| Message ID | 20170524154728.2492-1-punit.agrawal@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, 24 May 2017 16:47:28 +0100 Punit Agrawal <punit.agrawal@arm.com> wrote: > On failing to migrate a page, soft_offline_huge_page() performs the > necessary update to the hugepage ref-count. When > !hugepage_migration_supported() , unmap_and_move_hugepage() also > decrements the page ref-count for the hugepage. The combined behaviour > leaves the ref-count in an inconsistent state. > > This leads to soft lockups when running the overcommitted hugepage test > from mce-tests suite. > > Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 > soft offline: 0x83ed600: migration failed 1, type > 1fffc00000008008 (uptodate|head) > INFO: rcu_preempt detected stalls on CPUs/tasks: > Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 > (detected by 7, t=5254 jiffies, g=963, c=962, q=321) > thugetlb_overco R running task 0 2715 2685 0x00000008 > Call trace: > [<ffff000008089f90>] dump_backtrace+0x0/0x268 > [<ffff00000808a2d4>] show_stack+0x24/0x30 > [<ffff000008100d34>] sched_show_task+0x134/0x180 > [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c > [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 > [<ffff000008143a3c>] update_process_times+0x34/0x60 > [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 > [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 > [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 > [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 > [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 > [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 > [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 > [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 > [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 > > Fix this by dropping the ref-count decrement in > unmap_and_move_hugepage() when !hugepage_migration_supported(). > > Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") > Reported-by: Manoj Iyer <manoj.iyer@canonical.com> > Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> > Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> > Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> > Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> > Cc: Christoph Lameter <cl@linux.com> 32665f2bbfed was three years ago. Do you have any theory as to why this took so long to be detected? And do you believe a -stable backport is warranted?
On Wed, 24 May 2017, Punit Agrawal wrote: > On failing to migrate a page, soft_offline_huge_page() performs the > necessary update to the hugepage ref-count. When > !hugepage_migration_supported() , unmap_and_move_hugepage() also > decrements the page ref-count for the hugepage. The combined behaviour > leaves the ref-count in an inconsistent state. > > This leads to soft lockups when running the overcommitted hugepage test > from mce-tests suite. > > Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 > soft offline: 0x83ed600: migration failed 1, type > 1fffc00000008008 (uptodate|head) > INFO: rcu_preempt detected stalls on CPUs/tasks: > Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 > (detected by 7, t=5254 jiffies, g=963, c=962, q=321) > thugetlb_overco R running task 0 2715 2685 0x00000008 > Call trace: > [<ffff000008089f90>] dump_backtrace+0x0/0x268 > [<ffff00000808a2d4>] show_stack+0x24/0x30 > [<ffff000008100d34>] sched_show_task+0x134/0x180 > [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c > [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 > [<ffff000008143a3c>] update_process_times+0x34/0x60 > [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 > [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 > [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 > [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 > [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 > [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 > [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 > [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 > [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 > > Fix this by dropping the ref-count decrement in > unmap_and_move_hugepage() when !hugepage_migration_supported(). > > Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") > Reported-by: Manoj Iyer <manoj.iyer@canonical.com> > Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> > Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> > Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> > Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> > Cc: Christoph Lameter <cl@linux.com> > --- > Hi Andrew, > > We ran into this bug when working towards enabling memory corruption > on arm64. The patch was tested on an arm64 platform running v4.12-rc2 > with the series to enable memory corruption handling[0]. > > Please consider merging as a fix for the 4.12 release. > > Thanks, > Punit > I applied this patch applied to Ubuntu Zesty (4.10) kernel and tested on QDF2400 platform with mce-test without config migration enabled. == dmesg == [ 91.569358] Soft offlining page 0x1763c00 at 0x400000000000 [ 91.569364] soft offline: 0x1763c00: migration failed 1, type 100000000008008 [ 150.282911] Soft offlining page 0x1763c00 at 0x400000000000 [ 150.282917] soft offline: 0x1763c00: migration failed 1, type 100000000008008 The mce-test failed as expected but did not encounter the soft lockups. (The test case might have an error it is missing an 'echo' in failure case.) $ sudo ./run_hugepage_overcommit.sh *************************************************************************** Pay attention: This test checks that hugepage soft-offlining works under overcommitting. *************************************************************************** ------------------------------------- TestCase ./thugetlb_overcommit 1 FAIL: migration failed. Unpoisoning. Num of Executed Test Case: 1 Num of Failed Case: 1 Tested-By: Manoj Iyer <manoj.iyer@canonical.com> Thanks Manoj Iyer > [0] https://www.spinics.net/lists/arm-kernel/msg581657.html > --- > mm/migrate.c | 4 +--- > 1 file changed, 1 insertion(+), 3 deletions(-) > > diff --git a/mm/migrate.c b/mm/migrate.c > index 89a0a1707f4c..187abd1526df 100644 > --- a/mm/migrate.c > +++ b/mm/migrate.c > @@ -1201,10 +1201,8 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, > * tables or check whether the hugepage is pmd-based or not before > * kicking migration. > */ > - if (!hugepage_migration_supported(page_hstate(hpage))) { > - putback_active_hugepage(hpage); > + if (!hugepage_migration_supported(page_hstate(hpage))) > return -ENOSYS; > - } > > new_hpage = get_new_page(hpage, private, &result); > if (!new_hpage) > -- ============================ Manoj Iyer Ubuntu/Canonical ARM Servers - Cloud ============================
On Wed, May 24, 2017 at 04:47:28PM +0100, Punit Agrawal wrote: > On failing to migrate a page, soft_offline_huge_page() performs the > necessary update to the hugepage ref-count. When > !hugepage_migration_supported() , unmap_and_move_hugepage() also > decrements the page ref-count for the hugepage. The combined behaviour > leaves the ref-count in an inconsistent state. > > This leads to soft lockups when running the overcommitted hugepage test > from mce-tests suite. > > Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 > soft offline: 0x83ed600: migration failed 1, type > 1fffc00000008008 (uptodate|head) > INFO: rcu_preempt detected stalls on CPUs/tasks: > Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 > (detected by 7, t=5254 jiffies, g=963, c=962, q=321) > thugetlb_overco R running task 0 2715 2685 0x00000008 > Call trace: > [<ffff000008089f90>] dump_backtrace+0x0/0x268 > [<ffff00000808a2d4>] show_stack+0x24/0x30 > [<ffff000008100d34>] sched_show_task+0x134/0x180 > [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c > [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 > [<ffff000008143a3c>] update_process_times+0x34/0x60 > [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 > [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 > [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 > [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 > [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 > [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 > [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 > [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 > [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 > > Fix this by dropping the ref-count decrement in > unmap_and_move_hugepage() when !hugepage_migration_supported(). > > Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") > Reported-by: Manoj Iyer <manoj.iyer@canonical.com> > Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> > Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> > Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> > Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> > Cc: Christoph Lameter <cl@linux.com> > > -- > Hi Andrew, > > We ran into this bug when working towards enabling memory corruption > on arm64. The patch was tested on an arm64 platform running v4.12-rc2 > with the series to enable memory corruption handling[0]. > > Please consider merging as a fix for the 4.12 release. > > Thanks, > Punit > > [0] https://www.spinics.net/lists/arm-kernel/msg581657.html > --- > mm/migrate.c | 4 +--- > 1 file changed, 1 insertion(+), 3 deletions(-) > > diff --git a/mm/migrate.c b/mm/migrate.c > index 89a0a1707f4c..187abd1526df 100644 > --- a/mm/migrate.c > +++ b/mm/migrate.c > @@ -1201,10 +1201,8 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, > * tables or check whether the hugepage is pmd-based or not before > * kicking migration. > */ > - if (!hugepage_migration_supported(page_hstate(hpage))) { > - putback_active_hugepage(hpage); Thank you for reporting and suggestion, Punit, Manoj. Simply dropping this putback_active_hugepage() may resume the failure counting issue addressed in 32665f2bbfed, so I would recommend to call putback_movable_pages() in failure path in soft_offline_huge_page(). @@ -1600,7 +1600,8 @@ static int soft_offline_huge_page(struct page *page, int flags) * only one hugepage pointed to by hpage, so we need not * run through the pagelist here. */ - putback_active_hugepage(hpage); + if (!list_empty(&pagelist)) + putback_movable_pages(&pagelist); if (ret > 0) ret = -EIO; } else { Could you check this works for you? Thanks, Naoya Horiguchi
On Wed, May 24, 2017 at 12:56:10PM -0700, Andrew Morton wrote: > On Wed, 24 May 2017 16:47:28 +0100 Punit Agrawal <punit.agrawal@arm.com> wrote: > > > On failing to migrate a page, soft_offline_huge_page() performs the > > necessary update to the hugepage ref-count. When > > !hugepage_migration_supported() , unmap_and_move_hugepage() also > > decrements the page ref-count for the hugepage. The combined behaviour > > leaves the ref-count in an inconsistent state. > > > > This leads to soft lockups when running the overcommitted hugepage test > > from mce-tests suite. > > > > Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 > > soft offline: 0x83ed600: migration failed 1, type > > 1fffc00000008008 (uptodate|head) > > INFO: rcu_preempt detected stalls on CPUs/tasks: > > Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 > > (detected by 7, t=5254 jiffies, g=963, c=962, q=321) > > thugetlb_overco R running task 0 2715 2685 0x00000008 > > Call trace: > > [<ffff000008089f90>] dump_backtrace+0x0/0x268 > > [<ffff00000808a2d4>] show_stack+0x24/0x30 > > [<ffff000008100d34>] sched_show_task+0x134/0x180 > > [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c > > [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 > > [<ffff000008143a3c>] update_process_times+0x34/0x60 > > [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 > > [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 > > [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 > > [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 > > [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 > > [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 > > [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 > > [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 > > [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 > > > > Fix this by dropping the ref-count decrement in > > unmap_and_move_hugepage() when !hugepage_migration_supported(). > > > > Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") > > Reported-by: Manoj Iyer <manoj.iyer@canonical.com> > > Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> > > Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> > > Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> > > Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> > > Cc: Christoph Lameter <cl@linux.com> > > 32665f2bbfed was three years ago. Do you have any theory as to why > this took so long to be detected? My per-release testing only ran for "hugepage_migration_supported() == true" setting (i.e. x86 with CONFIG_HUGETLB_PAGE=y). I need extend the coverage. And other arch's developers recently have come to have interest in hugepage migration. > And do you believe a -stable > backport is warranted? I agree to send the fix to stable, so the stable tag is wanted. Cc: stable@kernel.org # v3.14+ Thanks, Naoya Horiguchi
Andrew Morton <akpm@linux-foundation.org> writes: > On Wed, 24 May 2017 16:47:28 +0100 Punit Agrawal <punit.agrawal@arm.com> wrote: > >> On failing to migrate a page, soft_offline_huge_page() performs the >> necessary update to the hugepage ref-count. When >> !hugepage_migration_supported() , unmap_and_move_hugepage() also >> decrements the page ref-count for the hugepage. The combined behaviour >> leaves the ref-count in an inconsistent state. >> >> This leads to soft lockups when running the overcommitted hugepage test >> from mce-tests suite. >> >> Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 >> soft offline: 0x83ed600: migration failed 1, type >> 1fffc00000008008 (uptodate|head) >> INFO: rcu_preempt detected stalls on CPUs/tasks: >> Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 >> (detected by 7, t=5254 jiffies, g=963, c=962, q=321) >> thugetlb_overco R running task 0 2715 2685 0x00000008 >> Call trace: >> [<ffff000008089f90>] dump_backtrace+0x0/0x268 >> [<ffff00000808a2d4>] show_stack+0x24/0x30 >> [<ffff000008100d34>] sched_show_task+0x134/0x180 >> [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c >> [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 >> [<ffff000008143a3c>] update_process_times+0x34/0x60 >> [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 >> [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 >> [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 >> [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 >> [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 >> [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 >> [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 >> [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 >> [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 >> >> Fix this by dropping the ref-count decrement in >> unmap_and_move_hugepage() when !hugepage_migration_supported(). >> >> Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") >> Reported-by: Manoj Iyer <manoj.iyer@canonical.com> >> Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> >> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> >> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> >> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> >> Cc: Christoph Lameter <cl@linux.com> > > 32665f2bbfed was three years ago. Do you have any theory as to why > this took so long to be detected? This only triggers on systems that enable memory failure handling (ARCH_SUPPORTS_MEMORY_FAILURE) but not hugepage migration (!ARCH_ENABLE_HUGEPAGE_MIGRATION). I imagine this wasn't triggered as there aren't many systems running this configuration. > And do you believe a -stable backport is warranted? I'll defer to Horiguchi-san's judgement here. > > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo@kvack.org. For more info on Linux MM, > see: http://www.linux-mm.org/ . > Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> writes: > On Wed, May 24, 2017 at 04:47:28PM +0100, Punit Agrawal wrote: >> On failing to migrate a page, soft_offline_huge_page() performs the >> necessary update to the hugepage ref-count. When >> !hugepage_migration_supported() , unmap_and_move_hugepage() also >> decrements the page ref-count for the hugepage. The combined behaviour >> leaves the ref-count in an inconsistent state. >> >> This leads to soft lockups when running the overcommitted hugepage test >> from mce-tests suite. >> >> Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 >> soft offline: 0x83ed600: migration failed 1, type >> 1fffc00000008008 (uptodate|head) >> INFO: rcu_preempt detected stalls on CPUs/tasks: >> Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 >> (detected by 7, t=5254 jiffies, g=963, c=962, q=321) >> thugetlb_overco R running task 0 2715 2685 0x00000008 >> Call trace: >> [<ffff000008089f90>] dump_backtrace+0x0/0x268 >> [<ffff00000808a2d4>] show_stack+0x24/0x30 >> [<ffff000008100d34>] sched_show_task+0x134/0x180 >> [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c >> [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 >> [<ffff000008143a3c>] update_process_times+0x34/0x60 >> [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 >> [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 >> [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 >> [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 >> [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 >> [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 >> [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 >> [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 >> [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 >> >> Fix this by dropping the ref-count decrement in >> unmap_and_move_hugepage() when !hugepage_migration_supported(). >> >> Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") >> Reported-by: Manoj Iyer <manoj.iyer@canonical.com> >> Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> >> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> >> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> >> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> >> Cc: Christoph Lameter <cl@linux.com> >> >> -- >> Hi Andrew, >> >> We ran into this bug when working towards enabling memory corruption >> on arm64. The patch was tested on an arm64 platform running v4.12-rc2 >> with the series to enable memory corruption handling[0]. >> >> Please consider merging as a fix for the 4.12 release. >> >> Thanks, >> Punit >> >> [0] https://www.spinics.net/lists/arm-kernel/msg581657.html >> --- >> mm/migrate.c | 4 +--- >> 1 file changed, 1 insertion(+), 3 deletions(-) >> >> diff --git a/mm/migrate.c b/mm/migrate.c >> index 89a0a1707f4c..187abd1526df 100644 >> --- a/mm/migrate.c >> +++ b/mm/migrate.c >> @@ -1201,10 +1201,8 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, >> * tables or check whether the hugepage is pmd-based or not before >> * kicking migration. >> */ >> - if (!hugepage_migration_supported(page_hstate(hpage))) { >> - putback_active_hugepage(hpage); > > Thank you for reporting and suggestion, Punit, Manoj. > > Simply dropping this putback_active_hugepage() may resume the failure > counting issue addressed in 32665f2bbfed, so I would recommend to call > putback_movable_pages() in failure path in soft_offline_huge_page(). > > @@ -1600,7 +1600,8 @@ static int soft_offline_huge_page(struct page *page, int flags) > * only one hugepage pointed to by hpage, so we need not > * run through the pagelist here. > */ > - putback_active_hugepage(hpage); > + if (!list_empty(&pagelist)) > + putback_movable_pages(&pagelist); > if (ret > 0) > ret = -EIO; > } else { > > Could you check this works for you? Using this sequence works as well. I'll send out an update shortly. Thanks > > Thanks, > Naoya Horiguchi > -- > To unsubscribe, send a message with 'unsubscribe linux-mm' in > the body to majordomo@kvack.org. For more info on Linux MM, > see: http://www.linux-mm.org/ . > Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
diff --git a/mm/migrate.c b/mm/migrate.c index 89a0a1707f4c..187abd1526df 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -1201,10 +1201,8 @@ static int unmap_and_move_huge_page(new_page_t get_new_page, * tables or check whether the hugepage is pmd-based or not before * kicking migration. */ - if (!hugepage_migration_supported(page_hstate(hpage))) { - putback_active_hugepage(hpage); + if (!hugepage_migration_supported(page_hstate(hpage))) return -ENOSYS; - } new_hpage = get_new_page(hpage, private, &result); if (!new_hpage)
On failing to migrate a page, soft_offline_huge_page() performs the necessary update to the hugepage ref-count. When !hugepage_migration_supported() , unmap_and_move_hugepage() also decrements the page ref-count for the hugepage. The combined behaviour leaves the ref-count in an inconsistent state. This leads to soft lockups when running the overcommitted hugepage test from mce-tests suite. Soft offlining pfn 0x83ed600 at process virtual address 0x400000000000 soft offline: 0x83ed600: migration failed 1, type 1fffc00000008008 (uptodate|head) INFO: rcu_preempt detected stalls on CPUs/tasks: Tasks blocked on level-0 rcu_node (CPUs 0-7): P2715 (detected by 7, t=5254 jiffies, g=963, c=962, q=321) thugetlb_overco R running task 0 2715 2685 0x00000008 Call trace: [<ffff000008089f90>] dump_backtrace+0x0/0x268 [<ffff00000808a2d4>] show_stack+0x24/0x30 [<ffff000008100d34>] sched_show_task+0x134/0x180 [<ffff0000081c90fc>] rcu_print_detail_task_stall_rnp+0x54/0x7c [<ffff00000813cfd4>] rcu_check_callbacks+0xa74/0xb08 [<ffff000008143a3c>] update_process_times+0x34/0x60 [<ffff0000081550e8>] tick_sched_handle.isra.7+0x38/0x70 [<ffff00000815516c>] tick_sched_timer+0x4c/0x98 [<ffff0000081442e0>] __hrtimer_run_queues+0xc0/0x300 [<ffff000008144fa4>] hrtimer_interrupt+0xac/0x228 [<ffff0000089a56d4>] arch_timer_handler_phys+0x3c/0x50 [<ffff00000812f1bc>] handle_percpu_devid_irq+0x8c/0x290 [<ffff0000081297fc>] generic_handle_irq+0x34/0x50 [<ffff000008129f00>] __handle_domain_irq+0x68/0xc0 [<ffff0000080816b4>] gic_handle_irq+0x5c/0xb0 Fix this by dropping the ref-count decrement in unmap_and_move_hugepage() when !hugepage_migration_supported(). Fixes: 32665f2bbfed ("mm/migrate: correct failure handling if !hugepage_migration_support()") Reported-by: Manoj Iyer <manoj.iyer@canonical.com> Signed-off-by: Punit Agrawal <punit.agrawal@arm.com> Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Wanpeng Li <liwanp@linux.vnet.ibm.com> Cc: Christoph Lameter <cl@linux.com> -- Hi Andrew, We ran into this bug when working towards enabling memory corruption on arm64. The patch was tested on an arm64 platform running v4.12-rc2 with the series to enable memory corruption handling[0]. Please consider merging as a fix for the 4.12 release. Thanks, Punit [0] https://www.spinics.net/lists/arm-kernel/msg581657.html --- mm/migrate.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)