Now that vmbus_driver is a platform pci driver,
cleanup the irq allocation mess by using the standard
irq allocation mechanisms.
Note that this patch generates an error when the checkpatch
script is run because of the IRQF_SAMPLE_RANDOM flag used in
request_irq() function. This interrupt is the only
external event this VM will get and consequently if this
flag (IRQF_SAMPLE_RANDOM) is not specified, experimentally
we have shown that the entropy in the VM will very very low.
Signed-off-by: K. Y. Srinivasan <[email protected]>
Signed-off-by: Haiyang Zhang <[email protected]>
Signed-off-by: Mike Sterling <[email protected]>
Signed-off-by: Abhishek Kane <[email protected]>
Signed-off-by: Hank Janssen <[email protected]>
---
drivers/staging/hv/vmbus_drv.c | 30 ++++++++++++++----------------
1 files changed, 14 insertions(+), 16 deletions(-)
diff --git a/drivers/staging/hv/vmbus_drv.c b/drivers/staging/hv/vmbus_drv.c
index e4855ac..4e45016 100644
--- a/drivers/staging/hv/vmbus_drv.c
+++ b/drivers/staging/hv/vmbus_drv.c
@@ -17,6 +17,8 @@
* Authors:
* Haiyang Zhang <[email protected]>
* Hank Janssen <[email protected]>
+ *
+ * 3/9/2011: K. Y. Srinivasan - Significant restructuring and cleanup
*/
#include <linux/init.h>
#include <linux/err.h>
@@ -37,10 +39,6 @@
#include "vmbus_private.h"
-/* FIXME! We need to do this dynamically for PIC and APIC system */
-#define VMBUS_IRQ 0x5
-#define VMBUS_IRQ_VECTOR IRQ5_VECTOR
-
static struct device *root_dev; /* Root device */
struct pci_dev *hv_pci_dev;
@@ -74,7 +72,6 @@ EXPORT_SYMBOL(vmbus_loglevel);
/* (ALL_MODULES << 16 | DEBUG_LVL_ENTEREXIT); */
/* (((VMBUS | VMBUS_DRV)<<16) | DEBUG_LVL_ENTEREXIT); */
-static int vmbus_irq = VMBUS_IRQ;
/* Set up per device attributes in /sys/bus/vmbus/devices/<bus device> */
static struct device_attribute vmbus_device_attrs[] = {
@@ -375,7 +372,7 @@ static ssize_t vmbus_show_device_attr(struct device *dev,
* - setup the vmbus root device
* - retrieve the channel offers
*/
-static int vmbus_bus_init(void)
+static int vmbus_bus_init(struct pci_dev *pdev)
{
struct vmbus_driver_context *vmbus_drv_ctx = &vmbus_drv;
int ret;
@@ -418,21 +415,23 @@ static int vmbus_bus_init(void)
}
/* Get the interrupt resource */
- ret = request_irq(vmbus_irq, vmbus_isr, IRQF_SAMPLE_RANDOM,
- driver_name, NULL);
+ ret = request_irq(pdev->irq, vmbus_isr,
+ IRQF_SHARED | IRQF_SAMPLE_RANDOM,
+ driver_name, pdev);
if (ret != 0) {
DPRINT_ERR(VMBUS_DRV, "ERROR - Unable to request IRQ %d",
- vmbus_irq);
+ pdev->irq);
bus_unregister(&vmbus_drv_ctx->bus);
ret = -1;
goto cleanup;
}
- vector = VMBUS_IRQ_VECTOR;
- DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", vmbus_irq, vector);
+ vector = IRQ0_VECTOR + pdev->irq;
+ DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", pdev->irq,
+ IRQ0_VECTOR + pdev->irq);
/* Register the root device */
root_dev = root_device_register(driver_name);
@@ -441,7 +440,7 @@ static int vmbus_bus_init(void)
DPRINT_ERR(VMBUS_DRV,
"ERROR - Unable to register vmbus root device");
- free_irq(vmbus_irq, NULL);
+ free_irq(pdev->irq, pdev);
bus_unregister(&vmbus_drv_ctx->bus);
ret = PTR_ERR(root_dev);
@@ -456,7 +455,7 @@ static int vmbus_bus_init(void)
ret = vmbus_connect();
if (ret) {
root_device_unregister(root_dev);
- free_irq(vmbus_irq, NULL);
+ free_irq(pdev->irq, pdev);
bus_unregister(&vmbus_drv_ctx->bus);
goto cleanup;
}
@@ -490,7 +489,7 @@ static void vmbus_bus_exit(void)
bus_unregister(&vmbus_drv_ctx->bus);
- free_irq(vmbus_irq, NULL);
+ free_irq(hv_pci_dev->irq, hv_pci_dev);
tasklet_kill(&vmbus_drv_ctx->msg_dpc);
tasklet_kill(&vmbus_drv_ctx->event_dpc);
@@ -902,7 +901,7 @@ static int __devinit hv_pci_probe(struct pci_dev *pdev,
if (err)
return err;
- err = vmbus_bus_init();
+ err = vmbus_bus_init(pdev);
if (err)
pci_disable_device(pdev);
@@ -942,7 +941,6 @@ static void __exit hv_pci_exit(void)
MODULE_LICENSE("GPL");
MODULE_VERSION(HV_DRV_VERSION);
-module_param(vmbus_irq, int, S_IRUGO);
module_param(vmbus_loglevel, int, S_IRUGO);
module_init(hv_pci_init);
--
1.5.5.6
On Thu, 10 Mar 2011, K. Y. Srinivasan wrote:
> Now that vmbus_driver is a platform pci driver,
> cleanup the irq allocation mess by using the standard
> irq allocation mechanisms.
>
> Note that this patch generates an error when the checkpatch
> script is run because of the IRQF_SAMPLE_RANDOM flag used in
> request_irq() function. This interrupt is the only
> external event this VM will get and consequently if this
> flag (IRQF_SAMPLE_RANDOM) is not specified, experimentally
> we have shown that the entropy in the VM will very very low.
Fair enough. We need to come up with some way to work around
this though.
> }
> - vector = VMBUS_IRQ_VECTOR;
>
> - DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", vmbus_irq, vector);
> + vector = IRQ0_VECTOR + pdev->irq;
> + DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", pdev->irq,
> + IRQ0_VECTOR + pdev->irq);
Why evaluating vector first and then not using it for that debug print
thingy?
Btw, are you going to replace that DPRINT_* stuff as well ?
Thanks,
tglx
On Mar 10, 2011, at 14:46, "Thomas Gleixner" <[email protected]> wrote:
>>
>
>> }
>> - vector = VMBUS_IRQ_VECTOR;
>>
>> - DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", vmbus_irq, vector);
>> + vector = IRQ0_VECTOR + pdev->irq;
>> + DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", pdev->irq,
>> + IRQ0_VECTOR + pdev->irq);
>
> Why evaluating vector first and then not using it for that debug print
> thingy?
>
> Btw, are you going to replace that DPRINT_* stuff as well ?
>
> Thanks,
>
>
Yes, that is in my next set of patches.
Hank
> -----Original Message-----
> From: Hank Janssen
> Sent: Thursday, March 10, 2011 5:54 PM
> To: Thomas Gleixner
> Cc: KY Srinivasan; [email protected]; [email protected];
> [email protected]; [email protected]; Haiyang Zhang; Mike
> Sterling; Abhishek Kane (Mindtree Consulting PVT LTD)
> Subject: Re: [PATCH 12/21] Staging: hv: Cleanup irq management
>
>
>
>
>
>
> On Mar 10, 2011, at 14:46, "Thomas Gleixner" <[email protected]> wrote:
> >>
> >
> >> }
> >> - vector = VMBUS_IRQ_VECTOR;
> >>
> >> - DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", vmbus_irq, vector);
> >> + vector = IRQ0_VECTOR + pdev->irq;
> >> + DPRINT_INFO(VMBUS_DRV, "irq 0x%x vector 0x%x", pdev->irq,
> >> + IRQ0_VECTOR + pdev->irq);
> >
> > Why evaluating vector first and then not using it for that debug print
> > thingy?
Good point; I will fix this before Hank gets rid of the DPRINT_INFO altogether.
Regards,
K. Y