Re: [PATCH] USB: xhci: fix lock-inversion problem

2017-05-02 Thread Mathias Nyman
On 02.05.2017 17:34, Alan Stern wrote: On Tue, 2 May 2017, Mathias Nyman wrote: On 01.05.2017 22:41, Alan Stern wrote: With threaded interrupts, bottom-half handlers are called with interrupts enabled. Therefore they can't safely use spin_lock(); they have to use spin_lock_irqsave(). Lockdep

Re: [PATCH] USB: xhci: fix lock-inversion problem

2017-05-02 Thread Alan Stern
On Tue, 2 May 2017, Mathias Nyman wrote: > On 01.05.2017 22:41, Alan Stern wrote: > > With threaded interrupts, bottom-half handlers are called with > > interrupts enabled. Therefore they can't safely use spin_lock(); they > > have to use spin_lock_irqsave(). Lockdep warns about a violation > >

Re: [PATCH] USB: xhci: fix lock-inversion problem

2017-05-02 Thread Mathias Nyman
On 01.05.2017 22:41, Alan Stern wrote: With threaded interrupts, bottom-half handlers are called with interrupts enabled. Therefore they can't safely use spin_lock(); they have to use spin_lock_irqsave(). Lockdep warns about a violation occurring in xhci_irq():

[PATCH] USB: xhci: fix lock-inversion problem

2017-05-01 Thread Alan Stern
With threaded interrupts, bottom-half handlers are called with interrupts enabled. Therefore they can't safely use spin_lock(); they have to use spin_lock_irqsave(). Lockdep warns about a violation occurring in xhci_irq(): = [ INFO: possibl