Netfilter Netdev ingress hook, skb->data point to packet(L3) and not to frame(L2)

Question

I'm using Netfilter (and next, add Netfilter_queue to send frames to the user-space app.) when I'm using the Netfilter hook with these values:

nfho.hook = hfunc;
nfho.hooknum = NF_NETDEV_INGRESS; // works for NF_INET_EGRESS;
nfho.pf = NFPROTO_NETDEV;// I tried this too:  NFPROTO_INET; and others.
nfho.priority = NF_IP_PRI_FIRST;
nfho.dev = dev_get_by_name(&init_net,"eth0");

when using ingress hooks, the skb->data points to the packet payload only, so I can see only the IP4/6/ Layer3 data etc (the skb->data doesn't point to the frame, which includes the src_mac dst_mac eth_type), but when changing to egress I can see the whole frame ( with the dmac-smac-ethType ). when using ingress the skb-> data points to the packet payload only, so I can see only the IP4 data etc (the skb doesn't point to the frame, which includes the mac etc'), but when changing to egress I can see the ETH header and the while frame.

I know I can get the eth type and MACs from the eth_header function, but my purpose is to send the packet to the user space and the packet uses only the layer3 data, but I want the payload include the frame too.

so my question is this: when using Netfilter ingress hook, does the Netfilter or any other kernel process, change the skb->data from pointing to the frame(L2) to point to the packet(L3)?

I'm not using xdp, because I want to use libnetfilter_queue next.

**THANK YOU FOR ANY HELP! THANK YOU FOR ANY HELP! **

my full code:

#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/netfilter.h>
#include <uapi/linux/netfilter_ipv4.h>
#include <linux/netfilter_bridge.h>
#include <linux/skbuff.h>
#include <linux/ip.h>
#include <net/ip.h>
#include <linux/if_ether.h>
#include <linux/netdevice.h>
static struct nf_hook_ops nfho; // Netfilter hook option struct
static void print_hex_dump(const char *prefix, const u8 *data, size_t len) {
 int i;
 printk(KERN_INFO "%s with len=%lu\n", prefix, len);
 for (i = 0; i < len; i++) {
 printk(KERN_CONT "%02X ", data[i]);
 if ((i + 1) % 16 == 0)
 printk(KERN_CONT "\n");
    }
 printk(KERN_CONT "\n");
}
static unsigned int hook_func(void *priv, struct sk_buff *skb, const struct nf_hook_state *state) {
 print_hex_dump("skb->data, skb->data_len",skb->data, skb->data_len);
 return NF_ACCEPT;
}
static int __init my_init_module(void) {
 nfho.dev = dev_get_by_name(&init_net, "eno0");
 nfho.hook = hook_func;
 nfho.hooknum = NF_NETDEV_INGRESS; 
 nfho.pf = NFPROTO_NETDEV; //NFPROTO_UNSPEC; //PF_PACKET; 
 nfho.priority = NF_IP_PRI_FIRST;
 
  nf_register_net_hook(&init_net, &nfho);
  printk(KERN_INFO "Hello \n");
 return 0;
}

static void __exit my_cleanup_module(void) {
 nf_unregister_net_hook(&init_net, &nfho); // Cleanup – unregister hook
 printk(KERN_INFO "BYE \n");
}

module_init(my_init_module);
module_exit(my_cleanup_module);

read all the netfilter docs. read all hooks docs, read about skb structs etc.