We must handle data access exception as well as memory address unaligned
exceptions from return from trap window fill faults, not just normal
TLB misses.
Otherwise we can get an OOPS that looks like this:
ld-linux.so.2(36808): Kernel bad sw trap 5 [#1]
CPU: 1 PID: 36808 Comm: ld-linux.so.2 Not tainted 4.6.0 #34
task: fff8000303be5c60 ti: fff8000301344000 task.ti: fff8000301344000
TSTATE: 0000004410001601 TPC: 0000000000a1a784 TNPC: 0000000000a1a788 Y: 00000002 Not tainted
TPC: <do_sparc64_fault+0x5c4/0x700>
g0: fff8000024fc8248 g1: 0000000000db04dc g2: 0000000000000000 g3: 0000000000000001
g4: fff8000303be5c60 g5: fff800030e672000 g6: fff8000301344000 g7: 0000000000000001
o0: 0000000000b95ee8 o1: 000000000000012b o2: 0000000000000000 o3: 0000000200b9b358
o4: 0000000000000000 o5: fff8000301344040 sp: fff80003013475c1 ret_pc: 0000000000a1a77c
RPC: <do_sparc64_fault+0x5bc/0x700>
l0: 00000000000007ff l1: 0000000000000000 l2: 000000000000005f l3: 0000000000000000
l4: fff8000301347e98 l5: fff8000024ff3060 l6: 0000000000000000 l7: 0000000000000000
i0: fff8000301347f60 i1: 0000000000102400 i2: 0000000000000000 i3: 0000000000000000
i4: 0000000000000000 i5: 0000000000000000 i6: fff80003013476a1 i7: 0000000000404d4c
I7: <user_rtt_fill_fixup+0x6c/0x7c>
Call Trace:
[0000000000404d4c] user_rtt_fill_fixup+0x6c/0x7c
The window trap handlers are slightly clever, the trap table entries for them are
composed of two pieces of code. First comes the code that actually performs
the window fill or spill trap handling, and then there are three instructions at
the end which are for exception processing.
The userland register window fill handler is:
add %sp, STACK_BIAS + 0x00, %g1; \
ldxa [%g1 + %g0] ASI, %l0; \
mov 0x08, %g2; \
mov 0x10, %g3; \
ldxa [%g1 + %g2] ASI, %l1; \
mov 0x18, %g5; \
ldxa [%g1 + %g3] ASI, %l2; \
ldxa [%g1 + %g5] ASI, %l3; \
add %g1, 0x20, %g1; \
ldxa [%g1 + %g0] ASI, %l4; \
ldxa [%g1 + %g2] ASI, %l5; \
ldxa [%g1 + %g3] ASI, %l6; \
ldxa [%g1 + %g5] ASI, %l7; \
add %g1, 0x20, %g1; \
ldxa [%g1 + %g0] ASI, %i0; \
ldxa [%g1 + %g2] ASI, %i1; \
ldxa [%g1 + %g3] ASI, %i2; \
ldxa [%g1 + %g5] ASI, %i3; \
add %g1, 0x20, %g1; \
ldxa [%g1 + %g0] ASI, %i4; \
ldxa [%g1 + %g2] ASI, %i5; \
ldxa [%g1 + %g3] ASI, %i6; \
ldxa [%g1 + %g5] ASI, %i7; \
restored; \
retry; nop; nop; nop; nop; \
b,a,pt %xcc, fill_fixup_dax; \
b,a,pt %xcc, fill_fixup_mna; \
b,a,pt %xcc, fill_fixup;
And the way this works is that if any of those memory accesses
generate an exception, the exception handler can revector to one of
those final three branch instructions depending upon which kind of
exception the memory access took. In this way, the fault handler
doesn't have to know if it was a spill or a fill that it's handling
the fault for. It just always branches to the last instruction in
the parent trap's handler.
For example, for a regular fault, the code goes:
winfix_trampoline:
rdpr %tpc, %g3
or %g3, 0x7c, %g3
wrpr %g3, %tnpc
done
All window trap handlers are 0x80 aligned, so if we "or" 0x7c into the
trap time program counter, we'll get that final instruction in the
trap handler.
On return from trap, we have to pull the register window in but we do
this by hand instead of just executing a "restore" instruction for
several reasons. The largest being that from Niagara and onward we
simply don't have enough levels in the trap stack to fully resolve all
possible exception cases of a window fault when we are already at
trap level 1 (which we enter to get ready to return from the original
trap).
This is executed inline via the FILL_*_RTRAP handlers. rtrap_64.S's
code branches directly to these to do the window fill by hand if
necessary. Now if you look at them, we'll see at the end:
ba,a,pt %xcc, user_rtt_fill_fixup;
ba,a,pt %xcc, user_rtt_fill_fixup;
ba,a,pt %xcc, user_rtt_fill_fixup;
And oops, all three cases are handled like a fault.
This doesn't work because each of these trap types (data access
exception, memory address unaligned, and faults) store their auxiliary
info in different registers to pass on to the C handler which does the
real work.
So in the case where the stack was unaligned, the unaligned trap
handler sets up the arg registers one way, and then we branched to
the fault handler which expects them setup another way.
So the FAULT_TYPE_* value ends up basically being garbage, and
randomly would generate the backtrace seen above.
Reported-by: Nick Alcock <nix@esperi.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 28a1f53 delays setting %pil to avoid potential
hardirq stack overflow in the common rtrap_irq path.
Setting %pil also needs to be delayed in the rtrap_nmi
path for the same reason.
Signed-off-by: Rob Gardner <rob.gardner@oracle.com>
Signed-off-by: Dave Aldridge <david.j.aldridge@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull irq cleanups from Ingo Molnar:
"This is a multi-arch cleanup series from Thomas Gleixner, which we
kept to near the end of the merge window, to not interfere with
architecture updates.
This series (motivated by the -rt kernel) unifies more aspects of IRQ
handling and generalizes PREEMPT_ACTIVE"
* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
preempt: Make PREEMPT_ACTIVE generic
sparc: Use preempt_schedule_irq
ia64: Use preempt_schedule_irq
m32r: Use preempt_schedule_irq
hardirq: Make hardirq bits generic
m68k: Simplify low level interrupt handling code
genirq: Prevent spurious detection for unconditionally polled interrupts
Mark the places when the system are in user or are in kernel.
This is used to make full dynticks system (tickless) --
CONFIG_NO_HZ_FULL dependence.
Signed-off-by: Kirill Tkhai <tkhai@yandex.ru>
CC: David Miller <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
The low level preemption code fiddles with the PREEMPT_ACTIVE bit for
no reason and calls schedule() with interrupts disabled, which is
wrong to begin with. Remove the PREEMPT_ACTIVE fiddling and call the
proper schedule_preempt_irq() function.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: David S. Miller <davem@davemloft.net>
Cc: sparclinux@vger.kernel.org
Link: http://lkml.kernel.org/r/20130917183628.966769884@linutronix.de
The invocation of softirq is now handled by irq_exit(), so there is no
need for sparc64 to invoke it on the trap-return path. In fact, doing so
is a bug because if the trap occurred in the idle loop, this invocation
can result in lockdep-RCU failures. The problem is that RCU ignores idle
CPUs, and the sparc64 trap-return path to the softirq handlers fails to
tell RCU that the CPU must be considered non-idle while those handlers
are executing. This means that RCU is ignoring any RCU read-side critical
sections in those handlers, which in turn means that RCU-protected data
can be yanked out from under those read-side critical sections.
The shiny new lockdep-RCU ability to detect RCU read-side critical sections
that RCU is ignoring located this problem.
The fix is straightforward: Make sparc64 stop manually invoking the
softirq handlers.
Reported-by: Meelis Roos <mroos@linux.ee>
Suggested-by: David Miller <davem@davemloft.net>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Tested-by: Meelis Roos <mroos@linux.ee>
Cc: stable@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
Analog of what commit 494486a1d2 had done
to alpha (another architecture with similar bug).
One note: in rtrap_32.S part clr %l6 has been a rudiment of left after
commit 28e6103665 (sparc: Fix debugger syscall
restart interactions) has killed %l6 use in there.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can overflow the hardirq stack if we set the %pil here
so early, just let the normal control flow do it.
This is fine as we are allowed to do the actual IRQ enable
at any point after we call trace_hardirqs_on.
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
sparc64: Make prom entry spinlock NMI safe.
sparc64: Kill off old sys_perfctr system call and state.
sparc: Update defconfigs.
sparc: Provide io{read,write}{16,32}be().
People should be using the perf events interfaces, and
the way these system call facilities used the %pcr conflicts
with the usage of the NMI watchdog and perf events.
Signed-off-by: David S. Miller <davem@davemloft.net>
Now that the return from alloc_percpu is compatible with the address
of per-cpu vars, it makes sense to hand around the address of per-cpu
variables. To make this sane, we remove the per_cpu__ prefix we used
created to stop people accidentally using these vars directly.
Now we have sparse, we can use that (next patch).
tj: * Updated to convert stuff which were missed by or added after the
original patch.
* Kill per_cpu_var() macro.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Christoph Lameter <cl@linux-foundation.org>
o Move all files from sparc64/kernel/ to sparc/kernel
- rename as appropriate
o Update sparc/Makefile to the changes
o Update sparc/kernel/Makefile to include the sparc64 files
NOTE: This commit changes link order on sparc64!
Link order had to change for either of sparc32 and sparc64.
And assuming sparc64 see more testing than sparc32 change link
order on sparc64 where issues will be caught faster.
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
