* [PATCH net-next 0/3] A own subdirectory for shared TCP code
@ 2017-10-03 22:22 Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory Richard Sailer
` (4 more replies)
0 siblings, 5 replies; 10+ messages in thread
From: Richard Sailer @ 2017-10-03 22:22 UTC (permalink / raw)
To: netdev, davem; +Cc: kuznet, yoshfuji
net/ipv4 currently contains around 100 source files containing the
IP implementation and lots of other functionality (UDP, TCP, xfrm, etc.)
# 1/3
To make the networking source tree more self documenting and well
structured 1/3 moves the 30 shared TCP source files to a own
subdirectory of net/ipv4/.
# 2/3
A huge part of the TCP code is congestion control algorithms.
With the same goal as above, this moves the cc algos to a own
subdirectory of net/ipv4/tcp/
# 3/3
Implementation of a suggestion of valdis, to have a new directory
net/ipv4v6_shared that contains all the code shared between
IPv4 and IPv6. This creates net/ipv4v6_shared and moves the
net/ipv4/tcp directory of shared TCP code there.
While every commit depends on its previous commits, it is
also possible and sensible to only apply (1) or (1,2) , depending
on how conservative one wants to decide regarding directory
structure changes.
[PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory
[PATCH net-next 2/3] tcp: Move cc algorithms to own subdirectory
[PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory
2017-10-03 22:22 [PATCH net-next 0/3] A own subdirectory for shared TCP code Richard Sailer
@ 2017-10-03 22:22 ` Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 2/3] tcp: Move cc algorithms " Richard Sailer
` (3 subsequent siblings)
4 siblings, 0 replies; 10+ messages in thread
From: Richard Sailer @ 2017-10-03 22:22 UTC (permalink / raw)
To: netdev, davem; +Cc: kuznet, yoshfuji
net/ipv4 contains around 100 source files containing the IP implementation
and various other functionality (UDP, TCP, xfrm, etc.). 30 of them make up
the TCP implementation.
I think moving the TCP implementation to a own subdirectory
makes the source tree more explorable and well structured.
Note: This only affects the TCP Code shared by IPv4 and IPv6,
the IPv4 specific part (tcp_ipv4.c and tcp_offload.c) is left in
net/ipv4/.
This creates the directory, moves the files and updates the Makefiles
acordingly.
Signed-off-by: Richard Sailer <richard_siegfried@systemli.org>
---
net/ipv4/Makefile | 25 ++-----------------------
net/ipv4/tcp/Makefile | 26 ++++++++++++++++++++++++++
net/ipv4/{ => tcp}/tcp.c | 0
net/ipv4/{ => tcp}/tcp_bbr.c | 0
net/ipv4/{ => tcp}/tcp_bic.c | 0
net/ipv4/{ => tcp}/tcp_cdg.c | 0
net/ipv4/{ => tcp}/tcp_cong.c | 0
net/ipv4/{ => tcp}/tcp_cubic.c | 0
net/ipv4/{ => tcp}/tcp_dctcp.c | 0
net/ipv4/{ => tcp}/tcp_diag.c | 0
net/ipv4/{ => tcp}/tcp_fastopen.c | 0
net/ipv4/{ => tcp}/tcp_highspeed.c | 0
net/ipv4/{ => tcp}/tcp_htcp.c | 0
net/ipv4/{ => tcp}/tcp_hybla.c | 0
net/ipv4/{ => tcp}/tcp_illinois.c | 0
net/ipv4/{ => tcp}/tcp_input.c | 0
net/ipv4/{ => tcp}/tcp_lp.c | 0
net/ipv4/{ => tcp}/tcp_metrics.c | 0
net/ipv4/{ => tcp}/tcp_minisocks.c | 0
net/ipv4/{ => tcp}/tcp_nv.c | 0
net/ipv4/{ => tcp}/tcp_output.c | 0
net/ipv4/{ => tcp}/tcp_probe.c | 0
net/ipv4/{ => tcp}/tcp_rate.c | 0
net/ipv4/{ => tcp}/tcp_recovery.c | 0
net/ipv4/{ => tcp}/tcp_scalable.c | 0
net/ipv4/{ => tcp}/tcp_timer.c | 0
net/ipv4/{ => tcp}/tcp_ulp.c | 0
net/ipv4/{ => tcp}/tcp_vegas.c | 0
net/ipv4/{ => tcp}/tcp_vegas.h | 0
net/ipv4/{ => tcp}/tcp_veno.c | 0
net/ipv4/{ => tcp}/tcp_westwood.c | 0
net/ipv4/{ => tcp}/tcp_yeah.c | 0
32 files changed, 28 insertions(+), 23 deletions(-)
create mode 100644 net/ipv4/tcp/Makefile
rename net/ipv4/{ => tcp}/tcp.c (100%)
rename net/ipv4/{ => tcp}/tcp_bbr.c (100%)
rename net/ipv4/{ => tcp}/tcp_bic.c (100%)
rename net/ipv4/{ => tcp}/tcp_cdg.c (100%)
rename net/ipv4/{ => tcp}/tcp_cong.c (100%)
rename net/ipv4/{ => tcp}/tcp_cubic.c (100%)
rename net/ipv4/{ => tcp}/tcp_dctcp.c (100%)
rename net/ipv4/{ => tcp}/tcp_diag.c (100%)
rename net/ipv4/{ => tcp}/tcp_fastopen.c (100%)
rename net/ipv4/{ => tcp}/tcp_highspeed.c (100%)
rename net/ipv4/{ => tcp}/tcp_htcp.c (100%)
rename net/ipv4/{ => tcp}/tcp_hybla.c (100%)
rename net/ipv4/{ => tcp}/tcp_illinois.c (100%)
rename net/ipv4/{ => tcp}/tcp_input.c (100%)
rename net/ipv4/{ => tcp}/tcp_lp.c (100%)
rename net/ipv4/{ => tcp}/tcp_metrics.c (100%)
rename net/ipv4/{ => tcp}/tcp_minisocks.c (100%)
rename net/ipv4/{ => tcp}/tcp_nv.c (100%)
rename net/ipv4/{ => tcp}/tcp_output.c (100%)
rename net/ipv4/{ => tcp}/tcp_probe.c (100%)
rename net/ipv4/{ => tcp}/tcp_rate.c (100%)
rename net/ipv4/{ => tcp}/tcp_recovery.c (100%)
rename net/ipv4/{ => tcp}/tcp_scalable.c (100%)
rename net/ipv4/{ => tcp}/tcp_timer.c (100%)
rename net/ipv4/{ => tcp}/tcp_ulp.c (100%)
rename net/ipv4/{ => tcp}/tcp_vegas.c (100%)
rename net/ipv4/{ => tcp}/tcp_vegas.h (100%)
rename net/ipv4/{ => tcp}/tcp_veno.c (100%)
rename net/ipv4/{ => tcp}/tcp_westwood.c (100%)
rename net/ipv4/{ => tcp}/tcp_yeah.c (100%)
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index afcb435adfbe..a106a1cf0e11 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -2,14 +2,11 @@
# Makefile for the Linux TCP/IP (INET) layer.
#
-obj-y := route.o inetpeer.o protocol.o \
+obj-y := tcp/ route.o inetpeer.o protocol.o \
ip_input.o ip_fragment.o ip_forward.o ip_options.o \
ip_output.o ip_sockglue.o inet_hashtables.o \
inet_timewait_sock.o inet_connection_sock.o \
- tcp.o tcp_input.o tcp_output.o tcp_timer.o tcp_ipv4.o \
- tcp_minisocks.o tcp_cong.o tcp_metrics.o tcp_fastopen.o \
- tcp_rate.o tcp_recovery.o tcp_ulp.o \
- tcp_offload.o datagram.o raw.o udp.o udplite.o \
+ datagram.o raw.o udp.o udplite.o tcp_offload.o tcp_ipv4.o \
udp_offload.o arp.o icmp.o devinet.o af_inet.o igmp.o \
fib_frontend.o fib_semantics.o fib_trie.o fib_notifier.o \
inet_fragment.o ping.o ip_tunnel_core.o gre_offload.o
@@ -39,26 +36,8 @@ obj-$(CONFIG_INET_XFRM_MODE_TUNNEL) += xfrm4_mode_tunnel.o
obj-$(CONFIG_IP_PNP) += ipconfig.o
obj-$(CONFIG_NETFILTER) += netfilter.o netfilter/
obj-$(CONFIG_INET_DIAG) += inet_diag.o
-obj-$(CONFIG_INET_TCP_DIAG) += tcp_diag.o
obj-$(CONFIG_INET_UDP_DIAG) += udp_diag.o
obj-$(CONFIG_INET_RAW_DIAG) += raw_diag.o
-obj-$(CONFIG_NET_TCPPROBE) += tcp_probe.o
-obj-$(CONFIG_TCP_CONG_BBR) += tcp_bbr.o
-obj-$(CONFIG_TCP_CONG_BIC) += tcp_bic.o
-obj-$(CONFIG_TCP_CONG_CDG) += tcp_cdg.o
-obj-$(CONFIG_TCP_CONG_CUBIC) += tcp_cubic.o
-obj-$(CONFIG_TCP_CONG_DCTCP) += tcp_dctcp.o
-obj-$(CONFIG_TCP_CONG_WESTWOOD) += tcp_westwood.o
-obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o
-obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o
-obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o
-obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o
-obj-$(CONFIG_TCP_CONG_NV) += tcp_nv.o
-obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o
-obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
-obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
-obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
-obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
obj-$(CONFIG_NETLABEL) += cipso_ipv4.o
obj-$(CONFIG_XFRM) += xfrm4_policy.o xfrm4_state.o xfrm4_input.o \
diff --git a/net/ipv4/tcp/Makefile b/net/ipv4/tcp/Makefile
new file mode 100644
index 000000000000..91d2c991a243
--- /dev/null
+++ b/net/ipv4/tcp/Makefile
@@ -0,0 +1,26 @@
+#
+# Makefile for Linux TCP
+#
+
+obj-y := tcp.o tcp_input.o tcp_output.o tcp_timer.o \
+ tcp_minisocks.o tcp_cong.o tcp_metrics.o tcp_fastopen.o \
+ tcp_rate.o tcp_recovery.o tcp_ulp.o
+
+obj-$(CONFIG_INET_TCP_DIAG) += tcp_diag.o
+obj-$(CONFIG_NET_TCPPROBE) += tcp_probe.o
+obj-$(CONFIG_TCP_CONG_BBR) += tcp_bbr.o
+obj-$(CONFIG_TCP_CONG_BIC) += tcp_bic.o
+obj-$(CONFIG_TCP_CONG_CDG) += tcp_cdg.o
+obj-$(CONFIG_TCP_CONG_CUBIC) += tcp_cubic.o
+obj-$(CONFIG_TCP_CONG_DCTCP) += tcp_dctcp.o
+obj-$(CONFIG_TCP_CONG_WESTWOOD) += tcp_westwood.o
+obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o
+obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o
+obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o
+obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o
+obj-$(CONFIG_TCP_CONG_NV) += tcp_nv.o
+obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o
+obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
+obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
+obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
+obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp/tcp.c
similarity index 100%
rename from net/ipv4/tcp.c
rename to net/ipv4/tcp/tcp.c
diff --git a/net/ipv4/tcp_bbr.c b/net/ipv4/tcp/tcp_bbr.c
similarity index 100%
rename from net/ipv4/tcp_bbr.c
rename to net/ipv4/tcp/tcp_bbr.c
diff --git a/net/ipv4/tcp_bic.c b/net/ipv4/tcp/tcp_bic.c
similarity index 100%
rename from net/ipv4/tcp_bic.c
rename to net/ipv4/tcp/tcp_bic.c
diff --git a/net/ipv4/tcp_cdg.c b/net/ipv4/tcp/tcp_cdg.c
similarity index 100%
rename from net/ipv4/tcp_cdg.c
rename to net/ipv4/tcp/tcp_cdg.c
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp/tcp_cong.c
similarity index 100%
rename from net/ipv4/tcp_cong.c
rename to net/ipv4/tcp/tcp_cong.c
diff --git a/net/ipv4/tcp_cubic.c b/net/ipv4/tcp/tcp_cubic.c
similarity index 100%
rename from net/ipv4/tcp_cubic.c
rename to net/ipv4/tcp/tcp_cubic.c
diff --git a/net/ipv4/tcp_dctcp.c b/net/ipv4/tcp/tcp_dctcp.c
similarity index 100%
rename from net/ipv4/tcp_dctcp.c
rename to net/ipv4/tcp/tcp_dctcp.c
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp/tcp_diag.c
similarity index 100%
rename from net/ipv4/tcp_diag.c
rename to net/ipv4/tcp/tcp_diag.c
diff --git a/net/ipv4/tcp_fastopen.c b/net/ipv4/tcp/tcp_fastopen.c
similarity index 100%
rename from net/ipv4/tcp_fastopen.c
rename to net/ipv4/tcp/tcp_fastopen.c
diff --git a/net/ipv4/tcp_highspeed.c b/net/ipv4/tcp/tcp_highspeed.c
similarity index 100%
rename from net/ipv4/tcp_highspeed.c
rename to net/ipv4/tcp/tcp_highspeed.c
diff --git a/net/ipv4/tcp_htcp.c b/net/ipv4/tcp/tcp_htcp.c
similarity index 100%
rename from net/ipv4/tcp_htcp.c
rename to net/ipv4/tcp/tcp_htcp.c
diff --git a/net/ipv4/tcp_hybla.c b/net/ipv4/tcp/tcp_hybla.c
similarity index 100%
rename from net/ipv4/tcp_hybla.c
rename to net/ipv4/tcp/tcp_hybla.c
diff --git a/net/ipv4/tcp_illinois.c b/net/ipv4/tcp/tcp_illinois.c
similarity index 100%
rename from net/ipv4/tcp_illinois.c
rename to net/ipv4/tcp/tcp_illinois.c
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp/tcp_input.c
similarity index 100%
rename from net/ipv4/tcp_input.c
rename to net/ipv4/tcp/tcp_input.c
diff --git a/net/ipv4/tcp_lp.c b/net/ipv4/tcp/tcp_lp.c
similarity index 100%
rename from net/ipv4/tcp_lp.c
rename to net/ipv4/tcp/tcp_lp.c
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp/tcp_metrics.c
similarity index 100%
rename from net/ipv4/tcp_metrics.c
rename to net/ipv4/tcp/tcp_metrics.c
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp/tcp_minisocks.c
similarity index 100%
rename from net/ipv4/tcp_minisocks.c
rename to net/ipv4/tcp/tcp_minisocks.c
diff --git a/net/ipv4/tcp_nv.c b/net/ipv4/tcp/tcp_nv.c
similarity index 100%
rename from net/ipv4/tcp_nv.c
rename to net/ipv4/tcp/tcp_nv.c
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp/tcp_output.c
similarity index 100%
rename from net/ipv4/tcp_output.c
rename to net/ipv4/tcp/tcp_output.c
diff --git a/net/ipv4/tcp_probe.c b/net/ipv4/tcp/tcp_probe.c
similarity index 100%
rename from net/ipv4/tcp_probe.c
rename to net/ipv4/tcp/tcp_probe.c
diff --git a/net/ipv4/tcp_rate.c b/net/ipv4/tcp/tcp_rate.c
similarity index 100%
rename from net/ipv4/tcp_rate.c
rename to net/ipv4/tcp/tcp_rate.c
diff --git a/net/ipv4/tcp_recovery.c b/net/ipv4/tcp/tcp_recovery.c
similarity index 100%
rename from net/ipv4/tcp_recovery.c
rename to net/ipv4/tcp/tcp_recovery.c
diff --git a/net/ipv4/tcp_scalable.c b/net/ipv4/tcp/tcp_scalable.c
similarity index 100%
rename from net/ipv4/tcp_scalable.c
rename to net/ipv4/tcp/tcp_scalable.c
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp/tcp_timer.c
similarity index 100%
rename from net/ipv4/tcp_timer.c
rename to net/ipv4/tcp/tcp_timer.c
diff --git a/net/ipv4/tcp_ulp.c b/net/ipv4/tcp/tcp_ulp.c
similarity index 100%
rename from net/ipv4/tcp_ulp.c
rename to net/ipv4/tcp/tcp_ulp.c
diff --git a/net/ipv4/tcp_vegas.c b/net/ipv4/tcp/tcp_vegas.c
similarity index 100%
rename from net/ipv4/tcp_vegas.c
rename to net/ipv4/tcp/tcp_vegas.c
diff --git a/net/ipv4/tcp_vegas.h b/net/ipv4/tcp/tcp_vegas.h
similarity index 100%
rename from net/ipv4/tcp_vegas.h
rename to net/ipv4/tcp/tcp_vegas.h
diff --git a/net/ipv4/tcp_veno.c b/net/ipv4/tcp/tcp_veno.c
similarity index 100%
rename from net/ipv4/tcp_veno.c
rename to net/ipv4/tcp/tcp_veno.c
diff --git a/net/ipv4/tcp_westwood.c b/net/ipv4/tcp/tcp_westwood.c
similarity index 100%
rename from net/ipv4/tcp_westwood.c
rename to net/ipv4/tcp/tcp_westwood.c
diff --git a/net/ipv4/tcp_yeah.c b/net/ipv4/tcp/tcp_yeah.c
similarity index 100%
rename from net/ipv4/tcp_yeah.c
rename to net/ipv4/tcp/tcp_yeah.c
--
2.14.1
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 2/3] tcp: Move cc algorithms to own subdirectory
2017-10-03 22:22 [PATCH net-next 0/3] A own subdirectory for shared TCP code Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory Richard Sailer
@ 2017-10-03 22:22 ` Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared Richard Sailer
` (2 subsequent siblings)
4 siblings, 0 replies; 10+ messages in thread
From: Richard Sailer @ 2017-10-03 22:22 UTC (permalink / raw)
To: netdev, davem; +Cc: kuznet, yoshfuji
Essentially the TCP code files consist of two groups:
* main code and some utilities (13 files)
* pluggable Congestion Control Algorithms (17 files)
Similar to the previous commit, this moves the Congestion control Algorithms to
a own subdirecty and updates the Makefiles accordingly to make the source tree
more well structured and self explaining.
Signed-off-by: Richard Sailer <richard_siegfried@systemli.org>
---
net/ipv4/tcp/Makefile | 18 ++----------------
net/ipv4/tcp/cc_algos/Makefile | 21 +++++++++++++++++++++
net/ipv4/tcp/{ => cc_algos}/tcp_bbr.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_bic.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_cdg.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_cubic.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_dctcp.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_highspeed.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_htcp.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_hybla.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_illinois.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_lp.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_nv.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_scalable.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_vegas.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_vegas.h | 0
net/ipv4/tcp/{ => cc_algos}/tcp_veno.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_westwood.c | 0
net/ipv4/tcp/{ => cc_algos}/tcp_yeah.c | 0
19 files changed, 23 insertions(+), 16 deletions(-)
create mode 100644 net/ipv4/tcp/cc_algos/Makefile
rename net/ipv4/tcp/{ => cc_algos}/tcp_bbr.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_bic.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_cdg.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_cubic.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_dctcp.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_highspeed.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_htcp.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_hybla.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_illinois.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_lp.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_nv.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_scalable.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_vegas.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_vegas.h (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_veno.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_westwood.c (100%)
rename net/ipv4/tcp/{ => cc_algos}/tcp_yeah.c (100%)
diff --git a/net/ipv4/tcp/Makefile b/net/ipv4/tcp/Makefile
index 91d2c991a243..60e96525a489 100644
--- a/net/ipv4/tcp/Makefile
+++ b/net/ipv4/tcp/Makefile
@@ -8,19 +8,5 @@ obj-y := tcp.o tcp_input.o tcp_output.o tcp_timer.o \
obj-$(CONFIG_INET_TCP_DIAG) += tcp_diag.o
obj-$(CONFIG_NET_TCPPROBE) += tcp_probe.o
-obj-$(CONFIG_TCP_CONG_BBR) += tcp_bbr.o
-obj-$(CONFIG_TCP_CONG_BIC) += tcp_bic.o
-obj-$(CONFIG_TCP_CONG_CDG) += tcp_cdg.o
-obj-$(CONFIG_TCP_CONG_CUBIC) += tcp_cubic.o
-obj-$(CONFIG_TCP_CONG_DCTCP) += tcp_dctcp.o
-obj-$(CONFIG_TCP_CONG_WESTWOOD) += tcp_westwood.o
-obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o
-obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o
-obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o
-obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o
-obj-$(CONFIG_TCP_CONG_NV) += tcp_nv.o
-obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o
-obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
-obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
-obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
-obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
+
+obj-$(CONFIG_TCP_CONG_ADVANCED) += cc_algos/
diff --git a/net/ipv4/tcp/cc_algos/Makefile b/net/ipv4/tcp/cc_algos/Makefile
new file mode 100644
index 000000000000..81999dd2a045
--- /dev/null
+++ b/net/ipv4/tcp/cc_algos/Makefile
@@ -0,0 +1,21 @@
+#
+# Makefile for all pluggable TCP congestion control algorithms
+#
+
+
+obj-$(CONFIG_TCP_CONG_BBR) += tcp_bbr.o
+obj-$(CONFIG_TCP_CONG_BIC) += tcp_bic.o
+obj-$(CONFIG_TCP_CONG_CDG) += tcp_cdg.o
+obj-$(CONFIG_TCP_CONG_CUBIC) += tcp_cubic.o
+obj-$(CONFIG_TCP_CONG_DCTCP) += tcp_dctcp.o
+obj-$(CONFIG_TCP_CONG_WESTWOOD) += tcp_westwood.o
+obj-$(CONFIG_TCP_CONG_HSTCP) += tcp_highspeed.o
+obj-$(CONFIG_TCP_CONG_HYBLA) += tcp_hybla.o
+obj-$(CONFIG_TCP_CONG_HTCP) += tcp_htcp.o
+obj-$(CONFIG_TCP_CONG_VEGAS) += tcp_vegas.o
+obj-$(CONFIG_TCP_CONG_NV) += tcp_nv.o
+obj-$(CONFIG_TCP_CONG_VENO) += tcp_veno.o
+obj-$(CONFIG_TCP_CONG_SCALABLE) += tcp_scalable.o
+obj-$(CONFIG_TCP_CONG_LP) += tcp_lp.o
+obj-$(CONFIG_TCP_CONG_YEAH) += tcp_yeah.o
+obj-$(CONFIG_TCP_CONG_ILLINOIS) += tcp_illinois.o
diff --git a/net/ipv4/tcp/tcp_bbr.c b/net/ipv4/tcp/cc_algos/tcp_bbr.c
similarity index 100%
rename from net/ipv4/tcp/tcp_bbr.c
rename to net/ipv4/tcp/cc_algos/tcp_bbr.c
diff --git a/net/ipv4/tcp/tcp_bic.c b/net/ipv4/tcp/cc_algos/tcp_bic.c
similarity index 100%
rename from net/ipv4/tcp/tcp_bic.c
rename to net/ipv4/tcp/cc_algos/tcp_bic.c
diff --git a/net/ipv4/tcp/tcp_cdg.c b/net/ipv4/tcp/cc_algos/tcp_cdg.c
similarity index 100%
rename from net/ipv4/tcp/tcp_cdg.c
rename to net/ipv4/tcp/cc_algos/tcp_cdg.c
diff --git a/net/ipv4/tcp/tcp_cubic.c b/net/ipv4/tcp/cc_algos/tcp_cubic.c
similarity index 100%
rename from net/ipv4/tcp/tcp_cubic.c
rename to net/ipv4/tcp/cc_algos/tcp_cubic.c
diff --git a/net/ipv4/tcp/tcp_dctcp.c b/net/ipv4/tcp/cc_algos/tcp_dctcp.c
similarity index 100%
rename from net/ipv4/tcp/tcp_dctcp.c
rename to net/ipv4/tcp/cc_algos/tcp_dctcp.c
diff --git a/net/ipv4/tcp/tcp_highspeed.c b/net/ipv4/tcp/cc_algos/tcp_highspeed.c
similarity index 100%
rename from net/ipv4/tcp/tcp_highspeed.c
rename to net/ipv4/tcp/cc_algos/tcp_highspeed.c
diff --git a/net/ipv4/tcp/tcp_htcp.c b/net/ipv4/tcp/cc_algos/tcp_htcp.c
similarity index 100%
rename from net/ipv4/tcp/tcp_htcp.c
rename to net/ipv4/tcp/cc_algos/tcp_htcp.c
diff --git a/net/ipv4/tcp/tcp_hybla.c b/net/ipv4/tcp/cc_algos/tcp_hybla.c
similarity index 100%
rename from net/ipv4/tcp/tcp_hybla.c
rename to net/ipv4/tcp/cc_algos/tcp_hybla.c
diff --git a/net/ipv4/tcp/tcp_illinois.c b/net/ipv4/tcp/cc_algos/tcp_illinois.c
similarity index 100%
rename from net/ipv4/tcp/tcp_illinois.c
rename to net/ipv4/tcp/cc_algos/tcp_illinois.c
diff --git a/net/ipv4/tcp/tcp_lp.c b/net/ipv4/tcp/cc_algos/tcp_lp.c
similarity index 100%
rename from net/ipv4/tcp/tcp_lp.c
rename to net/ipv4/tcp/cc_algos/tcp_lp.c
diff --git a/net/ipv4/tcp/tcp_nv.c b/net/ipv4/tcp/cc_algos/tcp_nv.c
similarity index 100%
rename from net/ipv4/tcp/tcp_nv.c
rename to net/ipv4/tcp/cc_algos/tcp_nv.c
diff --git a/net/ipv4/tcp/tcp_scalable.c b/net/ipv4/tcp/cc_algos/tcp_scalable.c
similarity index 100%
rename from net/ipv4/tcp/tcp_scalable.c
rename to net/ipv4/tcp/cc_algos/tcp_scalable.c
diff --git a/net/ipv4/tcp/tcp_vegas.c b/net/ipv4/tcp/cc_algos/tcp_vegas.c
similarity index 100%
rename from net/ipv4/tcp/tcp_vegas.c
rename to net/ipv4/tcp/cc_algos/tcp_vegas.c
diff --git a/net/ipv4/tcp/tcp_vegas.h b/net/ipv4/tcp/cc_algos/tcp_vegas.h
similarity index 100%
rename from net/ipv4/tcp/tcp_vegas.h
rename to net/ipv4/tcp/cc_algos/tcp_vegas.h
diff --git a/net/ipv4/tcp/tcp_veno.c b/net/ipv4/tcp/cc_algos/tcp_veno.c
similarity index 100%
rename from net/ipv4/tcp/tcp_veno.c
rename to net/ipv4/tcp/cc_algos/tcp_veno.c
diff --git a/net/ipv4/tcp/tcp_westwood.c b/net/ipv4/tcp/cc_algos/tcp_westwood.c
similarity index 100%
rename from net/ipv4/tcp/tcp_westwood.c
rename to net/ipv4/tcp/cc_algos/tcp_westwood.c
diff --git a/net/ipv4/tcp/tcp_yeah.c b/net/ipv4/tcp/cc_algos/tcp_yeah.c
similarity index 100%
rename from net/ipv4/tcp/tcp_yeah.c
rename to net/ipv4/tcp/cc_algos/tcp_yeah.c
--
2.14.1
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared
2017-10-03 22:22 [PATCH net-next 0/3] A own subdirectory for shared TCP code Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 2/3] tcp: Move cc algorithms " Richard Sailer
@ 2017-10-03 22:22 ` Richard Sailer
[not found] ` <CAOaVG14r2Uwm+r17Oa9jwKj50vRr8=SqwXEmmGAeW8rHj3FgXg@mail.gmail.com>
2017-10-03 22:42 ` [PATCH net-next 0/3] A own subdirectory for shared TCP code Eric Dumazet
2017-10-03 23:03 ` David Miller
4 siblings, 1 reply; 10+ messages in thread
From: Richard Sailer @ 2017-10-03 22:22 UTC (permalink / raw)
To: netdev, davem; +Cc: kuznet, yoshfuji
Currently a lot of Code shared between IPv4 and IPv6 resides in net/ipv4.
As an attempt to make things more modular and encapsulated + the source tree
more self documenting this commit:
* introduces net/ipv4v6shared
* moves the shared tcp code there
* updates the makefiles accordingly
* creates a new Kconfig file for the TCP code
* updates all other Kconfig files accordingly
Of course I left the ipv4 specific parts of tcp (tcp_ipv4.c and tcp_offload.c)
in net/ipv4.
Thanks valdis from #kernelnewbies for the hint of doing this, when already
isolating tcp code.
Signed-off-by: Richard Sailer <richard_siegfried@systemli.org>
---
net/Kconfig | 1 +
net/Makefile | 6 +-
net/ipv4/Kconfig | 321 --------------------
net/ipv4/Makefile | 4 +-
net/ipv4v6_shared/Makefile | 7 +
net/ipv4v6_shared/tcp/Kconfig | 324 +++++++++++++++++++++
net/{ipv4 => ipv4v6_shared}/tcp/Makefile | 0
net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/Makefile | 0
net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_bbr.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_bic.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_cdg.c | 0
.../tcp/cc_algos/tcp_cubic.c | 0
.../tcp/cc_algos/tcp_dctcp.c | 0
.../tcp/cc_algos/tcp_highspeed.c | 0
.../tcp/cc_algos/tcp_htcp.c | 0
.../tcp/cc_algos/tcp_hybla.c | 0
.../tcp/cc_algos/tcp_illinois.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_lp.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_nv.c | 0
.../tcp/cc_algos/tcp_scalable.c | 0
.../tcp/cc_algos/tcp_vegas.c | 0
.../tcp/cc_algos/tcp_vegas.h | 0
.../tcp/cc_algos/tcp_veno.c | 0
.../tcp/cc_algos/tcp_westwood.c | 0
.../tcp/cc_algos/tcp_yeah.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_cong.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_diag.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_fastopen.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_input.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_metrics.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_minisocks.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_output.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_probe.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_rate.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_recovery.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_timer.c | 0
net/{ipv4 => ipv4v6_shared}/tcp/tcp_ulp.c | 0
38 files changed, 339 insertions(+), 324 deletions(-)
create mode 100644 net/ipv4v6_shared/Makefile
create mode 100644 net/ipv4v6_shared/tcp/Kconfig
rename net/{ipv4 => ipv4v6_shared}/tcp/Makefile (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/Makefile (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_bbr.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_bic.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_cdg.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_cubic.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_dctcp.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_highspeed.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_htcp.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_hybla.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_illinois.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_lp.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_nv.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_scalable.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_vegas.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_vegas.h (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_veno.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_westwood.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/cc_algos/tcp_yeah.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_cong.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_diag.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_fastopen.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_input.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_metrics.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_minisocks.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_output.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_probe.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_rate.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_recovery.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_timer.c (100%)
rename net/{ipv4 => ipv4v6_shared}/tcp/tcp_ulp.c (100%)
diff --git a/net/Kconfig b/net/Kconfig
index 9dba2715919d..caa1c51c577d 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -86,6 +86,7 @@ config INET
if INET
source "net/ipv4/Kconfig"
+source "net/ipv4v6_shared/tcp/Kconfig"
source "net/ipv6/Kconfig"
source "net/netlabel/Kconfig"
diff --git a/net/Makefile b/net/Makefile
index ae2fe2283d2f..b089a88d3d5d 100644
--- a/net/Makefile
+++ b/net/Makefile
@@ -14,7 +14,11 @@ obj-$(CONFIG_NET) += $(tmp-y)
obj-$(CONFIG_LLC) += llc/
obj-$(CONFIG_NET) += ethernet/ 802/ sched/ netlink/ bpf/
obj-$(CONFIG_NETFILTER) += netfilter/
-obj-$(CONFIG_INET) += ipv4/
+
+# IPv6 depends on CONFIG_INET anyways, so this also includes the shared code
+# for IPv6 when needed.
+obj-$(CONFIG_INET) += ipv4/ ipv4v6_shared/
+
obj-$(CONFIG_TLS) += tls/
obj-$(CONFIG_XFRM) += xfrm/
obj-$(CONFIG_UNIX) += unix/
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 91a2557942fa..aa952b0c6a8c 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -261,42 +261,6 @@ config IP_PIMSM_V2
gated-5). This routing protocol is not used widely, so say N unless
you want to play with it.
-config SYN_COOKIES
- bool "IP: TCP syncookie support"
- ---help---
- Normal TCP/IP networking is open to an attack known as "SYN
- flooding". This denial-of-service attack prevents legitimate remote
- users from being able to connect to your computer during an ongoing
- attack and requires very little work from the attacker, who can
- operate from anywhere on the Internet.
-
- SYN cookies provide protection against this type of attack. If you
- say Y here, the TCP/IP stack will use a cryptographic challenge
- protocol known as "SYN cookies" to enable legitimate users to
- continue to connect, even when your machine is under attack. There
- is no need for the legitimate users to change their TCP/IP software;
- SYN cookies work transparently to them. For technical information
- about SYN cookies, check out <http://cr.yp.to/syncookies.html>.
-
- If you are SYN flooded, the source address reported by the kernel is
- likely to have been forged by the attacker; it is only reported as
- an aid in tracing the packets to their actual source and should not
- be taken as absolute truth.
-
- SYN cookies may prevent correct error reporting on clients when the
- server is really overloaded. If this happens frequently better turn
- them off.
-
- If you say Y here, you can disable SYN cookies at run time by
- saying Y to "/proc file system support" and
- "Sysctl support" below and executing the command
-
- echo 0 > /proc/sys/net/ipv4/tcp_syncookies
-
- after the /proc file system has been mounted.
-
- If unsure, say N.
-
config NET_IPVTI
tristate "Virtual (secure) IP: tunneling"
select INET_TUNNEL
@@ -465,288 +429,3 @@ config INET_DIAG_DESTROY
had been disconnected.
If unsure, say N.
-menuconfig TCP_CONG_ADVANCED
- bool "TCP: advanced congestion control"
- ---help---
- Support for selection of various TCP congestion control
- modules.
-
- Nearly all users can safely say no here, and a safe default
- selection will be made (CUBIC with new Reno as a fallback).
-
- If unsure, say N.
-
-if TCP_CONG_ADVANCED
-
-config TCP_CONG_BIC
- tristate "Binary Increase Congestion (BIC) control"
- default m
- ---help---
- BIC-TCP is a sender-side only change that ensures a linear RTT
- fairness under large windows while offering both scalability and
- bounded TCP-friendliness. The protocol combines two schemes
- called additive increase and binary search increase. When the
- congestion window is large, additive increase with a large
- increment ensures linear RTT fairness as well as good
- scalability. Under small congestion windows, binary search
- increase provides TCP friendliness.
- See http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/
-
-config TCP_CONG_CUBIC
- tristate "CUBIC TCP"
- default y
- ---help---
- This is version 2.0 of BIC-TCP which uses a cubic growth function
- among other techniques.
- See http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/cubic-paper.pdf
-
-config TCP_CONG_WESTWOOD
- tristate "TCP Westwood+"
- default m
- ---help---
- TCP Westwood+ is a sender-side only modification of the TCP Reno
- protocol stack that optimizes the performance of TCP congestion
- control. It is based on end-to-end bandwidth estimation to set
- congestion window and slow start threshold after a congestion
- episode. Using this estimation, TCP Westwood+ adaptively sets a
- slow start threshold and a congestion window which takes into
- account the bandwidth used at the time congestion is experienced.
- TCP Westwood+ significantly increases fairness wrt TCP Reno in
- wired networks and throughput over wireless links.
-
-config TCP_CONG_HTCP
- tristate "H-TCP"
- default m
- ---help---
- H-TCP is a send-side only modifications of the TCP Reno
- protocol stack that optimizes the performance of TCP
- congestion control for high speed network links. It uses a
- modeswitch to change the alpha and beta parameters of TCP Reno
- based on network conditions and in a way so as to be fair with
- other Reno and H-TCP flows.
-
-config TCP_CONG_HSTCP
- tristate "High Speed TCP"
- default n
- ---help---
- Sally Floyd's High Speed TCP (RFC 3649) congestion control.
- A modification to TCP's congestion control mechanism for use
- with large congestion windows. A table indicates how much to
- increase the congestion window by when an ACK is received.
- For more detail see http://www.icir.org/floyd/hstcp.html
-
-config TCP_CONG_HYBLA
- tristate "TCP-Hybla congestion control algorithm"
- default n
- ---help---
- TCP-Hybla is a sender-side only change that eliminates penalization of
- long-RTT, large-bandwidth connections, like when satellite legs are
- involved, especially when sharing a common bottleneck with normal
- terrestrial connections.
-
-config TCP_CONG_VEGAS
- tristate "TCP Vegas"
- default n
- ---help---
- TCP Vegas is a sender-side only change to TCP that anticipates
- the onset of congestion by estimating the bandwidth. TCP Vegas
- adjusts the sending rate by modifying the congestion
- window. TCP Vegas should provide less packet loss, but it is
- not as aggressive as TCP Reno.
-
-config TCP_CONG_NV
- tristate "TCP NV"
- default n
- ---help---
- TCP NV is a follow up to TCP Vegas. It has been modified to deal with
- 10G networks, measurement noise introduced by LRO, GRO and interrupt
- coalescence. In addition, it will decrease its cwnd multiplicatively
- instead of linearly.
-
- Note that in general congestion avoidance (cwnd decreased when # packets
- queued grows) cannot coexist with congestion control (cwnd decreased only
- when there is packet loss) due to fairness issues. One scenario when they
- can coexist safely is when the CA flows have RTTs << CC flows RTTs.
-
- For further details see http://www.brakmo.org/networking/tcp-nv/
-
-config TCP_CONG_SCALABLE
- tristate "Scalable TCP"
- default n
- ---help---
- Scalable TCP is a sender-side only change to TCP which uses a
- MIMD congestion control algorithm which has some nice scaling
- properties, though is known to have fairness issues.
- See http://www.deneholme.net/tom/scalable/
-
-config TCP_CONG_LP
- tristate "TCP Low Priority"
- default n
- ---help---
- TCP Low Priority (TCP-LP), a distributed algorithm whose goal is
- to utilize only the excess network bandwidth as compared to the
- ``fair share`` of bandwidth as targeted by TCP.
- See http://www-ece.rice.edu/networks/TCP-LP/
-
-config TCP_CONG_VENO
- tristate "TCP Veno"
- default n
- ---help---
- TCP Veno is a sender-side only enhancement of TCP to obtain better
- throughput over wireless networks. TCP Veno makes use of state
- distinguishing to circumvent the difficult judgment of the packet loss
- type. TCP Veno cuts down less congestion window in response to random
- loss packets.
- See <http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1177186>
-
-config TCP_CONG_YEAH
- tristate "YeAH TCP"
- select TCP_CONG_VEGAS
- default n
- ---help---
- YeAH-TCP is a sender-side high-speed enabled TCP congestion control
- algorithm, which uses a mixed loss/delay approach to compute the
- congestion window. It's design goals target high efficiency,
- internal, RTT and Reno fairness, resilience to link loss while
- keeping network elements load as low as possible.
-
- For further details look here:
- http://wil.cs.caltech.edu/pfldnet2007/paper/YeAH_TCP.pdf
-
-config TCP_CONG_ILLINOIS
- tristate "TCP Illinois"
- default n
- ---help---
- TCP-Illinois is a sender-side modification of TCP Reno for
- high speed long delay links. It uses round-trip-time to
- adjust the alpha and beta parameters to achieve a higher average
- throughput and maintain fairness.
-
- For further details see:
- http://www.ews.uiuc.edu/~shaoliu/tcpillinois/index.html
-
-config TCP_CONG_DCTCP
- tristate "DataCenter TCP (DCTCP)"
- default n
- ---help---
- DCTCP leverages Explicit Congestion Notification (ECN) in the network to
- provide multi-bit feedback to the end hosts. It is designed to provide:
-
- - High burst tolerance (incast due to partition/aggregate),
- - Low latency (short flows, queries),
- - High throughput (continuous data updates, large file transfers) with
- commodity, shallow-buffered switches.
-
- All switches in the data center network running DCTCP must support
- ECN marking and be configured for marking when reaching defined switch
- buffer thresholds. The default ECN marking threshold heuristic for
- DCTCP on switches is 20 packets (30KB) at 1Gbps, and 65 packets
- (~100KB) at 10Gbps, but might need further careful tweaking.
-
- For further details see:
- http://simula.stanford.edu/~alizade/Site/DCTCP_files/dctcp-final.pdf
-
-config TCP_CONG_CDG
- tristate "CAIA Delay-Gradient (CDG)"
- default n
- ---help---
- CAIA Delay-Gradient (CDG) is a TCP congestion control that modifies
- the TCP sender in order to:
-
- o Use the delay gradient as a congestion signal.
- o Back off with an average probability that is independent of the RTT.
- o Coexist with flows that use loss-based congestion control.
- o Tolerate packet loss unrelated to congestion.
-
- For further details see:
- D.A. Hayes and G. Armitage. "Revisiting TCP congestion control using
- delay gradients." In Networking 2011. Preprint: http://goo.gl/No3vdg
-
-config TCP_CONG_BBR
- tristate "BBR TCP"
- default n
- ---help---
-
- BBR (Bottleneck Bandwidth and RTT) TCP congestion control aims to
- maximize network utilization and minimize queues. It builds an explicit
- model of the the bottleneck delivery rate and path round-trip
- propagation delay. It tolerates packet loss and delay unrelated to
- congestion. It can operate over LAN, WAN, cellular, wifi, or cable
- modem links. It can coexist with flows that use loss-based congestion
- control, and can operate with shallow buffers, deep buffers,
- bufferbloat, policers, or AQM schemes that do not provide a delay
- signal. It requires the fq ("Fair Queue") pacing packet scheduler.
-
-choice
- prompt "Default TCP congestion control"
- default DEFAULT_CUBIC
- help
- Select the TCP congestion control that will be used by default
- for all connections.
-
- config DEFAULT_BIC
- bool "Bic" if TCP_CONG_BIC=y
-
- config DEFAULT_CUBIC
- bool "Cubic" if TCP_CONG_CUBIC=y
-
- config DEFAULT_HTCP
- bool "Htcp" if TCP_CONG_HTCP=y
-
- config DEFAULT_HYBLA
- bool "Hybla" if TCP_CONG_HYBLA=y
-
- config DEFAULT_VEGAS
- bool "Vegas" if TCP_CONG_VEGAS=y
-
- config DEFAULT_VENO
- bool "Veno" if TCP_CONG_VENO=y
-
- config DEFAULT_WESTWOOD
- bool "Westwood" if TCP_CONG_WESTWOOD=y
-
- config DEFAULT_DCTCP
- bool "DCTCP" if TCP_CONG_DCTCP=y
-
- config DEFAULT_CDG
- bool "CDG" if TCP_CONG_CDG=y
-
- config DEFAULT_BBR
- bool "BBR" if TCP_CONG_BBR=y
-
- config DEFAULT_RENO
- bool "Reno"
-endchoice
-
-endif
-
-config TCP_CONG_CUBIC
- tristate
- depends on !TCP_CONG_ADVANCED
- default y
-
-config DEFAULT_TCP_CONG
- string
- default "bic" if DEFAULT_BIC
- default "cubic" if DEFAULT_CUBIC
- default "htcp" if DEFAULT_HTCP
- default "hybla" if DEFAULT_HYBLA
- default "vegas" if DEFAULT_VEGAS
- default "westwood" if DEFAULT_WESTWOOD
- default "veno" if DEFAULT_VENO
- default "reno" if DEFAULT_RENO
- default "dctcp" if DEFAULT_DCTCP
- default "cdg" if DEFAULT_CDG
- default "bbr" if DEFAULT_BBR
- default "cubic"
-
-config TCP_MD5SIG
- bool "TCP: MD5 Signature Option support (RFC2385)"
- select CRYPTO
- select CRYPTO_MD5
- ---help---
- RFC2385 specifies a method of giving MD5 protection to TCP sessions.
- Its main (only?) use is to protect BGP sessions between core routers
- on the Internet.
-
- If unsure, say N.
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index a106a1cf0e11..d170d6e1417c 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -1,8 +1,8 @@
#
-# Makefile for the Linux TCP/IP (INET) layer.
+# Makefile for the Linux IP layer.
#
-obj-y := tcp/ route.o inetpeer.o protocol.o \
+obj-y := route.o inetpeer.o protocol.o \
ip_input.o ip_fragment.o ip_forward.o ip_options.o \
ip_output.o ip_sockglue.o inet_hashtables.o \
inet_timewait_sock.o inet_connection_sock.o \
diff --git a/net/ipv4v6_shared/Makefile b/net/ipv4v6_shared/Makefile
new file mode 100644
index 000000000000..41a67a0cf210
--- /dev/null
+++ b/net/ipv4v6_shared/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for shared networking code called/used by IPv4 and IPv6
+#
+# Note: Currently this only contains the shared part of the TCP Implementation
+# a lot of the shared code still resides in net/ipv4/
+
+obj-y := tcp/
diff --git a/net/ipv4v6_shared/tcp/Kconfig b/net/ipv4v6_shared/tcp/Kconfig
new file mode 100644
index 000000000000..dff4d046a42b
--- /dev/null
+++ b/net/ipv4v6_shared/tcp/Kconfig
@@ -0,0 +1,324 @@
+#
+# TCP configuration
+#
+config SYN_COOKIES
+ bool "IP: TCP syncookie support"
+ ---help---
+ Normal TCP/IP networking is open to an attack known as "SYN
+ flooding". This denial-of-service attack prevents legitimate remote
+ users from being able to connect to your computer during an ongoing
+ attack and requires very little work from the attacker, who can
+ operate from anywhere on the Internet.
+
+ SYN cookies provide protection against this type of attack. If you
+ say Y here, the TCP/IP stack will use a cryptographic challenge
+ protocol known as "SYN cookies" to enable legitimate users to
+ continue to connect, even when your machine is under attack. There
+ is no need for the legitimate users to change their TCP/IP software;
+ SYN cookies work transparently to them. For technical information
+ about SYN cookies, check out <http://cr.yp.to/syncookies.html>.
+
+ If you are SYN flooded, the source address reported by the kernel is
+ likely to have been forged by the attacker; it is only reported as
+ an aid in tracing the packets to their actual source and should not
+ be taken as absolute truth.
+
+ SYN cookies may prevent correct error reporting on clients when the
+ server is really overloaded. If this happens frequently better turn
+ them off.
+
+ If you say Y here, you can disable SYN cookies at run time by
+ saying Y to "/proc file system support" and
+ "Sysctl support" below and executing the command
+
+ echo 0 > /proc/sys/net/ipv4/tcp_syncookies
+
+ after the /proc file system has been mounted.
+
+ If unsure, say N.
+
+menuconfig TCP_CONG_ADVANCED
+ bool "TCP: advanced congestion control"
+ ---help---
+ Support for selection of various TCP congestion control
+ modules.
+
+ Nearly all users can safely say no here, and a safe default
+ selection will be made (CUBIC with new Reno as a fallback).
+
+ If unsure, say N.
+
+if TCP_CONG_ADVANCED
+
+config TCP_CONG_BIC
+ tristate "Binary Increase Congestion (BIC) control"
+ default m
+ ---help---
+ BIC-TCP is a sender-side only change that ensures a linear RTT
+ fairness under large windows while offering both scalability and
+ bounded TCP-friendliness. The protocol combines two schemes
+ called additive increase and binary search increase. When the
+ congestion window is large, additive increase with a large
+ increment ensures linear RTT fairness as well as good
+ scalability. Under small congestion windows, binary search
+ increase provides TCP friendliness.
+ See http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/
+
+config TCP_CONG_CUBIC
+ tristate "CUBIC TCP"
+ default y
+ ---help---
+ This is version 2.0 of BIC-TCP which uses a cubic growth function
+ among other techniques.
+ See http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/cubic-paper.pdf
+
+config TCP_CONG_WESTWOOD
+ tristate "TCP Westwood+"
+ default m
+ ---help---
+ TCP Westwood+ is a sender-side only modification of the TCP Reno
+ protocol stack that optimizes the performance of TCP congestion
+ control. It is based on end-to-end bandwidth estimation to set
+ congestion window and slow start threshold after a congestion
+ episode. Using this estimation, TCP Westwood+ adaptively sets a
+ slow start threshold and a congestion window which takes into
+ account the bandwidth used at the time congestion is experienced.
+ TCP Westwood+ significantly increases fairness wrt TCP Reno in
+ wired networks and throughput over wireless links.
+
+config TCP_CONG_HTCP
+ tristate "H-TCP"
+ default m
+ ---help---
+ H-TCP is a send-side only modifications of the TCP Reno
+ protocol stack that optimizes the performance of TCP
+ congestion control for high speed network links. It uses a
+ modeswitch to change the alpha and beta parameters of TCP Reno
+ based on network conditions and in a way so as to be fair with
+ other Reno and H-TCP flows.
+
+config TCP_CONG_HSTCP
+ tristate "High Speed TCP"
+ default n
+ ---help---
+ Sally Floyd's High Speed TCP (RFC 3649) congestion control.
+ A modification to TCP's congestion control mechanism for use
+ with large congestion windows. A table indicates how much to
+ increase the congestion window by when an ACK is received.
+ For more detail see http://www.icir.org/floyd/hstcp.html
+
+config TCP_CONG_HYBLA
+ tristate "TCP-Hybla congestion control algorithm"
+ default n
+ ---help---
+ TCP-Hybla is a sender-side only change that eliminates penalization of
+ long-RTT, large-bandwidth connections, like when satellite legs are
+ involved, especially when sharing a common bottleneck with normal
+ terrestrial connections.
+
+config TCP_CONG_VEGAS
+ tristate "TCP Vegas"
+ default n
+ ---help---
+ TCP Vegas is a sender-side only change to TCP that anticipates
+ the onset of congestion by estimating the bandwidth. TCP Vegas
+ adjusts the sending rate by modifying the congestion
+ window. TCP Vegas should provide less packet loss, but it is
+ not as aggressive as TCP Reno.
+
+config TCP_CONG_NV
+ tristate "TCP NV"
+ default n
+ ---help---
+ TCP NV is a follow up to TCP Vegas. It has been modified to deal with
+ 10G networks, measurement noise introduced by LRO, GRO and interrupt
+ coalescence. In addition, it will decrease its cwnd multiplicatively
+ instead of linearly.
+
+ Note that in general congestion avoidance (cwnd decreased when # packets
+ queued grows) cannot coexist with congestion control (cwnd decreased only
+ when there is packet loss) due to fairness issues. One scenario when they
+ can coexist safely is when the CA flows have RTTs << CC flows RTTs.
+
+ For further details see http://www.brakmo.org/networking/tcp-nv/
+
+config TCP_CONG_SCALABLE
+ tristate "Scalable TCP"
+ default n
+ ---help---
+ Scalable TCP is a sender-side only change to TCP which uses a
+ MIMD congestion control algorithm which has some nice scaling
+ properties, though is known to have fairness issues.
+ See http://www.deneholme.net/tom/scalable/
+
+config TCP_CONG_LP
+ tristate "TCP Low Priority"
+ default n
+ ---help---
+ TCP Low Priority (TCP-LP), a distributed algorithm whose goal is
+ to utilize only the excess network bandwidth as compared to the
+ ``fair share`` of bandwidth as targeted by TCP.
+ See http://www-ece.rice.edu/networks/TCP-LP/
+
+config TCP_CONG_VENO
+ tristate "TCP Veno"
+ default n
+ ---help---
+ TCP Veno is a sender-side only enhancement of TCP to obtain better
+ throughput over wireless networks. TCP Veno makes use of state
+ distinguishing to circumvent the difficult judgment of the packet loss
+ type. TCP Veno cuts down less congestion window in response to random
+ loss packets.
+ See <http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1177186>
+
+config TCP_CONG_YEAH
+ tristate "YeAH TCP"
+ select TCP_CONG_VEGAS
+ default n
+ ---help---
+ YeAH-TCP is a sender-side high-speed enabled TCP congestion control
+ algorithm, which uses a mixed loss/delay approach to compute the
+ congestion window. It's design goals target high efficiency,
+ internal, RTT and Reno fairness, resilience to link loss while
+ keeping network elements load as low as possible.
+
+ For further details look here:
+ http://wil.cs.caltech.edu/pfldnet2007/paper/YeAH_TCP.pdf
+
+config TCP_CONG_ILLINOIS
+ tristate "TCP Illinois"
+ default n
+ ---help---
+ TCP-Illinois is a sender-side modification of TCP Reno for
+ high speed long delay links. It uses round-trip-time to
+ adjust the alpha and beta parameters to achieve a higher average
+ throughput and maintain fairness.
+
+ For further details see:
+ http://www.ews.uiuc.edu/~shaoliu/tcpillinois/index.html
+
+config TCP_CONG_DCTCP
+ tristate "DataCenter TCP (DCTCP)"
+ default n
+ ---help---
+ DCTCP leverages Explicit Congestion Notification (ECN) in the network to
+ provide multi-bit feedback to the end hosts. It is designed to provide:
+
+ - High burst tolerance (incast due to partition/aggregate),
+ - Low latency (short flows, queries),
+ - High throughput (continuous data updates, large file transfers) with
+ commodity, shallow-buffered switches.
+
+ All switches in the data center network running DCTCP must support
+ ECN marking and be configured for marking when reaching defined switch
+ buffer thresholds. The default ECN marking threshold heuristic for
+ DCTCP on switches is 20 packets (30KB) at 1Gbps, and 65 packets
+ (~100KB) at 10Gbps, but might need further careful tweaking.
+
+ For further details see:
+ http://simula.stanford.edu/~alizade/Site/DCTCP_files/dctcp-final.pdf
+
+config TCP_CONG_CDG
+ tristate "CAIA Delay-Gradient (CDG)"
+ default n
+ ---help---
+ CAIA Delay-Gradient (CDG) is a TCP congestion control that modifies
+ the TCP sender in order to:
+
+ o Use the delay gradient as a congestion signal.
+ o Back off with an average probability that is independent of the RTT.
+ o Coexist with flows that use loss-based congestion control.
+ o Tolerate packet loss unrelated to congestion.
+
+ For further details see:
+ D.A. Hayes and G. Armitage. "Revisiting TCP congestion control using
+ delay gradients." In Networking 2011. Preprint: http://goo.gl/No3vdg
+
+config TCP_CONG_BBR
+ tristate "BBR TCP"
+ default n
+ ---help---
+
+ BBR (Bottleneck Bandwidth and RTT) TCP congestion control aims to
+ maximize network utilization and minimize queues. It builds an explicit
+ model of the the bottleneck delivery rate and path round-trip
+ propagation delay. It tolerates packet loss and delay unrelated to
+ congestion. It can operate over LAN, WAN, cellular, wifi, or cable
+ modem links. It can coexist with flows that use loss-based congestion
+ control, and can operate with shallow buffers, deep buffers,
+ bufferbloat, policers, or AQM schemes that do not provide a delay
+ signal. It requires the fq ("Fair Queue") pacing packet scheduler.
+
+choice
+ prompt "Default TCP congestion control"
+ default DEFAULT_CUBIC
+ help
+ Select the TCP congestion control that will be used by default
+ for all connections.
+
+ config DEFAULT_BIC
+ bool "Bic" if TCP_CONG_BIC=y
+
+ config DEFAULT_CUBIC
+ bool "Cubic" if TCP_CONG_CUBIC=y
+
+ config DEFAULT_HTCP
+ bool "Htcp" if TCP_CONG_HTCP=y
+
+ config DEFAULT_HYBLA
+ bool "Hybla" if TCP_CONG_HYBLA=y
+
+ config DEFAULT_VEGAS
+ bool "Vegas" if TCP_CONG_VEGAS=y
+
+ config DEFAULT_VENO
+ bool "Veno" if TCP_CONG_VENO=y
+
+ config DEFAULT_WESTWOOD
+ bool "Westwood" if TCP_CONG_WESTWOOD=y
+
+ config DEFAULT_DCTCP
+ bool "DCTCP" if TCP_CONG_DCTCP=y
+
+ config DEFAULT_CDG
+ bool "CDG" if TCP_CONG_CDG=y
+
+ config DEFAULT_BBR
+ bool "BBR" if TCP_CONG_BBR=y
+
+ config DEFAULT_RENO
+ bool "Reno"
+endchoice
+
+endif
+
+config TCP_CONG_CUBIC
+ tristate
+ depends on !TCP_CONG_ADVANCED
+ default y
+
+config DEFAULT_TCP_CONG
+ string
+ default "bic" if DEFAULT_BIC
+ default "cubic" if DEFAULT_CUBIC
+ default "htcp" if DEFAULT_HTCP
+ default "hybla" if DEFAULT_HYBLA
+ default "vegas" if DEFAULT_VEGAS
+ default "westwood" if DEFAULT_WESTWOOD
+ default "veno" if DEFAULT_VENO
+ default "reno" if DEFAULT_RENO
+ default "dctcp" if DEFAULT_DCTCP
+ default "cdg" if DEFAULT_CDG
+ default "bbr" if DEFAULT_BBR
+ default "cubic"
+
+config TCP_MD5SIG
+ bool "TCP: MD5 Signature Option support (RFC2385)"
+ select CRYPTO
+ select CRYPTO_MD5
+ ---help---
+ RFC2385 specifies a method of giving MD5 protection to TCP sessions.
+ Its main (only?) use is to protect BGP sessions between core routers
+ on the Internet.
+
+ If unsure, say N.
diff --git a/net/ipv4/tcp/Makefile b/net/ipv4v6_shared/tcp/Makefile
similarity index 100%
rename from net/ipv4/tcp/Makefile
rename to net/ipv4v6_shared/tcp/Makefile
diff --git a/net/ipv4/tcp/cc_algos/Makefile b/net/ipv4v6_shared/tcp/cc_algos/Makefile
similarity index 100%
rename from net/ipv4/tcp/cc_algos/Makefile
rename to net/ipv4v6_shared/tcp/cc_algos/Makefile
diff --git a/net/ipv4/tcp/cc_algos/tcp_bbr.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_bbr.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_bbr.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_bbr.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_bic.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_bic.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_bic.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_bic.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_cdg.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_cdg.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_cdg.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_cdg.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_cubic.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_cubic.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_cubic.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_cubic.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_dctcp.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_dctcp.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_dctcp.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_dctcp.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_highspeed.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_highspeed.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_highspeed.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_highspeed.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_htcp.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_htcp.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_htcp.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_htcp.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_hybla.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_hybla.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_hybla.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_hybla.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_illinois.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_illinois.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_illinois.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_illinois.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_lp.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_lp.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_lp.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_lp.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_nv.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_nv.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_nv.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_nv.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_scalable.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_scalable.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_scalable.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_scalable.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_vegas.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_vegas.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_vegas.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_vegas.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_vegas.h b/net/ipv4v6_shared/tcp/cc_algos/tcp_vegas.h
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_vegas.h
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_vegas.h
diff --git a/net/ipv4/tcp/cc_algos/tcp_veno.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_veno.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_veno.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_veno.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_westwood.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_westwood.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_westwood.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_westwood.c
diff --git a/net/ipv4/tcp/cc_algos/tcp_yeah.c b/net/ipv4v6_shared/tcp/cc_algos/tcp_yeah.c
similarity index 100%
rename from net/ipv4/tcp/cc_algos/tcp_yeah.c
rename to net/ipv4v6_shared/tcp/cc_algos/tcp_yeah.c
diff --git a/net/ipv4/tcp/tcp.c b/net/ipv4v6_shared/tcp/tcp.c
similarity index 100%
rename from net/ipv4/tcp/tcp.c
rename to net/ipv4v6_shared/tcp/tcp.c
diff --git a/net/ipv4/tcp/tcp_cong.c b/net/ipv4v6_shared/tcp/tcp_cong.c
similarity index 100%
rename from net/ipv4/tcp/tcp_cong.c
rename to net/ipv4v6_shared/tcp/tcp_cong.c
diff --git a/net/ipv4/tcp/tcp_diag.c b/net/ipv4v6_shared/tcp/tcp_diag.c
similarity index 100%
rename from net/ipv4/tcp/tcp_diag.c
rename to net/ipv4v6_shared/tcp/tcp_diag.c
diff --git a/net/ipv4/tcp/tcp_fastopen.c b/net/ipv4v6_shared/tcp/tcp_fastopen.c
similarity index 100%
rename from net/ipv4/tcp/tcp_fastopen.c
rename to net/ipv4v6_shared/tcp/tcp_fastopen.c
diff --git a/net/ipv4/tcp/tcp_input.c b/net/ipv4v6_shared/tcp/tcp_input.c
similarity index 100%
rename from net/ipv4/tcp/tcp_input.c
rename to net/ipv4v6_shared/tcp/tcp_input.c
diff --git a/net/ipv4/tcp/tcp_metrics.c b/net/ipv4v6_shared/tcp/tcp_metrics.c
similarity index 100%
rename from net/ipv4/tcp/tcp_metrics.c
rename to net/ipv4v6_shared/tcp/tcp_metrics.c
diff --git a/net/ipv4/tcp/tcp_minisocks.c b/net/ipv4v6_shared/tcp/tcp_minisocks.c
similarity index 100%
rename from net/ipv4/tcp/tcp_minisocks.c
rename to net/ipv4v6_shared/tcp/tcp_minisocks.c
diff --git a/net/ipv4/tcp/tcp_output.c b/net/ipv4v6_shared/tcp/tcp_output.c
similarity index 100%
rename from net/ipv4/tcp/tcp_output.c
rename to net/ipv4v6_shared/tcp/tcp_output.c
diff --git a/net/ipv4/tcp/tcp_probe.c b/net/ipv4v6_shared/tcp/tcp_probe.c
similarity index 100%
rename from net/ipv4/tcp/tcp_probe.c
rename to net/ipv4v6_shared/tcp/tcp_probe.c
diff --git a/net/ipv4/tcp/tcp_rate.c b/net/ipv4v6_shared/tcp/tcp_rate.c
similarity index 100%
rename from net/ipv4/tcp/tcp_rate.c
rename to net/ipv4v6_shared/tcp/tcp_rate.c
diff --git a/net/ipv4/tcp/tcp_recovery.c b/net/ipv4v6_shared/tcp/tcp_recovery.c
similarity index 100%
rename from net/ipv4/tcp/tcp_recovery.c
rename to net/ipv4v6_shared/tcp/tcp_recovery.c
diff --git a/net/ipv4/tcp/tcp_timer.c b/net/ipv4v6_shared/tcp/tcp_timer.c
similarity index 100%
rename from net/ipv4/tcp/tcp_timer.c
rename to net/ipv4v6_shared/tcp/tcp_timer.c
diff --git a/net/ipv4/tcp/tcp_ulp.c b/net/ipv4v6_shared/tcp/tcp_ulp.c
similarity index 100%
rename from net/ipv4/tcp/tcp_ulp.c
rename to net/ipv4v6_shared/tcp/tcp_ulp.c
--
2.14.1
^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 0/3] A own subdirectory for shared TCP code
2017-10-03 22:22 [PATCH net-next 0/3] A own subdirectory for shared TCP code Richard Sailer
` (2 preceding siblings ...)
2017-10-03 22:22 ` [PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared Richard Sailer
@ 2017-10-03 22:42 ` Eric Dumazet
2017-10-03 23:03 ` David Miller
4 siblings, 0 replies; 10+ messages in thread
From: Eric Dumazet @ 2017-10-03 22:42 UTC (permalink / raw)
To: Richard Sailer; +Cc: netdev, davem, kuznet, yoshfuji
On Wed, 2017-10-04 at 00:22 +0200, Richard Sailer wrote:
> net/ipv4 currently contains around 100 source files containing the
> IP implementation and lots of other functionality (UDP, TCP, xfrm, etc.)
>
> # 1/3
> To make the networking source tree more self documenting and well
> structured 1/3 moves the 30 shared TCP source files to a own
> subdirectory of net/ipv4/.
>
> # 2/3
> A huge part of the TCP code is congestion control algorithms.
> With the same goal as above, this moves the cc algos to a own
> subdirectory of net/ipv4/tcp/
>
> # 3/3
> Implementation of a suggestion of valdis, to have a new directory
> net/ipv4v6_shared that contains all the code shared between
> IPv4 and IPv6. This creates net/ipv4v6_shared and moves the
> net/ipv4/tcp directory of shared TCP code there.
>
> While every commit depends on its previous commits, it is
> also possible and sensible to only apply (1) or (1,2) , depending
> on how conservative one wants to decide regarding directory
> structure changes.
>
> [PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory
> [PATCH net-next 2/3] tcp: Move cc algorithms to own subdirectory
> [PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared
NACK for the whole series.
Simple reason :
This is going to be a maintenance nightmare, for people dealing with TCP
stack every day.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 0/3] A own subdirectory for shared TCP code
2017-10-03 22:22 [PATCH net-next 0/3] A own subdirectory for shared TCP code Richard Sailer
` (3 preceding siblings ...)
2017-10-03 22:42 ` [PATCH net-next 0/3] A own subdirectory for shared TCP code Eric Dumazet
@ 2017-10-03 23:03 ` David Miller
2017-10-04 18:54 ` Richard Siegfried
4 siblings, 1 reply; 10+ messages in thread
From: David Miller @ 2017-10-03 23:03 UTC (permalink / raw)
To: richard_siegfried; +Cc: netdev, kuznet, yoshfuji
From: Richard Sailer <richard_siegfried@systemli.org>
Date: Wed, 4 Oct 2017 00:22:10 +0200
> net/ipv4 currently contains around 100 source files containing the
> IP implementation and lots of other functionality (UDP, TCP, xfrm, etc.)
As someone who has to do backports regularly to -stable, there is no way
I am applying this.
Sorry.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared
[not found] ` <CAOaVG14r2Uwm+r17Oa9jwKj50vRr8=SqwXEmmGAeW8rHj3FgXg@mail.gmail.com>
@ 2017-10-04 8:57 ` Richard Siegfried
0 siblings, 0 replies; 10+ messages in thread
From: Richard Siegfried @ 2017-10-04 8:57 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev, David Miller, yoshfuji, kuznet
[-- Attachment #1.1: Type: text/plain, Size: 428 bytes --]
On 04/10/17 03:32, Stephen Hemminger wrote:
> The name ipv4v6_shared is longer than it needs to be maybe ip or ip_common
Well ip_common sounds nice.
But I don't fully understand what this means for the current situation.
I could do a v2, but would that make any sense since Dave and Eric
already said "no" to the whole concept.
Would it change anything to keep the tcp directory in net/ipv4/ (only
1/3 and 2/3)?
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 529 bytes --]
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 0/3] A own subdirectory for shared TCP code
2017-10-03 23:03 ` David Miller
@ 2017-10-04 18:54 ` Richard Siegfried
2017-10-04 20:27 ` Andrew Lunn
0 siblings, 1 reply; 10+ messages in thread
From: Richard Siegfried @ 2017-10-04 18:54 UTC (permalink / raw)
To: David Miller; +Cc: netdev
On 04/10/17 01:03, David Miller wrote:
> As someone who has to do backports regularly to -stable, there is no way
> I am applying this.
>
> Sorry.
Okay, I see.
Is grouping files into subdirectories something generally
unwanted/unlikely to be applied or is this specific to TCP / networking?
Because there are several other places in the source tree where I would
like to group things.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 0/3] A own subdirectory for shared TCP code
2017-10-04 18:54 ` Richard Siegfried
@ 2017-10-04 20:27 ` Andrew Lunn
2017-10-05 13:17 ` Richard Siegfried
0 siblings, 1 reply; 10+ messages in thread
From: Andrew Lunn @ 2017-10-04 20:27 UTC (permalink / raw)
To: Richard Siegfried; +Cc: David Miller, netdev
On Wed, Oct 04, 2017 at 08:54:17PM +0200, Richard Siegfried wrote:
> On 04/10/17 01:03, David Miller wrote:
> > As someone who has to do backports regularly to -stable, there is no way
> > I am applying this.
> >
> > Sorry.
> Okay, I see.
>
> Is grouping files into subdirectories something generally
> unwanted/unlikely to be applied or is this specific to TCP / networking?
>
> Because there are several other places in the source tree where I would
> like to group things.
Hi Richard
It is generally unwanted.
Have you tried back porting patches when the directory structure has
changed? Files have moved around? It makes it a lot harder to
do. Meaning patches are going to be back ported less often. Fixes
which could be security relevant might not get back ported, etc.
Kernel 4.4 is going to be supported until 2022. So moving files around
is going to make Greg Kroah-Hartman life more difficult for the next 5
years.
Andrew
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH net-next 0/3] A own subdirectory for shared TCP code
2017-10-04 20:27 ` Andrew Lunn
@ 2017-10-05 13:17 ` Richard Siegfried
0 siblings, 0 replies; 10+ messages in thread
From: Richard Siegfried @ 2017-10-05 13:17 UTC (permalink / raw)
To: Andrew Lunn; +Cc: netdev
[-- Attachment #1.1: Type: text/plain, Size: 539 bytes --]
On 04/10/17 22:27, Andrew Lunn wrote:
> Hi Richard
>
> It is generally unwanted.
>
> Have you tried back porting patches when the directory structure has
> changed? Files have moved around? It makes it a lot harder to
> do. Meaning patches are going to be back ported less often. Fixes
> which could be security relevant might not get back ported, etc.
>
> Kernel 4.4 is going to be supported until 2022. So moving files around
> is going to make Greg Kroah-Hartman life more difficult for the next 5
> years.
Ok, I see
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 529 bytes --]
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2017-10-05 13:17 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-10-03 22:22 [PATCH net-next 0/3] A own subdirectory for shared TCP code Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 1/3] net/ipv4: Move shared tcp code to own subdirectory Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 2/3] tcp: Move cc algorithms " Richard Sailer
2017-10-03 22:22 ` [PATCH net-next 3/3] Move shared tcp code to net/ipv4v6shared Richard Sailer
[not found] ` <CAOaVG14r2Uwm+r17Oa9jwKj50vRr8=SqwXEmmGAeW8rHj3FgXg@mail.gmail.com>
2017-10-04 8:57 ` Richard Siegfried
2017-10-03 22:42 ` [PATCH net-next 0/3] A own subdirectory for shared TCP code Eric Dumazet
2017-10-03 23:03 ` David Miller
2017-10-04 18:54 ` Richard Siegfried
2017-10-04 20:27 ` Andrew Lunn
2017-10-05 13:17 ` Richard Siegfried
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.