On Tue, Aug 16, 2016 at 3:32 AM, Robin Murphy <robin.mur...@arm.com> wrote: > Hi Guenter, > > On 16/08/16 00:19, Guenter Roeck wrote: >> Hi, >> >> we are having a problem with atomic accesses in pstore on some ARM >> CPUs (specifically rk3288 and rk3399). With those chips, atomic >> accesses fail with both pgprot_noncached and pgprot_writecombine >> memory. Atomic accesses do work when selecting PAGE_KERNEL protection. > > What's the pstore backed by? I'm guessing it's not normal DRAM. >
it is normal DRAM. >> Debugging on rk3399 shows the following crash. >> >> [ 0.912669] Bad mode in Error handler detected, code 0xbf000002 -- SError >> [ 0.920140] CPU: 4 PID: 1 Comm: swapper/0 Not tainted 4.4.14 #389 >> [ 0.926838] Hardware name: Google Kevin (DT) >> [ 0.931533] task: ffffffc0edfe0000 ti: ffffffc0edf7c000 task.ti: >> ffffffc0edf 7c000 >> [ 0.939780] PC is at __ll_sc___cmpxchg_case_mb_4+0x2c/0x5c >> [ 0.945811] LR is at 0x1 >> >> The "solution" for this problem in various Chrome OS releases is to >> disable atomic accesses in pstore entirely, which seems to be a bit >> brute-force. Question is what a proper upstream-acceptable solution > > From that, it sounds like the endpoint doesn't support exclusive > accesses. I imagine you get away with it through a PAGE_KERNEL mapping > by virtue of being write-back cacheable, such that the exclusives end up > being handled by the local monitor at L1 and don't go out to memory, but > I'm not sure even that's necessarily reliable. > > In general terms, not doing atomic accesses is probably the only 100% > safe solution. > Not sure I understand. Atomics work for the rest of the DRAM. Can't I just use the same access protection ? Thanks, Guenter > Robin. > >> might be. Introduce another memory type to select PAGE_KERNEL ? Is >> there some means to determine if atomic operations are supported with >> a given protection mask, maybe ? Anything else ? >> >> Thanks, >> Guenter >> >> _______________________________________________ >> linux-arm-kernel mailing list >> linux-arm-ker...@lists.infradead.org >> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel >> >
