Skip to contentSkip to conversation

irq-gic-v3: fix NULL dereference of disabled redist_base

PATCH Applied Heyi Guo 3 participants 6 messages Dec 16, 2019 – Feb 8, 2020

Subsystems (2)
lore
Share & tools
lore

Heyi Guo

author
#0 lore raw

PATCH 1/1 irq-gic-v3: fix NULL dereference of disabled redist_base

If we use ACPI MADT GICC structure to pass single redistributor base, and mark some GICC as disabled, we'll get below call trace during boot:

[    0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0
[    0.000000] GICv3: 256 SPIs implemented
[    0.000000] GICv3: 0 Extended SPIs implemented
[    0.000000] GICv3: Distributor has no Range Selector support
[    0.000000] Unable to handle kernel paging request at virtual address 000000000000ffe8
[    0.000000] Mem abort info:
[    0.000000]   ESR = 0x96000004
[    0.000000]   EC = 0x25: DABT (current EL), IL = 32 bits
[    0.000000]   SET = 0, FnV = 0
[    0.000000]   EA = 0, S1PTW = 0
[    0.000000] Data abort info:
[    0.000000]   ISV = 0, ISS = 0x00000004
[    0.000000]   CM = 0, WnR = 0
[    0.000000] [000000000000ffe8] user address but active_mm is swapper
[    0.000000] Internal error: Oops: 96000004 [#1] SMP
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.5.0-rc1 #5
[    0.000000] pstate: 20000085 (nzCv daIf -PAN -UAO)
[    0.000000] pc : gic_iterate_rdists+0x58/0x130
[    0.000000] lr : gic_iterate_rdists+0x80/0x130
[    0.000000] sp : ffff8000113d3cb0
[    0.000000] x29: ffff8000113d3cb0 x28: 0000000000000000
[    0.000000] x27: 0000000000000000 x26: 0000000000000018
[    0.000000] x25: 000000000000ffe8 x24: 000000000000003f
[    0.000000] x23: ffff800010588040 x22: 00000000000005e8
[    0.000000] x21: ffff8000113df7d0 x20: 0000030f00003f11
[    0.000000] x19: 0000000000000000 x18: ffffffffffffffff
[    0.000000] x17: 0000000014aeb8dc x16: 00000000c3ba0ccf
[    0.000000] x15: ffff8000113d9908 x14: ffff8000913d3a37
[    0.000000] x13: ffff8000113d3a45 x12: ffff800011402000
[    0.000000] x11: ffff8000113d39d0 x10: ffff8000113db980
[    0.000000] x9 : 00000000ffffffd0 x8 : ffff8000106dca98
[    0.000000] x7 : 000000000000005b x6 : 0000000000000000
[    0.000000] x5 : 0000000000000000 x4 : ffff8000128c0000
[    0.000000] x3 : ffff8000128a0000 x2 : ffff0003fc3c7000
[    0.000000] x1 : 0000000000000001 x0 : 000000000000ffe8
[    0.000000] Call trace:
[    0.000000]  gic_iterate_rdists+0x58/0x130
[    0.000000]  gic_init_bases+0x200/0x4b4
[    0.000000]  gic_acpi_init+0x148/0x284
[    0.000000]  acpi_match_madt+0x4c/0x84
[    0.000000]  acpi_table_parse_entries_array+0x188/0x278
[    0.000000]  acpi_table_parse_entries+0x70/0x98
[    0.000000]  acpi_table_parse_madt+0x40/0x50
[    0.000000]  __acpi_probe_device_table+0x88/0xe4
[    0.000000]  irqchip_init+0x38/0x40
[    0.000000]  init_IRQ+0x168/0x19c
[    0.000000]  start_kernel+0x328/0x508
[    0.000000] Code: f90017b6 9b3a7f16 f8766853 8b190260 (b9400000)
[    0.000000] ---[ end trace ae5cf232d924bfc1 ]---
[    0.000000] Kernel panic - not syncing: Fatal exception
[    0.000000] Rebooting in 3 seconds..

In this case, nr_redist_regions counts all GICC structures but only enabled ones have redistributor mapped. So add check to avoid NULL deference of redist_base.

Cc Thomas Gleixner <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Marc Zyngier <[email protected]>
---
 drivers/irqchip/irq-gic-v3.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/irqchip/irq-gic-v3.c +7 -0
index d6218012097b..bd9d55cadef9 100644
@@ -781,6 +781,13 @@ static int gic_iterate_rdists(int (*fn)(struct redist_region *, void __iomem *))
 		u64 typer;
 		u32 reg;
 
+		/*
+		 * redist_base may be NULL if we use single_redist and some GICC
+		 * structure is disabled.
+		 */
+		if (!ptr)
+			continue;
+
 		reg = readl_relaxed(ptr + GICR_PIDR2) & GIC_PIDR2_ARCH_MASK;
 		if (reg != GIC_PIDR2_ARCH_GICv3 &&
 		    reg != GIC_PIDR2_ARCH_GICv4) { /* We're in trouble... */
Signature · 1 line
2.19.1
maintainer
#1 lore raw

Hi Heyi,

Quoted 88 lines · Heyi Guo

If we use ACPI MADT GICC structure to pass single redistributor base, and mark some GICC as disabled, we'll get below call trace during boot:

[    0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0
[    0.000000] GICv3: 256 SPIs implemented
[    0.000000] GICv3: 0 Extended SPIs implemented
[    0.000000] GICv3: Distributor has no Range Selector support
[    0.000000] Unable to handle kernel paging request at virtual
address 000000000000ffe8
[    0.000000] Mem abort info:
[    0.000000]   ESR = 0x96000004
[    0.000000]   EC = 0x25: DABT (current EL), IL = 32 bits
[    0.000000]   SET = 0, FnV = 0
[    0.000000]   EA = 0, S1PTW = 0
[    0.000000] Data abort info:
[    0.000000]   ISV = 0, ISS = 0x00000004
[    0.000000]   CM = 0, WnR = 0
[    0.000000] [000000000000ffe8] user address but active_mm is swapper
[    0.000000] Internal error: Oops: 96000004 [#1] SMP
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.5.0-rc1 #5
[    0.000000] pstate: 20000085 (nzCv daIf -PAN -UAO)
[    0.000000] pc : gic_iterate_rdists+0x58/0x130
[    0.000000] lr : gic_iterate_rdists+0x80/0x130
[    0.000000] sp : ffff8000113d3cb0
[    0.000000] x29: ffff8000113d3cb0 x28: 0000000000000000
[    0.000000] x27: 0000000000000000 x26: 0000000000000018
[    0.000000] x25: 000000000000ffe8 x24: 000000000000003f
[    0.000000] x23: ffff800010588040 x22: 00000000000005e8
[    0.000000] x21: ffff8000113df7d0 x20: 0000030f00003f11
[    0.000000] x19: 0000000000000000 x18: ffffffffffffffff
[    0.000000] x17: 0000000014aeb8dc x16: 00000000c3ba0ccf
[    0.000000] x15: ffff8000113d9908 x14: ffff8000913d3a37
[    0.000000] x13: ffff8000113d3a45 x12: ffff800011402000
[    0.000000] x11: ffff8000113d39d0 x10: ffff8000113db980
[    0.000000] x9 : 00000000ffffffd0 x8 : ffff8000106dca98
[    0.000000] x7 : 000000000000005b x6 : 0000000000000000
[    0.000000] x5 : 0000000000000000 x4 : ffff8000128c0000
[    0.000000] x3 : ffff8000128a0000 x2 : ffff0003fc3c7000
[    0.000000] x1 : 0000000000000001 x0 : 000000000000ffe8
[    0.000000] Call trace:
[    0.000000]  gic_iterate_rdists+0x58/0x130
[    0.000000]  gic_init_bases+0x200/0x4b4
[    0.000000]  gic_acpi_init+0x148/0x284
[    0.000000]  acpi_match_madt+0x4c/0x84
[    0.000000]  acpi_table_parse_entries_array+0x188/0x278
[    0.000000]  acpi_table_parse_entries+0x70/0x98
[    0.000000]  acpi_table_parse_madt+0x40/0x50
[    0.000000]  __acpi_probe_device_table+0x88/0xe4
[    0.000000]  irqchip_init+0x38/0x40
[    0.000000]  init_IRQ+0x168/0x19c
[    0.000000]  start_kernel+0x328/0x508
[    0.000000] Code: f90017b6 9b3a7f16 f8766853 8b190260 (b9400000)
[    0.000000] ---[ end trace ae5cf232d924bfc1 ]---
[    0.000000] Kernel panic - not syncing: Fatal exception
[    0.000000] Rebooting in 3 seconds..

In this case, nr_redist_regions counts all GICC structures but only enabled ones have redistributor mapped. So add check to avoid NULL deference of redist_base.

Cc Thomas Gleixner <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Marc Zyngier <[email protected]>
---
 drivers/irqchip/irq-gic-v3.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/irqchip/irq-gic-v3.c +7 -0
index d6218012097b..bd9d55cadef9 100644
@@ -781,6 +781,13 @@ static int gic_iterate_rdists(int (*fn)(struct
 redist_region *, void __iomem *))
 		u64 typer;
 		u32 reg;
 
+		/*
+		 * redist_base may be NULL if we use single_redist and some GICC
+		 * structure is disabled.
+		 */
+		if (!ptr)
+			continue;
+
 		reg = readl_relaxed(ptr + GICR_PIDR2) & GIC_PIDR2_ARCH_MASK;
 		if (reg != GIC_PIDR2_ARCH_GICv3 &&
Quoted 1 line

reg != GIC_PIDR2_ARCH_GICv4) { /* We're in trouble... */

This feels like the wrong fix. The redistributor region array should be completely populated, and there is an assumption all over this driver that there is no junk in these structures.

You're seeing this because we don't track the number of *enabled* rdists, and allocate the number of regions based on the number of overall GICC entries instead of the number of enabled redistributors.

How about this instead?

        M.
drivers/irqchip/irq-gic-v3.c +7 -2
index 3a1866682dd0..9b26a860340b 100644
@@ -1861,6 +1861,7 @@ static struct
 	struct redist_region *redist_regs;
 	u32 nr_redist_regions;
 	bool single_redist;
+	int enabled_rdists;
 	u32 maint_irq;
 	int maint_irq_mode;
 	phys_addr_t vcpu_base;
@@ -1955,8 +1956,10 @@ static int __init gic_acpi_match_gicc(union acpi_subtable_headers *header,
 	 * If GICC is enabled and has valid gicr base address, then it means
 	 * GICR base is presented via GICC
 	 */
-	if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address)
+	if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address) {
+		acpi_data.enabled_rdists++;
 		return 0;
+	}
 
 	/*
 	 * It's perfectly valid firmware can pass disabled GICC entry, driver
@@ -1986,8 +1989,10 @@ static int __init gic_acpi_count_gicr_regions(void)
 
 	count = acpi_table_parse_madt(ACPI_MADT_TYPE_GENERIC_INTERRUPT,
 				      gic_acpi_match_gicc, 0);
-	if (count > 0)
+	if (count > 0) {
 		acpi_data.single_redist = true;
+		count = acpi_data.enabled_rdists;
+	}
 
 	return count;
 }
Signature · 1 line
Jazz is not dead. It just smells funny...
author
#2 lore raw
Quoted 82 lines · Marc Zyngier

Hi Heyi,

Quoted 79 lines · Heyi Guo

If we use ACPI MADT GICC structure to pass single redistributor base, and mark some GICC as disabled, we'll get below call trace during boot:

[    0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0 [    0.000000] GICv3: 256 SPIs implemented
[    0.000000] GICv3: 0 Extended SPIs implemented
[    0.000000] GICv3: Distributor has no Range Selector support [    0.000000] Unable to handle kernel paging request at virtual address 000000000000ffe8
[    0.000000] Mem abort info:
[    0.000000]   ESR = 0x96000004
[    0.000000]   EC = 0x25: DABT (current EL), IL = 32 bits [    0.000000]   SET = 0, FnV = 0
[    0.000000]   EA = 0, S1PTW = 0
[    0.000000] Data abort info:
[    0.000000]   ISV = 0, ISS = 0x00000004
[    0.000000]   CM = 0, WnR = 0
[    0.000000] [000000000000ffe8] user address but active_mm is swapper [    0.000000] Internal error: Oops: 96000004 [#1] SMP
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.5.0-rc1 #5 [    0.000000] pstate: 20000085 (nzCv daIf -PAN -UAO)
[    0.000000] pc : gic_iterate_rdists+0x58/0x130
[    0.000000] lr : gic_iterate_rdists+0x80/0x130
[    0.000000] sp : ffff8000113d3cb0
[    0.000000] x29: ffff8000113d3cb0 x28: 0000000000000000
[    0.000000] x27: 0000000000000000 x26: 0000000000000018
[    0.000000] x25: 000000000000ffe8 x24: 000000000000003f
[    0.000000] x23: ffff800010588040 x22: 00000000000005e8
[    0.000000] x21: ffff8000113df7d0 x20: 0000030f00003f11
[    0.000000] x19: 0000000000000000 x18: ffffffffffffffff
[    0.000000] x17: 0000000014aeb8dc x16: 00000000c3ba0ccf
[    0.000000] x15: ffff8000113d9908 x14: ffff8000913d3a37
[    0.000000] x13: ffff8000113d3a45 x12: ffff800011402000
[    0.000000] x11: ffff8000113d39d0 x10: ffff8000113db980
[    0.000000] x9 : 00000000ffffffd0 x8 : ffff8000106dca98
[    0.000000] x7 : 000000000000005b x6 : 0000000000000000
[    0.000000] x5 : 0000000000000000 x4 : ffff8000128c0000
[    0.000000] x3 : ffff8000128a0000 x2 : ffff0003fc3c7000
[    0.000000] x1 : 0000000000000001 x0 : 000000000000ffe8
[    0.000000] Call trace:
[    0.000000]  gic_iterate_rdists+0x58/0x130
[    0.000000]  gic_init_bases+0x200/0x4b4
[    0.000000]  gic_acpi_init+0x148/0x284
[    0.000000]  acpi_match_madt+0x4c/0x84
[    0.000000]  acpi_table_parse_entries_array+0x188/0x278
[    0.000000]  acpi_table_parse_entries+0x70/0x98
[    0.000000]  acpi_table_parse_madt+0x40/0x50
[    0.000000]  __acpi_probe_device_table+0x88/0xe4
[    0.000000]  irqchip_init+0x38/0x40
[    0.000000]  init_IRQ+0x168/0x19c
[    0.000000]  start_kernel+0x328/0x508
[    0.000000] Code: f90017b6 9b3a7f16 f8766853 8b190260 (b9400000) [    0.000000] ---[ end trace ae5cf232d924bfc1 ]---
[    0.000000] Kernel panic - not syncing: Fatal exception
[    0.000000] Rebooting in 3 seconds..

In this case, nr_redist_regions counts all GICC structures but only enabled ones have redistributor mapped. So add check to avoid NULL deference of redist_base.

Cc Thomas Gleixner <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Marc Zyngier <[email protected]>

---
 drivers/irqchip/irq-gic-v3.c | 7 +++++++

 1 file changed, 7 insertions(+)
drivers/irqchip/irq-gic-v3.c +0 -0
index d6218012097b..bd9d55cadef9 100644
@@ -781,6 +781,13 @@ static int gic_iterate_rdists(int (*fn)(struct
 redist_region *, void __iomem *))
          u64 typer;
          u32 reg;
 
Quoted 15 lines
Quoted 11 lines
drivers/irqchip/irq-gic-v3.c +2 -0
+        /*
+         * redist_base may be NULL if we use single_redist and some 

GICC
+         * structure is disabled.
+         */
+        if (!ptr)
+            continue;
+
         reg = readl_relaxed(ptr + GICR_PIDR2) & GIC_PIDR2_ARCH_MASK; if (reg != GIC_PIDR2_ARCH_GICv3 &&
             reg != GIC_PIDR2_ARCH_GICv4) { /* We're in trouble... */

This feels like the wrong fix. The redistributor region array should be completely populated, and there is an assumption all over this driver that there is no junk in these structures.

Oh, I thought the place holder for disabled GICR in nr_redist_regions were for some special reason, like CPU hotplug. Now I know I was wrong :)

Quoted 6 lines

You're seeing this because we don't track the number of *enabled* rdists, and allocate the number of regions based on the number of overall GICC entries instead of the number of enabled redistributors.

How about this instead?

It looks good to me, and works fine in my case.

Thanks,

Heyi

Quoted 42 lines

        M.

drivers/irqchip/irq-gic-v3.c +4 -1
index 3a1866682dd0..9b26a860340b 100644
@@ -1861,6 +1861,7 @@ static struct
      struct redist_region *redist_regs;
      u32 nr_redist_regions;
      bool single_redist;
+    int enabled_rdists;
      u32 maint_irq;
      int maint_irq_mode;
      phys_addr_t vcpu_base;
@@ -1955,8 +1956,10 @@ static int __init gic_acpi_match_gicc(union 
 acpi_subtable_headers *header,
       * If GICC is enabled and has valid gicr base address, then it means
       * GICR base is presented via GICC
       */
-    if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address)
+    if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address) {
+        acpi_data.enabled_rdists++;
          return 0;
+    }
 
      /*

      * It's perfectly valid firmware can pass disabled GICC entry, driver

drivers/irqchip/irq-gic-v3.c +3 -1
@@ -1986,8 +1989,10 @@ static int __init 
 gic_acpi_count_gicr_regions(void)
 
      count = acpi_table_parse_madt(ACPI_MADT_TYPE_GENERIC_INTERRUPT,
                        gic_acpi_match_gicc, 0);
-    if (count > 0)
+    if (count > 0) {
          acpi_data.single_redist = true;
+        count = acpi_data.enabled_rdists;
+    }
 
      return count;

 }

maintainer
#3 lore raw
Quoted 100 lines · Guoheyi
Quoted 82 lines · Marc Zyngier

Hi Heyi,

Quoted 79 lines · Heyi Guo

If we use ACPI MADT GICC structure to pass single redistributor base, and mark some GICC as disabled, we'll get below call trace during boot:

[    0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0 [    0.000000] GICv3: 256 SPIs implemented
[    0.000000] GICv3: 0 Extended SPIs implemented
[    0.000000] GICv3: Distributor has no Range Selector support [    0.000000] Unable to handle kernel paging request at virtual address 000000000000ffe8
[    0.000000] Mem abort info:
[    0.000000]   ESR = 0x96000004
[    0.000000]   EC = 0x25: DABT (current EL), IL = 32 bits [    0.000000]   SET = 0, FnV = 0
[    0.000000]   EA = 0, S1PTW = 0
[    0.000000] Data abort info:
[    0.000000]   ISV = 0, ISS = 0x00000004
[    0.000000]   CM = 0, WnR = 0
[    0.000000] [000000000000ffe8] user address but active_mm is swapper [    0.000000] Internal error: Oops: 96000004 [#1] SMP
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.5.0-rc1 #5 [    0.000000] pstate: 20000085 (nzCv daIf -PAN -UAO)
[    0.000000] pc : gic_iterate_rdists+0x58/0x130
[    0.000000] lr : gic_iterate_rdists+0x80/0x130
[    0.000000] sp : ffff8000113d3cb0
[    0.000000] x29: ffff8000113d3cb0 x28: 0000000000000000
[    0.000000] x27: 0000000000000000 x26: 0000000000000018
[    0.000000] x25: 000000000000ffe8 x24: 000000000000003f
[    0.000000] x23: ffff800010588040 x22: 00000000000005e8
[    0.000000] x21: ffff8000113df7d0 x20: 0000030f00003f11
[    0.000000] x19: 0000000000000000 x18: ffffffffffffffff
[    0.000000] x17: 0000000014aeb8dc x16: 00000000c3ba0ccf
[    0.000000] x15: ffff8000113d9908 x14: ffff8000913d3a37
[    0.000000] x13: ffff8000113d3a45 x12: ffff800011402000
[    0.000000] x11: ffff8000113d39d0 x10: ffff8000113db980
[    0.000000] x9 : 00000000ffffffd0 x8 : ffff8000106dca98
[    0.000000] x7 : 000000000000005b x6 : 0000000000000000
[    0.000000] x5 : 0000000000000000 x4 : ffff8000128c0000
[    0.000000] x3 : ffff8000128a0000 x2 : ffff0003fc3c7000
[    0.000000] x1 : 0000000000000001 x0 : 000000000000ffe8
[    0.000000] Call trace:
[    0.000000]  gic_iterate_rdists+0x58/0x130
[    0.000000]  gic_init_bases+0x200/0x4b4
[    0.000000]  gic_acpi_init+0x148/0x284
[    0.000000]  acpi_match_madt+0x4c/0x84
[    0.000000]  acpi_table_parse_entries_array+0x188/0x278
[    0.000000]  acpi_table_parse_entries+0x70/0x98
[    0.000000]  acpi_table_parse_madt+0x40/0x50
[    0.000000]  __acpi_probe_device_table+0x88/0xe4
[    0.000000]  irqchip_init+0x38/0x40
[    0.000000]  init_IRQ+0x168/0x19c
[    0.000000]  start_kernel+0x328/0x508
[    0.000000] Code: f90017b6 9b3a7f16 f8766853 8b190260 (b9400000) [    0.000000] ---[ end trace ae5cf232d924bfc1 ]---
[    0.000000] Kernel panic - not syncing: Fatal exception
[    0.000000] Rebooting in 3 seconds..

In this case, nr_redist_regions counts all GICC structures but only enabled ones have redistributor mapped. So add check to avoid NULL deference of redist_base.

Cc Thomas Gleixner <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Marc Zyngier <[email protected]>

---
 drivers/irqchip/irq-gic-v3.c | 7 +++++++

 1 file changed, 7 insertions(+)
drivers/irqchip/irq-gic-v3.c +0 -0
index d6218012097b..bd9d55cadef9 100644
@@ -781,6 +781,13 @@ static int gic_iterate_rdists(int (*fn)(struct
 redist_region *, void __iomem *))
          u64 typer;
          u32 reg;
 
Quoted 15 lines
Quoted 11 lines
drivers/irqchip/irq-gic-v3.c +2 -0
+        /*
+         * redist_base may be NULL if we use single_redist and 

some GICC
+         * structure is disabled.
+         */
+        if (!ptr)
+            continue;
+
         reg = readl_relaxed(ptr + GICR_PIDR2) & GIC_PIDR2_ARCH_MASK; if (reg != GIC_PIDR2_ARCH_GICv3 &&
             reg != GIC_PIDR2_ARCH_GICv4) { /* We're in trouble... */

This feels like the wrong fix. The redistributor region array should be completely populated, and there is an assumption all over this driver that there is no junk in these structures.

Quoted 3 lines

Oh, I thought the place holder for disabled GICR in nr_redist_regions were for some special reason, like CPU hotplug. Now I know I was wrong :)

CPU hotplug would imply that the redistributors are available. My interpretation of the ACPI MADT GICC subtable is that the redistributors are simply inaccessible when disabled.

Otherwise, it'd be legitimate to just map them and live with redistributors that do not have a corresponding CPU (which we otherwise do). See ebe2f8718007 for details.

If we need to support redistributors becoming enabled under our feet, then we'll have to handle this in a different way. We're not there yet.

Quoted 7 lines
Quoted 5 lines

You're seeing this because we don't track the number of *enabled* rdists, and allocate the number of regions based on the number of overall GICC entries instead of the number of enabled redistributors.

How about this instead?

It looks good to me, and works fine in my case.

Can I take this as a Tested-by: ?

Thanks,

        M.
Signature · 1 line
Jazz is not dead. It just smells funny...
author Tested-by
#4 lore raw
Quoted 109 lines · Marc Zyngier
Quoted 100 lines · Guoheyi
Quoted 82 lines · Marc Zyngier

Hi Heyi,

Quoted 79 lines · Heyi Guo

If we use ACPI MADT GICC structure to pass single redistributor base, and mark some GICC as disabled, we'll get below call trace during boot:

[    0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0 [    0.000000] GICv3: 256 SPIs implemented
[    0.000000] GICv3: 0 Extended SPIs implemented
[    0.000000] GICv3: Distributor has no Range Selector support [    0.000000] Unable to handle kernel paging request at virtual address 000000000000ffe8
[    0.000000] Mem abort info:
[    0.000000]   ESR = 0x96000004
[    0.000000]   EC = 0x25: DABT (current EL), IL = 32 bits [    0.000000]   SET = 0, FnV = 0
[    0.000000]   EA = 0, S1PTW = 0
[    0.000000] Data abort info:
[    0.000000]   ISV = 0, ISS = 0x00000004
[    0.000000]   CM = 0, WnR = 0
[    0.000000] [000000000000ffe8] user address but active_mm is swapper [    0.000000] Internal error: Oops: 96000004 [#1] SMP
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.5.0-rc1 #5 [    0.000000] pstate: 20000085 (nzCv daIf -PAN -UAO)
[    0.000000] pc : gic_iterate_rdists+0x58/0x130
[    0.000000] lr : gic_iterate_rdists+0x80/0x130
[    0.000000] sp : ffff8000113d3cb0
[    0.000000] x29: ffff8000113d3cb0 x28: 0000000000000000
[    0.000000] x27: 0000000000000000 x26: 0000000000000018
[    0.000000] x25: 000000000000ffe8 x24: 000000000000003f
[    0.000000] x23: ffff800010588040 x22: 00000000000005e8
[    0.000000] x21: ffff8000113df7d0 x20: 0000030f00003f11
[    0.000000] x19: 0000000000000000 x18: ffffffffffffffff
[    0.000000] x17: 0000000014aeb8dc x16: 00000000c3ba0ccf
[    0.000000] x15: ffff8000113d9908 x14: ffff8000913d3a37
[    0.000000] x13: ffff8000113d3a45 x12: ffff800011402000
[    0.000000] x11: ffff8000113d39d0 x10: ffff8000113db980
[    0.000000] x9 : 00000000ffffffd0 x8 : ffff8000106dca98
[    0.000000] x7 : 000000000000005b x6 : 0000000000000000
[    0.000000] x5 : 0000000000000000 x4 : ffff8000128c0000
[    0.000000] x3 : ffff8000128a0000 x2 : ffff0003fc3c7000
[    0.000000] x1 : 0000000000000001 x0 : 000000000000ffe8
[    0.000000] Call trace:
[    0.000000]  gic_iterate_rdists+0x58/0x130
[    0.000000]  gic_init_bases+0x200/0x4b4
[    0.000000]  gic_acpi_init+0x148/0x284
[    0.000000]  acpi_match_madt+0x4c/0x84
[    0.000000]  acpi_table_parse_entries_array+0x188/0x278
[    0.000000]  acpi_table_parse_entries+0x70/0x98
[    0.000000]  acpi_table_parse_madt+0x40/0x50
[    0.000000]  __acpi_probe_device_table+0x88/0xe4
[    0.000000]  irqchip_init+0x38/0x40
[    0.000000]  init_IRQ+0x168/0x19c
[    0.000000]  start_kernel+0x328/0x508
[    0.000000] Code: f90017b6 9b3a7f16 f8766853 8b190260 (b9400000) [    0.000000] ---[ end trace ae5cf232d924bfc1 ]---
[    0.000000] Kernel panic - not syncing: Fatal exception
[    0.000000] Rebooting in 3 seconds..

In this case, nr_redist_regions counts all GICC structures but only enabled ones have redistributor mapped. So add check to avoid NULL deference of redist_base.

Cc Thomas Gleixner <[email protected]>
Cc Jason Cooper <[email protected]>
Cc Marc Zyngier <[email protected]>

---
 drivers/irqchip/irq-gic-v3.c | 7 +++++++

 1 file changed, 7 insertions(+)
drivers/irqchip/irq-gic-v3.c +0 -0
index d6218012097b..bd9d55cadef9 100644
@@ -781,6 +781,13 @@ static int gic_iterate_rdists(int (*fn)(struct
 redist_region *, void __iomem *))
          u64 typer;
          u32 reg;
 
Quoted 15 lines
Quoted 11 lines
drivers/irqchip/irq-gic-v3.c +2 -0
+        /*
+         * redist_base may be NULL if we use single_redist and 

some GICC
+         * structure is disabled.
+         */
+        if (!ptr)
+            continue;
+
         reg = readl_relaxed(ptr + GICR_PIDR2) & GIC_PIDR2_ARCH_MASK; if (reg != GIC_PIDR2_ARCH_GICv3 &&
             reg != GIC_PIDR2_ARCH_GICv4) { /* We're in trouble... */

This feels like the wrong fix. The redistributor region array should be completely populated, and there is an assumption all over this driver that there is no junk in these structures.

Quoted 3 lines

Oh, I thought the place holder for disabled GICR in nr_redist_regions were for some special reason, like CPU hotplug. Now I know I was wrong :)

CPU hotplug would imply that the redistributors are available. My interpretation of the ACPI MADT GICC subtable is that the redistributors are simply inaccessible when disabled.

Quoted 7 lines

Otherwise, it'd be legitimate to just map them and live with redistributors that do not have a corresponding CPU (which we otherwise do). See ebe2f8718007 for details.

If we need to support redistributors becoming enabled under our feet, then we'll have to handle this in a different way. We're not there yet.

Got it.

Quoted 10 lines
Quoted 8 lines
Quoted 6 lines

You're seeing this because we don't track the number of *enabled* rdists, and allocate the number of regions based on the number of overall GICC entries instead of the number of enabled redistributors.

How about this instead?

It looks good to me, and works fine in my case.

Can I take this as a Tested-by: ?

Sure.

Thanks,

Heyi

Quoted 4 lines

Thanks,

        M.

merge bot Applied
#5 lore raw

[tip: irq/urgent] irqchip/gic-v3: Only provision redistributors that are enabled in ACPI

In reply to Heyi Guo
The following commit has been merged into the irq/urgent branch of tip:

The following commit has been merged into the irq/urgent branch of tip:

Commit-ID:     926b5dfa6b8dc666ff398044af6906b156e1d949
Gitweb:        https://git.kernel.org/tip/926b5dfa6b8dc666ff398044af6906b156e1d949
Author:        Marc Zyngier <[email protected]>
AuthorDate:    Mon, 16 Dec 2019 11:24:57 
Committer:     Marc Zyngier <[email protected]>
CommitterDate: Tue, 28 Jan 2020 13:17:46 

irqchip/gic-v3: Only provision redistributors that are enabled in ACPI

We currently allocate redistributor region structures for individual redistributors when ACPI doesn't present us with compact MMIO regions covering multiple redistributors.

It turns out that we allocate these structures even when the redistributor is flagged as disabled by ACPI. It works fine until someone actually tries to tarse one of these structures, and access the corresponding MMIO region.

Instead, track the number of enabled redistributors, and only allocate what is required. This makes sure that there is no invalid data to misuse.

---
 drivers/irqchip/irq-gic-v3.c |  9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/irqchip/irq-gic-v3.c +7 -2
index 286f982..c1f7af9 100644
@@ -1839,6 +1839,7 @@ static struct
 	struct redist_region *redist_regs;
 	u32 nr_redist_regions;
 	bool single_redist;
+	int enabled_rdists;
 	u32 maint_irq;
 	int maint_irq_mode;
 	phys_addr_t vcpu_base;
@@ -1933,8 +1934,10 @@ static int __init gic_acpi_match_gicc(union acpi_subtable_headers *header,
 	 * If GICC is enabled and has valid gicr base address, then it means
 	 * GICR base is presented via GICC
 	 */
-	if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address)
+	if ((gicc->flags & ACPI_MADT_ENABLED) && gicc->gicr_base_address) {
+		acpi_data.enabled_rdists++;
 		return 0;
+	}
 
 	/*
 	 * It's perfectly valid firmware can pass disabled GICC entry, driver
@@ -1964,8 +1967,10 @@ static int __init gic_acpi_count_gicr_regions(void)
 
 	count = acpi_table_parse_madt(ACPI_MADT_TYPE_GENERIC_INTERRUPT,
 				      gic_acpi_match_gicc, 0);
-	if (count > 0)
+	if (count > 0) {
 		acpi_data.single_redist = true;
+		count = acpi_data.enabled_rdists;
+	}
 
 	return count;
 }