VFs call adf_dev_stop() from a PF to VF interrupt bottom half.
This causes an oops "scheduling while atomic", because it tries
to acquire a mutex to un-register crypto algorithms.
This patch fixes the issue by calling adf_dev_stop() asynchronously.
Changes in v2:
- change kthread to a work queue.
Signed-off-by: Tadeusz Struk <[email protected]>
---
drivers/crypto/qat/qat_common/adf_common_drv.h | 2 +
drivers/crypto/qat/qat_common/adf_ctl_drv.c | 6 ++
drivers/crypto/qat/qat_common/adf_vf_isr.c | 59 +++++++++++++++++++++++-
3 files changed, 64 insertions(+), 3 deletions(-)
diff --git a/drivers/crypto/qat/qat_common/adf_common_drv.h b/drivers/crypto/qat/qat_common/adf_common_drv.h
index c9e4d46..fd096ed 100644
--- a/drivers/crypto/qat/qat_common/adf_common_drv.h
+++ b/drivers/crypto/qat/qat_common/adf_common_drv.h
@@ -144,6 +144,8 @@ void adf_disable_aer(struct adf_accel_dev *accel_dev);
void adf_dev_restore(struct adf_accel_dev *accel_dev);
int adf_init_aer(void);
void adf_exit_aer(void);
+int adf_init_vf_wq(void);
+void adf_exit_vf_wq(void);
int adf_init_admin_comms(struct adf_accel_dev *accel_dev);
void adf_exit_admin_comms(struct adf_accel_dev *accel_dev);
int adf_send_admin_init(struct adf_accel_dev *accel_dev);
diff --git a/drivers/crypto/qat/qat_common/adf_ctl_drv.c b/drivers/crypto/qat/qat_common/adf_ctl_drv.c
index 48a1248..116ddda 100644
--- a/drivers/crypto/qat/qat_common/adf_ctl_drv.c
+++ b/drivers/crypto/qat/qat_common/adf_ctl_drv.c
@@ -471,12 +471,17 @@ static int __init adf_register_ctl_device_driver(void)
if (adf_init_aer())
goto err_aer;
+ if (adf_init_vf_wq())
+ goto err_vf_wq;
+
if (qat_crypto_register())
goto err_crypto_register;
return 0;
err_crypto_register:
+ adf_exit_vf_wq();
+err_vf_wq:
adf_exit_aer();
err_aer:
adf_chr_drv_destroy();
@@ -489,6 +494,7 @@ static void __exit adf_unregister_ctl_device_driver(void)
{
adf_chr_drv_destroy();
adf_exit_aer();
+ adf_exit_vf_wq();
qat_crypto_unregister();
adf_clean_vf_map(false);
mutex_destroy(&adf_ctl_lock);
diff --git a/drivers/crypto/qat/qat_common/adf_vf_isr.c b/drivers/crypto/qat/qat_common/adf_vf_isr.c
index 09427b3..c3d5016 100644
--- a/drivers/crypto/qat/qat_common/adf_vf_isr.c
+++ b/drivers/crypto/qat/qat_common/adf_vf_isr.c
@@ -51,6 +51,7 @@
#include <linux/slab.h>
#include <linux/errno.h>
#include <linux/interrupt.h>
+#include <linux/workqueue.h>
#include "adf_accel_devices.h"
#include "adf_common_drv.h"
#include "adf_cfg.h"
@@ -64,6 +65,13 @@
#define ADF_VINTSOU_BUN BIT(0)
#define ADF_VINTSOU_PF2VF BIT(1)
+static struct workqueue_struct *adf_vf_stop_wq;
+
+struct adf_vf_stop_data {
+ struct adf_accel_dev *accel_dev;
+ struct work_struct work;
+};
+
static int adf_enable_msi(struct adf_accel_dev *accel_dev)
{
struct adf_accel_pci *pci_dev_info = &accel_dev->accel_pci_dev;
@@ -90,6 +98,20 @@ static void adf_disable_msi(struct adf_accel_dev *accel_dev)
pci_disable_msi(pdev);
}
+static void adf_dev_stop_async(struct work_struct *work)
+{
+ struct adf_vf_stop_data *stop_data =
+ container_of(work, struct adf_vf_stop_data, work);
+ struct adf_accel_dev *accel_dev = stop_data->accel_dev;
+
+ adf_dev_stop(accel_dev);
+ adf_dev_shutdown(accel_dev);
+
+ /* Re-enable PF2VF interrupts */
+ adf_enable_pf2vf_interrupts(accel_dev);
+ kfree(stop_data);
+}
+
static void adf_pf2vf_bh_handler(void *data)
{
struct adf_accel_dev *accel_dev = data;
@@ -107,11 +129,27 @@ static void adf_pf2vf_bh_handler(void *data)
goto err;
switch ((msg & ADF_PF2VF_MSGTYPE_MASK) >> ADF_PF2VF_MSGTYPE_SHIFT) {
- case ADF_PF2VF_MSGTYPE_RESTARTING:
+ case ADF_PF2VF_MSGTYPE_RESTARTING: {
+ struct adf_vf_stop_data *stop_data;
+
dev_dbg(&GET_DEV(accel_dev),
"Restarting msg received from PF 0x%x\n", msg);
- adf_dev_stop(accel_dev);
- break;
+
+ stop_data = kzalloc(sizeof(*stop_data), GFP_ATOMIC);
+ if (!stop_data) {
+ dev_err(&GET_DEV(accel_dev),
+ "Couldn't schedule stop for vf_%d\n",
+ accel_dev->accel_id);
+ return;
+ }
+ stop_data->accel_dev = accel_dev;
+ INIT_WORK(&stop_data->work, adf_dev_stop_async);
+ queue_work(adf_vf_stop_wq, &stop_data->work);
+ /* To ack, clear the PF2VFINT bit */
+ msg &= ~BIT(0);
+ ADF_CSR_WR(pmisc_bar_addr, hw_data->get_pf2vf_offset(0), msg);
+ return;
+ }
case ADF_PF2VF_MSGTYPE_VERSION_RESP:
dev_dbg(&GET_DEV(accel_dev),
"Version resp received from PF 0x%x\n", msg);
@@ -278,3 +316,18 @@ err_out:
return -EFAULT;
}
EXPORT_SYMBOL_GPL(adf_vf_isr_resource_alloc);
+
+int __init adf_init_vf_wq(void)
+{
+ adf_vf_stop_wq = create_workqueue("adf_vf_stop_wq");
+
+ return !adf_vf_stop_wq ? -EFAULT : 0;
+}
+
+void __exit adf_exit_vf_wq(void)
+{
+ if (adf_vf_stop_wq)
+ destroy_workqueue(adf_vf_stop_wq);
+
+ adf_vf_stop_wq = NULL;
+}
On Wed, Apr 06, 2016 at 11:01:54AM -0700, Tadeusz Struk wrote:
> VFs call adf_dev_stop() from a PF to VF interrupt bottom half.
> This causes an oops "scheduling while atomic", because it tries
> to acquire a mutex to un-register crypto algorithms.
> This patch fixes the issue by calling adf_dev_stop() asynchronously.
Applied.
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt