* mm/mprotect.c:140:48: sparse: sparse: context imbalance in 'change_pte_range' - different lock contexts for basic block
@ 2021-06-09 4:35 kernel test robot
0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2021-06-09 4:35 UTC (permalink / raw)
To: kbuild
[-- Attachment #1: Type: text/plain, Size: 16576 bytes --]
CC: kbuild-all(a)lists.01.org
CC: linux-kernel(a)vger.kernel.org
TO: Helge Deller <deller@gmx.de>
CC: John David Anglin <dave.anglin@bell.net>
tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head: 368094df48e680fa51cedb68537408cfa64b788e
commit: b7795074a04669d0a023babf786d29bf67c68783 parisc: Optimize per-pagetable spinlocks
date: 4 months ago
:::::: branch date: 11 hours ago
:::::: commit date: 4 months ago
config: parisc-randconfig-s032-20210607 (attached as .config)
compiler: hppa64-linux-gcc (GCC) 9.3.0
reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# apt-get install sparse
# sparse version: v0.6.3-341-g8af24329-dirty
# https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=b7795074a04669d0a023babf786d29bf67c68783
git remote add linus https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
git fetch --no-tags linus master
git checkout b7795074a04669d0a023babf786d29bf67c68783
# save the attached .config to linux build tree
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' W=1 ARCH=parisc
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
sparse warnings: (new ones prefixed by >>)
>> mm/mprotect.c:140:48: sparse: sparse: context imbalance in 'change_pte_range' - different lock contexts for basic block
--
mm/memory.c:698:1: sparse: sparse: context imbalance in 'copy_nonpresent_pte' - different lock contexts for basic block
mm/memory.c:900:9: sparse: sparse: context imbalance in 'copy_pte_range' - different lock contexts for basic block
mm/memory.c: note: in included file (through include/linux/pgtable.h, arch/parisc/include/asm/io.h, include/linux/io.h, ...):
arch/parisc/include/asm/pgtable.h:451:9: sparse: sparse: context imbalance in 'zap_pte_range' - different lock contexts for basic block
mm/memory.c:1623:16: sparse: sparse: context imbalance in '__get_locked_pte' - different lock contexts for basic block
mm/memory.c:1644:9: sparse: sparse: context imbalance in 'insert_page_into_pte_locked' - different lock contexts for basic block
mm/memory.c:1672:9: sparse: sparse: context imbalance in 'insert_page' - different lock contexts for basic block
mm/memory.c:1964:9: sparse: sparse: context imbalance in 'insert_pfn' - different lock contexts for basic block
mm/memory.c:2183:17: sparse: sparse: context imbalance in 'remap_pte_range' - different lock contexts for basic block
mm/memory.c:2419:17: sparse: sparse: context imbalance in 'apply_to_pte_range' - unexpected unlock
mm/memory.c:2912:17: sparse: sparse: context imbalance in 'wp_page_copy' - different lock contexts for basic block
mm/memory.c:3022:17: sparse: sparse: context imbalance in 'wp_pfn_shared' - unexpected unlock
mm/memory.c:3085:19: sparse: sparse: context imbalance in 'do_wp_page' - different lock contexts for basic block
mm/memory.c:3580:9: sparse: sparse: context imbalance in 'do_anonymous_page' - different lock contexts for basic block
mm/memory.c:3657:19: sparse: sparse: context imbalance in 'pte_alloc_one_map' - different lock contexts for basic block
mm/memory.c:3842:9: sparse: sparse: context imbalance in 'alloc_set_pte' - different lock contexts for basic block
mm/memory.c:3884:17: sparse: sparse: context imbalance in 'finish_fault' - unexpected unlock
mm/memory.c:3993:9: sparse: sparse: context imbalance in 'do_fault_around' - unexpected unlock
>> mm/memory.c:4209:32: sparse: sparse: context imbalance in 'do_numa_page' - different lock contexts for basic block
mm/memory.c:4426:9: sparse: sparse: context imbalance in 'handle_pte_fault' - different lock contexts for basic block
mm/memory.c:4712:5: sparse: sparse: context imbalance in 'follow_pte' - different lock contexts for basic block
mm/memory.c:4802:9: sparse: sparse: context imbalance in 'follow_pfn' - unexpected unlock
vim +/change_pte_range +140 mm/mprotect.c
36f881883c5794 Kirill A. Shutemov 2015-06-24 37
4b10e7d562c90d Mel Gorman 2012-10-25 38 static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
c1e6098b23bb46 Peter Zijlstra 2006-09-25 39 unsigned long addr, unsigned long end, pgprot_t newprot,
58705444c45b3c Peter Xu 2020-04-06 40 unsigned long cp_flags)
^1da177e4c3f41 Linus Torvalds 2005-04-16 41 {
0697212a411c1d Christoph Lameter 2006-06-23 42 pte_t *pte, oldpte;
705e87c0c3c384 Hugh Dickins 2005-10-29 43 spinlock_t *ptl;
7da4d641c58d20 Peter Zijlstra 2012-11-19 44 unsigned long pages = 0;
3e32158767b04d Andi Kleen 2016-12-12 45 int target_node = NUMA_NO_NODE;
58705444c45b3c Peter Xu 2020-04-06 46 bool dirty_accountable = cp_flags & MM_CP_DIRTY_ACCT;
58705444c45b3c Peter Xu 2020-04-06 47 bool prot_numa = cp_flags & MM_CP_PROT_NUMA;
292924b2602474 Peter Xu 2020-04-06 48 bool uffd_wp = cp_flags & MM_CP_UFFD_WP;
292924b2602474 Peter Xu 2020-04-06 49 bool uffd_wp_resolve = cp_flags & MM_CP_UFFD_WP_RESOLVE;
^1da177e4c3f41 Linus Torvalds 2005-04-16 50
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 51 /*
c1e8d7c6a7a682 Michel Lespinasse 2020-06-08 52 * Can be called with only the mmap_lock for reading by
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 53 * prot_numa so we must check the pmd isn't constantly
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 54 * changing from under us from pmd_none to pmd_trans_huge
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 55 * and/or the other way around.
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 56 */
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 57 if (pmd_trans_unstable(pmd))
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 58 return 0;
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 59
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 60 /*
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 61 * The pmd points to a regular pte so the pmd can't change
c1e8d7c6a7a682 Michel Lespinasse 2020-06-08 62 * from under us even if the mmap_lock is only hold for
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 63 * reading.
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 64 */
175ad4f1e7a29c Andrea Arcangeli 2017-02-22 65 pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
1ad9f620c3a22f Mel Gorman 2014-04-07 66
3e32158767b04d Andi Kleen 2016-12-12 67 /* Get target node for single threaded private VMAs */
3e32158767b04d Andi Kleen 2016-12-12 68 if (prot_numa && !(vma->vm_flags & VM_SHARED) &&
3e32158767b04d Andi Kleen 2016-12-12 69 atomic_read(&vma->vm_mm->mm_users) == 1)
3e32158767b04d Andi Kleen 2016-12-12 70 target_node = numa_node_id();
3e32158767b04d Andi Kleen 2016-12-12 71
3ea277194daaea Mel Gorman 2017-08-02 72 flush_tlb_batched_pending(vma->vm_mm);
6606c3e0da5360 Zachary Amsden 2006-09-30 73 arch_enter_lazy_mmu_mode();
^1da177e4c3f41 Linus Torvalds 2005-04-16 74 do {
0697212a411c1d Christoph Lameter 2006-06-23 75 oldpte = *pte;
0697212a411c1d Christoph Lameter 2006-06-23 76 if (pte_present(oldpte)) {
^1da177e4c3f41 Linus Torvalds 2005-04-16 77 pte_t ptent;
b191f9b106ea1a Mel Gorman 2015-03-25 78 bool preserve_write = prot_numa && pte_write(oldpte);
^1da177e4c3f41 Linus Torvalds 2005-04-16 79
e944fd67b625c0 Mel Gorman 2015-02-12 80 /*
e944fd67b625c0 Mel Gorman 2015-02-12 81 * Avoid trapping faults against the zero or KSM
e944fd67b625c0 Mel Gorman 2015-02-12 82 * pages. See similar comment in change_huge_pmd.
e944fd67b625c0 Mel Gorman 2015-02-12 83 */
e944fd67b625c0 Mel Gorman 2015-02-12 84 if (prot_numa) {
e944fd67b625c0 Mel Gorman 2015-02-12 85 struct page *page;
e944fd67b625c0 Mel Gorman 2015-02-12 86
a818f5363a0eba Huang Ying 2019-11-30 87 /* Avoid TLB flush if possible */
a818f5363a0eba Huang Ying 2019-11-30 88 if (pte_protnone(oldpte))
a818f5363a0eba Huang Ying 2019-11-30 89 continue;
a818f5363a0eba Huang Ying 2019-11-30 90
e944fd67b625c0 Mel Gorman 2015-02-12 91 page = vm_normal_page(vma, addr, oldpte);
e944fd67b625c0 Mel Gorman 2015-02-12 92 if (!page || PageKsm(page))
e944fd67b625c0 Mel Gorman 2015-02-12 93 continue;
10c1045f28e86a Mel Gorman 2015-02-12 94
859d4adc3415a6 Henry Willard 2018-01-31 95 /* Also skip shared copy-on-write pages */
859d4adc3415a6 Henry Willard 2018-01-31 96 if (is_cow_mapping(vma->vm_flags) &&
859d4adc3415a6 Henry Willard 2018-01-31 97 page_mapcount(page) != 1)
859d4adc3415a6 Henry Willard 2018-01-31 98 continue;
859d4adc3415a6 Henry Willard 2018-01-31 99
09a913a7a947fb Mel Gorman 2018-04-10 100 /*
09a913a7a947fb Mel Gorman 2018-04-10 101 * While migration can move some dirty pages,
09a913a7a947fb Mel Gorman 2018-04-10 102 * it cannot move them all from MIGRATE_ASYNC
09a913a7a947fb Mel Gorman 2018-04-10 103 * context.
09a913a7a947fb Mel Gorman 2018-04-10 104 */
9de4f22a60f731 Huang Ying 2020-04-06 105 if (page_is_file_lru(page) && PageDirty(page))
09a913a7a947fb Mel Gorman 2018-04-10 106 continue;
09a913a7a947fb Mel Gorman 2018-04-10 107
3e32158767b04d Andi Kleen 2016-12-12 108 /*
3e32158767b04d Andi Kleen 2016-12-12 109 * Don't mess with PTEs if page is already on the node
3e32158767b04d Andi Kleen 2016-12-12 110 * a single-threaded process is running on.
3e32158767b04d Andi Kleen 2016-12-12 111 */
3e32158767b04d Andi Kleen 2016-12-12 112 if (target_node == page_to_nid(page))
3e32158767b04d Andi Kleen 2016-12-12 113 continue;
e944fd67b625c0 Mel Gorman 2015-02-12 114 }
e944fd67b625c0 Mel Gorman 2015-02-12 115
04a8645304500b Aneesh Kumar K.V 2019-03-05 116 oldpte = ptep_modify_prot_start(vma, addr, pte);
04a8645304500b Aneesh Kumar K.V 2019-03-05 117 ptent = pte_modify(oldpte, newprot);
b191f9b106ea1a Mel Gorman 2015-03-25 118 if (preserve_write)
288bc54949fc26 Aneesh Kumar K.V 2017-02-24 119 ptent = pte_mk_savedwrite(ptent);
8a0516ed8b90c9 Mel Gorman 2015-02-12 120
292924b2602474 Peter Xu 2020-04-06 121 if (uffd_wp) {
292924b2602474 Peter Xu 2020-04-06 122 ptent = pte_wrprotect(ptent);
292924b2602474 Peter Xu 2020-04-06 123 ptent = pte_mkuffd_wp(ptent);
292924b2602474 Peter Xu 2020-04-06 124 } else if (uffd_wp_resolve) {
292924b2602474 Peter Xu 2020-04-06 125 /*
292924b2602474 Peter Xu 2020-04-06 126 * Leave the write bit to be handled
292924b2602474 Peter Xu 2020-04-06 127 * by PF interrupt handler, then
292924b2602474 Peter Xu 2020-04-06 128 * things like COW could be properly
292924b2602474 Peter Xu 2020-04-06 129 * handled.
292924b2602474 Peter Xu 2020-04-06 130 */
292924b2602474 Peter Xu 2020-04-06 131 ptent = pte_clear_uffd_wp(ptent);
292924b2602474 Peter Xu 2020-04-06 132 }
292924b2602474 Peter Xu 2020-04-06 133
8a0516ed8b90c9 Mel Gorman 2015-02-12 134 /* Avoid taking write faults for known dirty pages */
64e455079e1bd7 Peter Feiner 2014-10-13 135 if (dirty_accountable && pte_dirty(ptent) &&
64e455079e1bd7 Peter Feiner 2014-10-13 136 (pte_soft_dirty(ptent) ||
8a0516ed8b90c9 Mel Gorman 2015-02-12 137 !(vma->vm_flags & VM_SOFTDIRTY))) {
9d85d5863fa481 Aneesh Kumar K.V 2014-02-12 138 ptent = pte_mkwrite(ptent);
4b10e7d562c90d Mel Gorman 2012-10-25 139 }
04a8645304500b Aneesh Kumar K.V 2019-03-05 @140 ptep_modify_prot_commit(vma, addr, pte, oldpte, ptent);
7da4d641c58d20 Peter Zijlstra 2012-11-19 141 pages++;
f45ec5ff16a75f Peter Xu 2020-04-06 142 } else if (is_swap_pte(oldpte)) {
0697212a411c1d Christoph Lameter 2006-06-23 143 swp_entry_t entry = pte_to_swp_entry(oldpte);
f45ec5ff16a75f Peter Xu 2020-04-06 144 pte_t newpte;
0697212a411c1d Christoph Lameter 2006-06-23 145
0697212a411c1d Christoph Lameter 2006-06-23 146 if (is_write_migration_entry(entry)) {
0697212a411c1d Christoph Lameter 2006-06-23 147 /*
0697212a411c1d Christoph Lameter 2006-06-23 148 * A protection check is difficult so
0697212a411c1d Christoph Lameter 2006-06-23 149 * just be safe and disable write
0697212a411c1d Christoph Lameter 2006-06-23 150 */
0697212a411c1d Christoph Lameter 2006-06-23 151 make_migration_entry_read(&entry);
c3d16e16522fe3 Cyrill Gorcunov 2013-10-16 152 newpte = swp_entry_to_pte(entry);
c3d16e16522fe3 Cyrill Gorcunov 2013-10-16 153 if (pte_swp_soft_dirty(oldpte))
c3d16e16522fe3 Cyrill Gorcunov 2013-10-16 154 newpte = pte_swp_mksoft_dirty(newpte);
f45ec5ff16a75f Peter Xu 2020-04-06 155 if (pte_swp_uffd_wp(oldpte))
f45ec5ff16a75f Peter Xu 2020-04-06 156 newpte = pte_swp_mkuffd_wp(newpte);
f45ec5ff16a75f Peter Xu 2020-04-06 157 } else if (is_write_device_private_entry(entry)) {
5042db43cc26f5 Jérôme Glisse 2017-09-08 158 /*
5042db43cc26f5 Jérôme Glisse 2017-09-08 159 * We do not preserve soft-dirtiness. See
5042db43cc26f5 Jérôme Glisse 2017-09-08 160 * copy_one_pte() for explanation.
5042db43cc26f5 Jérôme Glisse 2017-09-08 161 */
5042db43cc26f5 Jérôme Glisse 2017-09-08 162 make_device_private_entry_read(&entry);
5042db43cc26f5 Jérôme Glisse 2017-09-08 163 newpte = swp_entry_to_pte(entry);
f45ec5ff16a75f Peter Xu 2020-04-06 164 if (pte_swp_uffd_wp(oldpte))
f45ec5ff16a75f Peter Xu 2020-04-06 165 newpte = pte_swp_mkuffd_wp(newpte);
f45ec5ff16a75f Peter Xu 2020-04-06 166 } else {
f45ec5ff16a75f Peter Xu 2020-04-06 167 newpte = oldpte;
f45ec5ff16a75f Peter Xu 2020-04-06 168 }
f45ec5ff16a75f Peter Xu 2020-04-06 169
f45ec5ff16a75f Peter Xu 2020-04-06 170 if (uffd_wp)
f45ec5ff16a75f Peter Xu 2020-04-06 171 newpte = pte_swp_mkuffd_wp(newpte);
f45ec5ff16a75f Peter Xu 2020-04-06 172 else if (uffd_wp_resolve)
f45ec5ff16a75f Peter Xu 2020-04-06 173 newpte = pte_swp_clear_uffd_wp(newpte);
5042db43cc26f5 Jérôme Glisse 2017-09-08 174
f45ec5ff16a75f Peter Xu 2020-04-06 175 if (!pte_same(oldpte, newpte)) {
f45ec5ff16a75f Peter Xu 2020-04-06 176 set_pte_at(vma->vm_mm, addr, pte, newpte);
5042db43cc26f5 Jérôme Glisse 2017-09-08 177 pages++;
5042db43cc26f5 Jérôme Glisse 2017-09-08 178 }
e920e14ca29b0b Mel Gorman 2013-10-07 179 }
^1da177e4c3f41 Linus Torvalds 2005-04-16 180 } while (pte++, addr += PAGE_SIZE, addr != end);
6606c3e0da5360 Zachary Amsden 2006-09-30 181 arch_leave_lazy_mmu_mode();
705e87c0c3c384 Hugh Dickins 2005-10-29 182 pte_unmap_unlock(pte - 1, ptl);
7da4d641c58d20 Peter Zijlstra 2012-11-19 183
7da4d641c58d20 Peter Zijlstra 2012-11-19 184 return pages;
^1da177e4c3f41 Linus Torvalds 2005-04-16 185 }
^1da177e4c3f41 Linus Torvalds 2005-04-16 186
:::::: The code at line 140 was first introduced by commit
:::::: 04a8645304500be88b3345b65fef7efe58016166 mm: update ptep_modify_prot_commit to take old pte value as arg
:::::: TO: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
:::::: CC: Linus Torvalds <torvalds@linux-foundation.org>
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 37093 bytes --]
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2021-06-09 4:35 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-06-09 4:35 mm/mprotect.c:140:48: sparse: sparse: context imbalance in 'change_pte_range' - different lock contexts for basic block kernel test robot
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.