Revision e2093926a098a8ccf0f1d10f6df8dad452cb28d3 authored by Ross Zwisler on 02 June 2017, 21:46:37 UTC, committed by Linus Torvalds on 02 June 2017, 22:07:37 UTC
We currently have two related PMD vs PTE races in the DAX code. These can both be easily triggered by having two threads reading and writing simultaneously to the same private mapping, with the key being that private mapping reads can be handled with PMDs but private mapping writes are always handled with PTEs so that we can COW. Here is the first race: CPU 0 CPU 1 (private mapping write) __handle_mm_fault() create_huge_pmd() - FALLBACK handle_pte_fault() passes check for pmd_devmap() (private mapping read) __handle_mm_fault() create_huge_pmd() dax_iomap_pmd_fault() inserts PMD dax_iomap_pte_fault() does a PTE fault, but we already have a DAX PMD installed in our page tables at this spot. Here's the second race: CPU 0 CPU 1 (private mapping read) __handle_mm_fault() passes check for pmd_none() create_huge_pmd() dax_iomap_pmd_fault() inserts PMD (private mapping write) __handle_mm_fault() create_huge_pmd() - FALLBACK (private mapping read) __handle_mm_fault() passes check for pmd_none() create_huge_pmd() handle_pte_fault() dax_iomap_pte_fault() inserts PTE dax_iomap_pmd_fault() inserts PMD, but we already have a PTE at this spot. The core of the issue is that while there is isolation between faults to the same range in the DAX fault handlers via our DAX entry locking, there is no isolation between faults in the code in mm/memory.c. This means for instance that this code in __handle_mm_fault() can run: if (pmd_none(*vmf.pmd) && transparent_hugepage_enabled(vma)) { ret = create_huge_pmd(&vmf); But by the time we actually get to run the fault handler called by create_huge_pmd(), the PMD is no longer pmd_none() because a racing PTE fault has installed a normal PMD here as a parent. This is the cause of the 2nd race. The first race is similar - there is the following check in handle_pte_fault(): } else { /* See comment in pte_alloc_one_map() */ if (pmd_devmap(*vmf->pmd) || pmd_trans_unstable(vmf->pmd)) return 0; So if a pmd_devmap() PMD (a DAX PMD) has been installed at vmf->pmd, we will bail and retry the fault. This is correct, but there is nothing preventing the PMD from being installed after this check but before we actually get to the DAX PTE fault handlers. In my testing these races result in the following types of errors: BUG: Bad rss-counter state mm:ffff8800a817d280 idx:1 val:1 BUG: non-zero nr_ptes on freeing mm: 15 Fix this issue by having the DAX fault handlers verify that it is safe to continue their fault after they have taken an entry lock to block other racing faults. [ross.zwisler@linux.intel.com: improve fix for colliding PMD & PTE entries] Link: http://lkml.kernel.org/r/20170526195932.32178-1-ross.zwisler@linux.intel.com Link: http://lkml.kernel.org/r/20170522215749.23516-2-ross.zwisler@linux.intel.com Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com> Reported-by: Pawel Lebioda <pawel.lebioda@intel.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: "Darrick J. Wong" <darrick.wong@oracle.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@lst.de> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: Matthew Wilcox <mawilcox@microsoft.com> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com> Cc: Pawel Lebioda <pawel.lebioda@intel.com> Cc: Dave Jiang <dave.jiang@intel.com> Cc: Xiong Zhou <xzhou@redhat.com> Cc: Eryu Guan <eguan@redhat.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent d0f0931
scsi_devinfo.h
#ifndef _SCSI_SCSI_DEVINFO_H
#define _SCSI_SCSI_DEVINFO_H
/*
* Flags for SCSI devices that need special treatment
*/
#define BLIST_NOLUN 0x001 /* Only scan LUN 0 */
#define BLIST_FORCELUN 0x002 /* Known to have LUNs, force scanning,
deprecated: Use max_luns=N */
#define BLIST_BORKEN 0x004 /* Flag for broken handshaking */
#define BLIST_KEY 0x008 /* unlock by special command */
#define BLIST_SINGLELUN 0x010 /* Do not use LUNs in parallel */
#define BLIST_NOTQ 0x020 /* Buggy Tagged Command Queuing */
#define BLIST_SPARSELUN 0x040 /* Non consecutive LUN numbering */
#define BLIST_MAX5LUN 0x080 /* Avoid LUNS >= 5 */
#define BLIST_ISROM 0x100 /* Treat as (removable) CD-ROM */
#define BLIST_LARGELUN 0x200 /* LUNs past 7 on a SCSI-2 device */
#define BLIST_INQUIRY_36 0x400 /* override additional length field */
#define BLIST_INQUIRY_58 0x800 /* ... for broken inquiry responses */
#define BLIST_NOSTARTONADD 0x1000 /* do not do automatic start on add */
#define BLIST_MS_SKIP_PAGE_08 0x2000 /* do not send ms page 0x08 */
#define BLIST_MS_SKIP_PAGE_3F 0x4000 /* do not send ms page 0x3f */
#define BLIST_USE_10_BYTE_MS 0x8000 /* use 10 byte ms before 6 byte ms */
#define BLIST_MS_192_BYTES_FOR_3F 0x10000 /* 192 byte ms page 0x3f request */
#define BLIST_REPORTLUN2 0x20000 /* try REPORT_LUNS even for SCSI-2 devs
(if HBA supports more than 8 LUNs) */
#define BLIST_NOREPORTLUN 0x40000 /* don't try REPORT_LUNS scan (SCSI-3 devs) */
#define BLIST_NOT_LOCKABLE 0x80000 /* don't use PREVENT-ALLOW commands */
#define BLIST_NO_ULD_ATTACH 0x100000 /* device is actually for RAID config */
#define BLIST_SELECT_NO_ATN 0x200000 /* select without ATN */
#define BLIST_RETRY_HWERROR 0x400000 /* retry HARDWARE_ERROR */
#define BLIST_MAX_512 0x800000 /* maximum 512 sector cdb length */
#define BLIST_ATTACH_PQ3 0x1000000 /* Scan: Attach to PQ3 devices */
#define BLIST_NO_DIF 0x2000000 /* Disable T10 PI (DIF) */
#define BLIST_SKIP_VPD_PAGES 0x4000000 /* Ignore SBC-3 VPD pages */
#define BLIST_SCSI3LUN 0x8000000 /* Scan more than 256 LUNs
for sequential scan */
#define BLIST_TRY_VPD_PAGES 0x10000000 /* Attempt to read VPD pages */
#define BLIST_NO_RSOC 0x20000000 /* don't try to issue RSOC */
#define BLIST_MAX_1024 0x40000000 /* maximum 1024 sector cdb length */
#define BLIST_SYNC_ALUA 0x80000000 /* Synchronous ALUA commands */
#endif
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...