Skip to contentSkip to conversation

make e1000 use the checksuming code

PATCH In review Gerd Hoffmann 2 participants 2 messages Jul 29, 2008

lore
Share & tools
lore

Gerd Hoffmann

author
#0 lore raw

PATCH 1/1 make e1000 use the checksuming code

Hi,

This patch makes the e1000 nic driver use the checksumming code added with the previous patch. That one better should be reviewed by someone knowing the e1000 better than I do. I think I got it right though ...

cheers,
Gerd

Signature · 1 line
http://kraxel.fedorapeople.org/xenner/
Quoted 1 line

From 6b61fa35c52092c6372e9d095e2131370cc151cd Mon Sep 17 00:00:00 2001

From: Gerd Hoffmann <[email protected]>
Date: Tue, 29 Jul 2008 13:39:32 +0200
Subject: [PATCH 05/22] e1000: use common checksumming code.

Signed-off-by Gerd Hoffmann <[email protected]>
---
 hw/e1000.c |   20 ++++++--------------
 1 files changed, 6 insertions(+), 14 deletions(-)
hw/e1000.c +6 -14
index 1a14051..d81d9be 100644
@@ -279,26 +279,18 @@ flash_eerd_read(E1000State *s, int x)
            E1000_EEPROM_RW_REG_DONE | r;
 }
 
-static unsigned int
-do_cksum(uint8_t *dp, uint8_t *de)
-{
-    unsigned int bsum[2] = {0, 0}, i, sum;
-
-    for (i = 1; dp < de; bsum[i^=1] += *dp++)
-        ;
-    sum = (bsum[0] << 8) + bsum[1];
-    sum = (sum >> 16) + (sum & 0xffff);
-    return ~(sum + (sum >> 16));
-}
-
 static void
 putsum(uint8_t *data, uint32_t n, uint32_t sloc, uint32_t css, uint32_t cse)
 {
+    uint32_t sum;
+
     if (cse && cse < n)
         n = cse + 1;
-    if (sloc < n-1)
+    if (sloc < n-1) {
+        sum = net_checksum_add(n-css, data+css);
         cpu_to_be16wu((uint16_t *)(data + sloc),
-                      do_cksum(data + css, data + n));
+                      net_checksum_finish(sum));
+    }
 }
 
 static void
Signature · 1 line
1.5.4.1
#1 lore raw
Quoted 6 lines · Gerd Hoffmann

Hi,

This patch makes the e1000 nic driver use the checksumming code added with the previous patch. That one better should be reviewed by someone knowing the e1000 better than I do. I think I got it right though ...

Applied this and the v2 of the checksumming patch. I reviewed the checksumming code carefully for the e1000 and AFAICT, it's identical to the old behaviour. It also tested for packet corruption and everything appears okay.

Thanks.

Regards,

Anthony Liguori

Quoted 4 lines

cheers,
Gerd