The patch titled
Subject: dax: provide diagnostics for pmd mapping failures
has been added to the -mm tree. Its filename is
dax-provide-diagnostics-for-pmd-mapping-failures.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/dax-provide-diagnostics-for-pmd-mapping-failures.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/dax-provide-diagnostics-for-pmd-mapping-failures.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Dan Williams <dan.j.williams@xxxxxxxxx>
Subject: dax: provide diagnostics for pmd mapping failures
There is a wide gamut of conditions that can trigger the dax pmd path to
fallback to pte mappings. Ideally we'd have a syscall interface to
determine mapping characteristics after the fact. In the meantime provide
debug messages.
Signed-off-by: Dan Williams <dan.j.williams@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/dax.c | 57 ++++++++++++++++++++++++++++++++++++++++++++---------
1 file changed, 48 insertions(+), 9 deletions(-)
diff -puN fs/dax.c~dax-provide-diagnostics-for-pmd-mapping-failures fs/dax.c
--- a/fs/dax.c~dax-provide-diagnostics-for-pmd-mapping-failures
+++ a/fs/dax.c
@@ -557,6 +557,14 @@ EXPORT_SYMBOL_GPL(dax_fault);
*/
#define PG_PMD_COLOUR ((PMD_SIZE >> PAGE_SHIFT) - 1)
+static void dax_pmd_dbg(struct block_device *bdev, unsigned long address,
+ const char *reason)
+{
+ pr_debug("%s%s dax_pmd: %s addr: %lx fallback: %s\n", bdev
+ ? dev_name(part_to_dev(bdev->bd_part)) : "", bdev
+ ? ": " : "", current->comm, address, reason);
+}
+
int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address,
pmd_t *pmd, unsigned int flags, get_block_t get_block,
dax_iodone_t complete_unwritten)
@@ -568,7 +576,7 @@ int __dax_pmd_fault(struct vm_area_struc
unsigned blkbits = inode->i_blkbits;
unsigned long pmd_addr = address & PMD_MASK;
bool write = flags & FAULT_FLAG_WRITE;
- struct block_device *bdev;
+ struct block_device *bdev = NULL;
pgoff_t size, pgoff;
sector_t block;
int result = 0;
@@ -580,21 +588,29 @@ int __dax_pmd_fault(struct vm_area_struc
/* Fall back to PTEs if we're going to COW */
if (write && !(vma->vm_flags & VM_SHARED)) {
split_huge_pmd(vma, pmd, address);
+ dax_pmd_dbg(bdev, address, "cow write");
return VM_FAULT_FALLBACK;
}
/* If the PMD would extend outside the VMA */
- if (pmd_addr < vma->vm_start)
+ if (pmd_addr < vma->vm_start) {
+ dax_pmd_dbg(bdev, address, "vma start unaligned");
return VM_FAULT_FALLBACK;
- if ((pmd_addr + PMD_SIZE) > vma->vm_end)
+ }
+ if ((pmd_addr + PMD_SIZE) > vma->vm_end) {
+ dax_pmd_dbg(bdev, address, "vma end unaligned");
return VM_FAULT_FALLBACK;
+ }
pgoff = linear_page_index(vma, pmd_addr);
size = (i_size_read(inode) + PAGE_SIZE - 1) >> PAGE_SHIFT;
if (pgoff >= size)
return VM_FAULT_SIGBUS;
/* If the PMD would cover blocks out of the file */
- if ((pgoff | PG_PMD_COLOUR) >= size)
+ if ((pgoff | PG_PMD_COLOUR) >= size) {
+ dax_pmd_dbg(bdev, address,
+ "offset + huge page size > file size");
return VM_FAULT_FALLBACK;
+ }
memset(&bh, 0, sizeof(bh));
block = (sector_t)pgoff << (PAGE_SHIFT - blkbits);
@@ -610,8 +626,10 @@ int __dax_pmd_fault(struct vm_area_struc
* just fall back to PTEs. Calling get_block 512 times in a loop
* would be silly.
*/
- if (!buffer_size_valid(&bh) || bh.b_size < PMD_SIZE)
+ if (!buffer_size_valid(&bh) || bh.b_size < PMD_SIZE) {
+ dax_pmd_dbg(bdev, address, "block allocation size invalid");
goto fallback;
+ }
/*
* If we allocated new storage, make sure no process has any
@@ -634,23 +652,33 @@ int __dax_pmd_fault(struct vm_area_struc
result = VM_FAULT_SIGBUS;
goto out;
}
- if ((pgoff | PG_PMD_COLOUR) >= size)
+ if ((pgoff | PG_PMD_COLOUR) >= size) {
+ dax_pmd_dbg(bdev, address, "pgoff unaligned");
goto fallback;
+ }
if (!write && !buffer_mapped(&bh) && buffer_uptodate(&bh)) {
spinlock_t *ptl;
pmd_t entry;
struct page *zero_page = get_huge_zero_page();
- if (unlikely(!zero_page))
+ if (unlikely(!zero_page)) {
+ dax_pmd_dbg(bdev, address, "no zero page");
goto fallback;
+ }
ptl = pmd_lock(vma->vm_mm, pmd);
if (!pmd_none(*pmd)) {
spin_unlock(ptl);
+ dax_pmd_dbg(bdev, address, "pmd already present");
goto fallback;
}
+ dev_dbg(part_to_dev(bdev->bd_part),
+ "%s: %s addr: %lx pfn: <zero> sect: %llx\n",
+ __func__, current->comm, address,
+ (unsigned long long) to_sector(&bh, inode));
+
entry = mk_pmd(zero_page, vma->vm_page_prot);
entry = pmd_mkhuge(entry);
set_pmd_at(vma->vm_mm, pmd_addr, pmd, entry);
@@ -667,8 +695,13 @@ int __dax_pmd_fault(struct vm_area_struc
result = VM_FAULT_SIGBUS;
goto out;
}
- if (length < PMD_SIZE
- || (pfn_t_to_pfn(dax.pfn) & PG_PMD_COLOUR)) {
+ if (length < PMD_SIZE) {
+ dax_pmd_dbg(bdev, address, "dax-length too small");
+ dax_unmap_atomic(bdev, &dax);
+ goto fallback;
+ }
+ if (pfn_t_to_pfn(dax.pfn) & PG_PMD_COLOUR) {
+ dax_pmd_dbg(bdev, address, "pfn unaligned");
dax_unmap_atomic(bdev, &dax);
goto fallback;
}
@@ -679,6 +712,7 @@ int __dax_pmd_fault(struct vm_area_struc
*/
if (pfn_t_has_page(dax.pfn)) {
dax_unmap_atomic(bdev, &dax);
+ dax_pmd_dbg(bdev, address, "pfn not in memmap");
goto fallback;
}
@@ -691,6 +725,11 @@ int __dax_pmd_fault(struct vm_area_struc
}
dax_unmap_atomic(bdev, &dax);
+ dev_dbg(part_to_dev(bdev->bd_part),
+ "%s: %s addr: %lx pfn: %lx sect: %llx\n",
+ __func__, current->comm, address,
+ pfn_t_to_pfn(dax.pfn),
+ (unsigned long long) dax.sector);
result |= vmf_insert_pfn_pmd(vma, address, pmd,
dax.pfn, write);
}
_
Patches currently in -mm which might be from dan.j.williams@xxxxxxxxx are
scatterlist-fix-sg_phys-masking.patch
pmem-dax-clean-up-clear_pmem.patch
dax-increase-granularity-of-dax_clear_blocks-operations.patch
dax-guarantee-page-aligned-results-from-bdev_direct_access.patch
dax-fix-lifetime-of-in-kernel-dax-mappings-with-dax_map_atomic.patch
dax-fix-lifetime-of-in-kernel-dax-mappings-with-dax_map_atomic-v3.patch
um-kill-pfn_t.patch
kvm-rename-pfn_t-to-kvm_pfn_t.patch
mm-dax-pmem-introduce-pfn_t.patch
mm-dax-pmem-introduce-pfn_t-v3.patch
mm-introduce-find_dev_pagemap.patch
x86-mm-introduce-vmem_altmap-to-augment-vmemmap_populate.patch
libnvdimm-pfn-pmem-allocate-memmap-array-in-persistent-memory.patch
avr32-convert-to-asm-generic-memory_modelh.patch
hugetlb-fix-compile-error-on-tile.patch
frv-fix-compiler-warning-from-definition-of-__pmd.patch
x86-mm-introduce-_page_devmap.patch
mm-dax-gpu-convert-vm_insert_mixed-to-pfn_t.patch
mm-dax-convert-vmf_insert_pfn_pmd-to-pfn_t.patch
list-introduce-list_del_poison.patch
libnvdimm-pmem-move-request_queue-allocation-earlier-in-probe.patch
mm-dax-pmem-introduce-getput_dev_pagemap-for-dax-gup.patch
mm-dax-dax-pmd-vs-thp-pmd-vs-hugetlbfs-pmd.patch
mm-x86-get_user_pages-for-dax-mappings.patch
dax-provide-diagnostics-for-pmd-mapping-failures.patch
dax-re-enable-dax-pmd-mappings.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
