Re: memory leak in hidg_set_alt

[Greg KH]

On Sat, May 30, 2020 at 05:18:06PM +0000, Kyungtae Kim wrote:
> We report a bug (in linux-5.6.11) found by FuzzUSB (a modified version
> of syzkaller)
> 
> kernel config: https://kt0755.github.io/etc/config_v5.6.11
> 
> An usb_request instance allocated for report in hidg_set_alt() leaked.
> 
> ==================================================================
> BUG: memory leak
> unreferenced object 0xffff88803af68400 (size 128):
>   comm "softirq", pid 0, jiffies 4294942450 (age 189.010s)
>   hex dump (first 32 bytes):
>     00 84 f6 3a 80 88 ff ff 00 84 f6 3a 80 88 ff ff  ...:.......:....
>     a8 83 3c 5e 80 88 ff ff 01 00 00 00 00 00 00 00  ..<^............
>   backtrace:
>     [<00000000c77ea9d0>] kmemleak_alloc_recursive include/linux/kmemleak.h:43 [inline]
>     [<00000000c77ea9d0>] slab_post_alloc_hook mm/slab.h:586 [inline]
>     [<00000000c77ea9d0>] slab_alloc_node mm/slub.c:2786 [inline]
>     [<00000000c77ea9d0>] slab_alloc mm/slub.c:2794 [inline]
>     [<00000000c77ea9d0>] kmem_cache_alloc_trace+0x15e/0x2d0 mm/slub.c:2811
>     [<0000000021977c74>] kmalloc include/linux/slab.h:555 [inline]
>     [<0000000021977c74>] kzalloc include/linux/slab.h:669 [inline]
>     [<0000000021977c74>] dummy_alloc_request+0xb4/0x1a0 drivers/usb/gadget/udc/dummy_hcd.c:662
>     [<0000000065bac76c>] usb_ep_alloc_request+0x69/0x2e0 drivers/usb/gadget/udc/core.c:178
>     [<00000000916a7f5e>] alloc_ep_req+0x27/0x1d0 drivers/usb/gadget/u_f.c:18
>     [<00000000a697106a>] hidg_alloc_ep_req drivers/usb/gadget/function/f_hid.c:458 [inline]
>     [<00000000a697106a>] hidg_set_alt+0x1ff/0xbe0 drivers/usb/gadget/function/f_hid.c:662
>     [<000000006a6a3007>] set_config drivers/usb/gadget/composite.c:838 [inline]
>     [<000000006a6a3007>] composite_setup+0x4231/0x6f10 drivers/usb/gadget/composite.c:1717
>     [<000000000c47b49d>] configfs_composite_setup+0x11a/0x170 drivers/usb/gadget/configfs.c:1466
>     [<00000000850a3c44>] dummy_timer+0xda5/0x33f0 drivers/usb/gadget/udc/dummy_hcd.c:1898
>     [<00000000aaf23b82>] call_timer_fn+0x20e/0x770 kernel/time/timer.c:1404
>     [<00000000e04e8038>] expire_timers kernel/time/timer.c:1449 [inline]
>     [<00000000e04e8038>] __run_timers kernel/time/timer.c:1773 [inline]
>     [<00000000e04e8038>] run_timer_softirq+0x63f/0x13c0 kernel/time/timer.c:1786
>     [<00000000deb0ec70>] __do_softirq+0x262/0xb46 kernel/softirq.c:292
>     [<00000000c245a5cd>] invoke_softirq kernel/softirq.c:373 [inline]
>     [<00000000c245a5cd>] irq_exit+0x161/0x1b0 kernel/softirq.c:413
>     [<00000000bacb5a5b>] exiting_irq arch/x86/include/asm/apic.h:546 [inline]
>     [<00000000bacb5a5b>] smp_apic_timer_interrupt+0x137/0x500 arch/x86/kernel/apic/apic.c:1146
>     [<000000001988f5a4>] apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:829
> ==================================================================

Great, can you send a patch fixing this please so you can get the proper
credit for finding and fixing the issue?

thanks,

greg k-h