B.A.T.M.A.N November 2016

b.a.t.m.a.n@lists.open-mesh.org

19 participants
48 discussions

[B.A.T.M.A.N.] PATCH alfred
by Jean-Jacques Sarton 01 Nov '16

01 Nov '16

I had problem with alfre which don't start. I'am using netfs. Swen has proposed a patch which remove the batadv_interface_check() function. May way is an other regarding the transitional period where the old an new feature are to be supported. Signed-off-by: jj.sarton(a)t-online.de diff --git a/batadv_query.c b/batadv_query.c index a671b79..c8732ef 100644 --- a/batadv_query.c +++ b/batadv_query.c @@ -136,7 +136,7 @@ int ipv6_to_mac(const struct in6_addr *addr, struct ether_addr *mac) return 0; } -int batadv_interface_check(const char *mesh_iface) +int batadv_interface_check_debugfs(const char *mesh_iface) { char full_path[MAX_PATH + 1]; FILE *f; @@ -166,6 +166,18 @@ int batadv_interface_check(const char *mesh_iface) return 0; } +int batadv_interface_check(const char *mesh_iface) +{ + int ret = 0; + enable_net_admin_capability(1); + ret = batadv_interface_check_netlink(mesh_iface); + enable_net_admin_capability(0); + + if ( ret < 0 ) + return batadv_interface_check_debugfs(mesh_iface); + return 0; +} + static int translate_mac_debugfs(const char *mesh_iface, const struct ether_addr *mac, struct ether_addr *mac_out) diff --git a/netlink.c b/netlink.c index 1b5695c..ca311e7 100644 --- a/netlink.c +++ b/netlink.c @@ -365,3 +365,30 @@ int get_tq_netlink(const char *mesh_iface, const struct ether_addr *mac, return 0; } + +int batadv_interface_check_netlink(const char *mesh_iface) +{ + struct get_tq_netlink_opts opts = { + .tq = 0, + .found = false, + .query_opts = { + .err = 0, + }, + }; + int ret = 0; + + memset(&opts.mac, 0, ETH_ALEN); + + ret = netlink_query_common(mesh_iface, BATADV_CMD_GET_ORIGINATORS, + get_tq_netlink_cb, &opts.query_opts); + if (ret < 0) + return ret; + + memset(&opts.mac, 0, ETH_ALEN); + ret = netlink_query_common(mesh_iface, BATADV_CMD_GET_ORIGINATORS, + get_tq_netlink_cb, &opts.query_opts); + + if (ret < 0) + return ret; + return 0; +} diff --git a/netlink.h b/netlink.h index b08e872..9bc75a1 100644 --- a/netlink.h +++ b/netlink.h @@ -49,6 +49,7 @@ int translate_mac_netlink(const char *mesh_iface, const struct ether_addr *mac, struct ether_addr *mac_out); int get_tq_netlink(const char *mesh_iface, const struct ether_addr *mac, uint8_t *tq); +int batadv_interface_check_netlink(const char *mesh_iface); extern struct nla_policy batadv_netlink_policy[];

2 1

how to config gateway ?
by johnzeng 01 Nov '16

01 Nov '16

Hello Dear Sir : after i succeed to connect hostA to hostB at wifi environment this afternoon , and i config gateway between hostA and hostB, i run avahi-autoipd bat0 at two host only ( not run dhcpcd ) and run batctl gw_mode server and batctl gw_mode server 5mbit/1mbit at hostA, and run batctl gw_mode client and batctl gw_mode client 20 at hostB Whether i need config dhcpcd at hostA ? if possible , please give me a example . Thanks --------------------------- HostB status -------------------------- [root@alarmpi ~]# batctl gwl Gateway (#/255) Nexthop [outgoingIF]: advertised uplink bandwidth ... [B.A.T.M.A.N. adv 2015.2, MainIF/MAC: wlan0/c8:3a:35:c9:ce:71 (bat0)] => 00:b0:8c:e7:25:3a (248) 00:b0:8c:e7:25:3a [ wlan0]: 5.0/1.0 MBit [root@alarmpi ~]# batctl gw_mode client (selection class: 20) [B.A.T.M.A.N. adv 2015.2, MainIF/MAC: wlan0/c8:3a:35:c9:ce:71 (bat0 BATMAN_IV)] Originator last-seen (#/255) Nexthop [outgoingIF]: Potential nexthops ... 00:b0:8c:e7:25:3a 0.840s (229) 00:b0:8c:e7:25:3a [ wlan0]: 00:b0:8c:e7:25:3a (229) --------------------------- HostA status -------------------------- [root@alarmpi ~]# batctl gw_mode server (announced bw: 5.0/1.0 MBit) [root@alarmpi ~]# batctl o [B.A.T.M.A.N. adv 2015.2, MainIF/MAC: wlan0/00:b0:8c:e7:25:3a (bat0 BATMAN_IV)] Originator last-seen (#/255) Nexthop [outgoingIF]: Potential nexthops ... c8:3a:35:c9:ce:71 0.440s (239) c8:3a:35:c9:ce:71 [ wlan0]: c8:3a:35:c9:ce:71 (239) [root@alarmpi ~]# batctl gwl Gateway (#/255) Nexthop [outgoingIF]: advertised uplink bandwidth ... [B.A.T.M.A.N. adv 2015.2, MainIF/MAC: wlan0/00:b0:8c:e7:25:3a (bat0)] No gateways in range ..

2 1

[B.A.T.M.A.N.] [PATCH] batman-adv: vis: Don't fail when debugfs cannot be mounted
by Sven Eckelmann 01 Nov '16

01 Nov '16

The vis server is supposed to use netlink when available. This allows to work on systems without debugfs and under network namespaces. Thus it is now not strictly required anymore to have debugfs available. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- vis/vis.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/vis/vis.c b/vis/vis.c index f4d13c7..51d7f5c 100644 --- a/vis/vis.c +++ b/vis/vis.c @@ -1120,13 +1120,7 @@ static struct globals *vis_init(int argc, char *argv[]) static int vis_server(struct globals *globals) { - char *debugfs_mnt; - - debugfs_mnt = debugfs_mount(NULL); - if (!debugfs_mnt) { - fprintf(stderr, "Error - can't mount or find debugfs\n"); - return EXIT_FAILURE; - } + debugfs_mount(NULL); globals->push = (struct alfred_push_data_v0 *) globals->buf; globals->vis_data = (struct vis_v1 *) (globals->buf + sizeof(*globals->push) + sizeof(struct alfred_data)); -- 2.10.1

1 0

[B.A.T.M.A.N.] [PATCH] alfred: Don't check for debugfs on startup
by Sven Eckelmann 01 Nov '16

01 Nov '16

debugfs is deprecated and netlink is used by the modern batman-adv module. Thus it can easily happen that debugfs was not enabled during the build of batman-adv. Also netns may not have access to debugfs. So drop the check for transtable_global and originators for the specified mesh interface. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- batadv_query.c | 30 ------------------------------ batadv_query.h | 1 - server.c | 4 ---- 3 files changed, 35 deletions(-) diff --git a/batadv_query.c b/batadv_query.c index a671b79..8815745 100644 --- a/batadv_query.c +++ b/batadv_query.c @@ -136,36 +136,6 @@ int ipv6_to_mac(const struct in6_addr *addr, struct ether_addr *mac) return 0; } -int batadv_interface_check(const char *mesh_iface) -{ - char full_path[MAX_PATH + 1]; - FILE *f; - - debugfs_make_path(DEBUG_BATIF_PATH_FMT "/" DEBUG_TRANSTABLE_GLOBAL, - mesh_iface, full_path, sizeof(full_path)); - f = fopen(full_path, "r"); - if (!f) { - fprintf(stderr, - "Could not find %s for interface %s. Make sure it is a valid batman-adv soft-interface\n", - DEBUG_TRANSTABLE_GLOBAL, mesh_iface); - return -1; - } - fclose(f); - - debugfs_make_path(DEBUG_BATIF_PATH_FMT "/" DEBUG_ORIGINATORS, - mesh_iface, full_path, sizeof(full_path)); - f = fopen(full_path, "r"); - if (!f) { - fprintf(stderr, - "Could not find %s for interface %s. Make sure it is a valid batman-adv soft-interface\n", - DEBUG_ORIGINATORS, mesh_iface); - return -1; - } - fclose(f); - - return 0; -} - static int translate_mac_debugfs(const char *mesh_iface, const struct ether_addr *mac, struct ether_addr *mac_out) diff --git a/batadv_query.h b/batadv_query.h index 320203b..907ed07 100644 --- a/batadv_query.h +++ b/batadv_query.h @@ -27,7 +27,6 @@ struct ether_addr *translate_mac(const char *mesh_iface, const struct ether_addr *mac); uint8_t get_tq(const char *mesh_iface, struct ether_addr *mac); -int batadv_interface_check(const char *mesh_iface); int mac_to_ipv6(const struct ether_addr *mac, struct in6_addr *addr); int ipv6_to_mac(const struct in6_addr *addr, struct ether_addr *mac); int is_ipv6_eui64(const struct in6_addr *addr); diff --git a/server.c b/server.c index 3a0ad8d..445e076 100644 --- a/server.c +++ b/server.c @@ -352,10 +352,6 @@ int alfred_server(struct globals *globals) return -1; } - if (strcmp(globals->mesh_iface, "none") != 0 && - batadv_interface_check(globals->mesh_iface) < 0) - return -1; - num_socks = netsock_open_all(globals); if (num_socks <= 0) { fprintf(stderr, "Failed to open interfaces\n"); -- 2.10.1

1 0

[B.A.T.M.A.N.] [PATCH] batman-adv: Deactivate __ro_after_init for kernels < 4.10
by Sven Eckelmann 01 Nov '16

01 Nov '16

The pre-4.10 kernels must write to batadv_netlink_family after the family was registered. Thus disabling __ro_after_init for now is required on these kernels. Fixes: c776b4a071d8 ("batman-adv: genetlink: mark families as __ro_after_init") Reported-by: Linus Lüssing <linus.luessing(a)c0d3.blue> Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- compat-include/linux/cache.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/compat-include/linux/cache.h b/compat-include/linux/cache.h index 93dff58..7b7069a 100644 --- a/compat-include/linux/cache.h +++ b/compat-include/linux/cache.h @@ -24,11 +24,13 @@ #include <linux/version.h> #include_next <linux/cache.h> -#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 6, 0) +#if LINUX_VERSION_CODE < KERNEL_VERSION(4, 10, 0) -#ifndef __ro_after_init -#define __ro_after_init +/* hack for netlink.c which marked the family ops as ro */ +#ifdef __ro_after_init +#undef __ro_after_init #endif +#define __ro_after_init #endif /* < KERNEL_VERSION(4, 6, 0) */ -- 2.10.2

2 2

[B.A.T.M.A.N.] [PATCH maint v3] batman-adv: Install module in $INSTALL_MOD_DIR/updates/net/batman-adv
by Sven Eckelmann 01 Nov '16

01 Nov '16

depmod may skip the batman-adv.ko module in the $INSTALL_MOD_DIR/update/ folder when it contains the word "build" as directory name. Avoid this extra directory "build" in the installation path by making the obj-y dir relative to build/ in the source directory. Reported-by: Jean-Jacques Sarton <jj.sarton(a)t-online.de> Fixes: 66ac14a09445 ("batman-adv: Introduce compat-patches support") Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- v3: - adjust commit message v2: - Remove clean target change which was accidentally added to this patch Makefile | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/Makefile b/Makefile index 1d2e01c..e9c499f 100644 --- a/Makefile +++ b/Makefile @@ -54,6 +54,7 @@ PATCH_FLAGS = --batch --fuzz=0 --forward --strip=1 --unified --version-control=n PATCH := patch $(PATCH_FLAGS) -i SPATCH_FLAGS := --in-place --relax-include-path --use-coccigrep --very-quiet CP := cp -fpR +LN := ln -sf SOURCE = $(wildcard net/batman-adv/*.[ch]) net/batman-adv/Makefile SOURCE_BUILD = $(wildcard $(BUILD_DIR)/net/batman-adv/*.[ch]) $(BUILD_DIR)/net/batman-adv/Makefile @@ -63,25 +64,25 @@ REVISION= $(shell if [ -d "$(PWD)/.git" ]; then \ echo $$(git --git-dir="$(PWD)/.git" describe --always --dirty --match "v*" |sed 's/^v//' 2> /dev/null || echo "[unknown]"); \ fi) NOSTDINC_FLAGS += \ - -I$(PWD)/compat-include/ \ - -I$(PWD)/include/ \ - -include $(PWD)/compat.h \ + -I$(PWD)/../compat-include/ \ + -I$(PWD)/../include/ \ + -include $(PWD)/../compat.h \ $(CFLAGS) ifneq ($(REVISION),) NOSTDINC_FLAGS += -DBATADV_SOURCE_VERSION=\"$(REVISION)\" endif -include $(PWD)/compat-sources/Makefile +-include $(PWD)/../compat-sources/Makefile -obj-y += build/net/batman-adv/ +obj-y += net/batman-adv/ export batman-adv-y BUILD_FLAGS := \ - M=$(PWD) \ - PWD=$(PWD) \ + M=$(BUILD_DIR) \ + PWD=$(BUILD_DIR) \ REVISION=$(REVISION) \ CONFIG_BATMAN_ADV=m \ CONFIG_BATMAN_ADV_DEBUG=$(CONFIG_BATMAN_ADV_DEBUG) \ @@ -109,6 +110,7 @@ config: $(SOURCE_STAMP): $(SOURCE) compat-patches/* $(MKDIR) $(BUILD_DIR)/net/batman-adv/ + @$(LN) ../Makefile $(BUILD_DIR)/Makefile @$(RM) $(SOURCE_BUILD) @$(CP) $(SOURCE) $(BUILD_DIR)/net/batman-adv/ @set -e; \

2 2

[B.A.T.M.A.N.] [PATCH] batman-adv: Add module alias for batadv netlink family
by Sven Eckelmann 01 Nov '16

01 Nov '16

The batman-adv module has to be loaded to fulfill genl request by the userspace. When it is not loaded then requests will fail. It is therefore useful to get the module automatically loaded when such a request is made. Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- net/batman-adv/main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/batman-adv/main.c b/net/batman-adv/main.c index 6b5dae6..d46415e 100644 --- a/net/batman-adv/main.c +++ b/net/batman-adv/main.c @@ -23,6 +23,7 @@ #include <linux/crc32c.h> #include <linux/errno.h> #include <linux/fs.h> +#include <linux/genetlink.h> #include <linux/if_ether.h> #include <linux/if_vlan.h> #include <linux/init.h> @@ -44,6 +45,7 @@ #include <linux/workqueue.h> #include <net/dsfield.h> #include <net/rtnetlink.h> +#include <uapi/linux/batman_adv.h> #include "bat_algo.h" #include "bat_iv_ogm.h" @@ -648,3 +650,4 @@ MODULE_DESCRIPTION(BATADV_DRIVER_DESC); MODULE_SUPPORTED_DEVICE(BATADV_DRIVER_DEVICE); MODULE_VERSION(BATADV_SOURCE_VERSION); MODULE_ALIAS_RTNL_LINK("batadv"); +MODULE_ALIAS_GENL_FAMILY(BATADV_NL_NAME); -- 2.10.1

1 1

[B.A.T.M.A.N.] [PATCH 1/3] batman-adv: genetlink: no longer support using static family IDs
by Sven Eckelmann 01 Nov '16

01 Nov '16

From: Johannes Berg <johannes.berg(a)intel.com> Static family IDs have never really been used, the only use case was the workaround I introduced for those users that assumed their family ID was also their multicast group ID. Additionally, because static family IDs would never be reserved by the generic netlink code, using a relatively low ID would only work for built-in families that can be registered immediately after generic netlink is started, which is basically only the control family (apart from the workaround code, which I also had to add code for so it would reserve those IDs) Thus, anything other than GENL_ID_GENERATE is flawed and luckily not used except in the cases I mentioned. Move those workarounds into a few lines of code, and then get rid of GENL_ID_GENERATE entirely, making it more robust. Signed-off-by: Johannes Berg <johannes.berg(a)intel.com> Signed-off-by: David S. Miller <davem(a)davemloft.net> Signed-off-by: Sven Eckelmann <sven(a)narfation.org> --- Patch already in net-next net/batman-adv/netlink.c | 1 - 1 file changed, 1 deletion(-) diff --git a/net/batman-adv/netlink.c b/net/batman-adv/netlink.c index aee20a3..a951e02 100644 --- a/net/batman-adv/netlink.c +++ b/net/batman-adv/netlink.c @@ -49,7 +49,6 @@ #include "translation-table.h" struct genl_family batadv_netlink_family = { - .id = GENL_ID_GENERATE, .hdrsize = 0, .name = BATADV_NL_NAME, .version = 1, -- 2.10.1

1 5

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

B.A.T.M.A.N November 2016