| Message ID | 20211020094826.3222052-1-mpe@ellerman.id.au |
|---|---|
| State | New |
| Headers | show |
| Series | powerpc/idle: Don't corrupt back chain when going idle | expand |
On Wed, 20 Oct 2021 20:48:26 +1100, Michael Ellerman wrote: > In isa206_idle_insn_mayloss() we store various registers into the stack > red zone, which is allowed. > > However inside the IDLE_STATE_ENTER_SEQ_NORET macro we save r2 again, > to 0(r1), which corrupts the stack back chain. > > We used to do the same in isa206_idle_insn_mayloss() itself, but we > fixed that in 73287caa9210 ("powerpc64/idle: Fix SP offsets when saving > GPRs"), however we missed that the macro also corrupts the back chain. > > [...] Applied to powerpc/fixes. [1/1] powerpc/idle: Don't corrupt back chain when going idle https://git.kernel.org/powerpc/c/496c5fe25c377ddb7815c4ce8ecfb676f051e9b6 cheers
diff --git a/arch/powerpc/kernel/idle_book3s.S b/arch/powerpc/kernel/idle_book3s.S index abb719b21cae..3d97fb833834 100644 --- a/arch/powerpc/kernel/idle_book3s.S +++ b/arch/powerpc/kernel/idle_book3s.S @@ -126,14 +126,16 @@ _GLOBAL(idle_return_gpr_loss) /* * This is the sequence required to execute idle instructions, as * specified in ISA v2.07 (and earlier). MSR[IR] and MSR[DR] must be 0. - * - * The 0(r1) slot is used to save r2 in isa206, so use that here. + * We have to store a GPR somewhere, ptesync, then reload it, and create + * a false dependency on the result of the load. It doesn't matter which + * GPR we store, or where we store it. We have already stored r2 to the + * stack at -8(r1) in isa206_idle_insn_mayloss, so use that. */ #define IDLE_STATE_ENTER_SEQ_NORET(IDLE_INST) \ /* Magic NAP/SLEEP/WINKLE mode enter sequence */ \ - std r2,0(r1); \ + std r2,-8(r1); \ ptesync; \ - ld r2,0(r1); \ + ld r2,-8(r1); \ 236: cmpd cr0,r2,r2; \ bne 236b; \ IDLE_INST; \
In isa206_idle_insn_mayloss() we store various registers into the stack red zone, which is allowed. However inside the IDLE_STATE_ENTER_SEQ_NORET macro we save r2 again, to 0(r1), which corrupts the stack back chain. We used to do the same in isa206_idle_insn_mayloss() itself, but we fixed that in 73287caa9210 ("powerpc64/idle: Fix SP offsets when saving GPRs"), however we missed that the macro also corrupts the back chain. Corrupting the back chain is bad for debuggability but doesn't necessarily cause a bug. However we recently changed the stack handling in some KVM code, and it now relies on the stack back chain being valid when it returns. The corruption causes that code to return with r1 pointing somewhere in kernel data, at some point LR is restored from the stack and we branch to NULL or somewhere else invalid. Only affects Power8 hosts running KVM guests, with dynamic_mt_modes enabled (which it is by default). The fixes tag below points to the commit that changed the KVM stack handling, exposing this bug. The actual corruption of the back chain has always existed since 948cf67c4726 ("powerpc: Add NAP mode support on Power7 in HV mode"). Fixes: 9b4416c5095c ("KVM: PPC: Book3S HV: Fix stack handling in idle_kvm_start_guest()") Signed-off-by: Michael Ellerman <mpe@ellerman.id.au> --- arch/powerpc/kernel/idle_book3s.S | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)