be called with the i_pages lock held.
> */
> -static void *get_unlocked_mapping_entry(struct address_space *mapping,
> - pgoff_t index, void ***slotp)
> +static void *__get_unlocked_mapping_entry(struct address_space *mapping,
> + pgoff_t index, void ***slotp, bool (*wait_fn)(void))
> {
> void *entry, **slot;
> struct wait_exceptional_entry_queue ewait;
> @@ -237,6 +237,8 @@ static void *get_unlocked_mapping_entry(struct address_space *mapping,
> ewait.wait.func = wake_exceptional_entry_func;
>
> for (;;) {
> + bool revalidate;
> +
> entry = __radix_tree_lookup(&mapping->i_pages, index, NULL,
> &slot);
> if (!entry ||
> @@ -251,14 +253,31 @@ static void *get_unlocked_mapping_entry(struct address_space *mapping,
> prepare_to_wait_exclusive(wq, &ewait.wait,
> TASK_UNINTERRUPTIBLE);
> xa_unlock_irq(&mapping->i_pages);
> - schedule();
> + revalidate = wait_fn();
> finish_wait(wq, &ewait.wait);
> xa_lock_irq(&mapping->i_pages);
> + if (revalidate)
> + return ERR_PTR(-EAGAIN);
> }
> }
>
> -static void dax_unlock_mapping_entry(struct address_space *mapping,
> - pgoff_t index)
> +static bool entry_wait(void)
> +{
> + schedule();
> + /*
> + * Never return an ERR_PTR() from
> + * __get_unlocked_mapping_entry(), just keep looping.
> + */
> + return false;
> +}
> +
> +static void *get_unlocked_mapping_entry(struct address_space *mapping,
> + pgoff_t index, void ***slotp)
> +{
> + return __get_unlocked_mapping_entry(mapping, index, slotp, entry_wait);
> +}
> +
> +static void unlock_mapping_entry(struct address_space *mapping, pgoff_t index)
> {
> void *entry, **slot;
>
> @@ -277,7 +296,7 @@ static void dax_unlock_mapping_entry(struct address_space *mapping,
> static void put_locked_mapping_entry(struct address_space *mapping,
> pgoff_t index)
> {
> - dax_unlock_mapping_entry(mapping, index);
> + unlock_mapping_entry(mapping, index);
> }
>
> /*
> @@ -374,6 +393,84 @@ static struct page *dax_busy_page(void *entry)
> return NULL;
> }
>
> +static bool entry_wait_revalidate(void)
> +{
> + rcu_read_unlock();
> + schedule();
> + rcu_read_lock();
> +
> + /*
> + * Tell __get_unlocked_mapping_entry() to take a break, we need
> + * to revalidate page->mapping after dropping locks
> + */
> + return true;
> +}
> +
> +bool dax_lock_mapping_entry(struct page *page)
> +{
> + pgoff_t index;
> + struct inode *inode;
> + bool did_lock = false;
> + void *entry = NULL, **slot;
> + struct address_space *mapping;
> +
> + rcu_read_lock();
> + for (;;) {
> + mapping = READ_ONCE(page->mapping);
> +
> + if (!dax_mapping(mapping))
> + break;
> +
> + /*
> + * In the device-dax case there's no need to lock, a
> + * struct dev_pagemap pin is sufficient to keep the
> + * inode alive, and we assume we have dev_pagemap pin
> + * otherwise we would not have a valid pfn_to_page()
> + * translation.
> + */
> + inode = mapping->host;
> + if (S_ISCHR(inode->i_mode)) {
> + did_lock = true;
> + break;
> + }
> +
> + xa_lock_irq(&mapping->i_pages);
> + if (mapping != page->mapping) {
> + xa_unlock_irq(&mapping->i_pages);
> + continue;
> + }
> + index = page->index;
> +
> + entry = __get_unlocked_mapping_entry(mapping, index, &slot,
> + entry_wait_revalidate);
> + if (!entry) {
> + xa_unlock_irq(&mapping->i_pages);
> + break;
> + } else if (IS_ERR(entry)) {
> + WARN_ON_ONCE(PTR_ERR(entry) != -EAGAIN);
> + continue;
> + }
> + lock_slot(mapping, slot);
> + did_lock = true;
> + xa_unlock_irq(&mapping->i_pages);
> + break;
> + }
> + rcu_read_unlock();
> +
> + return did_lock;
> +}
> +
> +void dax_unlock_mapping_entry(struct page *page)
> +{
> + struct address_space *mapping = page->mapping;
> + struct inode *inode = mapping->host;
> +
> + if (S_ISCHR(inode->i_mode))
> + return;
> +
> + unlock_mapping_entry(mapping, page->index);
> +}
> +
> /*
> * Find radix tree entry at given index. If it points to an exceptional entry,
> * return it with the radix tree entry locked. If the radix tree doesn't
> diff --git a/include/linux/dax.h b/include/linux/dax.h
> index 3855e3800f48..cf8ac51cf0d7 100644
> --- a/include/linux/dax.h
> +++ b/include/linux/dax.h
> @@ -88,6 +88,8 @@ int dax_writeback_mapping_range(struct address_space *mapping,
> struct block_device *bdev, struct writeback_control *wbc);
>
> struct page *dax_layout_busy_page(struct address_space *mapping);
> +bool dax_lock_mapping_entry(struct page *page);
> +void dax_unlock_mapping_entry(struct page *page);
> #else
> static inline bool bdev_dax_supported(struct block_device *bdev,
> int blocksize)
> @@ -119,6 +121,17 @@ static inline int dax_writeback_mapping_range(struct address_space *mapping,
> {
> return -EOPNOTSUPP;
> }
> +
> +static inline bool dax_lock_mapping_entry(struct page *page)
> +{
> + if (IS_DAX(page->mapping->host))
> + return true;
> + return false;
> +}
> +
> +static inline void dax_unlock_mapping_entry(struct page *page)
> +{
> +}
> #endif
>
> int dax_read_lock(void);
>
--
Jan Kara <[email protected]>
SUSE Labs, CR

2018-08-06 09:29:02

by Jan Kara

[permalink] [raw]

Subject: Re: [PATCH v6 10/13] mm, memory_failure: Teach memory_failure() about dev_pagemap pages

On Fri 13-07-18 21:50:21, Dan Williams wrote:
> mce: Uncorrected hardware memory error in user-access at af34214200
> {1}[Hardware Error]: It has been corrected by h/w and requires no further action
> mce: [Hardware Error]: Machine check events logged
> {1}[Hardware Error]: event severity: corrected
> Memory failure: 0xaf34214: reserved kernel page still referenced by 1 users
> [..]
> Memory failure: 0xaf34214: recovery action for reserved kernel page: Failed
> mce: Memory error not recovered
>
> In contrast to typical memory, dev_pagemap pages may be dax mapped. With
> dax there is no possibility to map in another page dynamically since dax
> establishes 1:1 physical address to file offset associations. Also
> dev_pagemap pages associated with NVDIMM / persistent memory devices can
> internal remap/repair addresses with poison. While memory_failure()
> assumes that it can discard typical poisoned pages and keep them
> unmapped indefinitely, dev_pagemap pages may be returned to service
> after the error is cleared.
>
> Teach memory_failure() to detect and handle MEMORY_DEVICE_HOST
> dev_pagemap pages that have poison consumed by userspace. Mark the
> memory as UC instead of unmapping it completely to allow ongoing access
> via the device driver (nd_pmem). Later, nd_pmem will grow support for
> marking the page back to WB when the error is cleared.
>
> Cc: Jan Kara <[email protected]>
> Cc: Christoph Hellwig <[email protected]>
> Cc: J?r?me Glisse <[email protected]>
> Cc: Matthew Wilcox <[email protected]>
> Cc: Naoya Horiguchi <[email protected]>
> Cc: Ross Zwisler <[email protected]>
> Signed-off-by: Dan Williams <[email protected]>

I'm not very familiar with memory failure code but at least from DAX POV
and from my rudimentary understanding of memory-failure the patch looks
sane to me.

Honza

> ---
> include/linux/mm.h | 1
> mm/memory-failure.c | 125 ++++++++++++++++++++++++++++++++++++++++++++++++++-
> 2 files changed, 124 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index a0fbb9ffe380..374e5e9284f7 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -2725,6 +2725,7 @@ enum mf_action_page_type {
> MF_MSG_TRUNCATED_LRU,
> MF_MSG_BUDDY,
> MF_MSG_BUDDY_2ND,
> + MF_MSG_DAX,
> MF_MSG_UNKNOWN,
> };
>
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index 8a81680d00dd..effaa7c7a1a4 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -55,6 +55,7 @@
> #include <linux/hugetlb.h>
> #include <linux/memory_hotplug.h>
> #include <linux/mm_inline.h>
> +#include <linux/memremap.h>
> #include <linux/kfifo.h>
> #include <linux/ratelimit.h>
> #include "internal.h"
> @@ -263,6 +264,40 @@ void shake_page(struct page *p, int access)
> }
> EXPORT_SYMBOL_GPL(shake_page);
>
> +static unsigned long dev_pagemap_mapping_size(struct page *page,
> + struct vm_area_struct *vma)
> +{
> + unsigned long address = vma_address(page, vma);
> + pgd_t *pgd;
> + p4d_t *p4d;
> + pud_t *pud;
> + pmd_t *pmd;
> + pte_t *pte;
> +
> + pgd = pgd_offset(vma->vm_mm, address);
> + if (!pgd_present(*pgd))
> + return 0;
> + p4d = p4d_offset(pgd, address);
> + if (!p4d_present(*p4d))
> + return 0;
> + pud = pud_offset(p4d, address);
> + if (!pud_present(*pud))
> + return 0;
> + if (pud_devmap(*pud))
> + return PUD_SIZE;
> + pmd = pmd_offset(pud, address);
> + if (!pmd_present(*pmd))
> + return 0;
> + if (pmd_devmap(*pmd))
> + return PMD_SIZE;
> + pte = pte_offset_map(pmd, address);
> + if (!pte_present(*pte))
> + return 0;
> + if (pte_devmap(*pte))
> + return PAGE_SIZE;
> + return 0;
> +}
> +
> /*
> * Failure handling: if we can't find or can't kill a process there's
> * not much we can do. We just print a message and ignore otherwise.
> @@ -292,7 +327,10 @@ static void add_to_kill(struct task_struct *tsk, struct page *p,
> }
> tk->addr = page_address_in_vma(p, vma);
> tk->addr_valid = 1;
> - tk->size_shift = compound_order(compound_head(p)) + PAGE_SHIFT;
> + if (is_zone_device_page(p))
> + tk->size_shift = ilog2(dev_pagemap_mapping_size(p, vma));
> + else
> + tk->size_shift = compound_order(compound_head(p)) + PAGE_SHIFT;
>
> /*
> * In theory we don't have to kill when the page was
> @@ -300,7 +338,7 @@ static void add_to_kill(struct task_struct *tsk, struct page *p,
> * likely very rare kill anyways just out of paranoia, but use
> * a SIGKILL because the error is not contained anymore.
> */
> - if (tk->addr == -EFAULT) {
> + if (tk->addr == -EFAULT || tk->size_shift == 0) {
> pr_info("Memory failure: Unable to find user space address %lx in %s\n",
> page_to_pfn(p), tsk->comm);
> tk->addr_valid = 0;
> @@ -514,6 +552,7 @@ static const char * const action_page_types[] = {
> [MF_MSG_TRUNCATED_LRU] = "already truncated LRU page",
> [MF_MSG_BUDDY] = "free buddy page",
> [MF_MSG_BUDDY_2ND] = "free buddy page (2nd try)",
> + [MF_MSG_DAX] = "dax page",
> [MF_MSG_UNKNOWN] = "unknown page",
> };
>
> @@ -1111,6 +1150,83 @@ static int memory_failure_hugetlb(unsigned long pfn, int flags)
> return res;
> }
>
> +static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
> + struct dev_pagemap *pgmap)
> +{
> + struct page *page = pfn_to_page(pfn);
> + const bool unmap_success = true;
> + unsigned long size = 0;
> + struct to_kill *tk;
> + LIST_HEAD(tokill);
> + int rc = -EBUSY;
> + loff_t start;
> +
> + /*
> + * Prevent the inode from being freed while we are interrogating
> + * the address_space, typically this would be handled by
> + * lock_page(), but dax pages do not use the page lock. This
> + * also prevents changes to the mapping of this pfn until
> + * poison signaling is complete.
> + */
> + if (!dax_lock_mapping_entry(page))
> + goto out;
> +
> + if (hwpoison_filter(page)) {
> + rc = 0;
> + goto unlock;
> + }
> +
> + switch (pgmap->type) {
> + case MEMORY_DEVICE_PRIVATE:
> + case MEMORY_DEVICE_PUBLIC:
> + /*
> + * TODO: Handle HMM pages which may need coordination
> + * with device-side memory.
> + */
> + goto unlock;
> + default:
> + break;
> + }
> +
> + /*
> + * Use this flag as an indication that the dax page has been
> + * remapped UC to prevent speculative consumption of poison.
> + */
> + SetPageHWPoison(page);
> +
> + /*
> + * Unlike System-RAM there is no possibility to swap in a
> + * different physical page at a given virtual address, so all
> + * userspace consumption of ZONE_DEVICE memory necessitates
> + * SIGBUS (i.e. MF_MUST_KILL)
> + */
> + flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> + collect_procs(page, &tokill, flags & MF_ACTION_REQUIRED);
> +
> + list_for_each_entry(tk, &tokill, nd)
> + if (tk->size_shift)
> + size = max(size, 1UL << tk->size_shift);
> + if (size) {
> + /*
> + * Unmap the largest mapping to avoid breaking up
> + * device-dax mappings which are constant size. The
> + * actual size of the mapping being torn down is
> + * communicated in siginfo, see kill_proc()
> + */
> + start = (page->index << PAGE_SHIFT) & ~(size - 1);
> + unmap_mapping_range(page->mapping, start, start + size, 0);
> + }
> + kill_procs(&tokill, flags & MF_MUST_KILL, !unmap_success, pfn, flags);
> + rc = 0;
> +unlock:
> + dax_unlock_mapping_entry(page);
> +out:
> + /* drop pgmap ref acquired in caller */
> + put_dev_pagemap(pgmap);
> + action_result(pfn, MF_MSG_DAX, rc ? MF_FAILED : MF_RECOVERED);
> + return rc;
> +}
> +
> /**
> * memory_failure - Handle memory failure of a page.
> * @pfn: Page Number of the corrupted page
> @@ -1133,6 +1249,7 @@ int memory_failure(unsigned long pfn, int flags)
> struct page *p;
> struct page *hpage;
> struct page *orig_head;
> + struct dev_pagemap *pgmap;
> int res;
> unsigned long page_flags;
>
> @@ -1145,6 +1262,10 @@ int memory_failure(unsigned long pfn, int flags)
> return -ENXIO;
> }
>
> + pgmap = get_dev_pagemap(pfn, NULL);
> + if (pgmap)
> + return memory_failure_dev_pagemap(pfn, flags, pgmap);
> +
> p = pfn_to_page(pfn);
> if (PageHuge(p))
> return memory_failure_hugetlb(pfn, flags);
>
--
Jan Kara <[email protected]>
SUSE Labs, CR