Skip to contentSkip to conversation

powerpc/xive: discard ESB load value when interrupt is invalid

PATCH In review Cédric Le Goater 3 participants 7 messages Jan 13, 2020 – Jan 16, 2020 PowerPC

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 powerpc/xive: discard ESB load value when interrupt is invalid

From: Frederic Barrat <[email protected]>

A load on an ESB page returning all 1's means that the underlying device has invalidated the access to the PQ state of the interrupt through mmio. It may happen, for example when querying a PHB interrupt while the PHB is in an error state.

In that case, we should consider the interrupt to be invalid when checking its state in the irq_get_irqchip_state() handler.

[ clg: - wrote a commit log
- introduced XIVE_ESB_INVALID ]

---
 arch/powerpc/include/asm/xive-regs.h |  1 +
 arch/powerpc/sysdev/xive/common.c    | 15 ++++++++++++---
 2 files changed, 13 insertions(+), 3 deletions(-)
arch/powerpc/include/asm/xive-regs.h +1 -0
index f2dfcd50a2d3..33aee7490cbb 100644
@@ -39,6 +39,7 @@
 
 #define XIVE_ESB_VAL_P		0x2
 #define XIVE_ESB_VAL_Q		0x1
+#define XIVE_ESB_INVALID	0xFF
 
 /*
  * Thread Management (aka "TM") registers
arch/powerpc/sysdev/xive/common.c +12 -3
index f5fadbd2533a..9651ca061828 100644
@@ -972,12 +972,21 @@ static int xive_get_irqchip_state(struct irq_data *data,
 				  enum irqchip_irq_state which, bool *state)
 {
 	struct xive_irq_data *xd = irq_data_get_irq_handler_data(data);
+	u8 pq;
 
 	switch (which) {
 	case IRQCHIP_STATE_ACTIVE:
-		*state = !xd->stale_p &&
-			 (xd->saved_p ||
-			  !!(xive_esb_read(xd, XIVE_ESB_GET) & XIVE_ESB_VAL_P));
+		pq = xive_esb_read(xd, XIVE_ESB_GET);
+
+		/*
+		 * The esb value being all 1's means we couldn't get
+		 * the PQ state of the interrupt through mmio. It may
+		 * happen, for example when querying a PHB interrupt
+		 * while the PHB is in an error state. We consider the
+		 * interrupt to be inactive in that case.
+		 */
+		*state = (pq != XIVE_ESB_INVALID) && !xd->stale_p &&
+			(xd->saved_p || !!(pq & XIVE_ESB_VAL_P));
 		return 0;
 	default:
 		return -EINVAL;
Signature · 1 line
2.21.1
#1 lore raw
Quoted 9 lines · Cédric Le Goater

From: Frederic Barrat <[email protected]>

A load on an ESB page returning all 1's means that the underlying device has invalidated the access to the PQ state of the interrupt through mmio. It may happen, for example when querying a PHB interrupt while the PHB is in an error state.

In that case, we should consider the interrupt to be invalid when checking its state in the irq_get_irqchip_state() handler.

and we need also these tags :

Fixes da15c03b047d ("powerpc/xive: Implement get_irqchip_state method for XIVE to fix shutdown race")
Quoted 52 lines

[ clg: - wrote a commit log
- introduced XIVE_ESB_INVALID ]

---
 arch/powerpc/include/asm/xive-regs.h |  1 +
 arch/powerpc/sysdev/xive/common.c    | 15 ++++++++++++---
 2 files changed, 13 insertions(+), 3 deletions(-)
arch/powerpc/include/asm/xive-regs.h +1 -0
index f2dfcd50a2d3..33aee7490cbb 100644
@@ -39,6 +39,7 @@
 
 #define XIVE_ESB_VAL_P		0x2
 #define XIVE_ESB_VAL_Q		0x1
+#define XIVE_ESB_INVALID	0xFF
 
 /*
  * Thread Management (aka "TM") registers
arch/powerpc/sysdev/xive/common.c +12 -3
index f5fadbd2533a..9651ca061828 100644
@@ -972,12 +972,21 @@ static int xive_get_irqchip_state(struct irq_data *data,
 				  enum irqchip_irq_state which, bool *state)
 {
 	struct xive_irq_data *xd = irq_data_get_irq_handler_data(data);
+	u8 pq;
 
 	switch (which) {
 	case IRQCHIP_STATE_ACTIVE:
-		*state = !xd->stale_p &&
-			 (xd->saved_p ||
-			  !!(xive_esb_read(xd, XIVE_ESB_GET) & XIVE_ESB_VAL_P));
+		pq = xive_esb_read(xd, XIVE_ESB_GET);
+
+		/*
+		 * The esb value being all 1's means we couldn't get
+		 * the PQ state of the interrupt through mmio. It may
+		 * happen, for example when querying a PHB interrupt
+		 * while the PHB is in an error state. We consider the
+		 * interrupt to be inactive in that case.
+		 */
+		*state = (pq != XIVE_ESB_INVALID) && !xd->stale_p &&
+			(xd->saved_p || !!(pq & XIVE_ESB_VAL_P));
 		return 0;
 	default:
 		return -EINVAL;
#2 lore raw
Quoted 8 lines · Cédric Le Goater
Quoted 7 lines · Cédric Le Goater

From: Frederic Barrat <[email protected]>

A load on an ESB page returning all 1's means that the underlying device has invalidated the access to the PQ state of the interrupt through mmio. It may happen, for example when querying a PHB interrupt while the PHB is in an error state.

Quoted 8 lines
Quoted 2 lines

In that case, we should consider the interrupt to be invalid when checking its state in the irq_get_irqchip_state() handler.

and we need also these tags :

Fixes da15c03b047d ("powerpc/xive: Implement get_irqchip_state method for XIVE to fix shutdown race")

I added those, although it's v5.4+, as the offending commit was first included in v5.4-rc1.

cheers

#3 lore raw
Quoted 13 lines · Michael Ellerman
Quoted 12 lines · Cédric Le Goater
Quoted 9 lines · Cédric Le Goater

From: Frederic Barrat <[email protected]>

A load on an ESB page returning all 1's means that the underlying device has invalidated the access to the PQ state of the interrupt through mmio. It may happen, for example when querying a PHB interrupt while the PHB is in an error state.

In that case, we should consider the interrupt to be invalid when checking its state in the irq_get_irqchip_state() handler.

Quoted 7 lines
Quoted 4 lines

and we need also these tags :

Fixes da15c03b047d ("powerpc/xive: Implement get_irqchip_state method for XIVE to fix shutdown race")

I added those, although it's v5.4+, as the offending commit was first included in v5.4-rc1.

Ah yes. I mistook the merge tag of the branch used for the PR (v5.3-rc2)

Thanks,

C.

#4 lore raw
Quoted 16 lines · On Tue, 14 Jan 2020 08:44:54 +0100 Cédric Le Goater <[email protected]> wrote:
Quoted 15 lines · Michael Ellerman
Quoted 14 lines · Cédric Le Goater
Quoted 9 lines · Cédric Le Goater

From: Frederic Barrat <[email protected]>

A load on an ESB page returning all 1's means that the underlying device has invalidated the access to the PQ state of the interrupt through mmio. It may happen, for example when querying a PHB interrupt while the PHB is in an error state.

In that case, we should consider the interrupt to be invalid when checking its state in the irq_get_irqchip_state() handler.

and we need also these tags :

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
Fixes da15c03b047d ("powerpc/xive: Implement get_irqchip_state method for XIVE to fix shutdown race")

I added those, although it's v5.4+, as the offending commit was first included in v5.4-rc1.

Ah yes. I mistook the merge tag of the branch used for the PR (v5.3-rc2)

You might want to use 'git tag --contains':

[greg@bahia kernel-linus]$ git tag --contains da15c03b047d for-linus
kvm-5.4-2
next-20191118
next-20191126
tags/kvm-5.4-1
tags/kvm-5.4-2
v5.4
v5.4-rc1
v5.4-rc2
v5.4-rc3
v5.4-rc4
v5.4-rc5
v5.4-rc6
v5.4-rc7
v5.4-rc8
v5.5-rc1

Quoted 4 lines

Thanks,

C.

#5 lore raw
Quoted 29 lines · Greg Kurz
Quoted 16 lines · On Tue, 14 Jan 2020 08:44:54 +0100 Cédric Le Goater <[email protected]> wrote:
Quoted 15 lines · Michael Ellerman
Quoted 14 lines · Cédric Le Goater
Quoted 9 lines · Cédric Le Goater

From: Frederic Barrat <[email protected]>

A load on an ESB page returning all 1's means that the underlying device has invalidated the access to the PQ state of the interrupt through mmio. It may happen, for example when querying a PHB interrupt while the PHB is in an error state.

In that case, we should consider the interrupt to be invalid when checking its state in the irq_get_irqchip_state() handler.

and we need also these tags :

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines
Fixes da15c03b047d ("powerpc/xive: Implement get_irqchip_state method for XIVE to fix shutdown race")

I added those, although it's v5.4+, as the offending commit was first included in v5.4-rc1.

Ah yes. I mistook the merge tag of the branch used for the PR (v5.3-rc2)

You might want to use 'git tag --contains':

Quoted 9 lines

[greg@bahia kernel-linus]$ git tag --contains da15c03b047d for-linus
kvm-5.4-2
next-20191118
next-20191126
tags/kvm-5.4-1
tags/kvm-5.4-2
v5.4
v5.4-rc1

Or:

  $ git describe --match "v[0-9]*" --contains da15c03b047d
  v5.4-rc1~99^2~134^2~17

cheers

#6 lore raw
Quoted 12 lines
Quoted 11 lines

You might want to use 'git tag --contains':

[greg@bahia kernel-linus]$ git tag --contains da15c03b047d for-linus
kvm-5.4-2
next-20191118
next-20191126
tags/kvm-5.4-1
tags/kvm-5.4-2
v5.4
v5.4-rc1

Quoted 4 lines

Or:

  $ git describe --match "v[0-9]*" --contains da15c03b047d
  v5.4-rc1~99^2~134^2~17

Nice. I am adding this command to my git aliases.

Thanks,

C.