* [PATCH v2 iproute2] htb: add support for direct_qlen attribute
From: Eric Dumazet @ 2013-09-13 16:10 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev
In-Reply-To: <1379086235.24408.41.camel@edumazet-glaptop>
From: Eric Dumazet <edumazet@google.com>
TCA_HTB_DIRECT_QLEN attribute is supported since linux-3.10
HTB classes use an internal pfifo queue, which limit was not reported
by tc, and value inherited from device tx_queue_len at setup time.
With this patch, tc displays the value and can change it.
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
tc/q_htb.c | 21 +++++++++++++++++++--
1 file changed, 19 insertions(+), 2 deletions(-)
diff --git a/tc/q_htb.c b/tc/q_htb.c
index 7b6f908..6737ddb 100644
--- a/tc/q_htb.c
+++ b/tc/q_htb.c
@@ -31,9 +31,11 @@
static void explain(void)
{
fprintf(stderr, "Usage: ... qdisc add ... htb [default N] [r2q N]\n"
+ " [direct_qlen P]\n"
" default minor id of class to which unclassified packets are sent {0}\n"
" r2q DRR quantums are computed as rate in Bps/r2q {10}\n"
" debug string of 16 numbers each 0-3 {0}\n\n"
+ " direct_qlen Limit of the direct queue {in packets}\n"
"... class add ... htb rate R1 [burst B1] [mpu B] [overhead O]\n"
" [prio P] [slot S] [pslot PS]\n"
" [ceil R2] [cburst B2] [mtu MTU] [quantum Q]\n"
@@ -108,6 +110,7 @@ static int htb_parse_class_opt(struct qdisc_util *qu, int argc, char **argv, str
unsigned mtu;
unsigned short mpu = 0;
unsigned short overhead = 0;
+ unsigned int direct_qlen = ~0U;
unsigned int linklayer = LINKLAYER_ETHERNET; /* Assume ethernet */
struct rtattr *tail;
@@ -125,6 +128,11 @@ static int htb_parse_class_opt(struct qdisc_util *qu, int argc, char **argv, str
if (get_u32(&mtu, *argv, 10)) {
explain1("mtu"); return -1;
}
+ } else if (matches(*argv, "direct_qlen") == 0) {
+ NEXT_ARG();
+ if (get_u32(&direct_qlen, *argv, 10)) {
+ explain1("direct_qlen"); return -1;
+ }
} else if (matches(*argv, "mpu") == 0) {
NEXT_ARG();
if (get_u16(&mpu, *argv, 10)) {
@@ -230,6 +238,9 @@ static int htb_parse_class_opt(struct qdisc_util *qu, int argc, char **argv, str
opt.cbuffer = tc_calc_xmittime(opt.ceil.rate, cbuffer);
tail = NLMSG_TAIL(n);
+ if (direct_qlen != ~0U)
+ addattr_l(n, 1024, TCA_HTB_DIRECT_QLEN,
+ &direct_qlen, sizeof(direct_qlen));
addattr_l(n, 1024, TCA_OPTIONS, NULL, 0);
addattr_l(n, 2024, TCA_HTB_PARMS, &opt, sizeof(opt));
addattr_l(n, 3024, TCA_HTB_RTAB, rtab, 1024);
@@ -240,7 +251,7 @@ static int htb_parse_class_opt(struct qdisc_util *qu, int argc, char **argv, str
static int htb_print_opt(struct qdisc_util *qu, FILE *f, struct rtattr *opt)
{
- struct rtattr *tb[TCA_HTB_RTAB+1];
+ struct rtattr *tb[TCA_HTB_MAX + 1];
struct tc_htb_opt *hopt;
struct tc_htb_glob *gopt;
double buffer,cbuffer;
@@ -253,7 +264,7 @@ static int htb_print_opt(struct qdisc_util *qu, FILE *f, struct rtattr *opt)
if (opt == NULL)
return 0;
- parse_rtattr_nested(tb, TCA_HTB_RTAB, opt);
+ parse_rtattr_nested(tb, TCA_HTB_MAX, opt);
if (tb[TCA_HTB_PARMS]) {
hopt = RTA_DATA(tb[TCA_HTB_PARMS]);
@@ -302,6 +313,12 @@ static int htb_print_opt(struct qdisc_util *qu, FILE *f, struct rtattr *opt)
if (show_details)
fprintf(f," ver %d.%d",gopt->version >> 16,gopt->version & 0xffff);
}
+ if (tb[TCA_HTB_DIRECT_QLEN] &&
+ RTA_PAYLOAD(tb[TCA_HTB_DIRECT_QLEN]) >= sizeof(__u32)) {
+ __u32 direct_qlen = rta_getattr_u32(tb[TCA_HTB_DIRECT_QLEN]);
+
+ fprintf(f, " direct_qlen %u", direct_qlen);
+ }
return 0;
}
^ permalink raw reply related
* Re: [PATCH] [Trivial] remove unnecessary header file inclusion
From: Ben Hutchings @ 2013-09-13 16:11 UTC (permalink / raw)
To: ZHAO Gang; +Cc: davem, netdev, linux-kernel
In-Reply-To: <20130913145718.GA2536@will>
On Fri, 2013-09-13 at 22:57 +0800, ZHAO Gang wrote:
> file linux/fib_rules.h doesn't exist, so remove it.
Do you think people wouldn't notice #include of a nonexistent file? gcc
is quite good at complaining about them...
It's now called include/uapi/linux/fib_rules.h.
Ben.
> Signed-off-by: ZHAO Gang <gamerh2o@gmail.com>
> ---
> include/net/fib_rules.h | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
> index 4b2b557..c816f91 100644
> --- a/include/net/fib_rules.h
> +++ b/include/net/fib_rules.h
> @@ -4,7 +4,6 @@
> #include <linux/types.h>
> #include <linux/slab.h>
> #include <linux/netdevice.h>
> -#include <linux/fib_rules.h>
> #include <net/flow.h>
> #include <net/rtnetlink.h>
>
--
Ben Hutchings, Staff Engineer, Solarflare
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* [PATCH] ath9k: mark wmi_event_swba as __packed
From: Chris Metcalf @ 2013-09-13 16:14 UTC (permalink / raw)
To: Luis R. Rodriguez, Jouni Malinen, Vasanthakumar Thiagarajan,
Senthil Balasubramanian, John W. Linville, linux-wireless,
ath9k-devel, netdev, linux-kernel
The other structures in wmi.h are already marked this way.
Without this marking, we get an unaliged access panic in the tilegx kernel:
Starting stack dump of tid 0, pid 0 (swapper) on cpu 35 at cycle 198675113844
frame 0: 0xfffffff7103ada90 ath9k_htc_swba+0x120/0x618 [ath9k_htc]
frame 1: 0xfffffff7103a4b10 ath9k_wmi_event_tasklet+0x1b0/0x270 [ath9k_htc]
frame 2: 0xfffffff700326570 tasklet_action+0x148/0x298
[...]
Signed-off-by: Chris Metcalf <cmetcalf@tilera.com>
---
drivers/net/wireless/ath/ath9k/wmi.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/ath/ath9k/wmi.h b/drivers/net/wireless/ath/ath9k/wmi.h
index fde6da6..0db37f2 100644
--- a/drivers/net/wireless/ath/ath9k/wmi.h
+++ b/drivers/net/wireless/ath/ath9k/wmi.h
@@ -39,7 +39,7 @@ struct wmi_fw_version {
struct wmi_event_swba {
__be64 tsf;
u8 beacon_pending;
-};
+} __packed;
/*
* 64 - HTC header - WMI header - 1 / txstatus
--
1.8.3.1
^ permalink raw reply related
* Re: [PATCH net-next] bridge: fix NULL pointer deref in br_handle_frame
From: Stephen Hemminger @ 2013-09-13 16:21 UTC (permalink / raw)
To: Hong Zhiguo; +Cc: netdev, davem, zhiguohong, eric.dumazet, vyasevic
In-Reply-To: <1379041787-14232-1-git-send-email-zhiguohong@tencent.com>
On Fri, 13 Sep 2013 11:09:47 +0800
Hong Zhiguo <honkiko@gmail.com> wrote:
> From: Hong Zhiguo <zhiguohong@tencent.com>
>
> I got an Oops on my box when br_handle_frame is called between these
> 2 lines of del_nbp:
> dev->priv_flags &= ~IFF_BRIDGE_PORT;
> /* --> br_handle_frame is called at this time */
> netdev_rx_handler_unregister(dev);
>
> In br_handle_frame the return of br_port_get_rcu(dev) is dereferenced
> without check but br_port_get_rcu(dev) returns NULL if:
> !(dev->priv_flags & IFF_BRIDGE_PORT)
>
> In my first fix I moved netdev_rx_handler_unregister up. Eric Dumazet
> pointed out the testing of IFF_BRIDGE_PORT is not necessary here since
> we're in rcu_read_lock and we have synchronize_net() in
> netdev_rx_handler_unregister. This fix removed the testing of
> IFF_BRIDGE_PORT.
>
> I tested the fix on my box with script doing "brctl addif" and "brctl
> delif" repeatedly while a lot of broadcast frame present on the LAN.
> I added msleep in del_nbp between setting of priv_flags and unregister
> so it's easy to reproduce the oops without the fix.
>
> I'll send another patch to net-next to take care of br_netfilter and
> ebtable if necessary(seems there's NULL check following but I'll
> have a look).
>
> The Oops(some lines omitted):
> BUG: unable to handle kernel NULL pointer dereference at 0000000000000021
> IP: [<ffffffff8150901d>] br_handle_frame+0xed/0x230
> Oops: 0000 [#1] PREEMPT SMP
> RIP: 0010:[<ffffffff8150901d>] [<ffffffff8150901d>] br_handle_frame+0xed/0x230
> RSP: 0018:ffff880030403c10 EFLAGS: 00010286
> Stack:
> ffff88002c945700 ffffffff81508f30 0000000000000000 ffff88002d41e000
> ffff880030403c98 ffffffff81477acb ffffffff81477821 ffff880030403c68
> ffffffff81090e10 00ff88002d545c80 ffff88002c945700 ffffffff81aa50c0
> Call Trace:
> <IRQ>
> [<ffffffff81508f30>] ? br_handle_frame_finish+0x300/0x300
> [<ffffffff81477acb>] __netif_receive_skb_core+0x39b/0x880
>
> Signed-off-by: Hong Zhiguo <zhiguohong@tencent.com>
> ---
> net/bridge/br_input.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
> index a2fd37e..2244049 100644
> --- a/net/bridge/br_input.c
> +++ b/net/bridge/br_input.c
> @@ -60,7 +60,7 @@ static int br_pass_frame_up(struct sk_buff *skb)
> int br_handle_frame_finish(struct sk_buff *skb)
> {
> const unsigned char *dest = eth_hdr(skb)->h_dest;
> - struct net_bridge_port *p = br_port_get_rcu(skb->dev);
> + struct net_bridge_port *p = rcu_dereference(skb->dev->rx_handler_data);
> struct net_bridge *br;
> struct net_bridge_fdb_entry *dst;
> struct net_bridge_mdb_entry *mdst;
> @@ -143,7 +143,7 @@ drop:
> /* note: already called with rcu_read_lock */
> static int br_handle_local_finish(struct sk_buff *skb)
> {
> - struct net_bridge_port *p = br_port_get_rcu(skb->dev);
> + struct net_bridge_port *p = rcu_dereference(skb->dev->rx_handler_data);
> u16 vid = 0;
>
> br_vlan_get_tag(skb, &vid);
> @@ -173,7 +173,7 @@ rx_handler_result_t br_handle_frame(struct sk_buff **pskb)
> if (!skb)
> return RX_HANDLER_CONSUMED;
>
> - p = br_port_get_rcu(skb->dev);
> + p = rcu_dereference(skb->dev->rx_handler_data);
>
> if (unlikely(is_link_local_ether_addr(dest))) {
> /*
There are more uses of br_port_get_rcu that have the same problem.
For example receiving an STP packet in that window.
I bet if you look at all the callers of br_port_get_rcu() you will
see the same issue, therefore either the check should be removed from br_port_get_rcu.
A little bit of history, the bridge code orginally did not use RCU,
and there was a flag on the device. Then RCU was added, then the receive
handler stuff was added.
^ permalink raw reply
* Re: ICMP rate limiting in IPv4 but not in IPv6
From: Hannes Frederic Sowa @ 2013-09-13 16:44 UTC (permalink / raw)
To: Andy Johnson; +Cc: netdev
In-Reply-To: <CAF0Lin2-ooC6iQCwjnf5196cfy0Xii6H5prS4CEsjeJVAFYcyA@mail.gmail.com>
On Fri, Sep 13, 2013 at 06:57:56PM +0300, Andy Johnson wrote:
> Hello,
> After probing into the RFC of ICMPv6, I am even more confused.
>
> RFC 4443 says:
>
> 2.4. Message Processing Rules
> ...
>
> (f) Finally, in order to limit the bandwidth and forwarding costs
> incurred by originating ICMPv6 error messages, an IPv6 node MUST
> limit the rate of ICMPv6 error messages it originates.
> ...
> The rate-limiting parameters SHOULD be configurable.
> ...
>
> Any ideas?
Well, nobody has implemented it because nobody missed the feature yet. I
don't believe there is another reason for that. If you come up with a
patch, I am sure it can go upstream.
Do you want to try to come up with a patch? We need to be a bit
careful regarding neighbor discovery but otherwise this should be
relativ straightforward. In the meantime you could also implement such
ratelimiting with netfilter.
Greetings,
Hannes
^ permalink raw reply
* [PATCH 1/1] be2net: missing variable initialization
From: Antonio Alecrim Jr @ 2013-09-13 17:05 UTC (permalink / raw)
To: Sathya Perla, netdev, linux-kernel; +Cc: Antonio Alecrim Jr
Signed-off-by: Antonio Alecrim Jr <antonio.alecrim@gmail.com>
---
drivers/net/ethernet/emulex/benet/be_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index 3224d28..100b528 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -2802,7 +2802,7 @@ static int be_vfs_if_create(struct be_adapter *adapter)
struct be_resources res = {0};
struct be_vf_cfg *vf_cfg;
u32 cap_flags, en_flags, vf;
- int status;
+ int status = 0;
cap_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST |
BE_IF_FLAGS_MULTICAST;
--
1.8.3.1
^ permalink raw reply related
* Re: ICMP rate limiting in IPv4 but not in IPv6
From: Loganaden Velvindron @ 2013-09-13 17:26 UTC (permalink / raw)
To: Andy Johnson, netdev
In-Reply-To: <20130913164404.GA32431@order.stressinduktion.org>
On Fri, Sep 13, 2013 at 8:44 PM, Hannes Frederic Sowa
<hannes@stressinduktion.org> wrote:
> On Fri, Sep 13, 2013 at 06:57:56PM +0300, Andy Johnson wrote:
>> Hello,
>> After probing into the RFC of ICMPv6, I am even more confused.
>>
>> RFC 4443 says:
>>
>> 2.4. Message Processing Rules
>> ...
>>
>> (f) Finally, in order to limit the bandwidth and forwarding costs
>> incurred by originating ICMPv6 error messages, an IPv6 node MUST
>> limit the rate of ICMPv6 error messages it originates.
>> ...
>> The rate-limiting parameters SHOULD be configurable.
>> ...
>>
>> Any ideas?
>
> Well, nobody has implemented it because nobody missed the feature yet. I
> don't believe there is another reason for that. If you come up with a
> patch, I am sure it can go upstream.
>
> Do you want to try to come up with a patch? We need to be a bit
> careful regarding neighbor discovery but otherwise this should be
> relativ straightforward. In the meantime you could also implement such
> ratelimiting with netfilter.
Careful ? Can you please elaborate ?
>
> Greetings,
>
> Hannes
>
> --
> To unsubscribe from this list: send the line "unsubscribe netdev" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
This message is strictly personal and the opinions expressed do not
represent those of my employers, either past or present.
^ permalink raw reply
* [RFC] iproute2: add json output to nstat
From: Stephen Hemminger @ 2013-09-13 17:35 UTC (permalink / raw)
To: netdev
New command line flag to output statistics in JSON format.
In our envrionment, we have scripts that parse output of commands.
It is better to use a format supported by existing parsers.
If this is popular, will add similar flags to ip and tc commands.
---
misc/nstat.c | 46 ++++++++++++++++++++++++++++++++++++++++------
1 file changed, 40 insertions(+), 6 deletions(-)
diff --git a/misc/nstat.c b/misc/nstat.c
index b685978..5a4cfe9 100644
--- a/misc/nstat.c
+++ b/misc/nstat.c
@@ -33,6 +33,7 @@ int dump_zeros = 0;
int reset_history = 0;
int ignore_history = 0;
int no_output = 0;
+int json_output = 0;
int no_update = 0;
int scan_interval = 0;
int time_constant = 0;
@@ -255,11 +256,17 @@ static void load_netstat(void)
}
}
+
static void dump_kern_db(FILE *fp, int to_hist)
{
struct nstat_ent *n, *h;
+ const char *eol = "\n";
+
h = hist_db;
- fprintf(fp, "#%s\n", info_source);
+ if (json_output)
+ fprintf(fp, "{ \"%s\": [", info_source);
+ else
+ fprintf(fp, "#%s\n", info_source);
for (n=kern_db; n; n=n->next) {
unsigned long long val = n->val;
if (!dump_zeros && !val && !n->rate)
@@ -276,15 +283,30 @@ static void dump_kern_db(FILE *fp, int to_hist)
}
}
}
- fprintf(fp, "%-32s%-16llu%6.1f\n", n->id, val, n->rate);
+
+ if (json_output) {
+ fprintf(fp, "%s { \"id\":\"%s\", \"val\":%llu,"
+ " \"rate\":%.1f }",
+ eol, n->id, val, n->rate);
+ eol = ",\n";
+ } else
+ fprintf(fp, "%-32s%-16llu%6.1f\n", n->id, val, n->rate);
}
+ if (json_output)
+ fprintf(fp, "\n] }\n");
}
static void dump_incr_db(FILE *fp)
{
struct nstat_ent *n, *h;
+ const char *eol = "\n";
+
h = hist_db;
- fprintf(fp, "#%s\n", info_source);
+ if (json_output)
+ fprintf(fp, "{ \"%s\": [", info_source);
+ else
+ fprintf(fp, "#%s\n", info_source);
+
for (n=kern_db; n; n=n->next) {
int ovfl = 0;
unsigned long long val = n->val;
@@ -304,9 +326,18 @@ static void dump_incr_db(FILE *fp)
continue;
if (!match(n->id))
continue;
- fprintf(fp, "%-32s%-16llu%6.1f%s\n", n->id, val,
- n->rate, ovfl?" (overflow)":"");
+
+ if (json_output) {
+ fprintf(fp, "%s { \"id\":\"%s\", \"val\":%llu,"
+ " \"rate\":%.1f, \"overflow\":%d }",
+ eol, n->id, val, n->rate, ovfl);
+ eol = ",\n";
+ } else
+ fprintf(fp, "%-32s%-16llu%6.1f%s\n", n->id, val,
+ n->rate, ovfl?" (overflow)":"");
}
+ if (json_output)
+ fprintf(fp, "\n] }\n");
}
static int children;
@@ -451,7 +482,7 @@ int main(int argc, char *argv[])
int ch;
int fd;
- while ((ch = getopt(argc, argv, "h?vVzrnasd:t:")) != EOF) {
+ while ((ch = getopt(argc, argv, "h?vVzrnasd:t:j")) != EOF) {
switch(ch) {
case 'z':
dump_zeros = 1;
@@ -478,6 +509,9 @@ int main(int argc, char *argv[])
exit(-1);
}
break;
+ case 'j':
+ json_output = 1;
+ break;
case 'v':
case 'V':
printf("nstat utility, iproute2-ss%s\n", SNAPSHOT);
--
1.7.10.4
^ permalink raw reply related
* Re: [PATCH 4/7] ptp: switch to use gpiolib
From: Krzysztof Halasa @ 2013-09-13 17:36 UTC (permalink / raw)
To: Richard Cochran
Cc: Linus Walleij, linux-gpio@vger.kernel.org, Imre Kaloz,
Alexandre Courbot, linux-arm-kernel@lists.infradead.org,
netdev@vger.kernel.org
In-Reply-To: <20130913133659.GA11394@localhost.localdomain>
Richard Cochran <richardcochran@gmail.com> writes:
> I was really talking about the IXP465 with its PTP function. The Intel
> devel board had a silicon bug making the PTP function useless. But
> even if the bug got fixed, still the PTP function was very limiting.
>
> I never got any feedback about the IXP465 PTP stuff, and so I doubt
> whether anyone is really using the Linux PTP driver on the IXP465, but
> you never know.
Ah, OK.
I have just checked, Intel's spec update says IXP46x steppings A1 and A2
are fixed. PTP on 46x-A0 is broken.
42x and 43x don't support PTP.
OTOH there is interesting stuff there:
Problem: The A-0 and A-1 stepping of the Intel® IXP45X/IXP46X Product Line of Network
Processors can enter a condition whereby the system locks up when certain
combinations of memory accesses occur nearly simultaneously.
After more extensive investigation, it has been identified that the lockup condition can
be caused by near-simultaneous accesses to 1 or 2 cache lines. The accesses to the
cache lines must be performed by the Intel XScale processor, South AHB, and/or North
AHB for the lockup to happen.
Hmmm...
--
Krzysztof Halasa
^ permalink raw reply
* [PATCH net 0/3] SLCAN/SLIP fixes and performance
From: Andre Naujoks @ 2013-09-13 17:37 UTC (permalink / raw)
To: davem; +Cc: linux-can, netdev, linux-kernel
Hi Dave,
these are some loosely related patches, that fix an ancient locking problem in
the slip and slcan drivers, add general ASCII-HEX to bin functions for
uppercase ASCII, fix the handling of CAN RTR frames in the slcan driver
and increase the performance for the slcan driver.
As these patches mainly contain fixes for the slip/slcan drivers that require
a tty layer fix included in 3.11, I would suggest to get the patches in via
the net tree for the 3.12 cycle. They should apply properly on the latest net
and mainline tree.
Best regards,
Andre
Andre Naujoks (3):
slip/slcan: added locking in wakeup function
lib: introduce upper case hex ascii helpers
slcan: rewrite of slc_bump and slc_encaps
drivers/net/can/slcan.c | 139 +++++++++++++++++++++++++++++++-----------------
drivers/net/slip/slip.c | 3 ++
include/linux/kernel.h | 11 ++++
lib/hexdump.c | 2 +
4 files changed, 106 insertions(+), 49 deletions(-)
--
1.8.4.rc3
^ permalink raw reply
* [PATCH net 3/3] slcan: rewrite of slc_bump and slc_encaps
From: Andre Naujoks @ 2013-09-13 17:37 UTC (permalink / raw)
To: davem, Wolfgang Grandegger, Marc Kleine-Budde, linux-can, netdev,
linux-kernel
In-Reply-To: <1379093833-4949-1-git-send-email-nautsch2@gmail.com>
The old implementation was heavy on str* functions and sprintf calls.
This version is more manual, but faster.
Profiling just the printing of a 3 char CAN-id resulted in 60 instructions
for the manual method and over 2000 for the sprintf method. Bear in
mind the profiling was done against libc and not the kernel sprintf.
Together with this rewrite an issue with sending and receiving of RTR frames
has been fixed by Oliver for the cases that the DLC is not zero.
Signed-off-by: Andre Naujoks <nautsch2@gmail.com>
Tested-by: Oliver Hartkopp <socketcan@hartkopp.net>
---
drivers/net/can/slcan.c | 136 +++++++++++++++++++++++++++++++-----------------
1 file changed, 87 insertions(+), 49 deletions(-)
diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c
index d571e2e..25377e5 100644
--- a/drivers/net/can/slcan.c
+++ b/drivers/net/can/slcan.c
@@ -76,6 +76,10 @@ MODULE_PARM_DESC(maxdev, "Maximum number of slcan interfaces");
/* maximum rx buffer len: extended CAN frame with timestamp */
#define SLC_MTU (sizeof("T1111222281122334455667788EA5F\r")+1)
+#define SLC_CMD_LEN 1
+#define SLC_SFF_ID_LEN 3
+#define SLC_EFF_ID_LEN 8
+
struct slcan {
int magic;
@@ -142,47 +146,63 @@ static void slc_bump(struct slcan *sl)
{
struct sk_buff *skb;
struct can_frame cf;
- int i, dlc_pos, tmp;
- unsigned long ultmp;
- char cmd = sl->rbuff[0];
-
- if ((cmd != 't') && (cmd != 'T') && (cmd != 'r') && (cmd != 'R'))
+ int i, tmp;
+ u32 tmpid;
+ char *cmd = sl->rbuff;
+
+ cf.can_id = 0;
+
+ switch (*cmd) {
+ case 'r':
+ cf.can_id = CAN_RTR_FLAG;
+ /* fallthrough */
+ case 't':
+ /* store dlc ASCII value and terminate SFF CAN ID string */
+ cf.can_dlc = sl->rbuff[SLC_CMD_LEN + SLC_SFF_ID_LEN];
+ sl->rbuff[SLC_CMD_LEN + SLC_SFF_ID_LEN] = 0;
+ /* point to payload data behind the dlc */
+ cmd += SLC_CMD_LEN + SLC_SFF_ID_LEN + 1;
+ break;
+ case 'R':
+ cf.can_id = CAN_RTR_FLAG;
+ /* fallthrough */
+ case 'T':
+ cf.can_id |= CAN_EFF_FLAG;
+ /* store dlc ASCII value and terminate EFF CAN ID string */
+ cf.can_dlc = sl->rbuff[SLC_CMD_LEN + SLC_EFF_ID_LEN];
+ sl->rbuff[SLC_CMD_LEN + SLC_EFF_ID_LEN] = 0;
+ /* point to payload data behind the dlc */
+ cmd += SLC_CMD_LEN + SLC_EFF_ID_LEN + 1;
+ break;
+ default:
return;
+ }
- if (cmd & 0x20) /* tiny chars 'r' 't' => standard frame format */
- dlc_pos = 4; /* dlc position tiiid */
- else
- dlc_pos = 9; /* dlc position Tiiiiiiiid */
-
- if (!((sl->rbuff[dlc_pos] >= '0') && (sl->rbuff[dlc_pos] < '9')))
+ if (kstrtou32(sl->rbuff + SLC_CMD_LEN, 16, &tmpid))
return;
- cf.can_dlc = sl->rbuff[dlc_pos] - '0'; /* get can_dlc from ASCII val */
+ cf.can_id |= tmpid;
- sl->rbuff[dlc_pos] = 0; /* terminate can_id string */
-
- if (kstrtoul(sl->rbuff+1, 16, &ultmp))
+ /* get can_dlc from sanitized ASCII value */
+ if (cf.can_dlc >= '0' && cf.can_dlc < '9')
+ cf.can_dlc -= '0';
+ else
return;
- cf.can_id = ultmp;
-
- if (!(cmd & 0x20)) /* NO tiny chars => extended frame format */
- cf.can_id |= CAN_EFF_FLAG;
-
- if ((cmd | 0x20) == 'r') /* RTR frame */
- cf.can_id |= CAN_RTR_FLAG;
-
*(u64 *) (&cf.data) = 0; /* clear payload */
- for (i = 0, dlc_pos++; i < cf.can_dlc; i++) {
- tmp = hex_to_bin(sl->rbuff[dlc_pos++]);
- if (tmp < 0)
- return;
- cf.data[i] = (tmp << 4);
- tmp = hex_to_bin(sl->rbuff[dlc_pos++]);
- if (tmp < 0)
- return;
- cf.data[i] |= tmp;
+ /* RTR frames may have a dlc > 0 but they never have any data bytes */
+ if (!(cf.can_id & CAN_RTR_FLAG)) {
+ for (i = 0; i < cf.can_dlc; i++) {
+ tmp = hex_to_bin(*cmd++);
+ if (tmp < 0)
+ return;
+ cf.data[i] = (tmp << 4);
+ tmp = hex_to_bin(*cmd++);
+ if (tmp < 0)
+ return;
+ cf.data[i] |= tmp;
+ }
}
skb = dev_alloc_skb(sizeof(struct can_frame) +
@@ -209,7 +229,6 @@ static void slc_bump(struct slcan *sl)
/* parse tty input stream */
static void slcan_unesc(struct slcan *sl, unsigned char s)
{
-
if ((s == '\r') || (s == '\a')) { /* CR or BEL ends the pdu */
if (!test_and_clear_bit(SLF_ERROR, &sl->flags) &&
(sl->rcount > 4)) {
@@ -236,27 +255,46 @@ static void slcan_unesc(struct slcan *sl, unsigned char s)
/* Encapsulate one can_frame and stuff into a TTY queue. */
static void slc_encaps(struct slcan *sl, struct can_frame *cf)
{
- int actual, idx, i;
- char cmd;
+ int actual, i;
+ unsigned char *pos;
+ unsigned char *endpos;
+ canid_t id = cf->can_id;
+
+ pos = sl->xbuff;
if (cf->can_id & CAN_RTR_FLAG)
- cmd = 'R'; /* becomes 'r' in standard frame format */
+ *pos = 'R'; /* becomes 'r' in standard frame format (SFF) */
else
- cmd = 'T'; /* becomes 't' in standard frame format */
+ *pos = 'T'; /* becomes 't' in standard frame format (SSF) */
- if (cf->can_id & CAN_EFF_FLAG)
- sprintf(sl->xbuff, "%c%08X%d", cmd,
- cf->can_id & CAN_EFF_MASK, cf->can_dlc);
- else
- sprintf(sl->xbuff, "%c%03X%d", cmd | 0x20,
- cf->can_id & CAN_SFF_MASK, cf->can_dlc);
+ /* determine number of chars for the CAN-identifier */
+ if (cf->can_id & CAN_EFF_FLAG) {
+ id &= CAN_EFF_MASK;
+ endpos = pos + SLC_EFF_ID_LEN;
+ } else {
+ *pos |= 0x20; /* convert R/T to lower case for SFF */
+ id &= CAN_SFF_MASK;
+ endpos = pos + SLC_SFF_ID_LEN;
+ }
+
+ /* build 3 (SFF) or 8 (EFF) digit CAN identifier */
+ pos++;
+ while (endpos >= pos) {
+ *endpos-- = hex_asc_upper[id & 0xf];
+ id >>= 4;
+ }
+
+ pos += (cf->can_id & CAN_EFF_FLAG) ? SLC_EFF_ID_LEN : SLC_SFF_ID_LEN;
- idx = strlen(sl->xbuff);
+ *pos++ = cf->can_dlc + '0';
- for (i = 0; i < cf->can_dlc; i++)
- sprintf(&sl->xbuff[idx + 2*i], "%02X", cf->data[i]);
+ /* RTR frames may have a dlc > 0 but they never have any data bytes */
+ if (!(cf->can_id & CAN_RTR_FLAG)) {
+ for (i = 0; i < cf->can_dlc; i++)
+ pos = hex_byte_pack_upper(pos, cf->data[i]);
+ }
- strcat(sl->xbuff, "\r"); /* add terminating character */
+ *pos++ = '\r';
/* Order of next two lines is *very* important.
* When we are sending a little amount of data,
@@ -267,8 +305,8 @@ static void slc_encaps(struct slcan *sl, struct can_frame *cf)
* 14 Oct 1994 Dmitry Gorodchanin.
*/
set_bit(TTY_DO_WRITE_WAKEUP, &sl->tty->flags);
- actual = sl->tty->ops->write(sl->tty, sl->xbuff, strlen(sl->xbuff));
- sl->xleft = strlen(sl->xbuff) - actual;
+ actual = sl->tty->ops->write(sl->tty, sl->xbuff, pos - sl->xbuff);
+ sl->xleft = (pos - sl->xbuff) - actual;
sl->xhead = sl->xbuff + actual;
sl->dev->stats.tx_bytes += cf->can_dlc;
}
--
1.8.4.rc3
^ permalink raw reply related
* [PATCH net 1/3] slip/slcan: added locking in wakeup function
From: Andre Naujoks @ 2013-09-13 17:37 UTC (permalink / raw)
To: davem, Wolfgang Grandegger, Marc Kleine-Budde, linux-can, netdev,
linux-kernel
In-Reply-To: <1379093833-4949-1-git-send-email-nautsch2@gmail.com>
The locking is needed, since the the internal buffer for the CAN frames is
changed during the wakeup call. This could cause buffer inconsistencies
under high loads, especially for the outgoing short CAN packet skbuffs.
The needed locks led to deadlocks before commit
"5ede52538ee2b2202d9dff5b06c33bfde421e6e4 tty: Remove extra wakeup from pty
write() path", which removed the direct callback to the wakeup function from the
tty layer.
As slcan.c is based on slip.c the issue in the original code is fixed, too.
Signed-off-by: Andre Naujoks <nautsch2@gmail.com>
---
drivers/net/can/slcan.c | 3 +++
drivers/net/slip/slip.c | 3 +++
2 files changed, 6 insertions(+)
diff --git a/drivers/net/can/slcan.c b/drivers/net/can/slcan.c
index 874188b..d571e2e 100644
--- a/drivers/net/can/slcan.c
+++ b/drivers/net/can/slcan.c
@@ -286,11 +286,13 @@ static void slcan_write_wakeup(struct tty_struct *tty)
if (!sl || sl->magic != SLCAN_MAGIC || !netif_running(sl->dev))
return;
+ spin_lock(&sl->lock);
if (sl->xleft <= 0) {
/* Now serial buffer is almost free & we can start
* transmission of another packet */
sl->dev->stats.tx_packets++;
clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
+ spin_unlock(&sl->lock);
netif_wake_queue(sl->dev);
return;
}
@@ -298,6 +300,7 @@ static void slcan_write_wakeup(struct tty_struct *tty)
actual = tty->ops->write(tty, sl->xhead, sl->xleft);
sl->xleft -= actual;
sl->xhead += actual;
+ spin_unlock(&sl->lock);
}
/* Send a can_frame to a TTY queue. */
diff --git a/drivers/net/slip/slip.c b/drivers/net/slip/slip.c
index a34d6bf..cc70ecf 100644
--- a/drivers/net/slip/slip.c
+++ b/drivers/net/slip/slip.c
@@ -429,11 +429,13 @@ static void slip_write_wakeup(struct tty_struct *tty)
if (!sl || sl->magic != SLIP_MAGIC || !netif_running(sl->dev))
return;
+ spin_lock(&sl->lock);
if (sl->xleft <= 0) {
/* Now serial buffer is almost free & we can start
* transmission of another packet */
sl->dev->stats.tx_packets++;
clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags);
+ spin_unlock(&sl->lock);
sl_unlock(sl);
return;
}
@@ -441,6 +443,7 @@ static void slip_write_wakeup(struct tty_struct *tty)
actual = tty->ops->write(tty, sl->xhead, sl->xleft);
sl->xleft -= actual;
sl->xhead += actual;
+ spin_unlock(&sl->lock);
}
static void sl_tx_timeout(struct net_device *dev)
--
1.8.4.rc3
^ permalink raw reply related
* [PATCH net 2/3] lib: introduce upper case hex ascii helpers
From: Andre Naujoks @ 2013-09-13 17:37 UTC (permalink / raw)
To: davem, Andrew Morton, Steven Rostedt, Rusty Russell,
Arnd Bergmann, Michael S. Tsirkin, Vladimir Kondratiev,
Jason Baron, Greg Kroah-Hartman, linux-kernel
Cc: linux-can, netdev
In-Reply-To: <1379093833-4949-1-git-send-email-nautsch2@gmail.com>
To be able to use the hex ascii functions in case sensitive environments
the array hex_asc_upper[] and the needed functions for hex_byte_pack_upper()
are introduced.
Signed-off-by: Andre Naujoks <nautsch2@gmail.com>
---
include/linux/kernel.h | 11 +++++++++++
lib/hexdump.c | 2 ++
2 files changed, 13 insertions(+)
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 482ad2d..672ddc4 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -439,6 +439,17 @@ static inline char *hex_byte_pack(char *buf, u8 byte)
return buf;
}
+extern const char hex_asc_upper[];
+#define hex_asc_upper_lo(x) hex_asc_upper[((x) & 0x0f)]
+#define hex_asc_upper_hi(x) hex_asc_upper[((x) & 0xf0) >> 4]
+
+static inline char *hex_byte_pack_upper(char *buf, u8 byte)
+{
+ *buf++ = hex_asc_upper_hi(byte);
+ *buf++ = hex_asc_upper_lo(byte);
+ return buf;
+}
+
static inline char * __deprecated pack_hex_byte(char *buf, u8 byte)
{
return hex_byte_pack(buf, byte);
diff --git a/lib/hexdump.c b/lib/hexdump.c
index 3f0494c..8499c81 100644
--- a/lib/hexdump.c
+++ b/lib/hexdump.c
@@ -14,6 +14,8 @@
const char hex_asc[] = "0123456789abcdef";
EXPORT_SYMBOL(hex_asc);
+const char hex_asc_upper[] = "0123456789ABCDEF";
+EXPORT_SYMBOL(hex_asc_upper);
/**
* hex_to_bin - convert a hex digit to its real value
--
1.8.4.rc3
^ permalink raw reply related
* [PATCH 1/1] isdn: hfcpci_softirq: get func return to suppress compiler warning
From: Antonio Alecrim Jr @ 2013-09-13 17:44 UTC (permalink / raw)
To: Karsten Keil, Masanari Iida, netdev, linux-kernel; +Cc: Antonio Alecrim Jr
Signed-off-by: Antonio Alecrim Jr <antonio.alecrim@gmail.com>
---
drivers/isdn/hardware/mISDN/hfcpci.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/isdn/hardware/mISDN/hfcpci.c b/drivers/isdn/hardware/mISDN/hfcpci.c
index 7f910c7..5b8df75b 100644
--- a/drivers/isdn/hardware/mISDN/hfcpci.c
+++ b/drivers/isdn/hardware/mISDN/hfcpci.c
@@ -2295,7 +2295,9 @@ _hfcpci_softirq(struct device *dev, void *arg)
static void
hfcpci_softirq(void *arg)
{
- (void) driver_for_each_device(&hfc_driver.driver, NULL, arg,
+ int ret;
+
+ ret = driver_for_each_device(&hfc_driver.driver, NULL, arg,
_hfcpci_softirq);
/* if next event would be in the past ... */
--
1.8.3.1
^ permalink raw reply related
* [PATCH] rtlwifi: rtl8192cu: Convert driver to use rtl_process_phyinfo()
From: Larry Finger @ 2013-09-13 17:44 UTC (permalink / raw)
To: linville-2XuSBdqkA4R54TAoqtyWWQ
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA, Larry Finger,
netdev-u79uwXL29TY76Z2rM5mHXA
Remove routine _rtl92c_process_phyinfo() by using the equivalent routine
in driver rtlwifi.
This change also allows the removal of 5 additional routines from rtl8192cu.
Signed-off-by: Larry Finger <Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
---
drivers/net/wireless/rtlwifi/rtl8192cu/mac.c | 187 +--------------------------
1 file changed, 2 insertions(+), 185 deletions(-)
diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/mac.c b/drivers/net/wireless/rtlwifi/rtl8192cu/mac.c
index da4f587..3936853 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192cu/mac.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192cu/mac.c
@@ -32,6 +32,7 @@
#include "../usb.h"
#include "../ps.h"
#include "../cam.h"
+#include "../stats.h"
#include "reg.h"
#include "def.h"
#include "phy.h"
@@ -738,16 +739,6 @@ static u8 _rtl92c_evm_db_to_percentage(char value)
return ret_val;
}
-static long _rtl92c_translate_todbm(struct ieee80211_hw *hw,
- u8 signal_strength_index)
-{
- long signal_power;
-
- signal_power = (long)((signal_strength_index + 1) >> 1);
- signal_power -= 95;
- return signal_power;
-}
-
static long _rtl92c_signal_scale_mapping(struct ieee80211_hw *hw,
long currsig)
{
@@ -913,180 +904,6 @@ static void _rtl92c_query_rxphystatus(struct ieee80211_hw *hw,
(hw, total_rssi /= rf_rx_num));
}
-static void _rtl92c_process_ui_rssi(struct ieee80211_hw *hw,
- struct rtl_stats *pstats)
-{
- struct rtl_priv *rtlpriv = rtl_priv(hw);
- struct rtl_phy *rtlphy = &(rtlpriv->phy);
- u8 rfpath;
- u32 last_rssi, tmpval;
-
- if (pstats->packet_toself || pstats->packet_beacon) {
- rtlpriv->stats.rssi_calculate_cnt++;
- if (rtlpriv->stats.ui_rssi.total_num++ >=
- PHY_RSSI_SLID_WIN_MAX) {
- rtlpriv->stats.ui_rssi.total_num =
- PHY_RSSI_SLID_WIN_MAX;
- last_rssi =
- rtlpriv->stats.ui_rssi.elements[rtlpriv->
- stats.ui_rssi.index];
- rtlpriv->stats.ui_rssi.total_val -= last_rssi;
- }
- rtlpriv->stats.ui_rssi.total_val += pstats->signalstrength;
- rtlpriv->stats.ui_rssi.elements[rtlpriv->stats.ui_rssi.
- index++] = pstats->signalstrength;
- if (rtlpriv->stats.ui_rssi.index >= PHY_RSSI_SLID_WIN_MAX)
- rtlpriv->stats.ui_rssi.index = 0;
- tmpval = rtlpriv->stats.ui_rssi.total_val /
- rtlpriv->stats.ui_rssi.total_num;
- rtlpriv->stats.signal_strength =
- _rtl92c_translate_todbm(hw, (u8) tmpval);
- pstats->rssi = rtlpriv->stats.signal_strength;
- }
- if (!pstats->is_cck && pstats->packet_toself) {
- for (rfpath = RF90_PATH_A; rfpath < rtlphy->num_total_rfpath;
- rfpath++) {
- if (!rtl8192_phy_check_is_legal_rfpath(hw, rfpath))
- continue;
- if (rtlpriv->stats.rx_rssi_percentage[rfpath] == 0) {
- rtlpriv->stats.rx_rssi_percentage[rfpath] =
- pstats->rx_mimo_signalstrength[rfpath];
- }
- if (pstats->rx_mimo_signalstrength[rfpath] >
- rtlpriv->stats.rx_rssi_percentage[rfpath]) {
- rtlpriv->stats.rx_rssi_percentage[rfpath] =
- ((rtlpriv->stats.
- rx_rssi_percentage[rfpath] *
- (RX_SMOOTH_FACTOR - 1)) +
- (pstats->rx_mimo_signalstrength[rfpath])) /
- (RX_SMOOTH_FACTOR);
-
- rtlpriv->stats.rx_rssi_percentage[rfpath] =
- rtlpriv->stats.rx_rssi_percentage[rfpath] +
- 1;
- } else {
- rtlpriv->stats.rx_rssi_percentage[rfpath] =
- ((rtlpriv->stats.
- rx_rssi_percentage[rfpath] *
- (RX_SMOOTH_FACTOR - 1)) +
- (pstats->rx_mimo_signalstrength[rfpath])) /
- (RX_SMOOTH_FACTOR);
- }
- }
- }
-}
-
-static void _rtl92c_update_rxsignalstatistics(struct ieee80211_hw *hw,
- struct rtl_stats *pstats)
-{
- struct rtl_priv *rtlpriv = rtl_priv(hw);
- int weighting = 0;
-
- if (rtlpriv->stats.recv_signal_power == 0)
- rtlpriv->stats.recv_signal_power = pstats->recvsignalpower;
- if (pstats->recvsignalpower > rtlpriv->stats.recv_signal_power)
- weighting = 5;
- else if (pstats->recvsignalpower < rtlpriv->stats.recv_signal_power)
- weighting = (-5);
- rtlpriv->stats.recv_signal_power =
- (rtlpriv->stats.recv_signal_power * 5 +
- pstats->recvsignalpower + weighting) / 6;
-}
-
-static void _rtl92c_process_pwdb(struct ieee80211_hw *hw,
- struct rtl_stats *pstats)
-{
- struct rtl_priv *rtlpriv = rtl_priv(hw);
- struct rtl_mac *mac = rtl_mac(rtl_priv(hw));
- long undec_sm_pwdb = 0;
-
- if (mac->opmode == NL80211_IFTYPE_ADHOC) {
- return;
- } else {
- undec_sm_pwdb = rtlpriv->dm.undec_sm_pwdb;
- }
- if (pstats->packet_toself || pstats->packet_beacon) {
- if (undec_sm_pwdb < 0)
- undec_sm_pwdb = pstats->rx_pwdb_all;
- if (pstats->rx_pwdb_all > (u32) undec_sm_pwdb) {
- undec_sm_pwdb = (((undec_sm_pwdb) *
- (RX_SMOOTH_FACTOR - 1)) +
- (pstats->rx_pwdb_all)) / (RX_SMOOTH_FACTOR);
- undec_sm_pwdb += 1;
- } else {
- undec_sm_pwdb = (((undec_sm_pwdb) *
- (RX_SMOOTH_FACTOR - 1)) +
- (pstats->rx_pwdb_all)) / (RX_SMOOTH_FACTOR);
- }
- rtlpriv->dm.undec_sm_pwdb = undec_sm_pwdb;
- _rtl92c_update_rxsignalstatistics(hw, pstats);
- }
-}
-
-static void _rtl92c_process_LINK_Q(struct ieee80211_hw *hw,
- struct rtl_stats *pstats)
-{
- struct rtl_priv *rtlpriv = rtl_priv(hw);
- u32 last_evm = 0, n_stream, tmpval;
-
- if (pstats->signalquality != 0) {
- if (pstats->packet_toself || pstats->packet_beacon) {
- if (rtlpriv->stats.LINK_Q.total_num++ >=
- PHY_LINKQUALITY_SLID_WIN_MAX) {
- rtlpriv->stats.LINK_Q.total_num =
- PHY_LINKQUALITY_SLID_WIN_MAX;
- last_evm =
- rtlpriv->stats.LINK_Q.elements
- [rtlpriv->stats.LINK_Q.index];
- rtlpriv->stats.LINK_Q.total_val -=
- last_evm;
- }
- rtlpriv->stats.LINK_Q.total_val +=
- pstats->signalquality;
- rtlpriv->stats.LINK_Q.elements
- [rtlpriv->stats.LINK_Q.index++] =
- pstats->signalquality;
- if (rtlpriv->stats.LINK_Q.index >=
- PHY_LINKQUALITY_SLID_WIN_MAX)
- rtlpriv->stats.LINK_Q.index = 0;
- tmpval = rtlpriv->stats.LINK_Q.total_val /
- rtlpriv->stats.LINK_Q.total_num;
- rtlpriv->stats.signal_quality = tmpval;
- rtlpriv->stats.last_sigstrength_inpercent = tmpval;
- for (n_stream = 0; n_stream < 2;
- n_stream++) {
- if (pstats->RX_SIGQ[n_stream] != -1) {
- if (!rtlpriv->stats.RX_EVM[n_stream]) {
- rtlpriv->stats.RX_EVM[n_stream]
- = pstats->RX_SIGQ[n_stream];
- }
- rtlpriv->stats.RX_EVM[n_stream] =
- ((rtlpriv->stats.RX_EVM
- [n_stream] *
- (RX_SMOOTH_FACTOR - 1)) +
- (pstats->RX_SIGQ
- [n_stream] * 1)) /
- (RX_SMOOTH_FACTOR);
- }
- }
- }
- } else {
- ;
- }
-}
-
-static void _rtl92c_process_phyinfo(struct ieee80211_hw *hw,
- u8 *buffer,
- struct rtl_stats *pcurrent_stats)
-{
- if (!pcurrent_stats->packet_matchbssid &&
- !pcurrent_stats->packet_beacon)
- return;
- _rtl92c_process_ui_rssi(hw, pcurrent_stats);
- _rtl92c_process_pwdb(hw, pcurrent_stats);
- _rtl92c_process_LINK_Q(hw, pcurrent_stats);
-}
^ permalink raw reply related
* [PATCH 0/7] rtlwifi: Patches to fix problems shown by smatch
From: Larry Finger @ 2013-09-13 17:44 UTC (permalink / raw)
To: linville-2XuSBdqkA4R54TAoqtyWWQ
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA, Larry Finger,
netdev-u79uwXL29TY76Z2rM5mHXA
*** BLURB HERE ***
Larry Finger (7):
rtlwifi: rtl8192du: Fix smatch errors in /rtl8192de/dm.c
rtlwifi: rtl8192de: Fix smatch warnings in rtl8192de/hw.c
rtlwifi: rtl8192cu: Fix smatch warning in rtl8192cu/trx.c
rtlwifi: rtl8192_common: Fix smatch errors and warnings in
rtl8192c/dm_common.c
rtlwifi: Fix smatch warning in pci.c
rtlwifi: Fix smatch warnings in usb.c
rtlwifi: rtl8188ee: Fix smatch warning in rtl8188ee/hw.c
drivers/net/wireless/rtlwifi/pci.c | 2 +-
drivers/net/wireless/rtlwifi/rtl8188ee/hw.c | 1 +
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c | 4 +-
drivers/net/wireless/rtlwifi/rtl8192cu/mac.c | 187 +---------------------
drivers/net/wireless/rtlwifi/rtl8192cu/trx.c | 2 +-
drivers/net/wireless/rtlwifi/rtl8192de/dm.c | 8 +-
drivers/net/wireless/rtlwifi/rtl8192de/hw.c | 2 +
drivers/net/wireless/rtlwifi/usb.c | 10 +-
8 files changed, 23 insertions(+), 193 deletions(-)
--
1.8.1.4
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* [PATCH 2/7] rtlwifi: rtl8192de: Fix smatch warnings in rtl8192de/hw.c
From: Larry Finger @ 2013-09-13 17:44 UTC (permalink / raw)
To: linville-2XuSBdqkA4R54TAoqtyWWQ
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA, Larry Finger,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
Smatch lists the following:
CHECK drivers/net/wireless/rtlwifi/rtl8192de/hw.c
drivers/net/wireless/rtlwifi/rtl8192de/hw.c:1200 rtl92de_set_qos() info: ignoring unreachable code.
drivers/net/wireless/rtlwifi/rtl8192de/hw.c:1200 rtl92de_set_qos() info: ignoring unreachable code.
Dead code is removed.
Signed-off-by: Larry Finger <Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
---
drivers/net/wireless/rtlwifi/rtl8192de/hw.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/wireless/rtlwifi/rtl8192de/hw.c b/drivers/net/wireless/rtlwifi/rtl8192de/hw.c
index 7dd8f6d..c9b0894 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192de/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192de/hw.c
@@ -1194,6 +1194,7 @@ void rtl92d_linked_set_reg(struct ieee80211_hw *hw)
* mac80211 will send pkt when scan */
void rtl92de_set_qos(struct ieee80211_hw *hw, int aci)
{
+/*
struct rtl_priv *rtlpriv = rtl_priv(hw);
rtl92d_dm_init_edca_turbo(hw);
return;
@@ -1213,6 +1214,7 @@ void rtl92de_set_qos(struct ieee80211_hw *hw, int aci)
RT_ASSERT(false, "invalid aci: %d !\n", aci);
break;
}
+ */
}
void rtl92de_enable_interrupt(struct ieee80211_hw *hw)
--
1.8.1.4
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 3/7] rtlwifi: rtl8192cu: Fix smatch warning in rtl8192cu/trx.c
From: Larry Finger @ 2013-09-13 17:45 UTC (permalink / raw)
To: linville-2XuSBdqkA4R54TAoqtyWWQ
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA, Larry Finger,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
Smatch lists the following:
CHECK drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
drivers/net/wireless/rtlwifi/rtl8192cu/trx.c:367 _rtl_rx_process() warn: assigning (-98) to unsigned variable 'stats.noise'
This warning is fixed by setting the value to 256-98.
Signed-off-by: Larry Finger <Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
---
drivers/net/wireless/rtlwifi/rtl8192cu/trx.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c b/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
index 763cf1d..18308b0 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192cu/trx.c
@@ -364,7 +364,7 @@ static void _rtl_rx_process(struct ieee80211_hw *hw, struct sk_buff *skb)
u8 *rxdesc;
struct rtl_stats stats = {
.signal = 0,
- .noise = -98,
+ .noise = 158, /* -98 dBm */
.rate = 0,
};
struct rx_fwinfo_92c *p_drvinfo;
--
1.8.1.4
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 6/7] rtlwifi: Fix smatch warnings in usb.c
From: Larry Finger @ 2013-09-13 17:45 UTC (permalink / raw)
To: linville-2XuSBdqkA4R54TAoqtyWWQ
Cc: linux-wireless-u79uwXL29TY76Z2rM5mHXA, Larry Finger,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
Smatch displays the following:
CHECK drivers/net/wireless/rtlwifi/usb.c
drivers/net/wireless/rtlwifi/usb.c:458 _rtl_usb_rx_process_agg() warn: assigning (-98) to unsigned variable 'stats.noise'
drivers/net/wireless/rtlwifi/usb.c:503 _rtl_usb_rx_process_noagg() warn: assigning (-98) to unsigned variable 'stats.noise'
drivers/net/wireless/rtlwifi/usb.c:596 _rtl_rx_get_padding() info: ignoring unreachable code.
drivers/net/wireless/rtlwifi/usb.c:596 _rtl_rx_get_padding() info: ignoring unreachable code.
The negative number to an unsigned quantity is fixed by adding 256 to -98
to get the equivalent negative number.
Signed-off-by: Larry Finger <Larry.Finger-tQ5ms3gMjBLk1uMJSBkQmQ@public.gmane.org>
---
drivers/net/wireless/rtlwifi/usb.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/drivers/net/wireless/rtlwifi/usb.c b/drivers/net/wireless/rtlwifi/usb.c
index e56778c..9f3dcb8 100644
--- a/drivers/net/wireless/rtlwifi/usb.c
+++ b/drivers/net/wireless/rtlwifi/usb.c
@@ -455,7 +455,7 @@ static void _rtl_usb_rx_process_agg(struct ieee80211_hw *hw,
struct ieee80211_rx_status rx_status = {0};
struct rtl_stats stats = {
.signal = 0,
- .noise = -98,
+ .noise = 158, /* -98 dBm */
.rate = 0,
};
@@ -498,7 +498,7 @@ static void _rtl_usb_rx_process_noagg(struct ieee80211_hw *hw,
struct ieee80211_rx_status rx_status = {0};
struct rtl_stats stats = {
.signal = 0,
- .noise = -98,
+ .noise = 158, /* -98 dBm */
.rate = 0,
};
@@ -582,12 +582,15 @@ static void _rtl_rx_work(unsigned long param)
static unsigned int _rtl_rx_get_padding(struct ieee80211_hdr *hdr,
unsigned int len)
{
+#if NET_IP_ALIGN != 0
unsigned int padding = 0;
+#endif
/* make function no-op when possible */
- if (NET_IP_ALIGN == 0 || len < sizeof(*hdr))
+ if (NET_IP_ALIGN == 0 || len < sizeof(struct ieee80211_hdr))
return 0;
+#if NET_IP_ALIGN != 0
/* alignment calculation as in lbtf_rx() / carl9170_rx_copy_data() */
/* TODO: deduplicate common code, define helper function instead? */
@@ -608,6 +611,7 @@ static unsigned int _rtl_rx_get_padding(struct ieee80211_hdr *hdr,
padding ^= NET_IP_ALIGN;
return padding;
+#endif
}
#define __RADIO_TAP_SIZE_RSV 32
--
1.8.1.4
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply related
* [PATCH 1/7] rtlwifi: rtl8192du: Fix smatch errors in /rtl8192de/dm.c
From: Larry Finger @ 2013-09-13 17:44 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, Larry Finger, netdev
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger@lwfinger.net>
Smatch lists the following:
CHECK drivers/net/wireless/rtlwifi/rtl8192de/dm.c
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1054 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'ofdm_index' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1056 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'ofdm_index' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1126 rtl92d_dm_txpower_tracking_callback_thermalmeter() debug: remove_pools: nr_children over 4000 (4596). (rtlpriv->dbg.global_debuglevel merged)
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1126 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1129 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1132 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1135 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1138 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1141 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1144 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1147 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch1ch13' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1151 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1154 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1157 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1160 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1163 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1166 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1169 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
drivers/net/wireless/rtlwifi/rtl8192de/dm.c:1172 rtl92d_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'cckswing_table_ch14' 33 <= 255
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
---
drivers/net/wireless/rtlwifi/rtl8192de/dm.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/drivers/net/wireless/rtlwifi/rtl8192de/dm.c b/drivers/net/wireless/rtlwifi/rtl8192de/dm.c
index 47875ba..eaeee77 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192de/dm.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192de/dm.c
@@ -840,9 +840,9 @@ static void rtl92d_dm_txpower_tracking_callback_thermalmeter(
bool internal_pa = false;
long ele_a = 0, ele_d, temp_cck, val_x, value32;
long val_y, ele_c = 0;
- u8 ofdm_index[2];
+ u8 ofdm_index[3];
s8 cck_index = 0;
- u8 ofdm_index_old[2] = {0, 0};
+ u8 ofdm_index_old[3] = {0, 0, 0};
s8 cck_index_old = 0;
u8 index;
int i;
@@ -1118,6 +1118,10 @@ static void rtl92d_dm_txpower_tracking_callback_thermalmeter(
val_x, val_y, ele_a, ele_c, ele_d,
val_x, val_y);
+ if (cck_index >= CCK_TABLE_SIZE)
+ cck_index = CCK_TABLE_SIZE - 1;
+ if (cck_index < 0)
+ cck_index = 0;
if (rtlhal->current_bandtype == BAND_ON_2_4G) {
/* Adjust CCK according to IQK result */
if (!rtlpriv->dm.cck_inch14) {
--
1.8.1.4
^ permalink raw reply related
* [PATCH 4/7] rtlwifi: rtl8192_common: Fix smatch errors and warnings in rtl8192c/dm_common.c
From: Larry Finger @ 2013-09-13 17:45 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, Larry Finger, netdev
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger@lwfinger.net>
Smatch lists the following:
CHECK drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:551 rtl92c_dm_pwdb_monitor() info: ignoring unreachable code.
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:551 rtl92c_dm_pwdb_monitor() info: ignoring unreachable code.
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:870 rtl92c_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'txpwr_level' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:870 rtl92c_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'txpwr_level' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:882 rtl92c_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'txpwr_level' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:883 rtl92c_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'txpwr_level' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:891 rtl92c_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'txpwr_level' 2 <= 2
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c:892 rtl92c_dm_txpower_tracking_callback_thermalmeter() error: buffer overflow 'txpwr_level' 2 <= 2
The unreachable code message is fixed by commenting out the code that follows a return.
The errors are fixed by increasing the size of txpwr_level.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
---
drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c b/drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c
index d2d57a2..0721756 100644
--- a/drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c
+++ b/drivers/net/wireless/rtlwifi/rtl8192c/dm_common.c
@@ -541,6 +541,7 @@ EXPORT_SYMBOL(rtl92c_dm_write_dig);
static void rtl92c_dm_pwdb_monitor(struct ieee80211_hw *hw)
{
+/*
struct rtl_priv *rtlpriv = rtl_priv(hw);
long tmpentry_max_pwdb = 0, tmpentry_min_pwdb = 0xff;
@@ -564,6 +565,7 @@ static void rtl92c_dm_pwdb_monitor(struct ieee80211_hw *hw)
h2c_parameter[0] = 0;
rtl92c_fill_h2c_cmd(hw, H2C_RSSI_REPORT, 3, h2c_parameter);
+ */
}
void rtl92c_dm_init_edca_turbo(struct ieee80211_hw *hw)
@@ -673,7 +675,7 @@ static void rtl92c_dm_txpower_tracking_callback_thermalmeter(struct ieee80211_hw
s8 cck_index = 0;
int i;
bool is2t = IS_92C_SERIAL(rtlhal->version);
- s8 txpwr_level[2] = {0, 0};
+ s8 txpwr_level[3] = {0, 0, 0};
u8 ofdm_min_index = 6, rf;
rtlpriv->dm.txpower_trackinginit = true;
--
1.8.1.4
^ permalink raw reply related
* [PATCH 5/7: rtlwifi: Fix smatch warning in pci.c
From: Larry Finger @ 2013-09-13 17:45 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, Larry Finger, netdev
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger@lwfinger.net>
Smatch reports the following:
CHECK drivers/net/wireless/rtlwifi/pci.c
drivers/net/wireless/rtlwifi/pci.c:739 _rtl_pci_rx_interrupt() warn: assigning (-98) to unsigned variable 'stats.noise'
This problem is fixed by changing the value to 256 - 98.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
---
drivers/net/wireless/rtlwifi/pci.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/wireless/rtlwifi/pci.c b/drivers/net/wireless/rtlwifi/pci.c
index 703f839..bf498f5 100644
--- a/drivers/net/wireless/rtlwifi/pci.c
+++ b/drivers/net/wireless/rtlwifi/pci.c
@@ -736,7 +736,7 @@ static void _rtl_pci_rx_interrupt(struct ieee80211_hw *hw)
struct rtl_stats stats = {
.signal = 0,
- .noise = -98,
+ .noise = 158, /* -98 dBm */
.rate = 0,
};
int index = rtlpci->rx_ring[rx_queue_idx].idx;
--
1.8.1.4
^ permalink raw reply related
* [PATCH 7/7] rtlwifi: rtl8188ee: Fix smatch warning in rtl8188ee/hw.c
From: Larry Finger @ 2013-09-13 17:45 UTC (permalink / raw)
To: linville; +Cc: linux-wireless, Larry Finger, netdev, Stable
In-Reply-To: <1379094304-22041-1-git-send-email-Larry.Finger@lwfinger.net>
Smatch lists the following:
CHECK drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
drivers/net/wireless/rtlwifi/rtl8188ee/hw.c:149 _rtl88ee_set_fw_clock_on() info: ignoring unreachable code.
drivers/net/wireless/rtlwifi/rtl8188ee/hw.c:149 _rtl88ee_set_fw_clock_on() info: ignoring unreachable code.
This info message is the result of a real error due to a missing break statement
in a "while (1)" loop.
Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Stable <stable@vger.kernel.org> [3.10+]
---
drivers/net/wireless/rtlwifi/rtl8188ee/hw.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c b/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
index b68cae3..e06971b 100644
--- a/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
+++ b/drivers/net/wireless/rtlwifi/rtl8188ee/hw.c
@@ -143,6 +143,7 @@ static void _rtl88ee_set_fw_clock_on(struct ieee80211_hw *hw,
} else {
rtlhal->fw_clk_change_in_progress = false;
spin_unlock_bh(&rtlpriv->locks.fw_ps_lock);
+ break;
}
}
--
1.8.1.4
^ permalink raw reply related
* Re: [PATCH 1/1] net: race condition when removing virtual net_device
From: Francesco Ruggeri @ 2013-09-13 17:54 UTC (permalink / raw)
To: Eric W. Biederman
Cc: David S. Miller, Eric Dumazet, Jiri Pirko, Alexander Duyck,
Cong Wang, netdev
In-Reply-To: <871u4t1d9t.fsf@xmission.com>
On Thu, Sep 12, 2013 at 10:50 PM, Eric W. Biederman
<ebiederm@xmission.com> wrote:
> Francesco Ruggeri <fruggeri@aristanetworks.com> writes:
>
>> That would be great. There would still be one scenario to take care of though:
>>
>> - veth interfaces v0 and v1 are in namespaces ns0 and ns1.
>> - process p0 unregisters v0, which also causes v1 to be unregistered.
>> When p0 enters netdev_run_todo both v0 and v1 are in net_todo_list and
>> have been unlisted from their namespaces.
>> - then in p0's netdev_run_todo:
>
> So I looked at this a little more and this problem appears largely
> specific to veth. In the normal case the caller of dellink has to hold
> a reference to the network namespace to find the device to delete.
>
> So I think the solution is just to warp the interface of the second
> device into the network namespace of the device we are actually
> deleting.
>
> I will buy that similar situations can happen with other virtual devices
> that have one foot in two network namespaces, and I expect the same
> solution will apply.
>
> So the patch below looks like the solution. If there is more than one
> device that needs this treatment perhaps the code should be moved
> into a helper function rather than expanded inline.
>
> Does this look like it will fix your issue?
>
> Eric
>
To summarize your proposal:
1) Remove re-broadcasting of NETDEV_UNREGISTER and
NETDEV_UNREGISTER_FINAL from netdev_wait_allrefs.
2) If unregistering a net_device causes unregistering of other virtual
devices (eg veth, macvlan, vlan) then move the virtual devices to the
namespace of the original net_device.
I do have some concerns about both correctness and feasibility of this approach.
About 2), namespace dependent operations triggered by unregistering
the virtual devices (eg rt_flush_dev, dst_dev_event/dst_ifdown and
probably more) would not be done in the namespaces where they should.
About the feasibility of 1), consider just as an example dst_dev_event
in NETDEV_UNREGISTER_FINAL. dst_dev_event will not process dst->dev
unless the dst_entry is already in dst_garbage.list, ie unless it has
already been dst_free'd or dst_release'd. But since destroying
resources is often delayed to workqueues or to one of several garbage
collection lists, can we guarantee that one broadcast of
NETDEV_UNREGISTER_FINAL is always enough? There may be similar cases
with NETDEV_UNREGISTER.
I do urge you to take a second look at the approach that I proposed.
All it does is make sure that a namespace loopback_dev is not removed
until any devices unlisted from that namespace are also disposed of.
That in turn prevents non-device pernet subsystems from exiting that
namespace in ops_exit_list.
Logically it is enforcing the right behavior (namely non-device pernet
subsystems should not exit a namespace until all devices in that
namespace - listed or unlisted - have been properly disposed of) and
it correctly supports existing code, such as rt_flush_dev and
dst_ifdown, which relies on the correct loopback_dev to be there.
Francesco
^ permalink raw reply
* Re: ICMP rate limiting in IPv4 but not in IPv6
From: Hannes Frederic Sowa @ 2013-09-13 18:17 UTC (permalink / raw)
To: Loganaden Velvindron; +Cc: Andy Johnson, netdev
In-Reply-To: <CAOp4FwQ6YWpMtZ9-64tVDc29RhtiUjwinQjZ32mjEu_pZfRHHA@mail.gmail.com>
On Fri, Sep 13, 2013 at 09:26:11PM +0400, Loganaden Velvindron wrote:
> On Fri, Sep 13, 2013 at 8:44 PM, Hannes Frederic Sowa
> <hannes@stressinduktion.org> wrote:
> > On Fri, Sep 13, 2013 at 06:57:56PM +0300, Andy Johnson wrote:
> >> Hello,
> >> After probing into the RFC of ICMPv6, I am even more confused.
> >>
> >> RFC 4443 says:
> >>
> >> 2.4. Message Processing Rules
> >> ...
> >>
> >> (f) Finally, in order to limit the bandwidth and forwarding costs
> >> incurred by originating ICMPv6 error messages, an IPv6 node MUST
> >> limit the rate of ICMPv6 error messages it originates.
> >> ...
> >> The rate-limiting parameters SHOULD be configurable.
> >> ...
> >>
> >> Any ideas?
> >
> > Well, nobody has implemented it because nobody missed the feature yet. I
> > don't believe there is another reason for that. If you come up with a
> > patch, I am sure it can go upstream.
> >
> > Do you want to try to come up with a patch? We need to be a bit
> > careful regarding neighbor discovery but otherwise this should be
> > relativ straightforward. In the meantime you could also implement such
> > ratelimiting with netfilter.
>
> Careful ? Can you please elaborate ?
>From my memory I knew we have rate limiting for redirect messages,
so I assuemed the rate limiter would be on the ndisc_send* path. But
ratelimiting ndisc redirects is a special case and only happens in
ndisc_send_redirect. I assuemd we would need to special case
icmpv6_xrlim_allow to always pass ndisc ns/na packets, sorry.
Maybe we could also switch to icmpv6_xrlim_allow for redirects to make
this configurable?
Btw. for sending redirects we seem to call inet_peer_xrlim_allow once
in the redirect function but also check it in ip6_forward. Maybe we can
drop the check in ip6_forward?
Greetings,
Hannes
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox