9:55 p.m.
This patchset is aimed to support shared pages tracking for fsdax.
Change from V2:
- Split 8th patch into other related to make it easy to review
- Other small fixes
Change from V1:
- Add the old memory-failure handler back for rolling back
- Add callback in MD's ->rmap() to support multiple mapping of dm device
- Add judgement for CONFIG_SYSFS
- Add pfn_valid() judgement in hwpoison_filter()
- Rebased to v5.11-rc5
This patchset moves owner tracking from dax_assocaite_entry() to pmem
device driver, by introducing an interface ->memory_failure() of struct
pagemap. This interface is called by memory_failure() in mm, and
implemented by pmem device. Then pmem device calls its ->corrupted_range()
to find the filesystem which the corrupted data located in, and call
filesystem handler to track files or metadata assocaited with this page.
Finally we are able to try to fix the corrupted data in filesystem and do
other necessary processing, such as killing processes who are using the
files affected.
The call trace is like this:
memory_failure()
pgmap->ops->memory_failure() => pmem_pgmap_memory_failure()
gendisk->fops->corrupted_range() => - pmem_corrupted_range()
- md_blk_corrupted_range()
sb->s_ops->currupted_range() => xfs_fs_corrupted_range()
xfs_rmap_query_range()
xfs_currupt_helper()
* corrupted on metadata
try to recover data, call xfs_force_shutdown()
* corrupted on file data
try to recover data, call mf_dax_mapping_kill_procs()
The fsdax & reflink support for XFS is not contained in this patchset.
(Rebased on v5.11-rc5)
==
Shiyang Ruan (11):
pagemap: Introduce ->memory_failure()
blk: Introduce ->corrupted_range() for block device
fs: Introduce ->corrupted_range() for superblock
block_dev: Introduce bd_corrupted_range()
mm, fsdax: Refactor memory-failure handler for dax mapping
mm, pmem: Implement ->memory_failure() in pmem driver
pmem: Implement ->corrupted_range() for pmem driver
dm: Introduce ->rmap() to find bdev offset
md: Implement ->corrupted_range()
xfs: Implement ->corrupted_range() for XFS
fs/dax: Remove useless functions
block/genhd.c | 6 ++
drivers/md/dm-linear.c | 20 ++++
drivers/md/dm.c | 61 +++++++++++
drivers/nvdimm/pmem.c | 45 ++++++++
fs/block_dev.c | 47 ++++++++-
fs/dax.c | 63 ++++-------
fs/xfs/xfs_fsops.c | 5 +
fs/xfs/xfs_mount.h | 1 +
fs/xfs/xfs_super.c | 112 ++++++++++++++++++++
include/linux/blkdev.h | 2 +
include/linux/dax.h | 1 +
include/linux/device-mapper.h | 5 +
include/linux/fs.h | 2 +
include/linux/genhd.h | 3 +
include/linux/memremap.h | 8 ++
include/linux/mm.h | 9 ++
mm/memory-failure.c | 190 +++++++++++++++++++++++-----------
17 files changed, 475 insertions(+), 105 deletions(-)
--
2.30.0
9:55 p.m.
New subject: [PATCH v3 01/11] pagemap: Introduce ->memory_failure()
When memory-failure occurs, we call this function which is implemented
by each kind of devices. For the fsdax case, pmem device driver
implements it. Pmem device driver will find out the block device where
the error page locates in, and try to get the filesystem on this block
device. And finally call filesystem handler to deal with the error.
The filesystem will try to recover the corrupted data if possiable.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
include/linux/memremap.h | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/include/linux/memremap.h b/include/linux/memremap.h
index 79c49e7f5c30..0bcf2b1e20bd 100644
--- a/include/linux/memremap.h
+++ b/include/linux/memremap.h
@@ -87,6 +87,14 @@ struct dev_pagemap_ops {
* the page back to a CPU accessible page.
*/
vm_fault_t (*migrate_to_ram)(struct vm_fault *vmf);
+
+ /*
+ * Handle the memory failure happens on one page. Notify the processes
+ * who are using this page, and try to recover the data on this page
+ * if necessary.
+ */
+ int (*memory_failure)(struct dev_pagemap *pgmap, unsigned long pfn,
+ int flags);
};
#define PGMAP_ALTMAP_VALID (1 << 0)
--
2.30.0
12:20 a.m.
New subject: [PATCH v3 01/11] pagemap: Introduce ->memory_failure()
On Mon, Feb 08, 2021 at 06:55:20PM +0800, Shiyang Ruan wrote:
When memory-failure occurs, we call this function which is
implemented
by each kind of devices. For the fsdax case, pmem device driver
implements it. Pmem device driver will find out the block device where
the error page locates in, and try to get the filesystem on this block
device. And finally call filesystem handler to deal with the error.
The filesystem will try to recover the corrupted data if possiable.
I'm not sure adding just a method without any of the support code
is a useful patch on its own.
7:36 a.m.
New subject: [PATCH v3 01/11] pagemap: Introduce ->memory_failure()
On Mon, Feb 8, 2021 at 2:55 AM Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com> wrote:
When memory-failure occurs, we call this function which is implemented
by each kind of devices. For the fsdax case, pmem device driver
implements it. Pmem device driver will find out the block device where
the error page locates in, and try to get the filesystem on this block
device. And finally call filesystem handler to deal with the error.
The filesystem will try to recover the corrupted data if possiable.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
include/linux/memremap.h | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/include/linux/memremap.h b/include/linux/memremap.h
index 79c49e7f5c30..0bcf2b1e20bd 100644
--- a/include/linux/memremap.h
+++ b/include/linux/memremap.h
@@ -87,6 +87,14 @@ struct dev_pagemap_ops {
* the page back to a CPU accessible page.
*/
vm_fault_t (*migrate_to_ram)(struct vm_fault *vmf);
+
+ /*
+ * Handle the memory failure happens on one page. Notify the processes
+ * who are using this page, and try to recover the data on this page
+ * if necessary.
+ */
+ int (*memory_failure)(struct dev_pagemap *pgmap, unsigned long pfn,
+ int flags);
};
After the conversation with Dave I don't see the point of this. If
there is a memory_failure() on a page, why not just call
memory_failure()? That already knows how to find the inode and the
filesystem can be notified from there.
Although memory_failure() is inefficient for large range failures, I'm
not seeing a better option, so I'm going to test calling
memory_failure() over a large range whenever an in-use dax-device is
hot-removed.
2:38 p.m.
New subject: [PATCH v3 01/11] pagemap: Introduce ->memory_failure()
On Mon, Feb 8, 2021 at 2:55 AM Shiyang Ruan
<ruansy.fnst(a)cn.fujitsu.com> wrote:
>
> When memory-failure occurs, we call this function which is implemented
> by each kind of devices. For the fsdax case, pmem device driver
> implements it. Pmem device driver will find out the block device where
> the error page locates in, and try to get the filesystem on this block
> device. And finally call filesystem handler to deal with the error.
> The filesystem will try to recover the corrupted data if possiable.
>
> Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
> ---
> include/linux/memremap.h | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> index 79c49e7f5c30..0bcf2b1e20bd 100644
> --- a/include/linux/memremap.h
> +++ b/include/linux/memremap.h
> @@ -87,6 +87,14 @@ struct dev_pagemap_ops {
> * the page back to a CPU accessible page.
> */
> vm_fault_t (*migrate_to_ram)(struct vm_fault *vmf);
> +
> + /*
> + * Handle the memory failure happens on one page. Notify the processes
> + * who are using this page, and try to recover the data on this page
> + * if necessary.
> + */
> + int (*memory_failure)(struct dev_pagemap *pgmap, unsigned long pfn,
> + int flags);
> };
After the conversation with Dave I don't see the point of this. If
there is a memory_failure() on a page, why not just call
memory_failure()? That already knows how to find the inode and the
filesystem can be notified from there.
We want memory_failure() supports reflinked files. In this case, we are not
able to track multiple files from a page(this broken page) because
page->mapping,page->index can only track one file. Thus, I introduce this
->memory_failure() implemented in pmem driver, to call ->corrupted_range()
upper level to upper level, and finally find out files who are
using(mmapping) this page.
Although memory_failure() is inefficient for large range failures, I'm
not seeing a better option, so I'm going to test calling
memory_failure() over a large range whenever an in-use dax-device is
hot-removed.
I did not test this for large range failure yet... I am not sure if it works
fine. But because of the complex tracking method, I think it would be more
inefficient in this case than before.
--
Thanks,
Ruan Shiyang.
9:55 p.m.
New subject: [PATCH v3 02/11] blk: Introduce ->corrupted_range() for block device
In fsdax mode, the memory failure happens on block device. So, it is
needed to introduce an interface for block devices. Each kind of block
device can handle the memory failure in ther own ways.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
include/linux/blkdev.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index f94ee3089e01..e0f5585aa06f 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1867,6 +1867,8 @@ struct block_device_operations {
int (*report_zones)(struct gendisk *, sector_t sector,
unsigned int nr_zones, report_zones_cb cb, void *data);
char *(*devnode)(struct gendisk *disk, umode_t *mode);
+ int (*corrupted_range)(struct gendisk *disk, struct block_device *bdev,
+ loff_t offset, size_t len, void *data);
struct module *owner;
const struct pr_ops *pr_ops;
};
--
2.30.0
12:21 a.m.
New subject: [PATCH v3 02/11] blk: Introduce ->corrupted_range() for block
device
On Mon, Feb 08, 2021 at 06:55:21PM +0800, Shiyang Ruan wrote:
In fsdax mode, the memory failure happens on block device. So, it
is
needed to introduce an interface for block devices. Each kind of block
device can handle the memory failure in ther own ways.
As told before: DAX operations please do not add anything to the block
device. We've been working very hard to decouple DAX from the block
device, and while we're not done regressing the split should not happen.
9:42 a.m.
New subject: [PATCH v3 02/11] blk: Introduce ->corrupted_range() for block
device
On Wed, Feb 10, 2021 at 02:21:39PM +0100, Christoph Hellwig wrote:
On Mon, Feb 08, 2021 at 06:55:21PM +0800, Shiyang Ruan wrote:
> In fsdax mode, the memory failure happens on block device. So, it is
> needed to introduce an interface for block devices. Each kind of block
> device can handle the memory failure in ther own ways.
As told before: DAX operations please do not add anything to the block
device. We've been working very hard to decouple DAX from the block
device, and while we're not done regressing the split should not happen.
I agree with you (Christoph) that (strictly speaking) within the scope of
the DAX work this isn't needed; xfs should be able to consume the
->memory_failure events directly and DTRT.
My vision here, however, is to establish upcalls for /both/ types of
stroage.
Regular block devices can use ->corrupted_range to push error
notifications upwards through the block stack to a filesystem, and we
can finally do a teensy bit more with scsi sense data about media
errors, or thinp wanting to warn the filesystem that it's getting low on
space and maybe this would be an agreeable time to self-FITRIM, or raid
noticing that a mirror is inconsistent and can the fs do something to
resolve the dispute, etc. Maybe we can use this mechanism to warn a
filesystem that someone did "echo 1 > /sys/block/sda/device/delete" and
we had better persist everything while we still can.
Memory devices will use ->memory_failure to tell us about ADR errors,
and I guess upcoming and past hotremove events. For fsdax you'd
probably have to send the announcement and invalidate the current ptes
to force filesystem pagefaults and the like.
Either way, I think this piece is fine, but I would change the dax
side to send the ->memory_failure events directly to xfs.
A gap here is that xfs can attach to rt/log devices but we don't
currently plumb in enough information that get_active_super can find
the correct filesystem.
I dunno, maybe we should add this to the thread here[1]?
[1]
https://lore.kernel.org/linux-xfs/CAPcyv4g3ZwbdLFx8bqMcNvXyrob8y6sBXXu=xP...
--D
5:10 p.m.
New subject: [PATCH v3 02/11] blk: Introduce ->corrupted_range() for block
device
On Thu, Mar 04, 2021 at 02:42:50PM -0800, Darrick J. Wong wrote:
My vision here, however, is to establish upcalls for /both/ types of
stroage.
I already have patches for doing these kinds of callbacks properly
for the block layer. They will be posted shortly.
9:55 p.m.
New subject: [PATCH v3 03/11] fs: Introduce ->corrupted_range() for superblock
Memory failure occurs in fsdax mode will finally be handled in
filesystem. We introduce this interface to find out files or metadata
affected by the corrupted range, and try to recover the corrupted data
if possiable.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
include/linux/fs.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index fd47deea7c17..4cc9ff9caa87 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1963,6 +1963,8 @@ struct super_operations {
struct shrink_control *);
long (*free_cached_objects)(struct super_block *,
struct shrink_control *);
+ int (*corrupted_range)(struct super_block *sb, struct block_device *bdev,
+ loff_t offset, size_t len, void *data);
};
/*
--
2.30.0
9:55 p.m.
New subject: [PATCH v3 04/11] block_dev: Introduce bd_corrupted_range() for block device
As is show in the call tree:
...
pgmap->ops->memory_failure()
gendisk->fops->corrupted_range()
sb->s_ops->currupted_range()
...
currupted_range() is called from disk(pmem) to superblock(filesystem).
Thus, we need to introduce a helper function to obtain the superblock
from a given disk.
Normally, a filesystem can be created by mkfs directly on a block device
or a parted disk. We can obtain the superblock by calling get_super().
But get_super() is not suitable for mapped device, such as LVM. To
obtain MD's superblock, we can iterate bd_holder_disks of a block device
to find out the MD that contains it. By this way, MD is able to call
corrupted_range() from its target block device.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
fs/block_dev.c | 47 ++++++++++++++++++++++++++++++++++++++++++-
include/linux/genhd.h | 2 ++
2 files changed, 48 insertions(+), 1 deletion(-)
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 3b8963e228a1..755b0d351479 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1079,6 +1079,28 @@ struct bd_holder_disk {
int refcnt;
};
+static int bd_disk_holder_corrupted_range(struct block_device *bdev, loff_t off,
+ size_t len, void *data)
+{
+ struct bd_holder_disk *holder;
+ struct gendisk *disk;
+ int rc = 0;
+
+ if (list_empty(&(bdev->bd_holder_disks)))
+ return -ENODEV;
+
+ list_for_each_entry(holder, &bdev->bd_holder_disks, list) {
+ disk = holder->disk;
+ if (disk->fops->corrupted_range) {
+ rc = disk->fops->corrupted_range(disk, bdev, off,
+ len, data);
+ if (rc != -ENODEV)
+ break;
+ }
+ }
+ return rc;
+}
+
static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev,
struct gendisk *disk)
{
@@ -1212,7 +1234,30 @@ void bd_unlink_disk_holder(struct block_device *bdev, struct
gendisk *disk)
mutex_unlock(&bdev->bd_mutex);
}
EXPORT_SYMBOL_GPL(bd_unlink_disk_holder);
-#endif
+#else
+static inline int bd_disk_holder_corrupted_range(struct block_device *bdev,
+ loff_t off, size_t len, void *data)
+{
+ return -EOPNOTSUPP;
+}
+#endif /* CONFIG_SYSFS */
+
+int bd_corrupted_range(struct block_device *bdev, loff_t disk_off,
+ loff_t bdev_off, size_t len, void *data)
+{
+ struct super_block *sb = get_super(bdev);
+ int rc = -EOPNOTSUPP;
+
+ if (!sb) {
+ rc = bd_disk_holder_corrupted_range(bdev, disk_off, len, data);
+ return rc;
+ } else if (sb->s_op->corrupted_range)
+ rc = sb->s_op->corrupted_range(sb, bdev, bdev_off, len, data);
+ drop_super(sb);
+
+ return rc;
+}
+EXPORT_SYMBOL(bd_corrupted_range);
static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part);
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 809aaa32d53c..751cbd559bba 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -314,6 +314,8 @@ void unregister_blkdev(unsigned int major, const char *name);
bool bdev_check_media_change(struct block_device *bdev);
int __invalidate_device(struct block_device *bdev, bool kill_dirty);
void set_capacity(struct gendisk *disk, sector_t size);
+int bd_corrupted_range(struct block_device *bdev, loff_t disk_off,
+ loff_t bdev_off, size_t len, void *data);
/* for drivers/char/raw.c: */
int blkdev_ioctl(struct block_device *, fmode_t, unsigned, unsigned long);
--
2.30.0
9:55 p.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler for dax mapping
The current memory_failure_dev_pagemap() can only handle single-mapped
dax page for fsdax mode. The dax page could be mapped by multiple files
and offsets if we let reflink feature & fsdax mode work together. So,
we refactor current implementation to support handle memory failure on
each file and offset.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
fs/dax.c | 21 ++++++++++
include/linux/dax.h | 1 +
include/linux/mm.h | 9 +++++
mm/memory-failure.c | 98 ++++++++++++++++++++++++++++++++++-----------
4 files changed, 105 insertions(+), 24 deletions(-)
diff --git a/fs/dax.c b/fs/dax.c
index 26d5dcd2d69e..c64c3a0e76a6 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -378,6 +378,27 @@ static struct page *dax_busy_page(void *entry)
return NULL;
}
+/*
+ * dax_load_pfn - Load pfn of the DAX entry corresponding to a page
+ * @mapping: The file whose entry we want to load
+ * @index: The offset where the DAX entry located in
+ *
+ * Return: pfn of the DAX entry
+ */
+unsigned long dax_load_pfn(struct address_space *mapping, unsigned long index)
+{
+ XA_STATE(xas, &mapping->i_pages, index);
+ void *entry;
+ unsigned long pfn;
+
+ xas_lock_irq(&xas);
+ entry = xas_load(&xas);
+ pfn = dax_to_pfn(entry);
+ xas_unlock_irq(&xas);
+
+ return pfn;
+}
+
/*
* dax_lock_mapping_entry - Lock the DAX entry corresponding to a page
* @page: The page whose entry we want to lock
diff --git a/include/linux/dax.h b/include/linux/dax.h
index b52f084aa643..89e56ceeffc7 100644
--- a/include/linux/dax.h
+++ b/include/linux/dax.h
@@ -150,6 +150,7 @@ int dax_writeback_mapping_range(struct address_space *mapping,
struct page *dax_layout_busy_page(struct address_space *mapping);
struct page *dax_layout_busy_page_range(struct address_space *mapping, loff_t start,
loff_t end);
+unsigned long dax_load_pfn(struct address_space *mapping, unsigned long index);
dax_entry_t dax_lock_page(struct page *page);
void dax_unlock_page(struct page *page, dax_entry_t cookie);
#else
diff --git a/include/linux/mm.h b/include/linux/mm.h
index ecdf8a8cd6ae..ab52bc633d84 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1157,6 +1157,14 @@ static inline bool is_device_private_page(const struct page *page)
page->pgmap->type == MEMORY_DEVICE_PRIVATE;
}
+static inline bool is_device_fsdax_page(const struct page *page)
+{
+ return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
+ IS_ENABLED(CONFIG_FS_DAX) &&
+ is_zone_device_page(page) &&
+ page->pgmap->type == MEMORY_DEVICE_FS_DAX;
+}
+
static inline bool is_pci_p2pdma_page(const struct page *page)
{
return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
@@ -3045,6 +3053,7 @@ enum mf_flags {
MF_MUST_KILL = 1 << 2,
MF_SOFT_OFFLINE = 1 << 3,
};
+extern int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t index, int
flags);
extern int memory_failure(unsigned long pfn, int flags);
extern void memory_failure_queue(unsigned long pfn, int flags);
extern void memory_failure_queue_kick(int cpu);
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index e9481632fcd1..158fe0c8e602 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -56,6 +56,7 @@
#include <linux/kfifo.h>
#include <linux/ratelimit.h>
#include <linux/page-isolation.h>
+#include <linux/dax.h>
#include "internal.h"
#include "ras/ras_event.h"
@@ -120,6 +121,13 @@ static int hwpoison_filter_dev(struct page *p)
if (PageSlab(p))
return -EINVAL;
+ if (pfn_valid(page_to_pfn(p))) {
+ if (is_device_fsdax_page(p))
+ return 0;
+ else
+ return -EINVAL;
+ }
+
mapping = page_mapping(p);
if (mapping == NULL || mapping->host == NULL)
return -EINVAL;
@@ -286,10 +294,9 @@ void shake_page(struct page *p, int access)
}
EXPORT_SYMBOL_GPL(shake_page);
-static unsigned long dev_pagemap_mapping_shift(struct page *page,
- struct vm_area_struct *vma)
+static unsigned long dev_pagemap_mapping_shift(struct vm_area_struct *vma,
+ unsigned long address)
{
- unsigned long address = vma_address(page, vma);
pgd_t *pgd;
p4d_t *p4d;
pud_t *pud;
@@ -329,9 +336,8 @@ static unsigned long dev_pagemap_mapping_shift(struct page *page,
* Schedule a process for later kill.
* Uses GFP_ATOMIC allocations to avoid potential recursions in the VM.
*/
-static void add_to_kill(struct task_struct *tsk, struct page *p,
- struct vm_area_struct *vma,
- struct list_head *to_kill)
+static void add_to_kill(struct task_struct *tsk, struct page *p, pgoff_t pgoff,
+ struct vm_area_struct *vma, struct list_head *to_kill)
{
struct to_kill *tk;
@@ -342,9 +348,12 @@ static void add_to_kill(struct task_struct *tsk, struct page *p,
}
tk->addr = page_address_in_vma(p, vma);
- if (is_zone_device_page(p))
- tk->size_shift = dev_pagemap_mapping_shift(p, vma);
- else
+ if (is_zone_device_page(p)) {
+ if (is_device_fsdax_page(p))
+ tk->addr = vma->vm_start +
+ ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+ tk->size_shift = dev_pagemap_mapping_shift(vma, tk->addr);
+ } else
tk->size_shift = page_shift(compound_head(p));
/*
@@ -492,7 +501,7 @@ static void collect_procs_anon(struct page *page, struct list_head
*to_kill,
if (!page_mapped_in_vma(page, vma))
continue;
if (vma->vm_mm == t->mm)
- add_to_kill(t, page, vma, to_kill);
+ add_to_kill(t, page, 0, vma, to_kill);
}
}
read_unlock(&tasklist_lock);
@@ -502,24 +511,19 @@ static void collect_procs_anon(struct page *page, struct list_head
*to_kill,
/*
* Collect processes when the error hit a file mapped page.
*/
-static void collect_procs_file(struct page *page, struct list_head *to_kill,
- int force_early)
+static void collect_procs_file(struct page *page, struct address_space *mapping,
+ pgoff_t pgoff, struct list_head *to_kill, int force_early)
{
struct vm_area_struct *vma;
struct task_struct *tsk;
- struct address_space *mapping = page->mapping;
- pgoff_t pgoff;
i_mmap_lock_read(mapping);
read_lock(&tasklist_lock);
- pgoff = page_to_pgoff(page);
for_each_process(tsk) {
struct task_struct *t = task_early_kill(tsk, force_early);
-
if (!t)
continue;
- vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff,
- pgoff) {
+ vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
/*
* Send early kill signal to tasks where a vma covers
* the page but the corrupted page is not necessarily
@@ -528,7 +532,7 @@ static void collect_procs_file(struct page *page, struct list_head
*to_kill,
* to be informed of all such data corruptions.
*/
if (vma->vm_mm == t->mm)
- add_to_kill(t, page, vma, to_kill);
+ add_to_kill(t, page, pgoff, vma, to_kill);
}
}
read_unlock(&tasklist_lock);
@@ -547,7 +551,8 @@ static void collect_procs(struct page *page, struct list_head
*tokill,
if (PageAnon(page))
collect_procs_anon(page, tokill, force_early);
else
- collect_procs_file(page, tokill, force_early);
+ collect_procs_file(page, page_mapping(page), page_to_pgoff(page),
+ tokill, force_early);
}
static const char *action_name[] = {
@@ -1214,6 +1219,50 @@ static int try_to_split_thp_page(struct page *page, const char
*msg)
return 0;
}
+int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t index, int flags)
+{
+ const bool unmap_success = true;
+ unsigned long pfn, size = 0;
+ struct to_kill *tk;
+ LIST_HEAD(to_kill);
+ int rc = -EBUSY;
+ loff_t start;
+
+ /* load the pfn of the dax mapping file */
+ pfn = dax_load_pfn(mapping, index);
+ if (!pfn)
+ return rc;
+ /*
+ * 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_file(pfn_to_page(pfn), mapping, index, &to_kill,
+ flags & MF_ACTION_REQUIRED);
+
+ list_for_each_entry(tk, &to_kill, 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 = (index << PAGE_SHIFT) & ~(size - 1);
+ unmap_mapping_range(mapping, start, start + size, 0);
+ }
+
+ kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success,
+ pfn, flags);
+ rc = 0;
+ return rc;
+}
+EXPORT_SYMBOL_GPL(mf_dax_mapping_kill_procs);
+
static int memory_failure_hugetlb(unsigned long pfn, int flags)
{
struct page *p = pfn_to_page(pfn);
@@ -1297,7 +1346,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
const bool unmap_success = true;
unsigned long size = 0;
struct to_kill *tk;
- LIST_HEAD(tokill);
+ LIST_HEAD(to_kill);
int rc = -EBUSY;
loff_t start;
dax_entry_t cookie;
@@ -1345,9 +1394,10 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int
flags,
* SIGBUS (i.e. MF_MUST_KILL)
*/
flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
- collect_procs(page, &tokill, flags & MF_ACTION_REQUIRED);
+ collect_procs_file(page, page->mapping, page->index, &to_kill,
+ flags & MF_ACTION_REQUIRED);
- list_for_each_entry(tk, &tokill, nd)
+ list_for_each_entry(tk, &to_kill, nd)
if (tk->size_shift)
size = max(size, 1UL << tk->size_shift);
if (size) {
@@ -1360,7 +1410,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
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);
+ kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success, pfn, flags);
rc = 0;
unlock:
dax_unlock_page(page, cookie);
--
2.30.0
12:33 a.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler
for dax mapping
+extern int mf_dax_mapping_kill_procs(struct address_space *mapping,
pgoff_t index, int flags);
No nee for the extern, please avoid the overly long line.
@@ -120,6 +121,13 @@ static int hwpoison_filter_dev(struct page *p)
if (PageSlab(p))
return -EINVAL;
+ if (pfn_valid(page_to_pfn(p))) {
+ if (is_device_fsdax_page(p))
+ return 0;
+ else
+ return -EINVAL;
+ }
+
This looks odd. For one there is no need for an else after a return.
But more importantly page_mapping() as called below pretty much assumes
a valid PFN, so something is fishy in this function.
+ if (is_zone_device_page(p)) {
+ if (is_device_fsdax_page(p))
+ tk->addr = vma->vm_start +
+ ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
The arithmetics here scream for a common helper, I suspect there might
be other places that could use the same helper.
+int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t
index, int flags)
Overly long line. Also the naming scheme with the mf_ seems rather
unusual. Maybe dax_kill_mapping_procs? Also please add a kerneldoc
comment describing the function given that it exported.
1:56 p.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler for
dax mapping
On 2021/2/10 äøå9:33, Christoph Hellwig wrote:
> +extern int mf_dax_mapping_kill_procs(struct address_space
*mapping, pgoff_t index, int flags);
No nee for the extern, please avoid the overly long line.
OK.
I'd like to confirm one thing... I have checked all of this patchset by
checkpatch.pl and it did not report the overly long line warning. So, I
should still obey the rule of 80 chars one line?
> @@ -120,6 +121,13 @@ static int hwpoison_filter_dev(struct page *p)
> if (PageSlab(p))
> return -EINVAL;
>
> + if (pfn_valid(page_to_pfn(p))) {
> + if (is_device_fsdax_page(p))
> + return 0;
> + else
> + return -EINVAL;
> + }
> +
This looks odd. For one there is no need for an else after a return.
But more importantly page_mapping() as called below pretty much assumes
a valid PFN, so something is fishy in this function.
Yes, a mistake here. I'll fix it.
> + if (is_zone_device_page(p)) {
> + if (is_device_fsdax_page(p))
> + tk->addr = vma->vm_start +
> + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
The arithmetics here scream for a common helper, I suspect there might
be other places that could use the same helper.
> +int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t index, int
flags)
Overly long line. Also the naming scheme with the mf_ seems rather
unusual. Maybe dax_kill_mapping_procs? Also please add a kerneldoc
comment describing the function given that it exported.
OK. Thanks for your guidance.
--
Thanks,
Ruan Shiyang.
7:32 p.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler
for dax mapping
On Wed, Feb 17, 2021 at 10:56:11AM +0800, Ruan Shiyang wrote:
I'd like to confirm one thing... I have checked all of this
patchset by
checkpatch.pl and it did not report the overly long line warning. So, I
should still obey the rule of 80 chars one line?
checkpatch.pl is completely broken, I would not rely on it.
Here is the quote from the coding style document:
"The preferred limit on the length of a single line is 80 columns.
Statements longer than 80 columns should be broken into sensible chunks,
unless exceeding 80 columns significantly increases readability and does
not hide information."
7:59 p.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler for
dax mapping
On 2021/2/18 äøå4:32, Christoph Hellwig wrote:
On Wed, Feb 17, 2021 at 10:56:11AM +0800, Ruan Shiyang wrote:
> I'd like to confirm one thing... I have checked all of this patchset by
> checkpatch.pl and it did not report the overly long line warning. So, I
> should still obey the rule of 80 chars one line?
checkpatch.pl is completely broken, I would not rely on it.
Here is the quote from the coding style document:
"The preferred limit on the length of a single line is 80 columns.
Statements longer than 80 columns should be broken into sensible chunks,
unless exceeding 80 columns significantly increases readability and does
not hide information."
OK. Got it. Thank you.
--
Ruan Shiyang.
2:21 p.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler for
dax mapping
On 2021/2/8 6:55 äøå, Shiyang Ruan wrote:
The current memory_failure_dev_pagemap() can only handle
single-mapped
dax page for fsdax mode. The dax page could be mapped by multiple files
and offsets if we let reflink feature & fsdax mode work together. So,
we refactor current implementation to support handle memory failure on
each file and offset.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
fs/dax.c | 21 ++++++++++
include/linux/dax.h | 1 +
include/linux/mm.h | 9 +++++
mm/memory-failure.c | 98 ++++++++++++++++++++++++++++++++++-----------
4 files changed, 105 insertions(+), 24 deletions(-)
diff --git a/fs/dax.c b/fs/dax.c
index 26d5dcd2d69e..c64c3a0e76a6 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -378,6 +378,27 @@ static struct page *dax_busy_page(void *entry)
return NULL;
}
+/*
+ * dax_load_pfn - Load pfn of the DAX entry corresponding to a page
+ * @mapping: The file whose entry we want to load
+ * @index: The offset where the DAX entry located in
+ *
+ * Return: pfn of the DAX entry
+ */
+unsigned long dax_load_pfn(struct address_space *mapping, unsigned long index)
+{
+ XA_STATE(xas, &mapping->i_pages, index);
+ void *entry;
+ unsigned long pfn;
+
+ xas_lock_irq(&xas);
+ entry = xas_load(&xas);
+ pfn = dax_to_pfn(entry);
+ xas_unlock_irq(&xas);
+
+ return pfn;
+}
+
/*
* dax_lock_mapping_entry - Lock the DAX entry corresponding to a page
* @page: The page whose entry we want to lock
diff --git a/include/linux/dax.h b/include/linux/dax.h
index b52f084aa643..89e56ceeffc7 100644
--- a/include/linux/dax.h
+++ b/include/linux/dax.h
@@ -150,6 +150,7 @@ int dax_writeback_mapping_range(struct address_space *mapping,
struct page *dax_layout_busy_page(struct address_space *mapping);
struct page *dax_layout_busy_page_range(struct address_space *mapping, loff_t start,
loff_t end);
+unsigned long dax_load_pfn(struct address_space *mapping, unsigned long index);
dax_entry_t dax_lock_page(struct page *page);
void dax_unlock_page(struct page *page, dax_entry_t cookie);
#else
diff --git a/include/linux/mm.h b/include/linux/mm.h
index ecdf8a8cd6ae..ab52bc633d84 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1157,6 +1157,14 @@ static inline bool is_device_private_page(const struct page
*page)
page->pgmap->type == MEMORY_DEVICE_PRIVATE;
}
+static inline bool is_device_fsdax_page(const struct page *page)
+{
+ return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
+ IS_ENABLED(CONFIG_FS_DAX) &&
+ is_zone_device_page(page) &&
+ page->pgmap->type == MEMORY_DEVICE_FS_DAX;
+}
+
static inline bool is_pci_p2pdma_page(const struct page *page)
{
return IS_ENABLED(CONFIG_DEV_PAGEMAP_OPS) &&
@@ -3045,6 +3053,7 @@ enum mf_flags {
MF_MUST_KILL = 1 << 2,
MF_SOFT_OFFLINE = 1 << 3,
};
+extern int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t index, int
flags);
extern int memory_failure(unsigned long pfn, int flags);
extern void memory_failure_queue(unsigned long pfn, int flags);
extern void memory_failure_queue_kick(int cpu);
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index e9481632fcd1..158fe0c8e602 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -56,6 +56,7 @@
#include <linux/kfifo.h>
#include <linux/ratelimit.h>
#include <linux/page-isolation.h>
+#include <linux/dax.h>
#include "internal.h"
#include "ras/ras_event.h"
@@ -120,6 +121,13 @@ static int hwpoison_filter_dev(struct page *p)
if (PageSlab(p))
return -EINVAL;
+ if (pfn_valid(page_to_pfn(p))) {
+ if (is_device_fsdax_page(p))
+ return 0;
+ else
+ return -EINVAL;
+ }
+
mapping = page_mapping(p);
if (mapping == NULL || mapping->host == NULL)
return -EINVAL;
@@ -286,10 +294,9 @@ void shake_page(struct page *p, int access)
}
EXPORT_SYMBOL_GPL(shake_page);
-static unsigned long dev_pagemap_mapping_shift(struct page *page,
- struct vm_area_struct *vma)
+static unsigned long dev_pagemap_mapping_shift(struct vm_area_struct *vma,
+ unsigned long address)
{
- unsigned long address = vma_address(page, vma);
pgd_t *pgd;
p4d_t *p4d;
pud_t *pud;
@@ -329,9 +336,8 @@ static unsigned long dev_pagemap_mapping_shift(struct page *page,
* Schedule a process for later kill.
* Uses GFP_ATOMIC allocations to avoid potential recursions in the VM.
*/
-static void add_to_kill(struct task_struct *tsk, struct page *p,
- struct vm_area_struct *vma,
- struct list_head *to_kill)
+static void add_to_kill(struct task_struct *tsk, struct page *p, pgoff_t pgoff,
+ struct vm_area_struct *vma, struct list_head *to_kill)
{
struct to_kill *tk;
@@ -342,9 +348,12 @@ static void add_to_kill(struct task_struct *tsk, struct page *p,
}
tk->addr = page_address_in_vma(p, vma);
- if (is_zone_device_page(p))
- tk->size_shift = dev_pagemap_mapping_shift(p, vma);
- else
+ if (is_zone_device_page(p)) {
+ if (is_device_fsdax_page(p))
+ tk->addr = vma->vm_start +
+ ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+ tk->size_shift = dev_pagemap_mapping_shift(vma, tk->addr);
+ } else
tk->size_shift = page_shift(compound_head(p));
/*
@@ -492,7 +501,7 @@ static void collect_procs_anon(struct page *page, struct list_head
*to_kill,
if (!page_mapped_in_vma(page, vma))
continue;
if (vma->vm_mm == t->mm)
- add_to_kill(t, page, vma, to_kill);
+ add_to_kill(t, page, 0, vma, to_kill);
}
}
read_unlock(&tasklist_lock);
@@ -502,24 +511,19 @@ static void collect_procs_anon(struct page *page, struct list_head
*to_kill,
/*
* Collect processes when the error hit a file mapped page.
*/
-static void collect_procs_file(struct page *page, struct list_head *to_kill,
- int force_early)
+static void collect_procs_file(struct page *page, struct address_space *mapping,
+ pgoff_t pgoff, struct list_head *to_kill, int force_early)
{
struct vm_area_struct *vma;
struct task_struct *tsk;
- struct address_space *mapping = page->mapping;
- pgoff_t pgoff;
i_mmap_lock_read(mapping);
read_lock(&tasklist_lock);
- pgoff = page_to_pgoff(page);
for_each_process(tsk) {
struct task_struct *t = task_early_kill(tsk, force_early);
-
if (!t)
continue;
- vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff,
- pgoff) {
+ vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
/*
* Send early kill signal to tasks where a vma covers
* the page but the corrupted page is not necessarily
@@ -528,7 +532,7 @@ static void collect_procs_file(struct page *page, struct list_head
*to_kill,
* to be informed of all such data corruptions.
*/
if (vma->vm_mm == t->mm)
- add_to_kill(t, page, vma, to_kill);
+ add_to_kill(t, page, pgoff, vma, to_kill);
}
}
read_unlock(&tasklist_lock);
@@ -547,7 +551,8 @@ static void collect_procs(struct page *page, struct list_head
*tokill,
if (PageAnon(page))
collect_procs_anon(page, tokill, force_early);
else
- collect_procs_file(page, tokill, force_early);
+ collect_procs_file(page, page_mapping(page), page_to_pgoff(page),
+ tokill, force_early);
}
static const char *action_name[] = {
@@ -1214,6 +1219,50 @@ static int try_to_split_thp_page(struct page *page, const char
*msg)
return 0;
}
+int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t index, int flags)
+{
+ const bool unmap_success = true;
+ unsigned long pfn, size = 0;
+ struct to_kill *tk;
+ LIST_HEAD(to_kill);
+ int rc = -EBUSY;
+ loff_t start;
+
+ /* load the pfn of the dax mapping file */
+ pfn = dax_load_pfn(mapping, index);
+ if (!pfn)
+ return rc;
+ /*
+ * 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;
MF_ACTION_REQUIRED only kill the current execution context. A page can be shared
when reflink file be mapped by different process. We can not kill all process
shared the page. Other process still can access the posioned page ?
Thanks,
zhong jiang
> + collect_procs_file(pfn_to_page(pfn), mapping, index, &to_kill,
> + flags & MF_ACTION_REQUIRED);
> +
> + list_for_each_entry(tk, &to_kill, 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 = (index << PAGE_SHIFT) & ~(size - 1);
> + unmap_mapping_range(mapping, start, start + size, 0);
> + }
> +
> + kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success,
> + pfn, flags);
> + rc = 0;
> + return rc;
> +}
> +EXPORT_SYMBOL_GPL(mf_dax_mapping_kill_procs);
> +
> static int memory_failure_hugetlb(unsigned long pfn, int flags)
> {
> struct page *p = pfn_to_page(pfn);
> @@ -1297,7 +1346,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int
flags,
> const bool unmap_success = true;
> unsigned long size = 0;
> struct to_kill *tk;
> - LIST_HEAD(tokill);
> + LIST_HEAD(to_kill);
> int rc = -EBUSY;
> loff_t start;
> dax_entry_t cookie;
> @@ -1345,9 +1394,10 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int
flags,
> * SIGBUS (i.e. MF_MUST_KILL)
> */
> flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> - collect_procs(page, &tokill, flags & MF_ACTION_REQUIRED);
> + collect_procs_file(page, page->mapping, page->index, &to_kill,
> + flags & MF_ACTION_REQUIRED);
>
> - list_for_each_entry(tk, &tokill, nd)
> + list_for_each_entry(tk, &to_kill, nd)
> if (tk->size_shift)
> size = max(size, 1UL << tk->size_shift);
> if (size) {
> @@ -1360,7 +1410,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int
flags,
> 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);
> + kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success, pfn, flags);
> rc = 0;
> unlock:
> dax_unlock_page(page, cookie);
2:46 p.m.
New subject: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler for
dax mapping
-----Original Message-----
From: zhong jiang <zhongjiang-ali(a)linux.alibaba.com>
Subject: Re: [PATCH v3 05/11] mm, fsdax: Refactor memory-failure handler for
dax mapping
> +int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t
> +index, int flags) {
> + const bool unmap_success = true;
> + unsigned long pfn, size = 0;
> + struct to_kill *tk;
> + LIST_HEAD(to_kill);
> + int rc = -EBUSY;
> + loff_t start;
> +
> + /* load the pfn of the dax mapping file */
> + pfn = dax_load_pfn(mapping, index);
> + if (!pfn)
> + return rc;
> + /*
> + * 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;
MF_ACTION_REQUIRED only kill the current execution context. A page can be
shared when reflink file be mapped by different process. We can not kill all
process shared the page. Other process still can access the posioned page ?
AFAIK, the other processes will receive a SIGBUS when accessing this corrupted range. But
I didn't add a testcase for this condition. I'll test it. Thanks for pointing
out.
--
Thanks,
Ruan Shiyang.
Thanks,
zhong jiang
> + collect_procs_file(pfn_to_page(pfn), mapping, index, &to_kill,
> + flags & MF_ACTION_REQUIRED);
> +
> + list_for_each_entry(tk, &to_kill, 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 = (index << PAGE_SHIFT) & ~(size - 1);
> + unmap_mapping_range(mapping, start, start + size, 0);
> + }
> +
> + kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success,
> + pfn, flags);
> + rc = 0;
> + return rc;
> +}
> +EXPORT_SYMBOL_GPL(mf_dax_mapping_kill_procs);
> +
> static int memory_failure_hugetlb(unsigned long pfn, int flags)
> {
> struct page *p = pfn_to_page(pfn);
> @@ -1297,7 +1346,7 @@ static int memory_failure_dev_pagemap(unsigned
long pfn, int flags,
> const bool unmap_success = true;
> unsigned long size = 0;
> struct to_kill *tk;
> - LIST_HEAD(tokill);
> + LIST_HEAD(to_kill);
> int rc = -EBUSY;
> loff_t start;
> dax_entry_t cookie;
> @@ -1345,9 +1394,10 @@ static int
memory_failure_dev_pagemap(unsigned long pfn, int flags,
> * SIGBUS (i.e. MF_MUST_KILL)
> */
> flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
> - collect_procs(page, &tokill, flags & MF_ACTION_REQUIRED);
> + collect_procs_file(page, page->mapping, page->index, &to_kill,
> + flags & MF_ACTION_REQUIRED);
>
> - list_for_each_entry(tk, &tokill, nd)
> + list_for_each_entry(tk, &to_kill, nd)
> if (tk->size_shift)
> size = max(size, 1UL << tk->size_shift);
> if (size) {
> @@ -1360,7 +1410,7 @@ static int memory_failure_dev_pagemap(unsigned
long pfn, int flags,
> 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);
> + kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success, pfn,
> +flags);
> rc = 0;
> unlock:
> dax_unlock_page(page, cookie);
9:55 p.m.
New subject: [PATCH v3 06/11] mm, pmem: Implement ->memory_failure() in pmem driver
Call the ->memory_failure() which is implemented by pmem driver, in
order to finally notify filesystem to handle the corrupted data. The
handler which collects and kills processes are moved into
mf_dax_mapping_kill_procs(), which will be called by filesystem.
Keep the old handler in order to roll back if driver/device/filesystem
does not support ->memory_failure()/->corrupted_range().
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
drivers/nvdimm/pmem.c | 26 +++++++++++
mm/memory-failure.c | 102 +++++++++++++++++++++++++-----------------
2 files changed, 87 insertions(+), 41 deletions(-)
diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
index 875076b0ea6c..c77c80e3d155 100644
--- a/drivers/nvdimm/pmem.c
+++ b/drivers/nvdimm/pmem.c
@@ -363,9 +363,35 @@ static void pmem_release_disk(void *__pmem)
put_disk(pmem->disk);
}
+static int pmem_pagemap_memory_failure(struct dev_pagemap *pgmap,
+ unsigned long pfn, int flags)
+{
+ struct pmem_device *pdev;
+ struct gendisk *disk;
+ loff_t disk_offset;
+ int rc = 0;
+ unsigned long size = page_size(pfn_to_page(pfn));
+
+ pdev = container_of(pgmap, struct pmem_device, pgmap);
+ disk = pdev->disk;
+ if (!disk)
+ return -ENXIO;
+
+ disk_offset = PFN_PHYS(pfn) - pdev->phys_addr - pdev->data_offset;
+ if (disk->fops->corrupted_range) {
+ rc = disk->fops->corrupted_range(disk, NULL, disk_offset,
+ size, &flags);
+ if (rc == -ENODEV)
+ rc = -ENXIO;
+ } else
+ rc = -EOPNOTSUPP;
+ return rc;
+}
+
static const struct dev_pagemap_ops fsdax_pagemap_ops = {
.kill = pmem_pagemap_kill,
.cleanup = pmem_pagemap_cleanup,
+ .memory_failure = pmem_pagemap_memory_failure,
};
static int pmem_attach_disk(struct device *dev,
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 158fe0c8e602..670e29cd263e 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1219,6 +1219,54 @@ static int try_to_split_thp_page(struct page *page, const char
*msg)
return 0;
}
+int mf_generic_kill_procs(unsigned long long pfn, int flags)
+{
+ struct page *page = pfn_to_page(pfn);
+ const bool unmap_success = true;
+ unsigned long size = 0;
+ struct to_kill *tk;
+ LIST_HEAD(to_kill);
+ loff_t start;
+ dax_entry_t cookie;
+
+ /*
+ * 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.
+ */
+ cookie = dax_lock_page(page);
+ if (!cookie)
+ return -EBUSY;
+ /*
+ * 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, &to_kill, flags & MF_ACTION_REQUIRED);
+
+ list_for_each_entry(tk, &to_kill, 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(&to_kill, flags & MF_MUST_KILL, !unmap_success, pfn, flags);
+
+ dax_unlock_page(page, cookie);
+ return 0;
+}
+
int mf_dax_mapping_kill_procs(struct address_space *mapping, pgoff_t index, int flags)
{
const bool unmap_success = true;
@@ -1343,13 +1391,7 @@ 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(to_kill);
int rc = -EBUSY;
- loff_t start;
- dax_entry_t cookie;
if (flags & MF_COUNT_INCREASED)
/*
@@ -1357,20 +1399,9 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int
flags,
*/
put_page(page);
- /*
- * 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.
- */
- cookie = dax_lock_page(page);
- if (!cookie)
- goto out;
-
if (hwpoison_filter(page)) {
rc = 0;
- goto unlock;
+ goto out;
}
if (pgmap->type == MEMORY_DEVICE_PRIVATE) {
@@ -1378,7 +1409,7 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
* TODO: Handle HMM pages which may need coordination
* with device-side memory.
*/
- goto unlock;
+ goto out;
}
/*
@@ -1388,32 +1419,21 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int
flags,
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)
+ * Call driver's implementation to handle the memory failure,
+ * otherwise roll back to generic handler.
*/
- flags |= MF_ACTION_REQUIRED | MF_MUST_KILL;
- collect_procs_file(page, page->mapping, page->index, &to_kill,
- flags & MF_ACTION_REQUIRED);
-
- list_for_each_entry(tk, &to_kill, nd)
- if (tk->size_shift)
- size = max(size, 1UL << tk->size_shift);
- if (size) {
+ if (pgmap->ops->memory_failure) {
+ rc = pgmap->ops->memory_failure(pgmap, pfn, flags);
/*
- * 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()
+ * Roll back to generic handler too if operation is not
+ * supported inside the driver/device/filesystem.
*/
- start = (page->index << PAGE_SHIFT) & ~(size - 1);
- unmap_mapping_range(page->mapping, start, start + size, 0);
+ if (rc != EOPNOTSUPP)
+ goto out;
}
- kill_procs(&to_kill, flags & MF_MUST_KILL, !unmap_success, pfn, flags);
- rc = 0;
-unlock:
- dax_unlock_page(page, cookie);
+
+ rc = mf_generic_kill_procs(pfn, flags);
+
out:
/* drop pgmap ref acquired in caller */
put_dev_pagemap(pgmap);
--
2.30.0
12:41 a.m.
New subject: [PATCH v3 06/11] mm, pmem: Implement ->memory_failure() in
pmem driver
+static int pmem_pagemap_memory_failure(struct dev_pagemap *pgmap,
+ unsigned long pfn, int flags)
+{
+ struct pmem_device *pdev;
+ struct gendisk *disk;
+ loff_t disk_offset;
+ int rc = 0;
+ unsigned long size = page_size(pfn_to_page(pfn));
+
+ pdev = container_of(pgmap, struct pmem_device, pgmap);
+ disk = pdev->disk;
Would be nice to initialize this at the time of declaration:
struct pmem_device *pdev =
container_of(pgmap, struct pmem_device, pgmap);
struct gendisk *disk = pdev->disk
unsigned long size = page_size(pfn_to_page(pfn));
+ if (!disk)
+ return -ENXIO;
+
+ disk_offset = PFN_PHYS(pfn) - pdev->phys_addr - pdev->data_offset;
+ if (disk->fops->corrupted_range) {
+ rc = disk->fops->corrupted_range(disk, NULL, disk_offset,
+ size, &flags);
+ if (rc == -ENODEV)
+ rc = -ENXIO;
+ } else
+ rc = -EOPNOTSUPP;
Why do we need the disk and fops check here? A pgmap registered by pmem.c
should always have a disk with pmem_fops. And more importantly this
has no business going through the block layer.
Instead the file system should deposit a callback when starting to use
the dax_device using fs_dax_get_by_bdev / dax_get_by_host and a private
data (the superblock), and we avoid all the lookup problems.
+int mf_generic_kill_procs(unsigned long long pfn, int flags)
This function seems to be only used inside of memory-failure.c, so it
could be marked static. Also I'd name it dax_generic_memory_failure
or something like that to match the naming of the ->memory_failure
pgmap operation.
Also maybe just splitting this out into a helper would be a nice prep
patch.
9:55 p.m.
New subject: [PATCH v3 07/11] pmem: Implement ->corrupted_range() for pmem driver
Obtain the superblock of a pmem disk, and call filesystem's
->corrupted_range() to handle the corrupted data.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
block/genhd.c | 6 ++++++
drivers/nvdimm/pmem.c | 19 +++++++++++++++++++
include/linux/genhd.h | 1 +
3 files changed, 26 insertions(+)
diff --git a/block/genhd.c b/block/genhd.c
index 419548e92d82..fd7cf03b65a8 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -936,6 +936,12 @@ struct block_device *bdget_disk(struct gendisk *disk, int partno)
return bdev;
}
+struct block_device *bdget_disk_sector(struct gendisk *disk, sector_t sector)
+{
+ return disk_map_sector_rcu(disk, sector);
+}
+EXPORT_SYMBOL(bdget_disk_sector);
+
/*
* print a full list of all partitions - intended for places where the root
* filesystem can't be mounted and thus to give the victim some idea of what
diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
index c77c80e3d155..e38b9f9c7d97 100644
--- a/drivers/nvdimm/pmem.c
+++ b/drivers/nvdimm/pmem.c
@@ -253,6 +253,24 @@ static int pmem_rw_page(struct block_device *bdev, sector_t sector,
return blk_status_to_errno(rc);
}
+static int pmem_corrupted_range(struct gendisk *disk, struct block_device *bdev,
+ loff_t disk_offset, size_t len, void *data)
+{
+ loff_t bdev_offset;
+ sector_t disk_sector = disk_offset >> SECTOR_SHIFT;
+ int rc = -ENODEV;
+
+ bdev = bdget_disk_sector(disk, disk_sector);
+ if (!bdev)
+ return rc;
+
+ bdev_offset = (disk_sector - get_start_sect(bdev)) << SECTOR_SHIFT;
+ rc = bd_corrupted_range(bdev, bdev_offset, bdev_offset, len, data);
+
+ bdput(bdev);
+ return rc;
+}
+
/* see "strong" declaration in tools/testing/nvdimm/pmem-dax.c */
__weak long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff,
long nr_pages, void **kaddr, pfn_t *pfn)
@@ -281,6 +299,7 @@ static const struct block_device_operations pmem_fops = {
.owner = THIS_MODULE,
.submit_bio = pmem_submit_bio,
.rw_page = pmem_rw_page,
+ .corrupted_range = pmem_corrupted_range,
};
static int pmem_dax_zero_page_range(struct dax_device *dax_dev, pgoff_t pgoff,
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 751cbd559bba..996f91b08d48 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -248,6 +248,7 @@ static inline void add_disk_no_queue_reg(struct gendisk *disk)
extern void del_gendisk(struct gendisk *gp);
extern struct block_device *bdget_disk(struct gendisk *disk, int partno);
+extern struct block_device *bdget_disk_sector(struct gendisk *disk, sector_t sector);
extern void set_disk_ro(struct gendisk *disk, int flag);
--
2.30.0
9:55 p.m.
New subject: [PATCH v3 08/11] dm: Introduce ->rmap() to find bdev offset
Pmem device could be a target of mapped device. In order to obtain
superblock on the mapped device, we introduce this to translate offset
from target device to md device.
Currently, we implement it on linear target, which is easy to do the
translation. Other targets will be supported in the future. However,
some targets may not support it because of the non-linear mapping.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
drivers/md/dm-linear.c | 20 ++++++++++++++++++++
include/linux/device-mapper.h | 5 +++++
2 files changed, 25 insertions(+)
diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c
index 00774b5d7668..90fdb4700afd 100644
--- a/drivers/md/dm-linear.c
+++ b/drivers/md/dm-linear.c
@@ -5,6 +5,7 @@
*/
#include "dm.h"
+#include "dm-core.h"
#include <linux/module.h>
#include <linux/init.h>
#include <linux/blkdev.h>
@@ -119,6 +120,24 @@ static void linear_status(struct dm_target *ti, status_type_t type,
}
}
+static int linear_rmap(struct dm_target *ti, sector_t offset,
+ rmap_callout_fn fn, void *data)
+{
+ struct linear_c *lc = (struct linear_c *) ti->private;
+ struct mapped_device *md = ti->table->md;
+ struct block_device *bdev;
+ sector_t disk_sect = offset - dm_target_offset(ti, lc->start);
+ int rc = -ENODEV;
+
+ bdev = bdget_disk_sector(md->disk, offset);
+ if (!bdev)
+ return rc;
+
+ rc = fn(ti, bdev, disk_sect, data);
+ bdput(bdev);
+ return rc;
+}
+
static int linear_prepare_ioctl(struct dm_target *ti, struct block_device **bdev)
{
struct linear_c *lc = (struct linear_c *) ti->private;
@@ -238,6 +257,7 @@ static struct target_type linear_target = {
.ctr = linear_ctr,
.dtr = linear_dtr,
.map = linear_map,
+ .rmap = linear_rmap,
.status = linear_status,
.prepare_ioctl = linear_prepare_ioctl,
.iterate_devices = linear_iterate_devices,
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 61a66fb8ebb3..c5cd1009a08d 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -58,6 +58,10 @@ typedef void (*dm_dtr_fn) (struct dm_target *ti);
* = 2: The target wants to push back the io
*/
typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio);
+typedef int (*rmap_callout_fn) (struct dm_target *ti, struct block_device *bdev,
+ sector_t sect, void *data);
+typedef int (*dm_rmap_fn) (struct dm_target *ti, sector_t offset,
+ rmap_callout_fn fn, void *data);
typedef int (*dm_clone_and_map_request_fn) (struct dm_target *ti,
struct request *rq,
union map_info *map_context,
@@ -175,6 +179,7 @@ struct target_type {
dm_ctr_fn ctr;
dm_dtr_fn dtr;
dm_map_fn map;
+ dm_rmap_fn rmap;
dm_clone_and_map_request_fn clone_and_map_rq;
dm_release_clone_request_fn release_clone_rq;
dm_endio_fn end_io;
--
2.30.0
9:55 p.m.
New subject: [PATCH v3 09/11] md: Implement ->corrupted_range()
With the support of ->rmap(), it is possible to obtain the superblock on
a mapped device.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
drivers/md/dm.c | 61 +++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 61 insertions(+)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 7bac564f3faa..31b0c340b695 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -507,6 +507,66 @@ static int dm_blk_report_zones(struct gendisk *disk, sector_t
sector,
#define dm_blk_report_zones NULL
#endif /* CONFIG_BLK_DEV_ZONED */
+struct corrupted_hit_info {
+ struct block_device *bdev;
+ sector_t offset;
+};
+
+static int dm_blk_corrupted_hit(struct dm_target *ti, struct dm_dev *dev,
+ sector_t start, sector_t count, void *data)
+{
+ struct corrupted_hit_info *bc = data;
+
+ return bc->bdev == (void *)dev->bdev &&
+ (start <= bc->offset && bc->offset < start + count);
+}
+
+struct corrupted_do_info {
+ size_t length;
+ void *data;
+};
+
+static int dm_blk_corrupted_do(struct dm_target *ti, struct block_device *bdev,
+ sector_t disk_sect, void *data)
+{
+ struct corrupted_do_info *bc = data;
+ loff_t disk_off = to_bytes(disk_sect);
+ loff_t bdev_off = to_bytes(disk_sect - get_start_sect(bdev));
+
+ return bd_corrupted_range(bdev, disk_off, bdev_off, bc->length, bc->data);
+}
+
+static int dm_blk_corrupted_range(struct gendisk *disk,
+ struct block_device *target_bdev,
+ loff_t target_offset, size_t len, void *data)
+{
+ struct mapped_device *md = disk->private_data;
+ struct dm_table *map;
+ struct dm_target *ti;
+ sector_t target_sect = to_sector(target_offset);
+ struct corrupted_hit_info hi = {target_bdev, target_sect};
+ struct corrupted_do_info di = {len, data};
+ int srcu_idx, i, rc = -ENODEV;
+
+ map = dm_get_live_table(md, &srcu_idx);
+ if (!map)
+ return rc;
+
+ for (i = 0; i < dm_table_get_num_targets(map); i++) {
+ ti = dm_table_get_target(map, i);
+ if (!(ti->type->iterate_devices && ti->type->rmap))
+ continue;
+ if (!ti->type->iterate_devices(ti, dm_blk_corrupted_hit, &hi))
+ continue;
+
+ rc = ti->type->rmap(ti, target_sect, dm_blk_corrupted_do, &di);
+ break;
+ }
+
+ dm_put_live_table(md, srcu_idx);
+ return rc;
+}
+
static int dm_prepare_ioctl(struct mapped_device *md, int *srcu_idx,
struct block_device **bdev)
{
@@ -3062,6 +3122,7 @@ static const struct block_device_operations dm_blk_dops = {
.getgeo = dm_blk_getgeo,
.report_zones = dm_blk_report_zones,
.pr_ops = &dm_pr_ops,
+ .corrupted_range = dm_blk_corrupted_range,
.owner = THIS_MODULE
};
--
2.30.0
9:55 p.m.
New subject: [PATCH v3 10/11] xfs: Implement ->corrupted_range() for XFS
This function is used to handle errors which may cause data lost in
filesystem. Such as memory failure in fsdax mode.
If the rmap feature of XFS enabled, we can query it to find files and
metadata which are associated with the corrupt data. For now all we do
is kill processes with that file mapped into their address spaces, but
future patches could actually do something about corrupt metadata.
After that, the memory failure needs to notify the processes who are
using those files.
Only support data device. Realtime device is not supported for now.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
fs/xfs/xfs_fsops.c | 5 ++
fs/xfs/xfs_mount.h | 1 +
fs/xfs/xfs_super.c | 112 +++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 118 insertions(+)
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 959ce91a3755..f03901a5c673 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -498,6 +498,11 @@ xfs_do_force_shutdown(
"Corruption of in-memory data detected. Shutting down filesystem");
if (XFS_ERRLEVEL_HIGH <= xfs_error_level)
xfs_stack_trace();
+ } else if (flags & SHUTDOWN_CORRUPT_META) {
+ xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_CORRUPT,
+"Corruption of on-disk metadata detected. Shutting down filesystem");
+ if (XFS_ERRLEVEL_HIGH <= xfs_error_level)
+ xfs_stack_trace();
} else if (logerror) {
xfs_alert_tag(mp, XFS_PTAG_SHUTDOWN_LOGERROR,
"Log I/O Error Detected. Shutting down filesystem");
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index dfa429b77ee2..8f0df67ffcc1 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -274,6 +274,7 @@ void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char
*fname,
#define SHUTDOWN_LOG_IO_ERROR 0x0002 /* write attempt to the log failed */
#define SHUTDOWN_FORCE_UMOUNT 0x0004 /* shutdown from a forced unmount */
#define SHUTDOWN_CORRUPT_INCORE 0x0008 /* corrupt in-memory data structures */
+#define SHUTDOWN_CORRUPT_META 0x0010 /* corrupt metadata on device */
/*
* Flags for xfs_mountfs
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 813be879a5e5..8906426a0f60 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -35,6 +35,11 @@
#include "xfs_refcount_item.h"
#include "xfs_bmap_item.h"
#include "xfs_reflink.h"
+#include "xfs_alloc.h"
+#include "xfs_rmap.h"
+#include "xfs_rmap_btree.h"
+#include "xfs_rtalloc.h"
+#include "xfs_bit.h"
#include <linux/magic.h>
#include <linux/fs_context.h>
@@ -1105,6 +1110,112 @@ xfs_fs_free_cached_objects(
return xfs_reclaim_inodes_nr(XFS_M(sb), sc->nr_to_scan);
}
+static int
+xfs_corrupt_helper(
+ struct xfs_btree_cur *cur,
+ struct xfs_rmap_irec *rec,
+ void *data)
+{
+ struct xfs_inode *ip;
+ struct address_space *mapping;
+ int rc = 0;
+ int *flags = data;
+
+ if (XFS_RMAP_NON_INODE_OWNER(rec->rm_owner) ||
+ (rec->rm_flags & (XFS_RMAP_ATTR_FORK | XFS_RMAP_BMBT_BLOCK))) {
+ // TODO check and try to fix metadata
+ rc = -EFSCORRUPTED;
+ xfs_force_shutdown(cur->bc_mp, SHUTDOWN_CORRUPT_META);
+ } else {
+ /*
+ * Get files that incore, filter out others that are not in use.
+ */
+ rc = xfs_iget(cur->bc_mp, cur->bc_tp, rec->rm_owner,
+ XFS_IGET_INCORE, 0, &ip);
+ if (rc || !ip)
+ return rc;
+ if (!VFS_I(ip)->i_mapping)
+ goto out;
+
+ mapping = VFS_I(ip)->i_mapping;
+ if (IS_DAX(VFS_I(ip)))
+ rc = mf_dax_mapping_kill_procs(mapping, rec->rm_offset,
+ *flags);
+ else {
+ rc = -EIO;
+ mapping_set_error(mapping, rc);
+ }
+
+ // TODO try to fix data
+out:
+ xfs_irele(ip);
+ }
+
+ return rc;
+}
+
+static int
+xfs_fs_corrupted_range(
+ struct super_block *sb,
+ struct block_device *bdev,
+ loff_t offset,
+ size_t len,
+ void *data)
+{
+ struct xfs_mount *mp = XFS_M(sb);
+ struct xfs_trans *tp = NULL;
+ struct xfs_btree_cur *cur = NULL;
+ struct xfs_rmap_irec rmap_low, rmap_high;
+ struct xfs_buf *agf_bp = NULL;
+ xfs_fsblock_t fsbno = XFS_B_TO_FSB(mp, offset);
+ xfs_filblks_t bcnt = XFS_B_TO_FSB(mp, len);
+ xfs_agnumber_t agno = XFS_FSB_TO_AGNO(mp, fsbno);
+ xfs_agblock_t agbno = XFS_FSB_TO_AGBNO(mp, fsbno);
+ int error = 0;
+
+ if (mp->m_rtdev_targp && mp->m_rtdev_targp->bt_bdev == bdev) {
+ xfs_warn(mp, "corrupted_range support not available for realtime device!");
+ return -EOPNOTSUPP;
+ }
+ if (mp->m_logdev_targp && mp->m_logdev_targp->bt_bdev == bdev
&&
+ mp->m_logdev_targp != mp->m_ddev_targp) {
+ xfs_err(mp, "ondisk log corrupt, shutting down fs!");
+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_META);
+ return -EFSCORRUPTED;
+ }
+
+ if (!xfs_sb_version_hasrmapbt(&mp->m_sb)) {
+ xfs_warn(mp, "corrupted_range needs rmapbt enabled!");
+ return -EOPNOTSUPP;
+ }
+
+ error = xfs_trans_alloc_empty(mp, &tp);
+ if (error)
+ return error;
+
+ error = xfs_alloc_read_agf(mp, tp, agno, 0, &agf_bp);
+ if (error)
+ goto out_cancel_tp;
+
+ cur = xfs_rmapbt_init_cursor(mp, tp, agf_bp, agno);
+
+ /* Construct a range for rmap query */
+ memset(&rmap_low, 0, sizeof(rmap_low));
+ memset(&rmap_high, 0xFF, sizeof(rmap_high));
+ rmap_low.rm_startblock = rmap_high.rm_startblock = agbno;
+ rmap_low.rm_blockcount = rmap_high.rm_blockcount = bcnt;
+
+ error = xfs_rmap_query_range(cur, &rmap_low, &rmap_high,
+ xfs_corrupt_helper, data);
+
+ xfs_btree_del_cursor(cur, error);
+ xfs_trans_brelse(tp, agf_bp);
+
+out_cancel_tp:
+ xfs_trans_cancel(tp);
+ return error;
+}
+
static const struct super_operations xfs_super_operations = {
.alloc_inode = xfs_fs_alloc_inode,
.destroy_inode = xfs_fs_destroy_inode,
@@ -1118,6 +1229,7 @@ static const struct super_operations xfs_super_operations = {
.show_options = xfs_fs_show_options,
.nr_cached_objects = xfs_fs_nr_cached_objects,
.free_cached_objects = xfs_fs_free_cached_objects,
+ .corrupted_range = xfs_fs_corrupted_range,
};
static int
--
2.30.0
12:44 a.m.
New subject: [PATCH v3 10/11] xfs: Implement ->corrupted_range() for XFS
+ if (XFS_RMAP_NON_INODE_OWNER(rec->rm_owner) ||
+ (rec->rm_flags & (XFS_RMAP_ATTR_FORK | XFS_RMAP_BMBT_BLOCK))) {
+ // TODO check and try to fix metadata
+ rc = -EFSCORRUPTED;
+ xfs_force_shutdown(cur->bc_mp, SHUTDOWN_CORRUPT_META);
Just return early here so that we can avoid the else later.
+ /*
+ * Get files that incore, filter out others that are not in use.
+ */
+ rc = xfs_iget(cur->bc_mp, cur->bc_tp, rec->rm_owner,
+ XFS_IGET_INCORE, 0, &ip);
Can we rename rc to error?
+ if (rc || !ip)
+ return rc;
No need to check for ip here.
+ if (!VFS_I(ip)->i_mapping)
+ goto out;
This can't happen either.
+
+ mapping = VFS_I(ip)->i_mapping;
+ if (IS_DAX(VFS_I(ip)))
+ rc = mf_dax_mapping_kill_procs(mapping, rec->rm_offset,
+ *flags);
+ else {
+ rc = -EIO;
+ mapping_set_error(mapping, rc);
+ }
By passing the method directly to the DAX device we should never get
this called for the non-DAX case.
9:55 p.m.
New subject: [PATCH v3 11/11] fs/dax: Remove useless functions
Since owner tarcking is triggerred by pmem device, these functions are
useless. So remove them.
Signed-off-by: Shiyang Ruan <ruansy.fnst(a)cn.fujitsu.com>
---
fs/dax.c | 46 ----------------------------------------------
1 file changed, 46 deletions(-)
diff --git a/fs/dax.c b/fs/dax.c
index c64c3a0e76a6..e20a5df03eec 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -323,48 +323,6 @@ static unsigned long dax_end_pfn(void *entry)
for (pfn = dax_to_pfn(entry); \
pfn < dax_end_pfn(entry); pfn++)
-/*
- * TODO: for reflink+dax we need a way to associate a single page with
- * multiple address_space instances at different linear_page_index()
- * offsets.
- */
-static void dax_associate_entry(void *entry, struct address_space *mapping,
- struct vm_area_struct *vma, unsigned long address)
-{
- unsigned long size = dax_entry_size(entry), pfn, index;
- int i = 0;
-
- if (IS_ENABLED(CONFIG_FS_DAX_LIMITED))
- return;
-
- index = linear_page_index(vma, address & ~(size - 1));
- for_each_mapped_pfn(entry, pfn) {
- struct page *page = pfn_to_page(pfn);
-
- WARN_ON_ONCE(page->mapping);
- page->mapping = mapping;
- page->index = index + i++;
- }
-}
-
-static void dax_disassociate_entry(void *entry, struct address_space *mapping,
- bool trunc)
-{
- unsigned long pfn;
-
- if (IS_ENABLED(CONFIG_FS_DAX_LIMITED))
- return;
-
- for_each_mapped_pfn(entry, pfn) {
- struct page *page = pfn_to_page(pfn);
-
- WARN_ON_ONCE(trunc && page_ref_count(page) > 1);
- WARN_ON_ONCE(page->mapping && page->mapping != mapping);
- page->mapping = NULL;
- page->index = 0;
- }
-}
-
static struct page *dax_busy_page(void *entry)
{
unsigned long pfn;
@@ -543,7 +501,6 @@ static void *grab_mapping_entry(struct xa_state *xas,
xas_lock_irq(xas);
}
- dax_disassociate_entry(entry, mapping, false);
xas_store(xas, NULL); /* undo the PMD join */
dax_wake_entry(xas, entry, true);
mapping->nrexceptional--;
@@ -680,7 +637,6 @@ static int __dax_invalidate_entry(struct address_space *mapping,
(xas_get_mark(&xas, PAGECACHE_TAG_DIRTY) ||
xas_get_mark(&xas, PAGECACHE_TAG_TOWRITE)))
goto out;
- dax_disassociate_entry(entry, mapping, trunc);
xas_store(&xas, NULL);
mapping->nrexceptional--;
ret = 1;
@@ -774,8 +730,6 @@ static void *dax_insert_entry(struct xa_state *xas,
if (dax_is_zero_entry(entry) || dax_is_empty_entry(entry)) {
void *old;
- dax_disassociate_entry(entry, mapping, false);
- dax_associate_entry(new_entry, mapping, vmf->vma, vmf->address);
/*
* Only swap our new entry into the page cache if the current
* entry is a zero page or an empty entry. If a normal PTE or
--
2.30.0
12:09 a.m.
New subject: [PATCH v3 11/11] fs/dax: Remove useless functions
On Mon, Feb 08, 2021 at 06:55:30PM +0800, Shiyang Ruan wrote:
Since owner tarcking is triggerred by pmem device, these functions
are
s/tarcking/tracking/
useless. So remove them.
Note that this patch does not apply for me when applying your two series
on top of 5.11-rc5.
550
days inactive
587
days old
26 comments
7 participants
participants (7)
-
Christoph Hellwig -
Dan Williams -
Darrick J. Wong -
Ruan Shiyang -
ruansy.fnstļ¼ fujitsu.com -
Shiyang Ruan
-
zhong jiang