[PATCH v2] irqchip/gic-v4.1: Ensure accessing the correct RD when writing INVALLR

[PATCH v2] irqchip/gic-v4.1: Ensure accessing the correct RD when writing INVALLR

[Zenghui Yu]

The GICv4.1 spec tells us that it's CONSTRAINED UNPREDICTABLE to issue a
register-based invalidation operation for a vPEID not mapped to that RD,
or another RD within the same CommonLPIAff group.

To follow this rule, commit f3a059219bc7 ("irqchip/gic-v4.1: Ensure mutual
exclusion between vPE affinity change and RD access") tried to address the
race between the RD accesses and the vPE affinity change, but somehow
forgot to take GICR_INVALLR into account. Let's take the vpe_lock before
evaluating vpe->col_idx to fix it.

Fixes: f3a059219bc7 ("irqchip/gic-v4.1: Ensure mutual exclusion between vPE affinity change and RD access")
Cc: stable@vger.kernel.org
Signed-off-by: Zenghui Yu <yuzenghui@huawei.com>
---
* From v1:
  - Add a proper Fixes: tag
  - Cc stable

 drivers/irqchip/irq-gic-v3-its.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
index beac4caefad9..5bf2263bdbcc 100644
--- a/drivers/irqchip/irq-gic-v3-its.c
+++ b/drivers/irqchip/irq-gic-v3-its.c
@@ -4087,18 +4087,22 @@ static void its_vpe_4_1_deschedule(struct its_vpe *vpe,
 static void its_vpe_4_1_invall(struct its_vpe *vpe)
 {
 	void __iomem *rdbase;
+	unsigned long flags;
 	u64 val;
+	int cpu;
 
 	val  = GICR_INVALLR_V;
 	val |= FIELD_PREP(GICR_INVALLR_VPEID, vpe->vpe_id);
 
 	/* Target the redistributor this vPE is currently known on */
-	raw_spin_lock(&gic_data_rdist_cpu(vpe->col_idx)->rd_lock);
-	rdbase = per_cpu_ptr(gic_rdists->rdist, vpe->col_idx)->rd_base;
+	cpu = vpe_to_cpuid_lock(vpe, &flags);
+	raw_spin_lock(&gic_data_rdist_cpu(cpu)->rd_lock);
+	rdbase = per_cpu_ptr(gic_rdists->rdist, cpu)->rd_base;
 	gic_write_lpir(val, rdbase + GICR_INVALLR);
 
 	wait_for_syncr(rdbase);
-	raw_spin_unlock(&gic_data_rdist_cpu(vpe->col_idx)->rd_lock);
+	raw_spin_unlock(&gic_data_rdist_cpu(cpu)->rd_lock);
+	vpe_to_cpuid_unlock(vpe, flags);
 }
 
 static int its_vpe_4_1_set_vcpu_affinity(struct irq_data *d, void *vcpu_info)
-- 
2.19.1

Re: [PATCH v2] irqchip/gic-v4.1: Ensure accessing the correct RD when writing INVALLR

[Marc Zyngier]

On Mon, 20 Jul 2020 17:23:28 +0800, Zenghui Yu wrote:
> The GICv4.1 spec tells us that it's CONSTRAINED UNPREDICTABLE to issue a
> register-based invalidation operation for a vPEID not mapped to that RD,
> or another RD within the same CommonLPIAff group.
> 
> To follow this rule, commit f3a059219bc7 ("irqchip/gic-v4.1: Ensure mutual
> exclusion between vPE affinity change and RD access") tried to address the
> race between the RD accesses and the vPE affinity change, but somehow
> forgot to take GICR_INVALLR into account. Let's take the vpe_lock before
> evaluating vpe->col_idx to fix it.

Applied to irq/irqchip-5.9, thanks!

[1/1] irqchip/gic-v4.1: Ensure accessing the correct RD when writing INVALLR
      commit: fdccf1d9d10395abfe082f50694f374997c6e101

Cheers,

	M.
-- 
Without deviation from the norm, progress is not possible.