[dpdk-dev,2/5] pci: Initialize common rte driver pointer
Checks
Commit Message
The existing code initializes a PCI driver pointer but not the common one.
As the result, ring_dma_zone_reserve() in drivers/net/bnx2x/bnx2x_rxtx.c
crashed as dev->device->driver==NULL.
This adds missing initialization.
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
---
lib/librte_eal/common/eal_common_pci.c | 1 +
1 file changed, 1 insertion(+)
Comments
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Alexey
> Kardashevskiy
> Sent: Thursday, April 20, 2017 8:24 AM
> To: dev@dpdk.org
> Cc: Alexey Kardashevskiy <aik@ozlabs.ru>; JPF@zurich.ibm.com;
> Gowrishankar Muthukrishnan <gowrishankar.m@in.ibm.com>
> Subject: [dpdk-dev] [PATCH dpdk 2/5] pci: Initialize common rte driver
> pointer
>
> The existing code initializes a PCI driver pointer but not the common one.
> As the result, ring_dma_zone_reserve() in drivers/net/bnx2x/bnx2x_rxtx.c
> crashed as dev->device->driver==NULL.
>
> This adds missing initialization.
>
> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
This and 3rd patch - could you please split them so that there are no unrelated patches in the patchset? I.e. a patchset for VFIO changes, a separate patch for PCI, and a patch for bnx2x driver, and then just make a note in the annotation that they are interdependent.
Thanks,
Anatoly
@@ -244,6 +244,7 @@ rte_eal_pci_probe_one_driver(struct rte_pci_driver *dr,
/* reference driver structure */
dev->driver = dr;
+ dev->device.driver = &dr->driver;
/* call the driver probe() function */
ret = dr->probe(dr, dev);