eth: bnxt: update header sizing defaults

JIRA: https://issues.redhat.com/browse/RHEL-106550

commit 99d028c
Author: Jakub Kicinski <kuba@kernel.org>
Date:   Sat Jan 18 18:05:17 2025 -0800

    eth: bnxt: update header sizing defaults

    300-400B RPC requests are fairly common. With the current default
    of 256B HDS threshold bnxt ends up splitting those, lowering PCIe
    bandwidth efficiency and increasing the number of memory allocation.

    Increase the HDS threshold to fit 4 buffers in a 4k page.
    This works out to 640B as the threshold on a typical kernel confing.
    This change increases the performance for a microbenchmark which
    receives 400B RPCs and sends empty responses by 4.5%.
    Admittedly this is just a single benchmark, but 256B works out to
    just 6 (so 2 more) packets per head page, because shinfo size
    dominates the headers.

    Now that we use page pool for the header pages I was also tempted
    to default rx_copybreak to 0, but in synthetic testing the copybreak
    size doesn't seem to make much difference.

    Reviewed-by: Michael Chan <michael.chan@broadcom.com>
    Link: https://patch.msgid.link/20250119020518.1962249-8-kuba@kernel.org
    Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Signed-off-by: Michal Schmidt <mschmidt@redhat.com>

Author: michich

drivers/net/ethernet/broadcom/bnxt/bnxt.c

@@ -4632,8 +4632,13 @@ void bnxt_set_tpa_flags(struct bnxt *bp)
 
 static void bnxt_init_ring_params(struct bnxt *bp)
 {
+	unsigned int rx_size;
+
 	bp->rx_copybreak = BNXT_DEFAULT_RX_COPYBREAK;
-	bp->dev->cfg->hds_thresh = BNXT_DEFAULT_RX_COPYBREAK;
+	/* Try to fit 4 chunks into a 4k page */
+	rx_size = SZ_1K -
+		NET_SKB_PAD - SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+	bp->dev->cfg->hds_thresh = max(BNXT_DEFAULT_RX_COPYBREAK, rx_size);
 }
 
 /* bp->rx_ring_size, bp->tx_ring_size, dev->mtu, BNXT_FLAG_{G|L}RO flags must