Skip to contentSkip to conversation

CCID 4: Use 1460 bytes as segment size in TCP

PATCH NAK Tommi Saviranta 3 participants 3 messages Sep 20, 2007 – Sep 22, 2007 Networking

lore
Share & tools
lore

Tommi Saviranta

author
#0 lore raw

PATCH 5/5 CCID 4: Use 1460 bytes as segment size in TCP

Uses hard-coded value of 1460 bytes in TCP throughput calculation.

Signed-off-by Tommi Saviranta <[email protected]>

---

net/dccp/ccids/ccid4.c +3 -3
diff -Nrup a/net/dccp/ccids/ccid4.c b/net/dccp/ccids/ccid4.c
@@ -459,7 +459,7 @@ static void ccid4_hc_tx_packet_recv(stru
 	}
 	/* perform step (4) of draft rfc3448bis, section 4.3 */
 	if (hctx->ccid4hctx_p > 0)
-		hctx->ccid4hctx_x_calc = tfrc_calc_x(hctx->ccid4hctx_s,
+		hctx->ccid4hctx_x_calc = tfrc_calc_x(1460,
 						     hctx->ccid4hctx_rtt,
 						     hctx->ccid4hctx_p);
 	ccid4_hc_tx_update_x(sk, &now);

@@ -740,7 +740,7 @@ static int ccid4_hc_rx_insert_options(st
 	return 0;
 }
 
-/** ccid4_first_li  -  Implements [RFC 3448, 6.3.1]
+/** ccid4_first_li  -  Implements [RFC 3448, 6.3.1], with s = 1460 [TFRC-SP]
  *
  * Determine the length of the first loss interval via inverse lookup.
  * Assume that X_recv can be computed by the throughput equation

@@ -770,7 +770,7 @@ static u32 ccid4_first_li(struct sock *s
 		}
 	}
 
-	fval = scaled_div(hcrx->ccid4hcrx_s, hcrx->ccid4hcrx_rtt);
+	fval = scaled_div(1460, hcrx->ccid4hcrx_rtt);
 	fval = scaled_div32(fval, x_recv);
 	p = tfrc_calc_x_reverse_lookup(fval);
Acked-by Leandro Sales on v1 5/5
#1 lore raw

Re: [PATCH 5/5] CCID 4: Use 1460 bytes as segment size in TCP throughput calculation

2007/9/20, Tommi Saviranta <[email protected]>:

Quoted 35 lines

Uses hard-coded value of 1460 bytes in TCP throughput calculation.

Signed-off-by Tommi Saviranta <[email protected]>

---
diff -Nrup a/net/dccp/ccids/ccid4.c b/net/dccp/ccids/ccid4.c

net/dccp/ccids/ccid4.c 2007-09-20 14:02:43.000000000 +0300 renamed from net/dccp/ccids/ccid4.c 2007-09-20 13:32:42.000000000 +0300 +3 -4
@@ -459,7 +459,7 @@ static void ccid4_hc_tx_packet_recv(stru
        }
        /* perform step (4) of draft rfc3448bis, section 4.3 */
        if (hctx->ccid4hctx_p > 0)
-               hctx->ccid4hctx_x_calc = tfrc_calc_x(hctx->ccid4hctx_s,
+               hctx->ccid4hctx_x_calc = tfrc_calc_x(1460,
                                                     hctx->ccid4hctx_rtt,
                                                     hctx->ccid4hctx_p);
        ccid4_hc_tx_update_x(sk, &now);

@@ -740,7 +740,7 @@ static int ccid4_hc_rx_insert_options(st
        return 0;
 }
 
-/** ccid4_first_li  -  Implements [RFC 3448, 6.3.1]
+/** ccid4_first_li  -  Implements [RFC 3448, 6.3.1], with s = 1460 [TFRC-SP]
  *
  * Determine the length of the first loss interval via inverse lookup.
  * Assume that X_recv can be computed by the throughput equation

@@ -770,7 +770,7 @@ static u32 ccid4_first_li(struct sock *s
                }
        }
 
-       fval = scaled_div(hcrx->ccid4hcrx_s, hcrx->ccid4hcrx_rtt);
+       fval = scaled_div(1460, hcrx->ccid4hcrx_rtt);
        fval = scaled_div32(fval, x_recv);
        p = tfrc_calc_x_reverse_lookup(fval);
-
Quoted 4 lines

To unsubscribe from this list: send the line "unsubscribe dccp" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Acked-by Leandro Sales <[email protected]>
#2 lore raw

Re: [PATCH 5/5] CCID 4: Use 1460 bytes as segment size in TCP throughput calculation

In reply to Tommi Saviranta
Quoted 1 line · Tommi Saviranta

+/** ccid4_first_li - Implements [RFC 3448, 6.3.1], with s = 1460 [TFRC-SP]

Put RFC4828 here instead of TFRC-SP.

Otherwise looks fine.

Signature · 3 lines
Web1: http://wand.net.nz/~iam4/
Web2: http://www.jandi.co.nz
Blog: http://iansblog.jandi.co.nz