On Wed, Aug 21, 2024 at 12:23:42AM +0000, Edgecombe, Rick P wrote: > On Mon, 2024-08-19 at 18:09 +0300, Nikolay Borisov wrote: > > > +/* > > > + * Some SEAMCALLs acquire the TDX module globally, and can fail with > > > + * TDX_OPERAND_BUSY. Use a global mutex to serialize these SEAMCALLs. > > > + */ > > > +static DEFINE_MUTEX(tdx_lock); > > > > The way this lock is used is very ugly. So it essentially mimics a lock > > which already lives in the tdx module. So why not simply gracefully > > handle the TDX_OPERAND_BUSY return value or change the interface of the > > module (yeah, it's probably late for this now) so expose the lock. This > > lock breaks one of the main rules of locking - "Lock data and not code" > > Hmm, we would have to make SEAMCALLs to spin on that lock, where as mutexes can > sleep. I suspect that is where it came from. But we are trying to make the code > simple and obviously correct and add optimizations later. This might fit that > pattern, especially since it is just used during VM creation and teardown. For handling the busy retries for SEAMCALL callers, we could just use iopoll.h read_poll_timeout(). I think it can handle toggling the resume bit while looping, need to test that though. See for example the smp_func_do_phymem_cache_wb() for toggling the resume variable. The overhead of a SEAMCALL may not be that bad in the retry case. Regards, Tony
