powerpc/64s: fix scv entry fallback flush vs interrupt

The L1D flush fallback functions are not recoverable vs interrupts,
yet the scv entry flush runs with MSR[EE]=1. This can result in a
timer (soft-NMI) or MCE or SRESET interrupt hitting here and overwriting
the EXRFI save area, which ends up corrupting userspace registers for
scv return.

Fix this by disabling RI and EE for the scv entry fallback flush.

Fixes: f79643787e ("powerpc/64s: flush L1D on kernel entry")
Cc: stable@vger.kernel.org # 5.9+ which also have flush L1D patch backport
Reported-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210111062408.287092-1-npiggin@gmail.com
This commit is contained in:
Nicholas Piggin 2021-01-11 16:24:08 +10:00 committed by Michael Ellerman
parent dd3a44c06f
commit 08685be776
6 changed files with 71 additions and 4 deletions

View file

@ -63,6 +63,12 @@
nop; \
nop;
#define SCV_ENTRY_FLUSH_SLOT \
SCV_ENTRY_FLUSH_FIXUP_SECTION; \
nop; \
nop; \
nop;
/*
* r10 must be free to use, r13 must be paca
*/
@ -70,6 +76,13 @@
STF_ENTRY_BARRIER_SLOT; \
ENTRY_FLUSH_SLOT
/*
* r10, ctr must be free to use, r13 must be paca
*/
#define SCV_INTERRUPT_TO_KERNEL \
STF_ENTRY_BARRIER_SLOT; \
SCV_ENTRY_FLUSH_SLOT
/*
* Macros for annotating the expected destination of (h)rfid
*