On Monday 09 January 2017 09:48 PM, Ferruh Yigit wrote:
On 1/9/2017 3:19 PM, Ferruh Yigit wrote:
On 12/26/2016 1:24 PM, Shreyansh Jain wrote:
These callbacks now act as first layer of PCI interfaces from the Bus.
Bus probe would enter the PMDs through the rte_driver->probe/remove
callbacks, falling to rte_xxx_driver->probe/remove (Currently, all the
drivers are rte_pci_driver).
This patch also changes QAT which is the only crypto PMD based on PCI.
All others would be changed in a separate patch focused on VDEV.
Signed-off-by: Shreyansh Jain <shreyansh.j...@nxp.com>
---
<...>
diff --git a/drivers/crypto/qat/rte_qat_cryptodev.c
b/drivers/crypto/qat/rte_qat_cryptodev.c
index 1e7ee61..bc1a9c6 100644
--- a/drivers/crypto/qat/rte_qat_cryptodev.c
+++ b/drivers/crypto/qat/rte_qat_cryptodev.c
@@ -120,6 +120,10 @@ crypto_qat_dev_init(__attribute__((unused)) struct
rte_cryptodev_driver *crypto_
static struct rte_cryptodev_driver rte_qat_pmd = {
.pci_drv = {
+ .driver = {
+ .probe = rte_eal_pci_probe,
+ .remove = rte_eal_pci_remove,
+ },
Since this part is common for all PCI drivers, why not make this part of
RTE_PMD_REGISTER_PCI macro?
I have seen your comment [1] which looks like better idea.
Thanks.
And, actually, it is mostly Thomas's idea :)
providing a rte_bus->probe(),
and rte_eal_bus_probe() calls rte_bus->probe()
for pci devices, rte_bus->probe = rte_eal_pci_probe ..
[1]
http://dpdk.org/ml/archives/dev/2017-January/054125.html
Thanks,
ferruh
