After calling irq_set_affinity_and_hint(), the cpumask pointer is
saved in desc->affinity_hint, and will be used later when reading
/proc/irq/<num>/affinity_hint. So the cpumask variable needs to be
persistent. Otherwise, we are accessing freed memory when reading
the affinity_hint file.
Also, need to clear affinity_hint before free_irq(), otherwise there
is a one-time warning and stack trace during module unloading:
[ 243.948687] WARNING: CPU: 10 PID: 1589 at kernel/irq/manage.c:1913 free_irq+0x318/0x360
...
[ 243.948753] Call Trace:
[ 243.948754] <TASK>
[ 243.948760] mana_gd_remove_irqs+0x78/0xc0 [mana]
[ 243.948767] mana_gd_remove+0x3e/0x80 [mana]
[ 243.948773] pci_device_remove+0x3d/0xb0
[ 243.948778] device_remove+0x46/0x70
[ 243.948782] device_release_driver_internal+0x1fe/0x280
[ 243.948785] driver_detach+0x4e/0xa0
[ 243.948787] bus_remove_driver+0x70/0xf0
[ 243.948789] driver_unregister+0x35/0x60
[ 243.948792] pci_unregister_driver+0x44/0x90
[ 243.948794] mana_driver_exit+0x14/0x3fe [mana]
[ 243.948800] __do_sys_delete_module.constprop.0+0x185/0x2f0
To fix the bug, use the persistent mask, cpumask_of(cpu#), and set
affinity_hint to NULL before freeing the IRQ, as required by free_irq().
Cc: [email protected]
Fixes: 71fa6887eeca ("net: mana: Assign interrupts to CPUs based on NUMA nodes")
Signed-off-by: Haiyang Zhang <[email protected]>
---
.../net/ethernet/microsoft/mana/gdma_main.c | 35 ++++++-------------
1 file changed, 10 insertions(+), 25 deletions(-)
diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c
index b144f2237748..a55d42332e20 100644
--- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
+++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
@@ -1218,8 +1218,6 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
struct gdma_context *gc = pci_get_drvdata(pdev);
struct gdma_irq_context *gic;
unsigned int max_irqs;
- u16 *cpus;
- cpumask_var_t req_mask;
int nvec, irq;
int err, i = 0, j;
@@ -1240,21 +1238,7 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
goto free_irq_vector;
}
- if (!zalloc_cpumask_var(&req_mask, GFP_KERNEL)) {
- err = -ENOMEM;
- goto free_irq;
- }
-
- cpus = kcalloc(nvec, sizeof(*cpus), GFP_KERNEL);
- if (!cpus) {
- err = -ENOMEM;
- goto free_mask;
- }
- for (i = 0; i < nvec; i++)
- cpus[i] = cpumask_local_spread(i, gc->numa_node);
-
for (i = 0; i < nvec; i++) {
- cpumask_set_cpu(cpus[i], req_mask);
gic = &gc->irq_contexts[i];
gic->handler = NULL;
gic->arg = NULL;
@@ -1269,17 +1253,16 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
irq = pci_irq_vector(pdev, i);
if (irq < 0) {
err = irq;
- goto free_mask;
+ goto free_irq;
}
err = request_irq(irq, mana_gd_intr, 0, gic->name, gic);
if (err)
- goto free_mask;
- irq_set_affinity_and_hint(irq, req_mask);
- cpumask_clear(req_mask);
+ goto free_irq;
+
+ irq_set_affinity_and_hint(irq, cpumask_of(cpumask_local_spread
+ (i, gc->numa_node)));
}
- free_cpumask_var(req_mask);
- kfree(cpus);
err = mana_gd_alloc_res_map(nvec, &gc->msix_resource);
if (err)
@@ -1290,13 +1273,12 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
return 0;
-free_mask:
- free_cpumask_var(req_mask);
- kfree(cpus);
free_irq:
for (j = i - 1; j >= 0; j--) {
irq = pci_irq_vector(pdev, j);
gic = &gc->irq_contexts[j];
+
+ irq_update_affinity_hint(irq, NULL);
free_irq(irq, gic);
}
@@ -1324,6 +1306,9 @@ static void mana_gd_remove_irqs(struct pci_dev *pdev)
continue;
gic = &gc->irq_contexts[i];
+
+ /* Need to clear the hint before free_irq */
+ irq_update_affinity_hint(irq, NULL);
free_irq(irq, gic);
}
--
2.25.1
From: LKML haiyangz <[email protected]> On Behalf Of Haiyang Zhang Sent: Wednesday, February 1, 2023 1:47 PM
>
> After calling irq_set_affinity_and_hint(), the cpumask pointer is
> saved in desc->affinity_hint, and will be used later when reading
> /proc/irq/<num>/affinity_hint. So the cpumask variable needs to be
> persistent. Otherwise, we are accessing freed memory when reading
> the affinity_hint file.
>
> Also, need to clear affinity_hint before free_irq(), otherwise there
> is a one-time warning and stack trace during module unloading:
>
> [ 243.948687] WARNING: CPU: 10 PID: 1589 at kernel/irq/manage.c:1913
> free_irq+0x318/0x360
> ...
> [ 243.948753] Call Trace:
> [ 243.948754] <TASK>
> [ 243.948760] mana_gd_remove_irqs+0x78/0xc0 [mana]
> [ 243.948767] mana_gd_remove+0x3e/0x80 [mana]
> [ 243.948773] pci_device_remove+0x3d/0xb0
> [ 243.948778] device_remove+0x46/0x70
> [ 243.948782] device_release_driver_internal+0x1fe/0x280
> [ 243.948785] driver_detach+0x4e/0xa0
> [ 243.948787] bus_remove_driver+0x70/0xf0
> [ 243.948789] driver_unregister+0x35/0x60
> [ 243.948792] pci_unregister_driver+0x44/0x90
> [ 243.948794] mana_driver_exit+0x14/0x3fe [mana]
> [ 243.948800] __do_sys_delete_module.constprop.0+0x185/0x2f0
>
> To fix the bug, use the persistent mask, cpumask_of(cpu#), and set
> affinity_hint to NULL before freeing the IRQ, as required by free_irq().
>
> Cc: [email protected]
> Fixes: 71fa6887eeca ("net: mana: Assign interrupts to CPUs based on NUMA nodes")
> Signed-off-by: Haiyang Zhang <[email protected]>
> ---
> .../net/ethernet/microsoft/mana/gdma_main.c | 35 ++++++-------------
> 1 file changed, 10 insertions(+), 25 deletions(-)
>
> diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c
> b/drivers/net/ethernet/microsoft/mana/gdma_main.c
> index b144f2237748..a55d42332e20 100644
> --- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
> +++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
> @@ -1218,8 +1218,6 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
> struct gdma_context *gc = pci_get_drvdata(pdev);
> struct gdma_irq_context *gic;
> unsigned int max_irqs;
> - u16 *cpus;
> - cpumask_var_t req_mask;
> int nvec, irq;
> int err, i = 0, j;
>
> @@ -1240,21 +1238,7 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
> goto free_irq_vector;
> }
>
> - if (!zalloc_cpumask_var(&req_mask, GFP_KERNEL)) {
> - err = -ENOMEM;
> - goto free_irq;
> - }
> -
> - cpus = kcalloc(nvec, sizeof(*cpus), GFP_KERNEL);
> - if (!cpus) {
> - err = -ENOMEM;
> - goto free_mask;
> - }
> - for (i = 0; i < nvec; i++)
> - cpus[i] = cpumask_local_spread(i, gc->numa_node);
> -
> for (i = 0; i < nvec; i++) {
> - cpumask_set_cpu(cpus[i], req_mask);
> gic = &gc->irq_contexts[i];
> gic->handler = NULL;
> gic->arg = NULL;
> @@ -1269,17 +1253,16 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
> irq = pci_irq_vector(pdev, i);
> if (irq < 0) {
> err = irq;
> - goto free_mask;
> + goto free_irq;
> }
>
> err = request_irq(irq, mana_gd_intr, 0, gic->name, gic);
> if (err)
> - goto free_mask;
> - irq_set_affinity_and_hint(irq, req_mask);
> - cpumask_clear(req_mask);
> + goto free_irq;
> +
> + irq_set_affinity_and_hint(irq, cpumask_of(cpumask_local_spread
> + (i, gc->numa_node)));
> }
> - free_cpumask_var(req_mask);
> - kfree(cpus);
>
> err = mana_gd_alloc_res_map(nvec, &gc->msix_resource);
> if (err)
> @@ -1290,13 +1273,12 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
>
> return 0;
>
> -free_mask:
> - free_cpumask_var(req_mask);
> - kfree(cpus);
> free_irq:
> for (j = i - 1; j >= 0; j--) {
> irq = pci_irq_vector(pdev, j);
> gic = &gc->irq_contexts[j];
> +
> + irq_update_affinity_hint(irq, NULL);
> free_irq(irq, gic);
> }
>
> @@ -1324,6 +1306,9 @@ static void mana_gd_remove_irqs(struct pci_dev *pdev)
> continue;
>
> gic = &gc->irq_contexts[i];
> +
> + /* Need to clear the hint before free_irq */
> + irq_update_affinity_hint(irq, NULL);
> free_irq(irq, gic);
> }
>
> --
> 2.25.1
Reviewed-by: Michael Kelley
On Wed, Feb 01, 2023 at 01:46:53PM -0800, Haiyang Zhang wrote:
> After calling irq_set_affinity_and_hint(), the cpumask pointer is
> saved in desc->affinity_hint, and will be used later when reading
> /proc/irq/<num>/affinity_hint. So the cpumask variable needs to be
> persistent. Otherwise, we are accessing freed memory when reading
> the affinity_hint file.
>
> Also, need to clear affinity_hint before free_irq(), otherwise there
> is a one-time warning and stack trace during module unloading:
>
> [ 243.948687] WARNING: CPU: 10 PID: 1589 at kernel/irq/manage.c:1913 free_irq+0x318/0x360
> ...
> [ 243.948753] Call Trace:
> [ 243.948754] <TASK>
> [ 243.948760] mana_gd_remove_irqs+0x78/0xc0 [mana]
> [ 243.948767] mana_gd_remove+0x3e/0x80 [mana]
> [ 243.948773] pci_device_remove+0x3d/0xb0
> [ 243.948778] device_remove+0x46/0x70
> [ 243.948782] device_release_driver_internal+0x1fe/0x280
> [ 243.948785] driver_detach+0x4e/0xa0
> [ 243.948787] bus_remove_driver+0x70/0xf0
> [ 243.948789] driver_unregister+0x35/0x60
> [ 243.948792] pci_unregister_driver+0x44/0x90
> [ 243.948794] mana_driver_exit+0x14/0x3fe [mana]
> [ 243.948800] __do_sys_delete_module.constprop.0+0x185/0x2f0
>
> To fix the bug, use the persistent mask, cpumask_of(cpu#), and set
> affinity_hint to NULL before freeing the IRQ, as required by free_irq().
>
> Cc: [email protected]
> Fixes: 71fa6887eeca ("net: mana: Assign interrupts to CPUs based on NUMA nodes")
> Signed-off-by: Haiyang Zhang <[email protected]>
> ---
> .../net/ethernet/microsoft/mana/gdma_main.c | 35 ++++++-------------
> 1 file changed, 10 insertions(+), 25 deletions(-)
>
Thanks,
Reviewed-by: Leon Romanovsky <[email protected]>
On Wed, 1 Feb 2023 13:46:53 -0800 Haiyang Zhang wrote:
> + irq_set_affinity_and_hint(irq, cpumask_of(cpumask_local_spread
> + (i, gc->numa_node)));
The line break here looks ugly.
Please use a local variable for the mask or the cpu.
> -----Original Message-----
> From: Jakub Kicinski <[email protected]>
> Sent: Thursday, February 2, 2023 2:38 PM
> To: Haiyang Zhang <[email protected]>
> Cc: [email protected]; [email protected]; Dexuan Cui
> <[email protected]>; KY Srinivasan <[email protected]>; Paul Rosswurm
> <[email protected]>; [email protected]; [email protected];
> [email protected]; [email protected]; [email protected]
> Subject: Re: [PATCH net,v2] net: mana: Fix accessing freed irq affinity_hint
>
> On Wed, 1 Feb 2023 13:46:53 -0800 Haiyang Zhang wrote:
> > + irq_set_affinity_and_hint(irq,
> cpumask_of(cpumask_local_spread
> > + (i, gc->numa_node)));
>
> The line break here looks ugly.
> Please use a local variable for the mask or the cpu.
Will do.