On Wed, Feb 09, 2022 at 01:33:02PM +0100, Ingo Rohloff wrote: > V2: > Incorporated Alan Sterns review comments: > Thanks for mentioning the READ_ONCE() semantics; I really had no clue. > > Note: > I think it's also correct to NOT use the "irqsave" variants > of spin_lock_irq/spin_unlock_irq in "usbfs_increase_memory_usage()". > > I am not sure if it's worth it? I checked the file, and it does look as though these routines are called only in contexts that can sleep. But using the _irq variants doesn't hurt much, so you might as well keep it that way. Alan Stern
