From: Jason Wang <jasowang@redhat.com>
To: peter.maydell@linaro.org, qemu-devel@nongnu.org
Cc: Dmitry Fleytman <dmitry.fleytman@ravellosystems.com>,
Leonid Bloch <leonid.bloch@ravellosystems.com>,
Jason Wang <jasowang@redhat.com>
Subject: [Qemu-devel] [PULL V3 08/20] net: Introduce Toeplitz hash calculator
Date: Thu, 26 May 2016 10:16:23 +0800 [thread overview]
Message-ID: <1464228995-26657-9-git-send-email-jasowang@redhat.com> (raw)
In-Reply-To: <1464228995-26657-1-git-send-email-jasowang@redhat.com>
From: Dmitry Fleytman <dmitry.fleytman@ravellosystems.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Dmitry Fleytman <dmitry.fleytman@ravellosystems.com>
Signed-off-by: Leonid Bloch <leonid.bloch@ravellosystems.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
include/net/checksum.h | 45 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 45 insertions(+)
diff --git a/include/net/checksum.h b/include/net/checksum.h
index 7de1acb..dd8b4f6 100644
--- a/include/net/checksum.h
+++ b/include/net/checksum.h
@@ -18,6 +18,7 @@
#ifndef QEMU_NET_CHECKSUM_H
#define QEMU_NET_CHECKSUM_H
+#include "qemu/bswap.h"
struct iovec;
uint32_t net_checksum_add_cont(int len, uint8_t *buf, int seq);
@@ -50,4 +51,48 @@ uint32_t net_checksum_add_iov(const struct iovec *iov,
const unsigned int iov_cnt,
uint32_t iov_off, uint32_t size);
+typedef struct toeplitz_key_st {
+ uint32_t leftmost_32_bits;
+ uint8_t *next_byte;
+} net_toeplitz_key;
+
+static inline
+void net_toeplitz_key_init(net_toeplitz_key *key, uint8_t *key_bytes)
+{
+ key->leftmost_32_bits = be32_to_cpu(*(uint32_t *)key_bytes);
+ key->next_byte = key_bytes + sizeof(uint32_t);
+}
+
+static inline
+void net_toeplitz_add(uint32_t *result,
+ uint8_t *input,
+ uint32_t len,
+ net_toeplitz_key *key)
+{
+ register uint32_t accumulator = *result;
+ register uint32_t leftmost_32_bits = key->leftmost_32_bits;
+ register uint32_t byte;
+
+ for (byte = 0; byte < len; byte++) {
+ register uint8_t input_byte = input[byte];
+ register uint8_t key_byte = *(key->next_byte++);
+ register uint8_t bit;
+
+ for (bit = 0; bit < 8; bit++) {
+ if (input_byte & (1 << 7)) {
+ accumulator ^= leftmost_32_bits;
+ }
+
+ leftmost_32_bits =
+ (leftmost_32_bits << 1) | ((key_byte & (1 << 7)) >> 7);
+
+ input_byte <<= 1;
+ key_byte <<= 1;
+ }
+ }
+
+ key->leftmost_32_bits = leftmost_32_bits;
+ *result = accumulator;
+}
+
#endif /* QEMU_NET_CHECKSUM_H */
--
2.7.4
next prev parent reply other threads:[~2016-05-26 2:17 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-26 2:16 [Qemu-devel] [PULL V3 00/20] Net patches Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 01/20] net/tap: Allocating Large sized arrays to heap Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 02/20] net: mipsnet: check packet length against buffer Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 03/20] msix: make msix_clr_pending() visible for clients Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 04/20] pci: Introduce define for PM capability version 1.1 Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 05/20] pcie: Add support for PCIe CAP v1 Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 06/20] pcie: Introduce function for DSN capability creation Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 07/20] vmxnet3: Use generic function for DSN capability definition Jason Wang
2016-05-26 2:16 ` Jason Wang [this message]
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 09/20] net: Add macros for MAC address tracing Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 10/20] vmxnet3: Use common MAC address tracing macros Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 11/20] net_pkt: Name vmxnet3 packet abstractions more generic Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 12/20] rtl8139: Move more TCP definitions to common header Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 13/20] net_pkt: Extend packet abstraction as required by e1000e functionality Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 14/20] vmxnet3: Use pci_dma_* API instead of cpu_physical_memory_* Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 15/20] e1000_regs: Add definitions for Intel 82574-specific bits Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 16/20] e1000: Move out code that will be reused in e1000e Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 17/20] net: Introduce e1000e device emulation Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 18/20] e1000e: Introduce qtest for e1000e device Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 19/20] net: vl: Move default_net to vl.c Jason Wang
2016-05-26 2:16 ` [Qemu-devel] [PULL V3 20/20] net/net: Add SocketReadState for reuse codes Jason Wang
2016-05-26 15:08 ` [Qemu-devel] [PULL V3 00/20] Net patches Peter Maydell
2016-05-26 15:20 ` Eric Blake
2016-05-26 15:28 ` Peter Maydell
2016-05-27 3:35 ` Jason Wang
2016-05-27 9:03 ` Peter Maydell
2016-05-29 15:22 ` Dmitry Fleytman
2016-05-29 16:45 ` Peter Maydell
2016-05-30 1:51 ` Jason Wang
2016-05-30 11:52 ` Peter Maydell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1464228995-26657-9-git-send-email-jasowang@redhat.com \
--to=jasowang@redhat.com \
--cc=dmitry.fleytman@ravellosystems.com \
--cc=leonid.bloch@ravellosystems.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).