Hi,
> -----Original Message-----
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Rosen Xu
> Sent: Friday, December 14, 2018 1:14 AM
> To: dev@dpdk.org
> Cc: Lu, Wenzhuo <wenzhuo...@intel.com>; Wu, Jingjing
> <jingjing...@intel.com>; Iremonger, Bernard <bernard.iremon...@intel.com>;
> Xu, Rosen <rosen...@intel.com>; Yigit, Ferruh <ferruh.yi...@intel.com>
> Subject: [dpdk-dev] [PATCH v2] app/test-pmd: add IFPGA AFU register
> read/write access for testpmd
>
> Currently register read/write of testpmd is only for PCI device, but more and
> more IFPGA based AFU devices need this feature to access registers, this patch
> will add support for it.
>
> Signed-off-by: Rosen Xu <rosen...@intel.com>
> - pci_len = pci_dev->mem_resource[0].len;
> - if (reg_off >= pci_len) {
> + if (reg_off >= len) {
> printf("Port %d: register offset %u (0x%X) out of port PCI "
Here log message mentions only PCI not ifpga device. Might need to edit the log.
> port_reg_bit_display(portid_t port_id, uint32_t reg_off, uint8_t bit_x) {
> uint32_t reg_v;
> -
> + const struct rte_bus *bus;
>
> if (port_id_is_invalid(port_id, ENABLED_WARN))
> return;
> @@ -935,7 +940,16 @@ void print_valid_ports(void)
> return;
> if (reg_bit_pos_is_invalid(bit_x))
> return;
> - reg_v = port_id_pci_reg_read(port_id, reg_off);
> +
> + bus = rte_bus_find_by_device(ports[port_id].dev_info.device);
> + if (bus && !strcmp(bus->name, "pci")) {
> + reg_v = port_id_pci_reg_read(port_id, reg_off);
> + } else if (bus && !strcmp(bus->name, "ifpga")) {
> + reg_v = port_id_afu_reg_read(port_id, reg_off);
> + } else {
> + printf("Not a PCI or AFU device\n");
> + return;
> + }
Here and in other places for reg_read , we have similar code i.e. finding the
device , checking its type, if ifpga call ifpga function else call pci
functions.
Can this common code be moved to new function say pci_read_reg like
port_reg_set() which we have already.
Also , again inside respective pci/ifpga reg read/write we are checking for pci
type. So can all this be simplified, to remove redundant code.
Thanks,
Reshma
