On Wed, Jul 07, 2021 at 01:38:32PM +0100, John Garry wrote: > Hi guys, > > Just a heads up in case it's a real issue - I was developing based on Linus' > tree at 79160a603bdb, and with KMEMLEAK and DEBUG_TEST_DRIVER_REMOVE configs > enabled, I get these warns: > DEBUG_TEST_DRIVER_REMOVE does odd things, which may well interfer with reference counting. I don't think this is a spurious issue (and I have no idea if it is a real issue), but at least for my part I don't plan to spend any time trying to track this down. Others, please feel free to jump in. Thanks, Guenter > root@(none)$ [ 859.608780] kmemleak: 3 new suspected memory leaks (see > /sys/kernel/debug/kmemleak) > [ 859.608780] kmemleak: 3 new suspected memory leaks (see > /sys/kernel/debug/kmemleak) > > root@(none)$ more /sys/kernel/debug/kmemleak > unreferenced object 0xffff0020f9810800 (size 1024): > comm "swapper/0", pid 1, jiffies 4294929730 (age 1071.904s) > hex dump (first 32 bytes): > 80 89 48 f8 20 00 ff ff 08 08 81 f9 20 00 ff ff ..H. ....... ... > 08 08 81 f9 20 00 ff ff 00 00 00 00 00 00 00 00 .... ........... > backtrace: > [<(____ptrval____)>] slab_post_alloc_hook+0x9c/0x270 > [<(____ptrval____)>] kmem_cache_alloc+0x198/0x2e0 > [<(____ptrval____)>] watchdog_dev_register+0x38/0x428 > [<(____ptrval____)>] __watchdog_register_device+0x13c/0x400 > [<(____ptrval____)>] watchdog_register_device+0x9c/0x108 > [<(____ptrval____)>] devm_watchdog_register_device+0x50/0xe0 > [<(____ptrval____)>] sbsa_gwdt_probe+0x278/0x488 > [<(____ptrval____)>] platform_probe+0x8c/0x108 > [<(____ptrval____)>] really_probe+0x130/0x558 > [<(____ptrval____)>] __driver_probe_device+0xb8/0x130 > [<(____ptrval____)>] driver_probe_device+0x60/0x150 > [<(____ptrval____)>] __driver_attach+0xa0/0x160 > [<(____ptrval____)>] bus_for_each_dev+0xec/0x160 > [<(____ptrval____)>] driver_attach+0x34/0x48 > [<(____ptrval____)>] bus_add_driver+0x1b8/0x2c0 > [<(____ptrval____)>] driver_register+0xc0/0x1e0 > unreferenced object 0xffff0020f8488980 (size 64): > comm "swapper/0", pid 1, jiffies 4294929730 (age 1071.904s) > hex dump (first 32 bytes): > 77 61 74 63 68 64 6f 67 30 00 00 00 00 00 00 00 watchdog0....... > 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ > backtrace: > [<(____ptrval____)>] slab_post_alloc_hook+0x9c/0x270 > [<(____ptrval____)>] __kmalloc_track_caller+0x190/0x2d0 > [<(____ptrval____)>] kvasprintf+0xe4/0x1a0 > [<(____ptrval____)>] kvasprintf_const+0xcc/0x178 > [<(____ptrval____)>] kobject_set_name_vargs+0x54/0xf0 > [<(____ptrval____)>] dev_set_name+0xa8/0xd8 > [<(____ptrval____)>] watchdog_dev_register+0x154/0x428 > [<(____ptrval____)>] __watchdog_register_device+0x13c/0x400 > [<(____ptrval____)>] watchdog_register_device+0x9c/0x108 > [<(____ptrval____)>] devm_watchdog_register_device+0x50/0xe0 > [<(____ptrval____)>] sbsa_gwdt_probe+0x278/0x488 > [<(____ptrval____)>] platform_probe+0x8c/0x108 > [<(____ptrval____)>] really_probe+0x130/0x558 > [<(____ptrval____)>] __driver_probe_device+0xb8/0x130 > [<(____ptrval____)>] driver_probe_device+0x60/0x150 > [<(____ptrval____)>] __driver_attach+0xa0/0x160 > unreferenced object 0xffff00208f9b6000 (size 256): > comm "swapper/0", pid 1, jiffies 4294929730 (age 1071.904s) > hex dump (first 32 bytes): > 00 00 00 00 00 00 00 00 08 60 9b 8f 20 00 ff ff .........`.. ... > 08 60 9b 8f 20 00 ff ff f8 47 b2 10 00 80 ff ff .`.. ....G...... > backtrace: > [<(____ptrval____)>] slab_post_alloc_hook+0x9c/0x270 > [<(____ptrval____)>] kmem_cache_alloc+0x198/0x2e0 > [<(____ptrval____)>] device_add+0x354/0xcf0 > [<(____ptrval____)>] cdev_device_add+0x74/0xc0 > [<(____ptrval____)>] watchdog_dev_register+0x1e0/0x428 > [<(____ptrval____)>] __watchdog_register_device+0x13c/0x400 > [<(____ptrval____)>] watchdog_register_device+0x9c/0x108 > [<(____ptrval____)>] devm_watchdog_register_device+0x50/0xe0 > [<(____ptrval____)>] sbsa_gwdt_probe+0x278/0x488 > [<(____ptrval____)>] platform_probe+0x8c/0x108 > [<(____ptrval____)>] really_probe+0x130/0x558 > [<(____ptrval____)>] __driver_probe_device+0xb8/0x130 > [<(____ptrval____)>] driver_probe_device+0x60/0x150 > [<(____ptrval____)>] __driver_attach+0xa0/0x160 > [<(____ptrval____)>] bus_for_each_dev+0xec/0x160 > [<(____ptrval____)>] driver_attach+0x34/0x48 > root@(none)$ > root@(none)$ > root@(none)$ > > I didn't see a report elsewhere. Maybe just a transient issue on Linus' > tree. > > Thanks, > John
