* Re: [PATCH net-next 4/5] net: dsa: add slave get master helper
From: Florian Fainelli @ 2017-10-12 23:02 UTC (permalink / raw)
To: Vivien Didelot, netdev; +Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <20171012225156.20758-5-vivien.didelot@savoirfairelinux.com>
On 10/12/2017 03:51 PM, Vivien Didelot wrote:
> Many part of the DSA slave code require to get the master device
> assigned to a slave device. Remove dsa_master_netdev() in favor of a
> dsa_slave_get_master() helper which does that.
>
> Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
> ---
> +static inline struct net_device *
> +dsa_slave_get_master(const struct net_device *dev)
> +{
> + struct dsa_port *dp = dsa_slave_to_port(dev);
> +
> + return dp->cpu_dp->netdev;
> +}
Nit: _get may convey the idea that a reference count may be incremented
when the function is called (balanced with a _put), so maybe name it
dsa_slave_to_master() which is more in line with dsa_slave_to_port() as
well? Other than that:
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
--
Florian
^ permalink raw reply
* Re: [PATCH net-next 5/5] net: dsa: split dsa_port's netdev member
From: Florian Fainelli @ 2017-10-12 23:05 UTC (permalink / raw)
To: Vivien Didelot, netdev; +Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <20171012225156.20758-6-vivien.didelot@savoirfairelinux.com>
On 10/12/2017 03:51 PM, Vivien Didelot wrote:
> The dsa_port structure has a "netdev" member, which can be used for
> either the master device, or the slave device, depending on its type.
>
> It is true that today, CPU port are not exposed to userspace, thus the
> port's netdev member can be used to point to its master interface.
>
> But it is still slightly confusing, so split it into more explicit
> "master" and "slave" members.
I do see some value in doing that, although I also see value in having
structure members be named after what they are, rather than their use
(oh well, it's all debatable anyway), see below for a suggestion on how
to reconcile the two:
> struct dsa_port {
> + /* Master device, physically connected if this is a CPU port */
> + struct net_device *master;
> +
> + /* Slave device, if this port is exposed to userspace */
> + struct net_device *slave;
> +
How about using:
union {
struct net_device *master;
struct net_device *slave;
} netdev;
Such that this serves both purposes of clearly communicating what the
structure member is, and it can be either one of the two, but not both
at the same time?
--
Florian
^ permalink raw reply
* Re: [RFC] Support for UNARP (RFC 1868)
From: Girish Moodalbail @ 2017-10-12 23:06 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev, davem, kuznet
In-Reply-To: <1507820236.31614.50.camel@edumazet-glaptop3.roam.corp.google.com>
Hello Eric,
The basic idea is to mark the ARP entry either FAILED or STALE as soon as we can
so that the subsequent packets that depend on that ARP entry will take the slow
path (neigh_resolve_output()).
Say, if base_reachable_time is 30 seconds, then an ARP entry will be in
reachable state somewhere between 15 to 45 seconds. Assuming the worst case, the
ARP entry will be in REACHABLE state for 45 seconds and the packets continue to
traverse the network towards the source machine and gets dropped their since the
VM has moved to destination machine.
Instead, based on the received UNARP packet if we mark the ARP entry
(a) FAILED
- we move to INCOMPLETE state and start the address resolution by sending
out ARP packets (up to allowed maximum number) until we get ARP response
back at which point we move the ARP entry state to reachable.
(b) STALE
- we move to DELAY state and set the next timer to DELAY_PROBE_TIME
(1 second) and continue to send queued packets in arp_queue.
- After 1 sec we move to PROBE state and start the address resolution like
in the case(a) above.
I was leaning towards (a). Please see in-line..
<snip>
>
> Hi Girish
>
> Your description (or patch title) is misleading. You apparently
> implement the receive side of the RFC.
You are right, it implements only the receive side of the RFC. If this RFC is
accepted, then we can change arping(8) to generate UNARP requests. We could also
add an option to ip-address(8) delete subcommand to generate UNARP whenever an
address is deleted from the interface.
> And the RFC had Proxy ARP in mind.
>
> What about security implications ?
Yes, this feature will extend the attack surface for L2 networks. However, the
attack vectors for this feature should be same as that of the gratuitous ARP,
right? The same attack mitigation techniques for gratuitous ARPs is equally
applicable here.
> Will TCP flows be terminated, instead
> of being smoothly migrated (TCP_REPAIR)
The TCP flows will not be terminated. Upon receiving UNARP packet, the ARP entry
will be marked FAILED. The subsequent TCP packets from the client (towards that
IP) will be queued (the first 3 packets in arp_queue and then other packets get
dropped) until the IP address is resolved again through the slow path
neigh_resolve_output().
The slow path marks the entry as INCOMPLETE and will start sending several ARP
requests (ucast_solicit + app_solicit + mcast_solicit) to resolve the IP. If the
resolution is successful, then the TCP packets will be sent out. If not, we will
invalidate the ARP entry and call arp_error_report() on the queued packets
(which will end up sending ICMP_HOST_UNREACH error). This behavior is same as
what will occur if TCP server disappears in the middle of a connection.
>
> What about IPv6 ? Or maybe more abruptly, do we still need to add
> features to IPv4 in 2017, 22 years after this RFC came ? ;)
Legit question :). Well one thing I have seen in Networking is that an old idea
circles back around later and turns out to be useful in new contexts and use
cases. Like I enumerated in my initial email there are certain use cases in
Cloud that might benefit from UNARP.
regards,
~Girish
>
> Thanks.
>
>
^ permalink raw reply
* Re: [PATCH] Add -target to clang switches while cross compiling.
From: Alexei Starovoitov @ 2017-10-12 23:16 UTC (permalink / raw)
To: Abhijit Ayarekar; +Cc: Abhijit Ayarekar, ast, daniel, netdev, linux-kernel, yhs
In-Reply-To: <20171012224322.GA7007@abhijit-ayarekar.cavium.com>
On Thu, Oct 12, 2017 at 03:43:22PM -0700, Abhijit Ayarekar wrote:
> On Thu, Oct 12, 2017 at 03:23:04PM -0700, Alexei Starovoitov wrote:
> > On Thu, Oct 12, 2017 at 01:45:57PM -0700, Abhijit Ayarekar wrote:
> > > Latest llvm update excludes assembly instructions.
> > > As a result __ASM_SYSREGS_H define is not required.
> > > -target switch includes appropriate target specific files.
> > >
> > > Tested on x86 and arm64 with llvm with git revision
> > > commit df6ca162269f9d756f8742bf4b658dcf690e3eb5
> > > Author: Yonghong Song <yhs@fb.com>
> > > Date: Thu Sep 28 02:46:11 2017 +0000
> > >
> > > bpf: add new insns for bswap_to_le and negation
> > >
> > > Signed-off-by: Abhijit Ayarekar <abhijit.ayarekar@caviumnetworks.com>
> > > ---
> > > samples/bpf/Makefile | 5 +++--
> > > 1 file changed, 3 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
> > > index ebc2ad6..81f9fcd 100644
> > > --- a/samples/bpf/Makefile
> > > +++ b/samples/bpf/Makefile
> > > @@ -180,6 +180,7 @@ CLANG ?= clang
> > > # Detect that we're cross compiling and use the cross compiler
> > > ifdef CROSS_COMPILE
> > > HOSTCC = $(CROSS_COMPILE)gcc
> > > +CLANG_ARCH_ARGS = -target $(ARCH)
> >
> > this is only need because you're crosscompiling, right?
> Yes
>
> > In native compilation it's unnecessary flag.
> > Only droping -D__ASM_SYSREG_H is enough, correct?
> >
> Yes. That is correct.
please update commit log then and reference proper llvm commit
that added asm support instead of 'add new insns for bswap_to_le'
which is unrelated.
^ permalink raw reply
* Re: [Patch net-next v2] tcp: add a tracepoint for tcp_retransmit_skb()
From: Alexei Starovoitov @ 2017-10-12 23:18 UTC (permalink / raw)
To: Cong Wang
Cc: netdev, Eric Dumazet, Hannes Frederic Sowa, Brendan Gregg,
Neal Cardwell
In-Reply-To: <20171012224807.1669-1-xiyou.wangcong@gmail.com>
On Thu, Oct 12, 2017 at 03:48:07PM -0700, Cong Wang wrote:
> We need a real-time notification for tcp retransmission
> for monitoring.
>
> Of course we could use ftrace to dynamically instrument this
> kernel function too, however we can't retrieve the connection
> information at the same time, for example perf-tools [1] reads
> /proc/net/tcp for socket details, which is slow when we have
> a lots of connections.
>
> Therefore, this patch adds a tracepoint for tcp_retransmit_skb()
> and exposes src/dst IP addresses and ports of the connection.
> This also makes it easier to integrate into perf.
>
> Note, I expose both IPv4 and IPv6 addresses at the same time:
> for a IPv4 socket, v4 mapped address is used as IPv6 addresses,
> for a IPv6 socket, LOOPBACK4_IPV6 is already filled by kernel.
> Also, add sk and skb pointers as they are useful for BPF.
>
> 1. https://github.com/brendangregg/perf-tools/blob/master/net/tcpretrans
>
> Cc: Eric Dumazet <edumazet@google.com>
> Cc: Alexei Starovoitov <alexei.starovoitov@gmail.com>
> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org>
> Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
> Cc: Neal Cardwell <ncardwell@google.com>
> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
> ---
> include/trace/events/tcp.h | 68 ++++++++++++++++++++++++++++++++++++++++++++++
> net/core/net-traces.c | 1 +
> net/ipv4/tcp_output.c | 3 ++
> 3 files changed, 72 insertions(+)
> create mode 100644 include/trace/events/tcp.h
>
> diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h
> new file mode 100644
> index 000000000000..749f93c542ab
> --- /dev/null
> +++ b/include/trace/events/tcp.h
> @@ -0,0 +1,68 @@
> +#undef TRACE_SYSTEM
> +#define TRACE_SYSTEM tcp
> +
> +#if !defined(_TRACE_TCP_H) || defined(TRACE_HEADER_MULTI_READ)
> +#define _TRACE_TCP_H
> +
> +#include <linux/ipv6.h>
> +#include <linux/tcp.h>
> +#include <linux/tracepoint.h>
> +#include <net/ipv6.h>
> +
> +TRACE_EVENT(tcp_retransmit_skb,
> +
> + TP_PROTO(struct sock *sk, struct sk_buff *skb, int segs),
> +
> + TP_ARGS(sk, skb, segs),
> +
> + TP_STRUCT__entry(
> + __field(void *, skbaddr)
> + __field(void *, skaddr)
> + __field(__u16, sport)
> + __field(__u16, dport)
> + __array(__u8, saddr, 4)
> + __array(__u8, daddr, 4)
> + __array(__u8, saddr_v6, 16)
> + __array(__u8, daddr_v6, 16)
> + ),
...
> if (likely(!err)) {
> TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS;
> + trace_tcp_retransmit_skb(sk, skb, segs);
looks great to me, but why 'segs' is there?
It's unused.
^ permalink raw reply
* Re: [PATCH net-next 5/5] net: dsa: split dsa_port's netdev member
From: Vivien Didelot @ 2017-10-12 23:21 UTC (permalink / raw)
To: Florian Fainelli, netdev
Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <5b637ea3-6c4b-1d93-ed43-09038af473f9@gmail.com>
Hi Florian,
Florian Fainelli <f.fainelli@gmail.com> writes:
> On 10/12/2017 03:51 PM, Vivien Didelot wrote:
>> The dsa_port structure has a "netdev" member, which can be used for
>> either the master device, or the slave device, depending on its type.
>>
>> It is true that today, CPU port are not exposed to userspace, thus the
>> port's netdev member can be used to point to its master interface.
>>
>> But it is still slightly confusing, so split it into more explicit
>> "master" and "slave" members.
>
> I do see some value in doing that, although I also see value in having
> structure members be named after what they are, rather than their use
> (oh well, it's all debatable anyway), see below for a suggestion on how
> to reconcile the two:
>
>> struct dsa_port {
>> + /* Master device, physically connected if this is a CPU port */
>> + struct net_device *master;
>> +
>> + /* Slave device, if this port is exposed to userspace */
>> + struct net_device *slave;
>> +
>
> How about using:
>
> union {
> struct net_device *master;
> struct net_device *slave;
> } netdev;
>
> Such that this serves both purposes of clearly communicating what the
> structure member is, and it can be either one of the two, but not both
> at the same time?
I love that! It makes clear that master is not available for a non-CPU
port. Using this union is correct for the moment because DSA and CPU
ports don't have a slave device attached to them. If this becomes true
one day (unlikely), we'll remove the union.
Thanks,
Vivien
^ permalink raw reply
* Re: [PATCH net-next 4/5] net: dsa: add slave get master helper
From: Vivien Didelot @ 2017-10-12 23:23 UTC (permalink / raw)
To: Florian Fainelli, netdev
Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <e3e7001e-8618-0b31-cd28-ebcdad561f7b@gmail.com>
Hi Florian,
Florian Fainelli <f.fainelli@gmail.com> writes:
> On 10/12/2017 03:51 PM, Vivien Didelot wrote:
>> Many part of the DSA slave code require to get the master device
>> assigned to a slave device. Remove dsa_master_netdev() in favor of a
>> dsa_slave_get_master() helper which does that.
>>
>> Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
>> ---
>
>> +static inline struct net_device *
>> +dsa_slave_get_master(const struct net_device *dev)
>> +{
>> + struct dsa_port *dp = dsa_slave_to_port(dev);
>> +
>> + return dp->cpu_dp->netdev;
>> +}
>
> Nit: _get may convey the idea that a reference count may be incremented
> when the function is called (balanced with a _put), so maybe name it
> dsa_slave_to_master() which is more in line with dsa_slave_to_port() as
> well? Other than that:
>
> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Perfect, I was hesitating myself between the two, but I didn't have a
good argument for any of them. Now I have one! I'll change that for
dsa_slave_to_master and add your tag.
Thanks,
Vivien
^ permalink raw reply
* [PATCH] wanxl: use m68k-linux-gnu-as if available
From: Adam Borowski @ 2017-10-12 23:24 UTC (permalink / raw)
To: Krzysztof Halasa, netdev; +Cc: Adam Borowski
This fixes build failure on Debian based systems: GNU as is the only m68k
assembler available in the archive (package binutils-m68k-linux-gnu).
Signed-off-by: Adam Borowski <kilobyte@angband.pl>
---
I have no relevant hardware, thus I can't check whether the built firmware
actually works. Some opcodes are translated differently, thus it might be
possible that some extra options are required. Or possibly the assembler
has long since changed -- the prebuilt copy hasn't been updated anywhere
within recorded history.
In any case, I admit I don't really care about this driver -- it's merely
a notorious cause of randconfig failures.
drivers/net/wan/Makefile | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/wan/Makefile b/drivers/net/wan/Makefile
index 73c2326603fc..fa17b18765b0 100644
--- a/drivers/net/wan/Makefile
+++ b/drivers/net/wan/Makefile
@@ -38,14 +38,20 @@ obj-$(CONFIG_SLIC_DS26522) += slic_ds26522.o
clean-files := wanxlfw.inc
$(obj)/wanxl.o: $(obj)/wanxlfw.inc
+HAVE_GNU_M68K_AS := $(shell command -v m68k-linux-gnu-as 2> /dev/null)
ifeq ($(CONFIG_WANXL_BUILD_FIRMWARE),y)
ifeq ($(ARCH),m68k)
AS68K = $(AS)
LD68K = $(LD)
+else
+ifdef HAVE_GNU_M68K_AS
+ AS68K = m68k-linux-gnu-as
+ LD68K = m68k-linux-gnu-ld
else
AS68K = as68k
LD68K = ld68k
endif
+endif
quiet_cmd_build_wanxlfw = BLD FW $@
cmd_build_wanxlfw = \
--
2.15.0.rc0
^ permalink raw reply related
* [PATCH] Add -target to clang switch while cross compiling.
From: Abhijit Ayarekar @ 2017-10-12 23:45 UTC (permalink / raw)
To: ast; +Cc: daniel, netdev, linux-kernel, Abhijit Ayarekar
In-Reply-To: <1507841157-10487-1-git-send-email-abhijit.ayarekar@caviumnetworks.com>
Update to llvm excludes assembly instructions.
llvm git revision is below
commit 2865ab6996164e7854d55c9e21c065fad7c26569
Author: Yonghong Song <yhs@fb.com>
Date: Mon Sep 18 23:29:36 2017 +0000
bpf: add inline-asm support
Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@313593 91177308-0d34-0410-b5e6-96231b3b80d8
__ASM_SYSREG_H define is not required for native compile.
-target switch includes appropriate target specific files
while cross compiling
Tested on x86 and arm64.
Signed-off-by: Abhijit Ayarekar <abhijit.ayarekar@caviumnetworks.com>
---
samples/bpf/Makefile | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index ebc2ad6..81f9fcd 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -180,6 +180,7 @@ CLANG ?= clang
# Detect that we're cross compiling and use the cross compiler
ifdef CROSS_COMPILE
HOSTCC = $(CROSS_COMPILE)gcc
+CLANG_ARCH_ARGS = -target $(ARCH)
endif
# Trick to allow make to be run from this directory
@@ -229,9 +230,9 @@ $(obj)/tracex5_kern.o: $(obj)/syscall_nrs.h
$(obj)/%.o: $(src)/%.c
$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) -I$(obj) \
-I$(srctree)/tools/testing/selftests/bpf/ \
- -D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
+ -D__KERNEL__ -Wno-unused-value -Wno-pointer-sign \
-D__TARGET_ARCH_$(ARCH) -Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
- -Wno-unknown-warning-option \
+ -Wno-unknown-warning-option $(CLANG_ARCH_ARGS) \
-O2 -emit-llvm -c $< -o -| $(LLC) -march=bpf -filetype=obj -o $@
--
2.7.4
^ permalink raw reply related
* [ANNOUNCE] nftables 0.8 release
From: Pablo Neira Ayuso @ 2017-10-12 23:49 UTC (permalink / raw)
To: netfilter-devel; +Cc: netdev, netfilter, netfilter-announce, lwn
[-- Attachment #1: Type: text/plain, Size: 9719 bytes --]
Hi!
The Netfilter project proudly presents:
nftables 0.8
This release contains new features available up to the (upcoming)
Linux 4.14 kernel release:
* Support for stateful objects, these objects are uniquely identified by
a user-defined name, you can refer to them from rules, and there is a
well established interface to operate with them, eg.
# nft add counter filter test
This creates a quota object whose name is 'test'.
# nft list counters
table ip filter {
counter test {
packets 0 bytes 0
}
You can then refer to these objects from maps:
# nft add table filter
# nft add chain filter input { type filter hook input priority 0\; }
# nft add map filter badguys { type ipv4_addr : counter \; }
# nft add rule filter input counter name ip saddr map @badguys
# nft add counter filter badguy1
# nft add counter filter badguy2
# nft add element filter badguys { 192.168.2.3 : "badguy1" }
# nft add element filter badguys { 192.168.2.4 : "badguy2" }
Implicit map definitions are supported too:
table ip filter {
counter http-traffic {
packets 8 bytes 672
}
chain input {
type filter hook input priority 0; policy accept;
counter name tcp dport map { 80 : "http-traffic", 443 : "http-traffic"}
}
}
You can atomically dump and reset these objects:
# nft reset counter ip filter badguy1
counter test {
packets 1024 bytes 100000
}
# nft reset counter ip filter badguy1
counter test {
packets 0 bytes 0
}
Currently: counters, quota and limit are supported. Note: limit is
available starting 4.14-rc.
* Sort set elements when listing them, from lower to largest, eg.
# nft add table x
# nft add set x y { type ipv4_addr\; }
# nft add element x y { 192.168.1.2, 192.168.1.1, 192.168.1.4, 192.168.1.3 }
# nft list ruleset
table ip x {
set y {
type ipv4_addr
elements = { 192.168.1.1, 192.168.1.2,
192.168.1.3, 192.168.1.4 }
}
}
When listing very large sets, nft takes almost the same time as
before, so impact of this new feature is negligible.
* TCP option matching and mangling support. This includes TCP maximum
segment size mangling, eg.
# nft add rule mangle forward tcp flags syn tcp option maxseg size set rt mtu
People that own routers with ppp interfaces, you have no excuses to
migrate to nftables, this is your replacement for the TCPMSS target ;-)
* Add new `-s' option for listings without stateful information:
# nft -s list ruleset
table ip filter {
chain output {
type filter hook output priority 0; policy accept;
tcp dport https counter
tcp dport https quota 25 mbytes
}
}
* Add new -c/--check option for nft, to tests if your ruleset loads fine,
into the kernel, this is a dry run mode, eg.
# nft -c ruleset.nft
You can also use it in incremental rule updates scenarios:
# nft -c add rule x y counter
* Connection tracking helper support, eg.
table ip filter {
ct helper ftp-standard {
type "ftp" protocol tcp
}
chain y {
tcp dport ftp ct helper set "ftp-standard"
}
}
Note for iptables users: In nftables, you have to specify what helper
you want to enable specifically, then set it from rules, given the
former automatic helper assignment approach is deprecated, see for
more info: https://home.regit.org/netfilter-en/secure-use-of-helpers/
* Add --echo option, to print the handle that the kernel allocates to
uniquely identify rules, eg.
# nft --echo --handle add rule ip t c tcp dport {22, 80} accept
add rule ip t c tcp dport { ssh, http } accept # handle 2
* Conntrack zone support, eg.
table raw {
chain pre {
type filter hook prerouting priority -300;
iif eth3 ct zone set 23
}
chain out {
type filter hook output priority -300;
oif eth3 ct zone set 23
}
}
* Symmetric hash support, eg.
# nft add rule ip nat prerouting ct mark set symhash mod 2
* Add support to include directories from nft natives scripts, files are
loaded in alphanumerical order, eg.
include "/foo/*.nft"
Assuming the following content on such folder:
/foo
/foo/02_rules.nft
/foo/01_rules.nft
"01_rules.nft" is loaded before "02_rules.nft".
* Allow to check if IPv6 extension header or TCP option exists or is
missing, eg.
# nft add rule ip6 x y exthdr frag exists drop
# nft add rule inet x y tcp option window missing drop
Same for fib lookups, eg.
# nft add rule x y fib daddr oif exists accept
* Extend quota support to display wasted bytes, eg.
table ip x {
chain y {
type filter hook forward priority 0; policy accept;
quota over 200 mbytes used 1143 kbytes drop
}
}
This is useful to restore quota between reboots.
* Add ct average matching, to match average bytes per packet a connection has
transferred so far, to map the existing feature available in the
iptables connbytes match.
eg. match average pkt in both directions:
# nft add rule x y ct avgpkt gt 100
eg. match avgpkt in original direction:
# nft add rule x y ct original avgpkt gt 200
* Allow to flush maps and flow tables, eg.
# nft flush map filter map1
# nft flush flow table filter ft-https
* Allow to embed set definition into an existing set, eg.
# nft -f ruleset.nft
define myset = {
1.1.1.0,
2.2.2.0,
}
add rule ip saddr { $myset, 3.3.3.0 }
And scripting like now works too:
define dnat_ports = { 1234-1567 }
define port_allow = {
53, # dns
$dnat_ports, # dnat
}
* Slightly better error reporting when ruleset loaded via nft -f, eg.
# nft -f /home/test/x
/home/test/x:4:17-70: Error: Could not process rule: No such file or directory
ip saddr { 1.1.1.1, 2.2.2.2, 3.3.3.3, 4.4.4.4 } jump x
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
In the example above, the chain 'x' that this jump refers to does not exist.
We can still do better and make it even more fine grain...
* Don't wraparound set elements when listing, eg.
% nft list ruleset -nnn
table ip t {
set s {
type inet_service
elements = { 1, 2, 3, 4, 10,
432, 433, 434, 435, 436,
437, 438, 439, 440, 441,
442, 443, 444, 445, 446,
447, 448, 449, 450, 12345 }
}
}
Current approach is rather simple, improvable since the number of
elements to fit in is not calculated based of the number of terminal
columns, but we can make it at some point.
* Allow to list sets in a table, eg.
# nft list sets table inet filter
* Conntrack event filtering support via rule, eg.
# nft add rule x y ct event new counter
... lots of bugfixes too:
* Fix wrong bytecode generation for IPv6 that may result in mismatching
packets due to extensions placed before the transport header.
* Always print range expressions numerically. This has been a problem
with network services, where character '-' is possible as part of the
name.
* Replace getaddrinfo() by inet_service internal lookup table, we don't
rely on /etc/services anymore, so we have a predictable input from the
parser side.
* Do not round consumed quota bytes when listing then.
* Fix print of ip dnat address.
* Fix icmp matching in inet tables.
* Fix display of set keys with host byte endian byteorder.
* Fix nested set merge size adjustment, eg.
# nft add rule ip saddr { { 1.1.1.0/24, 3.3.3.0/24 }, 2.2.2.0/24 }
* Fix missing seed option in hash expression in listings.
* Generate a random seed in hash expression if seed option is empty, eg.
# nft add rule x y ct mark set jhash ip saddr mod 2
generates a random seed internally, that is not displayed. If you want
an explicit seed, you have to specify it.
* Allow to match ICMPv6 packets from IPv4, this is a firewall software so
we should allow to match even things that don't make sense.
* Fix printing of range elements in named sets in monitor mode.
* Fix printing of optional attributes 'flags', 'gc-interval' and
'timeout' in set declarations.
* Fix parsing of IPv4-Mapped IPv6 addresses.
* Fix lots of memory leaks.
... and lots of documentation updates! and regression tests!
Resources
=========
The nftables code can be obtained from:
* http://netfilter.org/projects/nftables/downloads.html
* ftp://ftp.netfilter.org/pub/nftables
* git://git.netfilter.org/nftables
To build the code, libnftnl 1.0.8 and libmnl >= 1.0.2 are required:
* http://netfilter.org/projects/libnftnl/index.html
* http://netfilter.org/projects/libmnl/index.html
Visit our wikipage for user documentation at:
* http://wiki.nftables.org
For the manpage reference, check man(8) nft.
In case of bugs and feature request, file them via:
* https://bugzilla.netfilter.org
Make sure you create no duplicates already, thanks!
Happy firewalling!
[-- Attachment #2: changes-nftables-0.8.txt --]
[-- Type: text/plain, Size: 16203 bytes --]
Anatole Denis (2):
scanner: fix search_in_include_path test
erec: Fix input descriptors for included files
Anders K. Pedersen (1):
netlink_linearize: skip set element expression in set statement key
Arturo Borrero Gonzalez (5):
payload: explicit network ctx assignment for icmp/icmp6 in special families
expression: print sets and maps in pretty format
evaluate: avoid reference to multiple src data in statements which set values
monitor: add debug messages
monitor: Fix printing of range elements in named sets
Bertrand Bonnefoy-Claudet (1):
doc: Fix typo in manpage
Daniel Kahn Gillmor (1):
man: use https for wiki link
Duncan Roe (3):
datatype: printf format warnings on 32-bit build
doc: nft.8 add chain synopsis
doc: nft.8 Document rule replace
Elise Lennion (27):
src: sort set elements in netlink_get_setelems()
tests: shell: add maps tests
src: Allow to list ruleset without stateful information
doc: Include stateless option
tests: py: Use stateless option on tests
tests: shell: Add test for set element leak bug
src: Honor obligatory stateless printing of flow tables
tests: shell: Add test for flush bug on sets with size
src: Allow reset single stateful object
src: Allow list single stateful object
evaluate: Evaluate table name before reset stateful objects in a table
src: Allow list stateful objects in a table
tests: py: Add suport for stateful objects in python tests
tests: py: Add tests for stateful objects
main: Validate the number of numeric options
src: Always print range expressions numerically
configure: Require newer version of libxtables
datatype: Replace getaddrinfo() by internal lookup table
doc: Document stateful objects
doc: Document sets
statement: Avoid rounding bytes in get_rate()
doc: Document maps
parser_bison: Allow flushing flow tables
parser_bison: Allow flushing maps
doc: Document add / delete element operations of sets and maps
src: Make flush command selective of the set structure type
tests: Add test for flush command selectiveness on sets
Eric Leblond (12):
src: fix memory leak when listing rules
parser: fix memory leak in set creation
parser: fix bison warnings
parser: error if needed at EOF
evaluate: fix build with clang
scanner: free filename when destroying scanner
cli: fix heap buffer overflow
mnl: fix error handling in mnl_batch_talk
src: add nft_ctx_new() and nft_ctx_free()
src: add nft_run_cmd_*() functions
src: release caches from nft_ctx_free() path
src: add flags fo nft_ctx_new
Florian Westphal (75):
parser_bison: remove obsolete tokens
evaluate: fix export length and data corruption
statement: fix print of ip dnat address
payload: insert implicit meta tcp dependency when matching tcp options
src: add conntrack zone support
ct: refactor print function so it can be re-used for ct statement
src: support zone set statement with optional direction
tests: add test entries for conntrack zones
netlink: BUG when object type is unknown
doc: ct zone set support
files: provide 'raw' table equivalent
doc: fix ct mark set example
src: add initial ct helper support
evaluate: refactor CMD_OBJ_QUOTA/COUNTER handling
src: allow listing all ct helpers
src: implement add/create/delete for ct helper objects
tests: py: add ct helper tests
tests: add insert-failure test
doc: ct helper objects and helper set support
exthdr: avoid crash with older kernels
src: fix build warning on i686
ct: add conntrack event mask support
parser: allow listing sets in one table
netlink_delinearize: don't kill dependencies accross statements
tests: don't kill dependencies accross statements
netlink_delink_delinearize: don't store dependency unless relop checks is eq check
src: add a comment wrt. reject dependency insertion
netlink_delinearize: reject: remove dependency for tcp-resets
tests: add ip reject with tcp and check for mark too
payload: split ll proto dependency into helper
src: allow update of net base w. meta l4proto icmpv6
src: ipv6: switch implicit dependencies to meta l4proto
payload: enforce ip/ip6 protocol depending on icmp or icmpv6
tests: fix up meta l4proto change for ip6 family
src: ip: switch implicit dependencies to meta l4proto too
tests: fix up meta l4proto change for ip family
Merge branch 'meta_l4_dependency'
ct: fix inet/bridge/netdev family handling for saddr/daddr
meta: permit meta nfproto ip in ip family
parser: allow ct eventmask set new,related
netlink_delinearize: prefer ct event set foo,bar over 'set foo|bar'
src: rename ct eventmask to event
tests: restrict ct saddr test to inet family
tests: remove two non-sensical rules
tests: restrict meta nfproto test cases to inet family
evaluate: reject meta nfproto outside of inet family
tcpopt: make tcptopt structs static
exhtdr: pacify compiler warning
parser: compact list of rhs keyword expressions
rename struct ct to ct_helper
tests: fix arp chain test
evaluate: shift immediate value when adjusting size for csum fixup
tests: add test case for ttl/protocol set
src: add tcp options set support
rt: add path mtu support
tests: add tcp option set support / tcp mss mangling test cases
files: add arp filter and add in/output to nat skeleton
evaluate: convert expr_rt byteorder when evaluating statment arg
evaluate: prepare to store expr key rather than datatype
src: store expression as set key instead of data type
src: ct: remove unused functions
src: prepare for future ct timeout policy support
src: parser_bison: rename ct_block to ct_helper_block
bison: permit keywords in list_stmt_expressions
src: add alternate syntax for ct saddr
src: ct: store proto base of ct key, if any
src: ct: add eval part to inject dependencies for ct saddr/daddr
src: unifiy meta and ct postprocessing
tests: update inet/bridge icmp test case
src: ct: print nfproto name for some header fields
tests: ct: adjust test case commands
src: rt: add keyword distinction for nexthop vs nexthop6
tests: rt: fix test cases
doc: update man page
Merge branch 'ct_rt_syntax_06'
Harsha Sharma (5):
INSTALL: Update dependency list and configure with libxtables support
test: shell: execute shell/run-tests.sh from any directory
tests: shell: add testcases for named objects
test: shell: update shell/run-tests.sh to refer to relative path of testcase
nftables: make pointers in string arrays constant
Ismo Puustinen (8):
scanner: add support for include directories
tests: test include directories
scanner: add files in include dirs in alphabetical order.
man: add include directory documentation.
tests: added tests for ordering files in include dirs.
scanner: support for wildcards in include statements.
tests: update include directory tests to support wildcard syntax.
man: add include statement documentation.
JingPiao Chen (1):
include: remove outdated comment
Laura Garcia Liebana (2):
src: hash: support of symmetric hash
src: hash: fix seed attribute not listed
Liping Zhang (4):
ct: add average bytes per packet counter support
src: fix crash when inputting an incomplete set add command
hash: generate a random seed if seed option is empty
src: delete the old cache when dumping is interrupted
Manuel Messner (7):
evaluate: fix typo
include: linux: netfilter: nf_tables: copy file from nf-next
exthdr: prepare for tcp support
exthdr: prepare exthdr_gen_dependency for tcp support
src: add TCP option matching
payload: automatically kill dependencies for exthdr and tcpopt
tests: py: Add basic tests for ip, ip6 and inet
Oleksandr Natalenko (2):
utils: provide array allocation wrapper
segtree: allocate memory for arrays on heap
Pablo M. Bermudo Garay (5):
src: add new generic context structure nft_ctx
src: add --check option flag
parser: fix typo
src: add stateful object support for limit
tests: py: add tests for limit stateful object
Pablo Neira Ayuso (65):
include: fetch nf_tables.h updates
src: remove SET_F_* flag definitions
src: add used quota support
src: listing of stateful objects
src: add/create/delete stateful objects
src: reset internal stateful objects
parser_bison: allow RESET token from rhs
src: add stateful object reference expression
src: add support for stateful object maps
src: support for stateful object monitoring
mnl: use nftnl_nlmsg_build_hdr()
tests: shell: cover cache flush on flush ruleset
netlink: don't bail out on dump errors
rule: check for errors from cache_init_objects() for stateful objects
tests: shell: validate set size
include: refresh linux/netfilter/nf_tables.h
tests: shell: remove nft_set_bitmap on each run
include: add tcpopt.h to Makefile.am
evaluate: store byteorder for set keys
netlink: store set byteorder in NFTA_SET_USERDATA
netlink_delinearize: remove integer_type_postprocess()
tests: py: fix incorrect bytecode in numgen and hash mappings
datatype: add DTYPE_F_CLONE flag
evaluate: set byteorder as lhs expression context in stmt_evaluate_arg()
src: rename set_keytype_alloc() to set_datatype_alloc()
netlink: rework NFTNL_SET_USERDATA to accomodate new attributes
src: store byteorder for set data
src: revisit tcp options support
netlink: use nftnl_udata_put_u32()/nftnl_udata_get_u32()
tests: shell: test comments with small sets
doc: revisit fib examples
tests: shell: delete element and chain in transaction
rule: adjust set expression size accordingly with intervals
segtree: reset element size counter before adding intervals to set
netlink: add size description for constant sets
tests: py: flush stdout on each test run
expression: don't trim off unary expression on delinearization
src: remove global nftnl_batch structure in mnl layer
doc: nft: describe -I override behaviour when path starts by ./ and /
src: display default directory for file inclusion in -h/--help
src: error reporting for nested ruleset representation
evaluate: merge nested set flags
evaluate: Better error reporting for bad set references
include: Pass nf_sock where needed as parameter
src: handle rule tracing as an monitor object
src: make netlink sequence number non-static
rule: remove sequence number from struct eval_ctx
src: add include_paths to struct nft_ctx
src: add maximum number of parser errors to struct nft_ctx
src: remove ifdef DEBUG pollution
src: add struct mnl_ctx
mnl: pass struct netlink_ctx to mnl_nft_socket_sendmsg()
src: add debugging mask to context structure
src: call nft_init() and nft_exit() from context routines
src: move nf_sock into nft_ctx structure
netlink: remove nfsock_open()
src: add nft_ctx_netlink_init()
mnl: do not set NLM_F_CREATE in deletion requests
parser_bison: consolidate stmt_expr rule
parser_bison: use keywords in ct expression
parser_bison: allow helper keyword in ct object kind
mnl: fix broken sequence number allocation
parser_bison: allow to used named limit from dictionaries too
scanner: IPv4-Mapped IPv6 addresses support
configure: Bump version to v0.8
Phil Sutter (56):
Introduce boolean datatype and boolean expression
exthdr: Add support for exthdr specific flags
exthdr: Implement existence check
fib: Support existence check
tests: Adjust for changed exthdr debug output
doc: Document boolean type and applications
proto: Add some exotic ICMPv6 types
doc: Describe ICMP(v6) expression and types
evaluate: set: Allow for set elems to be sets
evaluate: set: Fix nested set merge size adjustment
tests: Add test cases for nested anonymous sets
tests: shell: netns/0003many_0: Fix cleanup after error
sets: Fix for missing space after last element
nft.8: Enhance NAT documentation
segtree: Fix expr_value_cmp()
src: Allow passing the parent set to set_expr_alloc()
segtree: Introduce flag for half-open range elements
tests: Add basic monitor testing framework
monitor: Print NEWGEN events
tests/monitor: Ignore newgen messages in output
tests/monitor: Add a small README
monitor: Fix printing of set declarations
tests/monitor: Simplify testcases
tests/monitor: Clear ruleset after testing
mnl: Consolidate mnl_batch_talk() parameters
netlink: Pass nlmsg flags from rule.c
Implement --echo option
tests: Add a simple test suite for --echo option
nft.8: Document operations on ruleset
nft.8: Describe conntrack types
nft.8: Review reject statement description
nft.8: Add note about supported hooks for bridge family
nft.8: Describe base chain details
nft.8: Complete short description of arp family
mnl: Drop --echo support for non-batch calls
netlink: Fix segfault when using --echo flag
tests: Merge monitor and echo test suites
echo: Fix for added delays in rule updates
src: Keep cache in struct nft_ctx
src: Fix debug_proto_ctx()
scanner: Fix for memleak due to unclosed file pointer
scanner: Fix for wrong parameter type of scanner_destroy()
scanner: Make use of yylex_init_extra()
parser: Fix for memleak when commands fail
parser: Fix memleaks for STRING token (and derived ones)
doc: nft: Fix and enhance synopsis section
main: Fix for return of uninitialized variable in nft_run_cmd_from_filename()
monitor: Fix for incorrect debug_mask
rule: Use C99-style initializer in cache_init()
exthdr: Simplify tcp option printing a bit
erec_print: Pass output FILE pointer to netlink_dump_expr()
rule: Refactor chain_print_declaration()
src: get rid of printf
main: Flush output from nft_gmp_print()
main: Drop stdout hack to expose nft_print() implementation issues
evaluate: Fix debug output
Shyam Saini (4):
tests: py: Add test for ambiguity while setting the value
tests: shell: Add test for incomplete set add set command
tests: shell: Test input descriptors for included files
tests: py: Fail test forcefully when bug is not fixed
Timothy Redaelli (1):
tests: py: fix some typos in README
Tobias Klauser (1):
build: add missing backslash to list of CFLAGS
Varsha Rao (18):
src: Remove expire information from list stateless ruleset.
src: Pass stateless, numeric, ip2name and handle variables as structure members.
include: Remove datatype_register().
src: Remove __init and __exit macro definitions.
src: netlink: Remove variable nf_mon_sock.
src: nft monitor ruleset
man: Document nft monitor ruleset.
src: netlink: Subscribe nft monitor and nft monitor trace to respective groups.
src: introduce struct nft_cache
man: Fix typo.
src: Remove xt_stmt_() functions.
src: gmputil: Remove mpz_get_be64() function.
src: cli: Remove cli_display() function.
src: netlink: Remove unused functions.
src: mnl: Remove unused functions.
src: Initialize struct stmt in _match and _target functions.
tests: shell: Add tests for chain rename.
tests: files: Remove old tests for chain rename.
^ permalink raw reply
* [PATCH] Add -target to clang switch while cross compiling.
From: Abhijit Ayarekar @ 2017-10-12 23:58 UTC (permalink / raw)
To: ast; +Cc: daniel, netdev, linux-kernel, Abhijit Ayarekar
In-Reply-To: <20171012231615.ugya5fa4mvrze6hw@ast-mbp>
Update to llvm excludes assembly instructions.
llvm git revision is below
commit 2865ab6996164e7854d55c9e21c065fad7c26569
Author: Yonghong Song <yhs@fb.com>
Date: Mon Sep 18 23:29:36 2017 +0000
bpf: add inline-asm support
Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@313593 91177308-0d34-0410-b5e6-96231b3b80d8
__ASM_SYSREG_H define is not required for native compile.
-target switch includes appropriate target specific files
while cross compiling
Tested on x86 and arm64.
Signed-off-by: Abhijit Ayarekar <abhijit.ayarekar@caviumnetworks.com>
---
samples/bpf/Makefile | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index ebc2ad6..81f9fcd 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -180,6 +180,7 @@ CLANG ?= clang
# Detect that we're cross compiling and use the cross compiler
ifdef CROSS_COMPILE
HOSTCC = $(CROSS_COMPILE)gcc
+CLANG_ARCH_ARGS = -target $(ARCH)
endif
# Trick to allow make to be run from this directory
@@ -229,9 +230,9 @@ $(obj)/tracex5_kern.o: $(obj)/syscall_nrs.h
$(obj)/%.o: $(src)/%.c
$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) -I$(obj) \
-I$(srctree)/tools/testing/selftests/bpf/ \
- -D__KERNEL__ -D__ASM_SYSREG_H -Wno-unused-value -Wno-pointer-sign \
+ -D__KERNEL__ -Wno-unused-value -Wno-pointer-sign \
-D__TARGET_ARCH_$(ARCH) -Wno-compare-distinct-pointer-types \
-Wno-gnu-variable-sized-type-not-at-end \
-Wno-address-of-packed-member -Wno-tautological-compare \
- -Wno-unknown-warning-option \
+ -Wno-unknown-warning-option $(CLANG_ARCH_ARGS) \
-O2 -emit-llvm -c $< -o -| $(LLC) -march=bpf -filetype=obj -o $@
--
2.7.4
^ permalink raw reply related
* Waiting for your response to my numerous un-replied emails to you concerning your family inheritance fund ($7.5 million dollars). I seek your assistance and I assured of your capability to champion this business opportunity to remit $7.5 million dollars into your account, if you are interested let me know so that I can send you the details of this transaction. I agreed that 55% of this money will be for you 5% will be set aside for expenses incurred during the business and 40% would be for me. And if you are not willing to handle the transaction don’t border to reply me.
From: Collins Ogbor @ 2017-10-13 0:32 UTC (permalink / raw)
^ permalink raw reply
* [PATCH net-next v2 0/2] add UniPhier AVE ethernet support
From: Kunihiko Hayashi @ 2017-10-13 0:35 UTC (permalink / raw)
To: netdev, Andrew Lunn, Florian Fainelli
Cc: Rob Herring, Mark Rutland, linux-arm-kernel, linux-kernel,
devicetree, Masahiro Yamada, Masami Hiramatsu, Jassi Brar,
Kunihiko Hayashi
This series adds support for Socionext AVE ethernet controller implemented
on UniPhier SoCs. This driver supports RGMII/RMII modes.
v1: http://www.spinics.net/lists/netdev/msg454292.html
The PHY patch included in v1 has already separated in:
http://www.spinics.net/lists/netdev/msg454595.html
Changes since v1:
- add/remove devicetree properties and sub-node
- remove "internal-phy-interrupt" and "desc-bits" property
- add SoC data structures based on compatible strings
- add node operation to apply "mdio" sub-node
- add support for features
- add support for {get,set}_pauseparam and pause frame operations
- add support for ndo_get_stats64 instead of ndo_get_stats
- replace with desiable functions
- replace check for valid phy_mode with phy_interface{_mode}_is_rgmii()
- replace phy attach message with phy_attached_info()
- replace 32bit operation with {upper,lower}_32_bits() on ave_wdesc_addr()
- replace nway_reset and get_link with generic functions
- move operations to proper functions
- move phy_start_aneg() to ndo_open,
and remove unnecessary PHY interrupt operations
See http://www.spinics.net/lists/netdev/msg454590.html
- move irq initialization and descriptor memory allocation to ndo_open
- move initialization of reset and clock and mdiobus to ndo_init
- fix skbuffer operations
- fix skb alignment operations and add Rx buffer adjustment for descriptor
See http://www.spinics.net/lists/netdev/msg456014.html
- add error returns when dma_map_single() failed
- clean up code structures
- clean up wait-loop and wake-queue conditions
- add ave_wdesc_addr() and offset definitions
- add ave_macaddr_init() to clean up mac-address operation
- fix checking whether Tx entry is not enough
- fix supported features of phydev
- add necessary free/disable operations
- add phydev check on ave_{get,set}_wol()
- remove netif_carrier functions, phydev initializer, and Tx budget check
- change obsolate codes
- replace ndev->{base_addr,irq} with the members of ave_private
- rename goto labels and mask definitions, and remove unused codes
Kunihiko Hayashi (2):
dt-bindings: net: add DT bindings for Socionext UniPhier AVE
net: ethernet: socionext: add AVE ethernet driver
.../bindings/net/socionext,uniphier-ave4.txt | 53 +
drivers/net/ethernet/Kconfig | 1 +
drivers/net/ethernet/Makefile | 1 +
drivers/net/ethernet/socionext/Kconfig | 22 +
drivers/net/ethernet/socionext/Makefile | 4 +
drivers/net/ethernet/socionext/sni_ave.c | 1773 ++++++++++++++++++++
6 files changed, 1854 insertions(+)
create mode 100644 Documentation/devicetree/bindings/net/socionext,uniphier-ave4.txt
create mode 100644 drivers/net/ethernet/socionext/Kconfig
create mode 100644 drivers/net/ethernet/socionext/Makefile
create mode 100644 drivers/net/ethernet/socionext/sni_ave.c
--
2.7.4
^ permalink raw reply
* [PATCH net-next v2 1/2] dt-bindings: net: add DT bindings for Socionext UniPhier AVE
From: Kunihiko Hayashi @ 2017-10-13 0:35 UTC (permalink / raw)
To: netdev, Andrew Lunn, Florian Fainelli
Cc: Rob Herring, Mark Rutland, linux-arm-kernel, linux-kernel,
devicetree, Masahiro Yamada, Masami Hiramatsu, Jassi Brar,
Kunihiko Hayashi
In-Reply-To: <1507854928-4357-1-git-send-email-hayashi.kunihiko@socionext.com>
DT bindings for the AVE ethernet controller found on Socionext's
UniPhier platforms.
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Signed-off-by: Jassi Brar <jaswinder.singh@linaro.org>
---
.../bindings/net/socionext,uniphier-ave4.txt | 53 ++++++++++++++++++++++
1 file changed, 53 insertions(+)
create mode 100644 Documentation/devicetree/bindings/net/socionext,uniphier-ave4.txt
diff --git a/Documentation/devicetree/bindings/net/socionext,uniphier-ave4.txt b/Documentation/devicetree/bindings/net/socionext,uniphier-ave4.txt
new file mode 100644
index 0000000..25f4d92
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/socionext,uniphier-ave4.txt
@@ -0,0 +1,53 @@
+* Socionext AVE ethernet controller
+
+This describes the devicetree bindings for AVE ethernet controller
+implemented on Socionext UniPhier SoCs.
+
+Required properties:
+ - compatible: Should be
+ - "socionext,uniphier-pro4-ave4" : for Pro4 SoC
+ - "socionext,uniphier-pxs2-ave4" : for PXs2 SoC
+ - "socionext,uniphier-ld20-ave4" : for LD20 SoC
+ - "socionext,uniphier-ld11-ave4" : for LD11 SoC
+ - reg: Address where registers are mapped and size of region.
+ - interrupts: Should contain the MAC interrupt.
+ - phy-mode: See ethernet.txt in the same directory. Allow to choose
+ "rgmii", "rmii", or "mii" according to the PHY.
+ - pinctrl-names: List of assigned state names, see pinctrl
+ binding documentation.
+ - pinctrl-0: List of phandles to configure the GPIO pin,
+ see pinctrl binding documentation. Choose this appropriately
+ according to phy-mode.
+ - <&pinctrl_ether_rgmii> if phy-mode is "rgmii".
+ - <&pinctrl_ether_rmii> if phy-mode is "rmii".
+ - <&pinctrl_ether_mii> if phy-mode is "mii".
+ - phy-handle: Should point to the external phy device.
+ See ethernet.txt file in the same directory.
+ - mdio subnode: Should be device tree subnode with the following required
+ properties:
+ - #address-cells: Must be <1>.
+ - #size-cells: Must be <0>.
+ - reg: phy ID number, usually a small integer.
+
+Optional properties:
+ - local-mac-address: See ethernet.txt in the same directory.
+
+Example:
+
+ ether: ethernet@65000000 {
+ compatible = "socionext,uniphier-ld20-ave4";
+ reg = <0x65000000 0x8500>;
+ interrupts = <0 66 4>;
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_ether_rgmii>;
+ phy-mode = "rgmii";
+ phy-handle = <ðphy>;
+ local-mac-address = [00 00 00 00 00 00];
+ mdio {
+ #address-cells = <1>;
+ #size-cells = <0>;
+ ethphy: ethphy@1 {
+ reg = <1>;
+ };
+ };
+ };
--
2.7.4
^ permalink raw reply related
* [PATCH net-next v2 2/2] net: ethernet: socionext: add AVE ethernet driver
From: Kunihiko Hayashi @ 2017-10-13 0:35 UTC (permalink / raw)
To: netdev, Andrew Lunn, Florian Fainelli
Cc: Rob Herring, Mark Rutland, linux-arm-kernel, linux-kernel,
devicetree, Masahiro Yamada, Masami Hiramatsu, Jassi Brar,
Kunihiko Hayashi
In-Reply-To: <1507854928-4357-1-git-send-email-hayashi.kunihiko@socionext.com>
The UniPhier platform from Socionext provides the AVE ethernet
controller that includes MAC and MDIO bus supporting RGMII/RMII
modes. The controller is named AVE.
Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@socionext.com>
Signed-off-by: Jassi Brar <jaswinder.singh@linaro.org>
---
drivers/net/ethernet/Kconfig | 1 +
drivers/net/ethernet/Makefile | 1 +
drivers/net/ethernet/socionext/Kconfig | 22 +
drivers/net/ethernet/socionext/Makefile | 4 +
drivers/net/ethernet/socionext/sni_ave.c | 1773 ++++++++++++++++++++++++++++++
5 files changed, 1801 insertions(+)
create mode 100644 drivers/net/ethernet/socionext/Kconfig
create mode 100644 drivers/net/ethernet/socionext/Makefile
create mode 100644 drivers/net/ethernet/socionext/sni_ave.c
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index c604213..d50519e 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -170,6 +170,7 @@ source "drivers/net/ethernet/sis/Kconfig"
source "drivers/net/ethernet/sfc/Kconfig"
source "drivers/net/ethernet/sgi/Kconfig"
source "drivers/net/ethernet/smsc/Kconfig"
+source "drivers/net/ethernet/socionext/Kconfig"
source "drivers/net/ethernet/stmicro/Kconfig"
source "drivers/net/ethernet/sun/Kconfig"
source "drivers/net/ethernet/tehuti/Kconfig"
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index a0a03d4..9f55b36 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -81,6 +81,7 @@ obj-$(CONFIG_SFC) += sfc/
obj-$(CONFIG_SFC_FALCON) += sfc/falcon/
obj-$(CONFIG_NET_VENDOR_SGI) += sgi/
obj-$(CONFIG_NET_VENDOR_SMSC) += smsc/
+obj-$(CONFIG_NET_VENDOR_SOCIONEXT) += socionext/
obj-$(CONFIG_NET_VENDOR_STMICRO) += stmicro/
obj-$(CONFIG_NET_VENDOR_SUN) += sun/
obj-$(CONFIG_NET_VENDOR_TEHUTI) += tehuti/
diff --git a/drivers/net/ethernet/socionext/Kconfig b/drivers/net/ethernet/socionext/Kconfig
new file mode 100644
index 0000000..3a1829e
--- /dev/null
+++ b/drivers/net/ethernet/socionext/Kconfig
@@ -0,0 +1,22 @@
+config NET_VENDOR_SOCIONEXT
+ bool "Socionext ethernet drivers"
+ default y
+ ---help---
+ Option to select ethernet drivers for Socionext platforms.
+
+ Note that the answer to this question doesn't directly affect the
+ kernel: saying N will just cause the configurator to skip all
+ the questions about Socionext devices. If you say Y, you will be asked
+ for your specific card in the following questions.
+
+if NET_VENDOR_SOCIONEXT
+
+config SNI_AVE
+ tristate "Socionext AVE ethernet support"
+ depends on (ARCH_UNIPHIER || COMPILE_TEST) && OF
+ select PHYLIB
+ ---help---
+ Driver for gigabit ethernet MACs, called AVE, in the
+ Socionext UniPhier family.
+
+endif #NET_VENDOR_SOCIONEXT
diff --git a/drivers/net/ethernet/socionext/Makefile b/drivers/net/ethernet/socionext/Makefile
new file mode 100644
index 0000000..0356341
--- /dev/null
+++ b/drivers/net/ethernet/socionext/Makefile
@@ -0,0 +1,4 @@
+#
+# Makefile for all ethernet ip drivers on Socionext platforms
+#
+obj-$(CONFIG_SNI_AVE) += sni_ave.o
diff --git a/drivers/net/ethernet/socionext/sni_ave.c b/drivers/net/ethernet/socionext/sni_ave.c
new file mode 100644
index 0000000..7e399fc
--- /dev/null
+++ b/drivers/net/ethernet/socionext/sni_ave.c
@@ -0,0 +1,1773 @@
+/**
+ * sni_ave.c - Socionext UniPhier AVE ethernet driver
+ *
+ * Copyright 2014 Panasonic Corporation
+ * Copyright 2015-2017 Socionext Inc.
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 of
+ * the License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/bitops.h>
+#include <linux/clk.h>
+#include <linux/etherdevice.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/mii.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/of_net.h>
+#include <linux/of_mdio.h>
+#include <linux/of_platform.h>
+#include <linux/phy.h>
+#include <linux/reset.h>
+#include <linux/types.h>
+#include <linux/u64_stats_sync.h>
+
+/* General Register Group */
+#define AVE_IDR 0x0 /* ID */
+#define AVE_VR 0x4 /* Version */
+#define AVE_GRR 0x8 /* Global Reset */
+#define AVE_CFGR 0xc /* Configuration */
+
+/* Interrupt Register Group */
+#define AVE_GIMR 0x100 /* Global Interrupt Mask */
+#define AVE_GISR 0x104 /* Global Interrupt Status */
+
+/* MAC Register Group */
+#define AVE_TXCR 0x200 /* TX Setup */
+#define AVE_RXCR 0x204 /* RX Setup */
+#define AVE_RXMAC1R 0x208 /* MAC address (lower) */
+#define AVE_RXMAC2R 0x20c /* MAC address (upper) */
+#define AVE_MDIOCTR 0x214 /* MDIO Control */
+#define AVE_MDIOAR 0x218 /* MDIO Address */
+#define AVE_MDIOWDR 0x21c /* MDIO Data */
+#define AVE_MDIOSR 0x220 /* MDIO Status */
+#define AVE_MDIORDR 0x224 /* MDIO Rd Data */
+
+/* Descriptor Control Register Group */
+#define AVE_DESCC 0x300 /* Descriptor Control */
+#define AVE_TXDC 0x304 /* TX Descriptor Configuration */
+#define AVE_RXDC0 0x308 /* RX Descriptor Ring0 Configuration */
+#define AVE_IIRQC 0x34c /* Interval IRQ Control */
+
+/* Packet Filter Register Group */
+#define AVE_PKTF_BASE 0x800 /* PF Base Address */
+#define AVE_PFMBYTE_BASE 0xd00 /* PF Mask Byte Base Address */
+#define AVE_PFMBIT_BASE 0xe00 /* PF Mask Bit Base Address */
+#define AVE_PFSEL_BASE 0xf00 /* PF Selector Base Address */
+#define AVE_PFEN 0xffc /* Packet Filter Enable */
+#define AVE_PKTF(ent) (AVE_PKTF_BASE + (ent) * 0x40)
+#define AVE_PFMBYTE(ent) (AVE_PFMBYTE_BASE + (ent) * 8)
+#define AVE_PFMBIT(ent) (AVE_PFMBIT_BASE + (ent) * 4)
+#define AVE_PFSEL(ent) (AVE_PFSEL_BASE + (ent) * 4)
+
+/* 64bit descriptor memory */
+#define AVE_DESC_SIZE_64 12 /* Descriptor Size */
+
+#define AVE_TXDM_64 0x1000 /* Tx Descriptor Memory */
+#define AVE_RXDM_64 0x1c00 /* Rx Descriptor Memory */
+
+#define AVE_TXDM_SIZE_64 0x0ba0 /* Tx Descriptor Memory Size 3KB */
+#define AVE_RXDM_SIZE_64 0x6000 /* Rx Descriptor Memory Size 24KB */
+
+/* 32bit descriptor memory */
+#define AVE_DESC_SIZE_32 8 /* Descriptor Size */
+
+#define AVE_TXDM_32 0x1000 /* Tx Descriptor Memory */
+#define AVE_RXDM_32 0x1800 /* Rx Descriptor Memory */
+
+#define AVE_TXDM_SIZE_32 0x07c0 /* Tx Descriptor Memory Size 2KB */
+#define AVE_RXDM_SIZE_32 0x4000 /* Rx Descriptor Memory Size 16KB */
+
+/* RMII Bridge Register Group */
+#define AVE_RSTCTRL 0x8028 /* Reset control */
+#define AVE_RSTCTRL_RMIIRST BIT(16)
+#define AVE_LINKSEL 0x8034 /* Link speed setting */
+#define AVE_LINKSEL_100M BIT(0)
+
+/* AVE_GRR */
+#define AVE_GRR_RXFFR BIT(5) /* Reset RxFIFO */
+#define AVE_GRR_PHYRST BIT(4) /* Reset external PHY */
+#define AVE_GRR_GRST BIT(0) /* Reset all MAC */
+
+/* AVE_GISR (common with GIMR) */
+#define AVE_GI_PHY BIT(24) /* PHY interrupt */
+#define AVE_GI_TX BIT(16) /* Tx complete */
+#define AVE_GI_RXERR BIT(8) /* Receive frame more than max size */
+#define AVE_GI_RXOVF BIT(7) /* Overflow at the RxFIFO */
+#define AVE_GI_RXDROP BIT(6) /* Drop packet */
+#define AVE_GI_RXIINT BIT(5) /* Interval interrupt */
+
+/* AVE_TXCR */
+#define AVE_TXCR_FLOCTR BIT(18) /* Flow control */
+#define AVE_TXCR_TXSPD_1G BIT(17)
+#define AVE_TXCR_TXSPD_100 BIT(16)
+
+/* AVE_RXCR */
+#define AVE_RXCR_RXEN BIT(30) /* Rx enable */
+#define AVE_RXCR_FDUPEN BIT(22) /* Interface mode */
+#define AVE_RXCR_FLOCTR BIT(21) /* Flow control */
+#define AVE_RXCR_AFEN BIT(19) /* MAC address filter */
+#define AVE_RXCR_DRPEN BIT(18) /* Drop pause frame */
+#define AVE_RXCR_MPSIZ_MASK GENMASK(10, 0)
+
+/* AVE_MDIOCTR */
+#define AVE_MDIOCTR_RREQ BIT(3) /* Read request */
+#define AVE_MDIOCTR_WREQ BIT(2) /* Write request */
+
+/* AVE_MDIOSR */
+#define AVE_MDIOSR_STS BIT(0) /* access status */
+
+/* AVE_DESCC */
+#define AVE_DESCC_TD BIT(0) /* Enable Tx descriptor */
+#define AVE_DESCC_RDSTP BIT(4) /* Pause Rx descriptor */
+#define AVE_DESCC_RD0 BIT(8) /* Enable Rx descriptor Ring0 */
+#define AVE_DESCC_STATUS_MASK GENMASK(31, 16)
+
+/* AVE_TXDC */
+#define AVE_TXDC_SIZE GENMASK(27, 16) /* Size of Tx descriptor */
+#define AVE_TXDC_ADDR GENMASK(11, 0) /* Start address */
+#define AVE_TXDC_ADDR_START 0
+
+/* AVE_RXDC0 */
+#define AVE_RXDC0_SIZE GENMASK(30, 16) /* Size of Rx descriptor */
+#define AVE_RXDC0_ADDR GENMASK(14, 0) /* Start address */
+#define AVE_RXDC0_ADDR_START 0
+
+/* AVE_IIRQC */
+#define AVE_IIRQC_EN0 BIT(27) /* Enable interval interrupt Ring0 */
+#define AVE_IIRQC_BSCK GENMASK(15, 0) /* Interval count unit */
+
+/* Command status for Descriptor */
+#define AVE_STS_OWN BIT(31) /* Descriptor ownership */
+#define AVE_STS_INTR BIT(29) /* Request for interrupt */
+#define AVE_STS_OK BIT(27) /* Normal transmit */
+/* TX */
+#define AVE_STS_NOCSUM BIT(28) /* No use HW checksum */
+#define AVE_STS_1ST BIT(26) /* Head of buffer chain */
+#define AVE_STS_LAST BIT(25) /* Tail of buffer chain */
+#define AVE_STS_OWC BIT(21) /* Out of window,Late Collision */
+#define AVE_STS_EC BIT(20) /* Excess collision occurred */
+#define AVE_STS_PKTLEN_TX_MASK GENMASK(15, 0)
+/* RX */
+#define AVE_STS_CSSV BIT(21) /* Checksum check performed */
+#define AVE_STS_CSER BIT(20) /* Checksum error detected */
+#define AVE_STS_PKTLEN_RX_MASK GENMASK(10, 0)
+
+/* AVE_CFGR */
+#define AVE_CFGR_FLE BIT(31) /* Filter Function */
+#define AVE_CFGR_CHE BIT(30) /* Checksum Function */
+#define AVE_CFGR_MII BIT(27) /* Func mode (1:MII/RMII, 0:RGMII) */
+#define AVE_CFGR_IPFCEN BIT(24) /* IP fragment sum Enable */
+
+#define AVE_MAX_ETHFRAME 1518
+
+/* Packet filter size */
+#define AVE_PF_SIZE 17 /* Number of all packet filter */
+#define AVE_PF_MULTICAST_SIZE 7 /* Number of multicast filter */
+
+/* Packet filter definition */
+#define AVE_PFNUM_FILTER 0 /* No.0 */
+#define AVE_PFNUM_UNICAST 1 /* No.1 */
+#define AVE_PFNUM_BROADCAST 2 /* No.2 */
+#define AVE_PFNUM_MULTICAST 11 /* No.11-17 */
+
+/* NETIF Message control */
+#define AVE_DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | \
+ NETIF_MSG_PROBE | \
+ NETIF_MSG_LINK | \
+ NETIF_MSG_TIMER | \
+ NETIF_MSG_IFDOWN | \
+ NETIF_MSG_IFUP | \
+ NETIF_MSG_RX_ERR | \
+ NETIF_MSG_TX_ERR)
+
+/* Parameter for descriptor */
+#define AVE_NR_TXDESC 32 /* Tx descriptor */
+#define AVE_NR_RXDESC 64 /* Rx descriptor */
+
+#define AVE_DESC_OFS_CMDSTS 0
+#define AVE_DESC_OFS_ADDRL 4
+#define AVE_DESC_OFS_ADDRU 8
+
+/* Parameter for interrupt */
+#define AVE_INTM_COUNT 20
+#define AVE_FORCE_TXINTCNT 1
+
+#define IS_DESC_64BIT(p) ((p)->data->is_desc_64bit)
+
+enum desc_id {
+ AVE_DESCID_TX = 0,
+ AVE_DESCID_RX,
+};
+
+enum desc_state {
+ AVE_DESC_STOP = 0,
+ AVE_DESC_START,
+ AVE_DESC_RX_SUSPEND,
+ AVE_DESC_RX_PERMIT,
+};
+
+struct ave_desc {
+ struct sk_buff *skbs;
+ dma_addr_t skbs_dma;
+ size_t skbs_dmalen;
+};
+
+struct ave_desc_info {
+ u32 ndesc; /* number of descriptor */
+ u32 daddr; /* start address of descriptor */
+ u32 proc_idx; /* index of processing packet */
+ u32 done_idx; /* index of processed packet */
+ struct ave_desc *desc; /* skb info related descriptor */
+};
+
+struct ave_soc_data {
+ bool is_desc_64bit;
+};
+
+struct ave_stats {
+ struct u64_stats_sync syncp;
+ u64 packets;
+ u64 bytes;
+ u64 errors;
+ u64 dropped;
+ u64 collisions;
+ u64 fifo_errors;
+};
+
+struct ave_private {
+ void __iomem *base;
+ int irq;
+ int phy_id;
+ unsigned int desc_size;
+ u32 msg_enable;
+ struct clk *clk;
+ struct reset_control *rst;
+ phy_interface_t phy_mode;
+ struct phy_device *phydev;
+ struct mii_bus *mdio;
+
+ /* stats */
+ struct ave_stats stats_rx;
+ struct ave_stats stats_tx;
+
+ /* NAPI support */
+ struct net_device *ndev;
+ struct napi_struct napi_rx;
+ struct napi_struct napi_tx;
+
+ /* descriptor */
+ struct ave_desc_info rx;
+ struct ave_desc_info tx;
+
+ /* flow control */
+ int pause_auto;
+ int pause_rx;
+ int pause_tx;
+
+ const struct ave_soc_data *data;
+};
+
+static inline u32 ave_r32(struct net_device *ndev, u32 offset)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ return readl_relaxed(priv->base + offset);
+}
+
+static inline void ave_w32(struct net_device *ndev, u32 offset, u32 value)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ writel_relaxed(value, priv->base + offset);
+}
+
+static inline u32 ave_rdesc(struct net_device *ndev, enum desc_id id,
+ int entry, int offset)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 addr = (id == AVE_DESCID_TX) ? priv->tx.daddr : priv->rx.daddr;
+
+ addr += entry * priv->desc_size + offset;
+
+ return ave_r32(ndev, addr);
+}
+
+static inline u32 ave_rdesc_cmdsts(struct net_device *ndev, enum desc_id id,
+ int entry)
+{
+ return ave_rdesc(ndev, id, entry, AVE_DESC_OFS_CMDSTS);
+}
+
+static inline void ave_wdesc(struct net_device *ndev, enum desc_id id,
+ int entry, int offset, u32 val)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 addr = (id == AVE_DESCID_TX) ? priv->tx.daddr : priv->rx.daddr;
+
+ addr += entry * priv->desc_size + offset;
+
+ ave_w32(ndev, addr, val);
+}
+
+static inline void ave_wdesc_cmdsts(struct net_device *ndev, enum desc_id id,
+ int entry, u32 val)
+{
+ ave_wdesc(ndev, id, entry, AVE_DESC_OFS_CMDSTS, val);
+}
+
+static void ave_wdesc_addr(struct net_device *ndev, enum desc_id id,
+ int entry, dma_addr_t paddr)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ ave_wdesc(ndev, id, entry, AVE_DESC_OFS_ADDRL, lower_32_bits(paddr));
+ if (IS_DESC_64BIT(priv))
+ ave_wdesc(ndev, id,
+ entry, AVE_DESC_OFS_ADDRU, upper_32_bits(paddr));
+ else if ((u64)paddr > (u64)U32_MAX)
+ netdev_warn(ndev, "DMA address exceeds the address space\n");
+}
+
+static void ave_get_fwversion(struct net_device *ndev, char *buf, int len)
+{
+ u32 major, minor;
+
+ major = (ave_r32(ndev, AVE_VR) & GENMASK(15, 8)) >> 8;
+ minor = (ave_r32(ndev, AVE_VR) & GENMASK(7, 0));
+ snprintf(buf, len, "v%u.%u", major, minor);
+}
+
+static void ave_get_drvinfo(struct net_device *ndev,
+ struct ethtool_drvinfo *info)
+{
+ struct device *dev = ndev->dev.parent;
+
+ strlcpy(info->driver, dev->driver->name, sizeof(info->driver));
+ strlcpy(info->bus_info, dev_name(dev), sizeof(info->bus_info));
+ ave_get_fwversion(ndev, info->fw_version, sizeof(info->fw_version));
+}
+
+static u32 ave_get_msglevel(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ return priv->msg_enable;
+}
+
+static void ave_set_msglevel(struct net_device *ndev, u32 val)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ priv->msg_enable = val;
+}
+
+static void ave_get_wol(struct net_device *ndev,
+ struct ethtool_wolinfo *wol)
+{
+ wol->supported = 0;
+ wol->wolopts = 0;
+
+ if (ndev->phydev)
+ phy_ethtool_get_wol(ndev->phydev, wol);
+}
+
+static int ave_set_wol(struct net_device *ndev,
+ struct ethtool_wolinfo *wol)
+{
+ int ret;
+
+ if (!ndev->phydev ||
+ (wol->wolopts & (WAKE_ARP | WAKE_MAGICSECURE)))
+ return -EOPNOTSUPP;
+
+ ret = phy_ethtool_set_wol(ndev->phydev, wol);
+ if (!ret)
+ device_set_wakeup_enable(&ndev->dev, !!wol->wolopts);
+
+ return ret;
+}
+
+static void ave_get_pauseparam(struct net_device *ndev,
+ struct ethtool_pauseparam *pause)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ pause->autoneg = priv->pause_auto;
+ pause->rx_pause = priv->pause_rx;
+ pause->tx_pause = priv->pause_tx;
+}
+
+static int ave_set_pauseparam(struct net_device *ndev,
+ struct ethtool_pauseparam *pause)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ struct phy_device *phydev = ndev->phydev;
+
+ if (!phydev)
+ return -EINVAL;
+
+ priv->pause_auto = pause->autoneg;
+ priv->pause_rx = pause->rx_pause;
+ priv->pause_tx = pause->tx_pause;
+
+ phydev->advertising &= ~(ADVERTISED_Pause | ADVERTISED_Asym_Pause);
+ if (pause->rx_pause)
+ phydev->advertising |= ADVERTISED_Pause | ADVERTISED_Asym_Pause;
+ if (pause->tx_pause)
+ phydev->advertising ^= ADVERTISED_Asym_Pause;
+
+ if (pause->autoneg) {
+ if (netif_running(ndev))
+ phy_start_aneg(phydev);
+ }
+
+ return 0;
+}
+
+static const struct ethtool_ops ave_ethtool_ops = {
+ .get_link_ksettings = phy_ethtool_get_link_ksettings,
+ .set_link_ksettings = phy_ethtool_set_link_ksettings,
+ .get_drvinfo = ave_get_drvinfo,
+ .nway_reset = phy_ethtool_nway_reset,
+ .get_link = ethtool_op_get_link,
+ .get_msglevel = ave_get_msglevel,
+ .set_msglevel = ave_set_msglevel,
+ .get_wol = ave_get_wol,
+ .set_wol = ave_set_wol,
+ .get_pauseparam = ave_get_pauseparam,
+ .set_pauseparam = ave_set_pauseparam,
+};
+
+static int ave_pfsel_start(struct net_device *ndev, unsigned int entry)
+{
+ u32 val;
+
+ if (WARN_ON(entry > AVE_PF_SIZE))
+ return -EINVAL;
+
+ val = ave_r32(ndev, AVE_PFEN);
+ ave_w32(ndev, AVE_PFEN, val | BIT(entry));
+
+ return 0;
+}
+
+static int ave_pfsel_stop(struct net_device *ndev, unsigned int entry)
+{
+ u32 val;
+
+ if (WARN_ON(entry > AVE_PF_SIZE))
+ return -EINVAL;
+
+ val = ave_r32(ndev, AVE_PFEN);
+ ave_w32(ndev, AVE_PFEN, val & ~BIT(entry));
+
+ return 0;
+}
+
+static int ave_pfsel_macaddr_set(struct net_device *ndev,
+ unsigned int entry, unsigned char *mac_addr,
+ unsigned int set_size)
+{
+ u32 val;
+
+ if (WARN_ON(entry > AVE_PF_SIZE))
+ return -EINVAL;
+ if (WARN_ON(set_size > 6))
+ return -EINVAL;
+
+ ave_pfsel_stop(ndev, entry);
+
+ /* set MAC address for the filter */
+ val = mac_addr[0] | (mac_addr[1] << 8) | (mac_addr[2] << 16)
+ | (mac_addr[3] << 24);
+ ave_w32(ndev, AVE_PKTF(entry), val);
+ val = mac_addr[4] | (mac_addr[5] << 8);
+ ave_w32(ndev, AVE_PKTF(entry) + 4, val);
+
+ /* set byte mask */
+ ave_w32(ndev, AVE_PFMBYTE(entry), GENMASK(31, set_size) & ~0xc0);
+ ave_w32(ndev, AVE_PFMBYTE(entry) + 4, 0xFFFFFFFF);
+
+ /* set bit mask filter */
+ ave_w32(ndev, AVE_PFMBIT(entry), 0x0000FFFF);
+
+ /* set selector to ring 0 */
+ ave_w32(ndev, AVE_PFSEL(entry), 0);
+
+ /* restart filter */
+ ave_pfsel_start(ndev, entry);
+
+ return 0;
+}
+
+static int ave_mdio_busywait(struct net_device *ndev)
+{
+ int ret = 0, loop = 100;
+ u32 mdiosr;
+
+ /* wait until completion */
+ while (--loop) {
+ mdiosr = ave_r32(ndev, AVE_MDIOSR);
+ if (!(mdiosr & AVE_MDIOSR_STS))
+ break;
+
+ usleep_range(10, 20);
+ }
+
+ if (!loop) {
+ netdev_err(ndev,
+ "failed to read from MDIO (status:0x%08x)\n",
+ mdiosr);
+ ret = -ETIMEDOUT;
+ }
+
+ return ret;
+}
+
+static int ave_mdiobus_read(struct mii_bus *bus, int phyid, int regnum)
+{
+ struct net_device *ndev = bus->priv;
+ u32 mdioctl;
+ int ret;
+
+ /* write address */
+ ave_w32(ndev, AVE_MDIOAR, (phyid << 8) | regnum);
+
+ /* read request */
+ mdioctl = ave_r32(ndev, AVE_MDIOCTR);
+ ave_w32(ndev, AVE_MDIOCTR,
+ (mdioctl | AVE_MDIOCTR_RREQ) & ~AVE_MDIOCTR_WREQ);
+
+ ret = ave_mdio_busywait(ndev);
+ if (ret) {
+ netdev_err(ndev, "phy-%d reg-%x read failed\n",
+ phyid, regnum);
+ return ret;
+ }
+
+ return ave_r32(ndev, AVE_MDIORDR) & GENMASK(15, 0);
+}
+
+static int ave_mdiobus_write(struct mii_bus *bus,
+ int phyid, int regnum, u16 val)
+{
+ struct net_device *ndev = bus->priv;
+ u32 mdioctl;
+ int ret;
+
+ /* write address */
+ ave_w32(ndev, AVE_MDIOAR, (phyid << 8) | regnum);
+
+ /* write data */
+ ave_w32(ndev, AVE_MDIOWDR, val);
+
+ /* write request */
+ mdioctl = ave_r32(ndev, AVE_MDIOCTR);
+ ave_w32(ndev, AVE_MDIOCTR,
+ (mdioctl | AVE_MDIOCTR_WREQ) & ~AVE_MDIOCTR_RREQ);
+
+ ret = ave_mdio_busywait(ndev);
+ if (ret)
+ netdev_err(ndev, "phy-%d reg-%x write failed\n",
+ phyid, regnum);
+
+ return ret;
+}
+
+static dma_addr_t ave_dma_map(struct net_device *ndev, struct ave_desc *desc,
+ void *ptr, size_t len,
+ enum dma_data_direction dir)
+{
+ dma_addr_t paddr;
+
+ paddr = dma_map_single(ndev->dev.parent, ptr, len, dir);
+ if (unlikely(dma_mapping_error(ndev->dev.parent, paddr))) {
+ paddr = (dma_addr_t)-ENOMEM;
+ } else {
+ desc->skbs_dma = paddr;
+ desc->skbs_dmalen = len;
+ }
+
+ return paddr;
+}
+
+static void ave_dma_unmap(struct net_device *ndev, struct ave_desc *desc,
+ enum dma_data_direction dir)
+{
+ if (!desc->skbs_dma)
+ return;
+
+ dma_unmap_single(ndev->dev.parent,
+ desc->skbs_dma, desc->skbs_dmalen, dir);
+ desc->skbs_dma = 0;
+}
+
+/* Set Rx descriptor and memory */
+static int ave_set_rxdesc(struct net_device *ndev, int entry)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ struct sk_buff *skb;
+ dma_addr_t paddr;
+ int ret = 0;
+
+ skb = priv->rx.desc[entry].skbs;
+ if (!skb) {
+ skb = netdev_alloc_skb_ip_align(ndev,
+ AVE_MAX_ETHFRAME);
+ if (!skb) {
+ netdev_err(ndev, "can't allocate skb for Rx\n");
+ return -ENOMEM;
+ }
+ }
+
+ /* set disable to cmdsts */
+ ave_wdesc_cmdsts(ndev, AVE_DESCID_RX, entry,
+ AVE_STS_INTR | AVE_STS_OWN);
+
+ /* map Rx buffer
+ * Rx buffer set to the Rx descriptor has two restrictions:
+ * - Rx buffer address is 4 byte aligned.
+ * - Rx buffer begins with 2 byte headroom, and data will be put from
+ * (buffer + 2).
+ * To satisfy this, specify the address to put back the buffer
+ * pointer advanced by NET_IP_ALIGN by netdev_alloc_skb_ip_align(),
+ * and expand the map size by NET_IP_ALIGN.
+ */
+ paddr = ave_dma_map(ndev, &priv->rx.desc[entry],
+ skb->data - NET_IP_ALIGN,
+ AVE_MAX_ETHFRAME + NET_IP_ALIGN, DMA_FROM_DEVICE);
+ if (paddr == -ENOMEM) {
+ netdev_err(ndev, "can't map skb for Rx\n");
+ dev_kfree_skb_any(skb);
+ return -ENOMEM;
+ }
+ priv->rx.desc[entry].skbs = skb;
+
+ /* set buffer pointer */
+ ave_wdesc_addr(ndev, AVE_DESCID_RX, entry, paddr);
+
+ /* set enable to cmdsts */
+ ave_wdesc_cmdsts(ndev, AVE_DESCID_RX, entry,
+ AVE_STS_INTR | AVE_MAX_ETHFRAME);
+
+ return ret;
+}
+
+/* Switch state of descriptor */
+static int ave_desc_switch(struct net_device *ndev, enum desc_state state)
+{
+ int counter;
+ int ret = 0;
+ u32 val;
+
+ switch (state) {
+ case AVE_DESC_START:
+ ave_w32(ndev, AVE_DESCC, AVE_DESCC_TD | AVE_DESCC_RD0);
+ break;
+
+ case AVE_DESC_STOP:
+ ave_w32(ndev, AVE_DESCC, 0);
+ counter = 100;
+ while (--counter) {
+ usleep_range(100, 150);
+ if (!ave_r32(ndev, AVE_DESCC))
+ break;
+ }
+ if (!counter) {
+ netdev_err(ndev, "can't stop descriptor\n");
+ ret = -EBUSY;
+ }
+ break;
+
+ case AVE_DESC_RX_SUSPEND:
+ val = ave_r32(ndev, AVE_DESCC);
+ val |= AVE_DESCC_RDSTP;
+ val &= ~AVE_DESCC_STATUS_MASK;
+ ave_w32(ndev, AVE_DESCC, val);
+
+ counter = 1000;
+ while (--counter) {
+ usleep_range(100, 150);
+ val = ave_r32(ndev, AVE_DESCC);
+ if (val & (AVE_DESCC_RDSTP << 16))
+ break;
+ }
+ if (!counter) {
+ netdev_err(ndev, "can't suspend descriptor\n");
+ ret = -EBUSY;
+ }
+ break;
+
+ case AVE_DESC_RX_PERMIT:
+ val = ave_r32(ndev, AVE_DESCC);
+ val &= ~AVE_DESCC_RDSTP;
+ val &= ~AVE_DESCC_STATUS_MASK;
+ ave_w32(ndev, AVE_DESCC, val);
+ break;
+
+ default:
+ ret = -EINVAL;
+ break;
+ }
+
+ return ret;
+}
+
+static int ave_tx_completion(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 proc_idx, done_idx, ndesc, cmdsts;
+ unsigned int nr_freebuf = 0;
+ unsigned int tx_packets = 0;
+ unsigned int tx_bytes = 0;
+
+ proc_idx = priv->tx.proc_idx;
+ done_idx = priv->tx.done_idx;
+ ndesc = priv->tx.ndesc;
+
+ /* free pre stored skb from done to proc */
+ while (proc_idx != done_idx) {
+ /* get cmdsts */
+ cmdsts = ave_rdesc_cmdsts(ndev, AVE_DESCID_TX, done_idx);
+
+ /* do nothing if owner is HW (==1 for Tx) */
+ if (cmdsts & AVE_STS_OWN)
+ break;
+
+ /* check Tx status and updates statistics */
+ if (cmdsts & AVE_STS_OK) {
+ tx_bytes += cmdsts & AVE_STS_PKTLEN_TX_MASK;
+ /* success */
+ if (cmdsts & AVE_STS_LAST)
+ tx_packets++;
+ } else {
+ /* error */
+ if (cmdsts & AVE_STS_LAST) {
+ priv->stats_tx.errors++;
+ if (cmdsts & (AVE_STS_OWC | AVE_STS_EC))
+ priv->stats_tx.collisions++;
+ }
+ }
+
+ /* release skb */
+ if (priv->tx.desc[done_idx].skbs) {
+ ave_dma_unmap(ndev, &priv->tx.desc[done_idx],
+ DMA_TO_DEVICE);
+ dev_consume_skb_any(priv->tx.desc[done_idx].skbs);
+ priv->tx.desc[done_idx].skbs = NULL;
+ nr_freebuf++;
+ }
+ done_idx = (done_idx + 1) % ndesc;
+ }
+
+ priv->tx.done_idx = done_idx;
+
+ /* update stats */
+ u64_stats_update_begin(&priv->stats_tx.syncp);
+ priv->stats_tx.packets += tx_packets;
+ priv->stats_tx.bytes += tx_bytes;
+ u64_stats_update_end(&priv->stats_tx.syncp);
+
+ /* wake queue for freeing buffer */
+ if (unlikely(netif_queue_stopped(ndev)) && nr_freebuf)
+ netif_wake_queue(ndev);
+
+ return nr_freebuf;
+}
+
+static int ave_rx(struct net_device *ndev, int num)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ struct sk_buff *skb;
+ u32 proc_idx, done_idx, ndesc, cmdsts;
+ int restpkt, npkts;
+ unsigned int pktlen;
+ unsigned int rx_packets = 0;
+ unsigned int rx_bytes = 0;
+
+ proc_idx = priv->rx.proc_idx;
+ done_idx = priv->rx.done_idx;
+ ndesc = priv->rx.ndesc;
+ restpkt = ((proc_idx + ndesc - 1) - done_idx) % ndesc;
+
+ for (npkts = 0; npkts < num; npkts++) {
+ /* we can't receive more packet, so fill desc quickly */
+ if (--restpkt < 0)
+ break;
+
+ cmdsts = ave_rdesc_cmdsts(ndev, AVE_DESCID_RX, proc_idx);
+
+ /* do nothing if owner is HW (==0 for Rx) */
+ if (!(cmdsts & AVE_STS_OWN))
+ break;
+
+ if (!(cmdsts & AVE_STS_OK)) {
+ priv->stats_rx.errors++;
+ proc_idx = (proc_idx + 1) % ndesc;
+ continue;
+ }
+
+ pktlen = cmdsts & AVE_STS_PKTLEN_RX_MASK;
+
+ /* get skbuff for rx */
+ skb = priv->rx.desc[proc_idx].skbs;
+ priv->rx.desc[proc_idx].skbs = NULL;
+
+ ave_dma_unmap(ndev, &priv->rx.desc[proc_idx], DMA_FROM_DEVICE);
+
+ skb->dev = ndev;
+ skb_put(skb, pktlen);
+ skb->protocol = eth_type_trans(skb, ndev);
+
+ if ((cmdsts & AVE_STS_CSSV) && (!(cmdsts & AVE_STS_CSER)))
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+ rx_packets++;
+ rx_bytes += pktlen;
+
+ netif_receive_skb(skb);
+
+ proc_idx = (proc_idx + 1) % ndesc;
+ }
+
+ priv->rx.proc_idx = proc_idx;
+
+ /* update stats */
+ u64_stats_update_begin(&priv->stats_rx.syncp);
+ priv->stats_rx.packets += rx_packets;
+ priv->stats_rx.bytes += rx_bytes;
+ u64_stats_update_end(&priv->stats_rx.syncp);
+
+ /* refill the Rx buffers */
+ while (proc_idx != done_idx) {
+ if (ave_set_rxdesc(ndev, done_idx))
+ break;
+ done_idx = (done_idx + 1) % ndesc;
+ }
+
+ priv->rx.done_idx = done_idx;
+
+ return npkts;
+}
+
+static inline u32 ave_irq_disable_all(struct net_device *ndev)
+{
+ u32 ret;
+
+ ret = ave_r32(ndev, AVE_GIMR);
+ ave_w32(ndev, AVE_GIMR, 0);
+
+ return ret;
+}
+
+static inline void ave_irq_restore(struct net_device *ndev, u32 val)
+{
+ ave_w32(ndev, AVE_GIMR, val);
+}
+
+static inline void ave_irq_enable(struct net_device *ndev, u32 bitflag)
+{
+ ave_w32(ndev, AVE_GIMR, ave_r32(ndev, AVE_GIMR) | bitflag);
+ ave_w32(ndev, AVE_GISR, bitflag);
+}
+
+static inline void ave_irq_disable(struct net_device *ndev, u32 bitflag)
+{
+ ave_w32(ndev, AVE_GIMR, ave_r32(ndev, AVE_GIMR) & ~bitflag);
+}
+
+static void ave_global_reset(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 val;
+
+ /* set config register */
+ val = AVE_CFGR_FLE | AVE_CFGR_IPFCEN | AVE_CFGR_CHE;
+ if (!phy_interface_mode_is_rgmii(priv->phy_mode))
+ val |= AVE_CFGR_MII;
+ ave_w32(ndev, AVE_CFGR, val);
+
+ /* reset RMII register */
+ val = ave_r32(ndev, AVE_RSTCTRL);
+ val &= ~AVE_RSTCTRL_RMIIRST;
+ ave_w32(ndev, AVE_RSTCTRL, val);
+
+ /* assert reset */
+ ave_w32(ndev, AVE_GRR, AVE_GRR_GRST | AVE_GRR_PHYRST);
+ msleep(20);
+
+ /* 1st, negate PHY reset only */
+ ave_w32(ndev, AVE_GRR, AVE_GRR_GRST);
+ msleep(40);
+
+ /* negate reset */
+ ave_w32(ndev, AVE_GRR, 0);
+ msleep(40);
+
+ /* negate RMII register */
+ val = ave_r32(ndev, AVE_RSTCTRL);
+ val |= AVE_RSTCTRL_RMIIRST;
+ ave_w32(ndev, AVE_RSTCTRL, val);
+
+ /* disable all interrupt */
+ ave_irq_disable_all(ndev);
+}
+
+static void ave_rxf_reset(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 rxcr_org;
+
+ /* save and disable MAC receive op */
+ rxcr_org = ave_r32(ndev, AVE_RXCR);
+ ave_w32(ndev, AVE_RXCR, rxcr_org & (~AVE_RXCR_RXEN));
+
+ /* suspend Rx descriptor */
+ ave_desc_switch(ndev, AVE_DESC_RX_SUSPEND);
+
+ /* receive all packets before descriptor starts */
+ ave_rx(ndev, priv->rx.ndesc);
+
+ /* assert reset */
+ ave_w32(ndev, AVE_GRR, AVE_GRR_RXFFR);
+ usleep_range(40, 50);
+
+ /* negate reset */
+ ave_w32(ndev, AVE_GRR, 0);
+ usleep_range(10, 20);
+
+ /* negate interrupt status */
+ ave_w32(ndev, AVE_GISR, AVE_GI_RXOVF);
+
+ /* permit descriptor */
+ ave_desc_switch(ndev, AVE_DESC_RX_PERMIT);
+
+ /* restore MAC reccieve op */
+ ave_w32(ndev, AVE_RXCR, rxcr_org);
+}
+
+static irqreturn_t ave_interrupt(int irq, void *netdev)
+{
+ struct net_device *ndev = (struct net_device *)netdev;
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 gimr_val, gisr_val;
+
+ gimr_val = ave_irq_disable_all(ndev);
+
+ /* get interrupt status */
+ gisr_val = ave_r32(ndev, AVE_GISR);
+
+ /* PHY */
+ if (gisr_val & AVE_GI_PHY)
+ ave_w32(ndev, AVE_GISR, AVE_GI_PHY);
+
+ /* check exceeding packet */
+ if (gisr_val & AVE_GI_RXERR) {
+ ave_w32(ndev, AVE_GISR, AVE_GI_RXERR);
+ netdev_err(ndev, "receive a packet exceeding frame buffer\n");
+ }
+
+ gisr_val &= gimr_val;
+ if (!gisr_val)
+ goto exit_isr;
+
+ /* RxFIFO overflow */
+ if (gisr_val & AVE_GI_RXOVF) {
+ priv->stats_rx.fifo_errors++;
+ ave_rxf_reset(ndev);
+ goto exit_isr;
+ }
+
+ /* Rx drop */
+ if (gisr_val & AVE_GI_RXDROP) {
+ priv->stats_rx.dropped++;
+ ave_w32(ndev, AVE_GISR, AVE_GI_RXDROP);
+ }
+
+ /* Rx interval */
+ if (gisr_val & AVE_GI_RXIINT) {
+ napi_schedule(&priv->napi_rx);
+ /* still force to disable Rx interrupt until NAPI finishes */
+ gimr_val &= ~AVE_GI_RXIINT;
+ }
+
+ /* Tx completed */
+ if (gisr_val & AVE_GI_TX) {
+ napi_schedule(&priv->napi_tx);
+ /* still force to disable Tx interrupt until NAPI finishes */
+ gimr_val &= ~AVE_GI_TX;
+ }
+
+exit_isr:
+ ave_irq_restore(ndev, gimr_val);
+
+ return IRQ_HANDLED;
+}
+
+static int ave_poll_rx(struct napi_struct *napi, int budget)
+{
+ struct ave_private *priv;
+ struct net_device *ndev;
+ int num;
+
+ priv = container_of(napi, struct ave_private, napi_rx);
+ ndev = priv->ndev;
+
+ num = ave_rx(ndev, budget);
+ if (num < budget) {
+ napi_complete_done(napi, num);
+
+ /* enable Rx interrupt when NAPI finishes */
+ ave_irq_enable(ndev, AVE_GI_RXIINT);
+ }
+
+ return num;
+}
+
+static int ave_poll_tx(struct napi_struct *napi, int budget)
+{
+ struct ave_private *priv;
+ struct net_device *ndev;
+ int num;
+
+ priv = container_of(napi, struct ave_private, napi_tx);
+ ndev = priv->ndev;
+
+ num = ave_tx_completion(ndev);
+ napi_complete(napi);
+
+ /* enable Tx interrupt when NAPI finishes */
+ ave_irq_enable(ndev, AVE_GI_TX);
+
+ return num;
+}
+
+static void ave_pfsel_promisc_set(struct net_device *ndev,
+ unsigned int entry, u32 rxring)
+{
+ if (WARN_ON(entry > AVE_PF_SIZE))
+ return;
+
+ ave_pfsel_stop(ndev, entry);
+
+ /* set byte mask */
+ ave_w32(ndev, AVE_PFMBYTE(entry), 0xFFFFFFFF);
+ ave_w32(ndev, AVE_PFMBYTE(entry) + 4, 0xFFFFFFFF);
+
+ /* set bit mask filter */
+ ave_w32(ndev, AVE_PFMBIT(entry), 0x0000FFFF);
+
+ /* set selector to rxring */
+ ave_w32(ndev, AVE_PFSEL(entry), rxring);
+
+ ave_pfsel_start(ndev, entry);
+}
+
+static void ave_pfsel_init(struct net_device *ndev)
+{
+ int i;
+ unsigned char bcast_mac[ETH_ALEN];
+
+ eth_broadcast_addr(bcast_mac);
+
+ /* Stop all filter */
+ for (i = 0; i < AVE_PF_SIZE; i++)
+ ave_pfsel_stop(ndev, i);
+
+ /* promiscious entry, select ring 0 */
+ ave_pfsel_promisc_set(ndev, AVE_PFNUM_FILTER, 0);
+
+ /* unicast entry */
+ ave_pfsel_macaddr_set(ndev, AVE_PFNUM_UNICAST, ndev->dev_addr, 6);
+
+ /* broadcast entry */
+ ave_pfsel_macaddr_set(ndev, AVE_PFNUM_BROADCAST, bcast_mac, 6);
+}
+
+static void ave_adjust_link(struct net_device *ndev)
+{
+ struct phy_device *phydev = ndev->phydev;
+ u32 val, txcr, rxcr, rxcr_org;
+ u16 rmt_adv = 0, lcl_adv = 0;
+ u8 cap;
+
+ /* set RGMII speed */
+ val = ave_r32(ndev, AVE_TXCR);
+ val &= ~(AVE_TXCR_TXSPD_100 | AVE_TXCR_TXSPD_1G);
+
+ if (phy_interface_is_rgmii(phydev) && phydev->speed == SPEED_1000)
+ val |= AVE_TXCR_TXSPD_1G;
+ else if (phydev->speed == SPEED_100)
+ val |= AVE_TXCR_TXSPD_100;
+
+ ave_w32(ndev, AVE_TXCR, val);
+
+ /* set RMII speed (100M/10M only) */
+ if (!phy_interface_is_rgmii(phydev)) {
+ val = ave_r32(ndev, AVE_LINKSEL);
+ if (phydev->speed == SPEED_10)
+ val &= ~AVE_LINKSEL_100M;
+ else
+ val |= AVE_LINKSEL_100M;
+ ave_w32(ndev, AVE_LINKSEL, val);
+ }
+
+ /* check current RXCR/TXCR */
+ rxcr = ave_r32(ndev, AVE_RXCR);
+ txcr = ave_r32(ndev, AVE_TXCR);
+ rxcr_org = rxcr;
+
+ if (phydev->duplex) {
+ rxcr |= AVE_RXCR_FDUPEN;
+
+ if (phydev->pause)
+ rmt_adv |= LPA_PAUSE_CAP;
+ if (phydev->asym_pause)
+ rmt_adv |= LPA_PAUSE_ASYM;
+ if (phydev->advertising & ADVERTISED_Pause)
+ lcl_adv |= ADVERTISE_PAUSE_CAP;
+ if (phydev->advertising & ADVERTISED_Asym_Pause)
+ lcl_adv |= ADVERTISE_PAUSE_ASYM;
+
+ cap = mii_resolve_flowctrl_fdx(lcl_adv, rmt_adv);
+ if (cap & FLOW_CTRL_TX)
+ txcr |= AVE_TXCR_FLOCTR;
+ else
+ txcr &= ~AVE_TXCR_FLOCTR;
+ if (cap & FLOW_CTRL_RX)
+ rxcr |= AVE_RXCR_FLOCTR;
+ else
+ rxcr &= ~AVE_RXCR_FLOCTR;
+ } else {
+ rxcr &= ~AVE_RXCR_FDUPEN;
+ rxcr &= ~AVE_RXCR_FLOCTR;
+ txcr &= ~AVE_TXCR_FLOCTR;
+ }
+
+ if (rxcr_org != rxcr) {
+ /* disable Rx mac */
+ ave_w32(ndev, AVE_RXCR, rxcr & ~AVE_RXCR_RXEN);
+ /* change and enable TX/Rx mac */
+ ave_w32(ndev, AVE_TXCR, txcr);
+ ave_w32(ndev, AVE_RXCR, rxcr);
+ }
+
+ phy_print_status(phydev);
+}
+
+static void ave_macaddr_init(struct net_device *ndev)
+{
+ unsigned char *mac_addr;
+ u32 val;
+
+ /* set macaddr */
+ mac_addr = ndev->dev_addr;
+ val = mac_addr[0] | (mac_addr[1] << 8) | (mac_addr[2] << 16)
+ | (mac_addr[3] << 24);
+ ave_w32(ndev, AVE_RXMAC1R, val);
+ val = mac_addr[4] | (mac_addr[5] << 8);
+ ave_w32(ndev, AVE_RXMAC2R, val);
+
+ /* pfsel unicast entry */
+ ave_pfsel_macaddr_set(ndev, AVE_PFNUM_UNICAST, ndev->dev_addr, 6);
+}
+
+static int ave_init(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ struct device *dev = ndev->dev.parent;
+ struct device_node *np = dev->of_node, *mdio_np;
+ struct ethtool_wolinfo wol = { .cmd = ETHTOOL_GWOL };
+ struct phy_device *phydev;
+ int ret = 0;
+
+ /* enable clk because of hw access until ndo_open */
+ clk_prepare_enable(priv->clk);
+ reset_control_deassert(priv->rst);
+
+ /* reset MAC and PHY */
+ ave_global_reset(ndev);
+
+ /* register MDIO bus */
+ mdio_np = of_get_child_by_name(np, "mdio");
+ if (!mdio_np) {
+ dev_err(dev, "mdio node not found\n");
+ ret = -EINVAL;
+ goto out_clk_disable;
+ }
+ ret = of_mdiobus_register(priv->mdio, mdio_np);
+ of_node_put(mdio_np);
+ if (ret) {
+ dev_err(dev, "failed to register mdiobus\n");
+ goto out_clk_disable;
+ }
+
+ /* attach PHY with MAC */
+ phydev = of_phy_get_and_connect(ndev, np, ave_adjust_link);
+ if (!phydev) {
+ dev_err(dev, "could not attach to PHY\n");
+ ret = -ENODEV;
+ goto out_mdio_unregister;
+ }
+
+ priv->phydev = phydev;
+
+ phy_ethtool_get_wol(phydev, &wol);
+ device_set_wakeup_capable(&ndev->dev, !!wol.supported);
+
+ if (!phy_interface_is_rgmii(phydev)) {
+ phydev->supported &= ~PHY_GBIT_FEATURES;
+ phydev->supported |= PHY_BASIC_FEATURES;
+ }
+ phydev->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
+
+ phy_attached_info(phydev);
+
+ return ret;
+
+out_mdio_unregister:
+ mdiobus_unregister(priv->mdio);
+out_clk_disable:
+ reset_control_assert(priv->rst);
+ clk_disable_unprepare(priv->clk);
+
+ return ret;
+}
+
+static void ave_uninit(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+
+ phy_disconnect(priv->phydev);
+
+ /* disable clk because of hw access after ndo_stop */
+ reset_control_assert(priv->rst);
+ clk_disable_unprepare(priv->clk);
+}
+
+static int ave_open(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ int entry;
+ int ret;
+ u32 val;
+
+ /* request interrupt */
+ ret = request_irq(priv->irq, ave_interrupt, IRQF_SHARED, ndev->name,
+ ndev);
+ if (ret)
+ return ret;
+
+ /* allocate descriptors */
+ priv->tx.desc = kcalloc(priv->tx.ndesc, sizeof(*priv->tx.desc),
+ GFP_KERNEL);
+ if (!priv->tx.desc) {
+ ret = -ENOMEM;
+ goto out_free_irq;
+ }
+
+ priv->rx.desc = kcalloc(priv->rx.ndesc, sizeof(*priv->rx.desc),
+ GFP_KERNEL);
+ if (!priv->rx.desc) {
+ kfree(priv->tx.desc);
+ ret = -ENOMEM;
+ goto out_free_irq;
+ }
+
+ /* initialize Tx work and descriptor */
+ priv->tx.proc_idx = 0;
+ priv->tx.done_idx = 0;
+ for (entry = 0; entry < priv->tx.ndesc; entry++) {
+ ave_wdesc_cmdsts(ndev, AVE_DESCID_TX, entry, 0);
+ ave_wdesc_addr(ndev, AVE_DESCID_TX, entry, 0);
+ }
+ ave_w32(ndev, AVE_TXDC, AVE_TXDC_ADDR_START
+ | (((priv->tx.ndesc * priv->desc_size) << 16) & AVE_TXDC_SIZE));
+
+ /* initialize Rx work and descriptor */
+ priv->rx.proc_idx = 0;
+ priv->rx.done_idx = 0;
+ for (entry = 0; entry < priv->rx.ndesc; entry++) {
+ if (ave_set_rxdesc(ndev, entry))
+ break;
+ }
+ ave_w32(ndev, AVE_RXDC0, AVE_RXDC0_ADDR_START
+ | (((priv->rx.ndesc * priv->desc_size) << 16) & AVE_RXDC0_SIZE));
+
+ /* enable descriptor */
+ ave_desc_switch(ndev, AVE_DESC_START);
+
+ /* initialize filter */
+ ave_pfsel_init(ndev);
+
+ /* initialize macaddr */
+ ave_macaddr_init(ndev);
+
+ /* set Rx configuration */
+ /* full duplex, enable pause drop, enalbe flow control */
+ val = AVE_RXCR_RXEN | AVE_RXCR_FDUPEN | AVE_RXCR_DRPEN |
+ AVE_RXCR_FLOCTR | (AVE_MAX_ETHFRAME & AVE_RXCR_MPSIZ_MASK);
+ ave_w32(ndev, AVE_RXCR, val);
+
+ /* set Tx configuration */
+ /* enable flow control, disable loopback */
+ ave_w32(ndev, AVE_TXCR, AVE_TXCR_FLOCTR);
+
+ /* enable timer, clear EN,INTM, and mask interval unit(BSCK) */
+ val = ave_r32(ndev, AVE_IIRQC) & AVE_IIRQC_BSCK;
+ val |= AVE_IIRQC_EN0 | (AVE_INTM_COUNT << 16);
+ ave_w32(ndev, AVE_IIRQC, val);
+
+ /* set interrupt mask */
+ val = AVE_GI_RXIINT | AVE_GI_RXOVF | AVE_GI_TX;
+ ave_irq_restore(ndev, val);
+
+ napi_enable(&priv->napi_rx);
+ napi_enable(&priv->napi_tx);
+
+ phy_start(ndev->phydev);
+ phy_start_aneg(ndev->phydev);
+ netif_start_queue(ndev);
+
+ return 0;
+
+out_free_irq:
+ disable_irq(priv->irq);
+ free_irq(priv->irq, ndev);
+
+ return ret;
+}
+
+static int ave_stop(struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ int entry;
+
+ /* disable all interrupt */
+ ave_irq_disable_all(ndev);
+ disable_irq(priv->irq);
+ free_irq(priv->irq, ndev);
+
+ netif_tx_disable(ndev);
+ phy_stop(ndev->phydev);
+ napi_disable(&priv->napi_tx);
+ napi_disable(&priv->napi_rx);
+
+ /* disable descriptor */
+ ave_desc_switch(ndev, AVE_DESC_STOP);
+
+ /* free Tx buffer */
+ for (entry = 0; entry < priv->tx.ndesc; entry++) {
+ if (!priv->tx.desc[entry].skbs)
+ continue;
+
+ ave_dma_unmap(ndev, &priv->tx.desc[entry], DMA_TO_DEVICE);
+ dev_kfree_skb_any(priv->tx.desc[entry].skbs);
+ priv->tx.desc[entry].skbs = NULL;
+ }
+ priv->tx.proc_idx = 0;
+ priv->tx.done_idx = 0;
+
+ /* free Rx buffer */
+ for (entry = 0; entry < priv->rx.ndesc; entry++) {
+ if (!priv->rx.desc[entry].skbs)
+ continue;
+
+ ave_dma_unmap(ndev, &priv->rx.desc[entry], DMA_FROM_DEVICE);
+ dev_kfree_skb_any(priv->rx.desc[entry].skbs);
+ priv->rx.desc[entry].skbs = NULL;
+ }
+ priv->rx.proc_idx = 0;
+ priv->rx.done_idx = 0;
+
+ /* free descriptors */
+ kfree(priv->tx.desc);
+ kfree(priv->rx.desc);
+
+ return 0;
+}
+
+static int ave_start_xmit(struct sk_buff *skb, struct net_device *ndev)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ u32 proc_idx, done_idx, ndesc, cmdsts;
+ int freepkt;
+ dma_addr_t paddr;
+
+ proc_idx = priv->tx.proc_idx;
+ done_idx = priv->tx.done_idx;
+ ndesc = priv->tx.ndesc;
+ freepkt = ((done_idx + ndesc - 1) - proc_idx) % ndesc;
+
+ /* stop queue when not enough entry */
+ if (unlikely(freepkt < 1)) {
+ netif_stop_queue(ndev);
+ return NETDEV_TX_BUSY;
+ }
+
+ /* add padding for short packet */
+ if (skb_put_padto(skb, ETH_ZLEN)) {
+ priv->stats_tx.dropped++;
+ return NETDEV_TX_OK;
+ }
+
+ /* map Tx buffer
+ * Tx buffer set to the Tx descriptor doesn't have any restriction.
+ */
+ paddr = ave_dma_map(ndev, &priv->tx.desc[proc_idx],
+ skb->data,
+ skb->len, DMA_TO_DEVICE);
+ if (paddr == -ENOMEM) {
+ dev_kfree_skb_any(skb);
+ priv->stats_tx.dropped++;
+ return NETDEV_TX_OK;
+ }
+
+ priv->tx.desc[proc_idx].skbs = skb;
+
+ /* set buffer address to descriptor */
+ ave_wdesc_addr(ndev, AVE_DESCID_TX, proc_idx, paddr);
+
+ /* set flag and length to send */
+ cmdsts = AVE_STS_OWN | AVE_STS_1ST | AVE_STS_LAST
+ | (skb->len & AVE_STS_PKTLEN_TX_MASK);
+
+ /* set interrupt per AVE_FORCE_TXINTCNT or when queue is stopped */
+ if (!(proc_idx % AVE_FORCE_TXINTCNT) || netif_queue_stopped(ndev))
+ cmdsts |= AVE_STS_INTR;
+
+ /* disable checksum calculation when skb doesn't calurate checksum */
+ if (skb->ip_summed == CHECKSUM_NONE ||
+ skb->ip_summed == CHECKSUM_UNNECESSARY)
+ cmdsts |= AVE_STS_NOCSUM;
+
+ /* set cmdsts */
+ ave_wdesc_cmdsts(ndev, AVE_DESCID_TX, proc_idx, cmdsts);
+
+ priv->tx.proc_idx = (proc_idx + 1) % ndesc;
+
+ return NETDEV_TX_OK;
+}
+
+static int ave_ioctl(struct net_device *ndev, struct ifreq *ifr, int cmd)
+{
+ return phy_mii_ioctl(ndev->phydev, ifr, cmd);
+}
+
+static void ave_set_rx_mode(struct net_device *ndev)
+{
+ int count, mc_cnt = netdev_mc_count(ndev);
+ struct netdev_hw_addr *hw_adr;
+ u32 val;
+ u8 v4multi_macadr[6] = { 0x01, 0x00, 0x00, 0x00, 0x00, 0x00 };
+ u8 v6multi_macadr[6] = { 0x33, 0x00, 0x00, 0x00, 0x00, 0x00 };
+
+ /* MAC addr filter enable for promiscious mode */
+ val = ave_r32(ndev, AVE_RXCR);
+ if (ndev->flags & IFF_PROMISC || !mc_cnt)
+ val &= ~AVE_RXCR_AFEN;
+ else
+ val |= AVE_RXCR_AFEN;
+ ave_w32(ndev, AVE_RXCR, val);
+
+ /* set all multicast address */
+ if ((ndev->flags & IFF_ALLMULTI) || mc_cnt > AVE_PF_MULTICAST_SIZE) {
+ ave_pfsel_macaddr_set(ndev, AVE_PFNUM_MULTICAST,
+ v4multi_macadr, 1);
+ ave_pfsel_macaddr_set(ndev, AVE_PFNUM_MULTICAST + 1,
+ v6multi_macadr, 1);
+ } else {
+ /* stop all multicast filter */
+ for (count = 0; count < AVE_PF_MULTICAST_SIZE; count++)
+ ave_pfsel_stop(ndev, AVE_PFNUM_MULTICAST + count);
+
+ /* set multicast addresses */
+ count = 0;
+ netdev_for_each_mc_addr(hw_adr, ndev) {
+ if (count == mc_cnt)
+ break;
+ ave_pfsel_macaddr_set(ndev, AVE_PFNUM_MULTICAST + count,
+ hw_adr->addr, 6);
+ count++;
+ }
+ }
+}
+
+static void ave_get_stats64(struct net_device *ndev,
+ struct rtnl_link_stats64 *stats)
+{
+ struct ave_private *priv = netdev_priv(ndev);
+ unsigned int start;
+
+ do {
+ start = u64_stats_fetch_begin_irq(&priv->stats_rx.syncp);
+ stats->rx_packets = priv->stats_rx.packets;
+ stats->rx_bytes = priv->stats_rx.bytes;
+ } while (u64_stats_fetch_retry_irq(&priv->stats_rx.syncp, start));
+
+ do {
+ start = u64_stats_fetch_begin_irq(&priv->stats_tx.syncp);
+ stats->tx_packets = priv->stats_tx.packets;
+ stats->tx_bytes = priv->stats_tx.bytes;
+ } while (u64_stats_fetch_retry_irq(&priv->stats_tx.syncp, start));
+
+ stats->rx_errors = priv->stats_rx.errors;
+ stats->tx_errors = priv->stats_tx.errors;
+ stats->rx_dropped = priv->stats_rx.dropped;
+ stats->tx_dropped = priv->stats_tx.dropped;
+ stats->rx_fifo_errors = priv->stats_rx.fifo_errors;
+ stats->collisions = priv->stats_tx.collisions;
+}
+
+static int ave_set_mac_address(struct net_device *ndev, void *p)
+{
+ int ret = eth_mac_addr(ndev, p);
+
+ if (ret)
+ return ret;
+
+ ave_macaddr_init(ndev);
+
+ return 0;
+}
+
+static const struct net_device_ops ave_netdev_ops = {
+ .ndo_init = ave_init,
+ .ndo_uninit = ave_uninit,
+ .ndo_open = ave_open,
+ .ndo_stop = ave_stop,
+ .ndo_start_xmit = ave_start_xmit,
+ .ndo_do_ioctl = ave_ioctl,
+ .ndo_set_rx_mode = ave_set_rx_mode,
+ .ndo_get_stats64 = ave_get_stats64,
+ .ndo_set_mac_address = ave_set_mac_address,
+};
+
+static int ave_probe(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+ struct device_node *np = dev->of_node;
+ u32 ave_id;
+ struct ave_private *priv;
+ const struct ave_soc_data *data;
+ phy_interface_t phy_mode;
+ struct net_device *ndev;
+ struct resource *res;
+ void __iomem *base;
+ int irq, ret = 0;
+ char buf[ETHTOOL_FWVERS_LEN];
+ const void *mac_addr;
+
+ data = of_device_get_match_data(dev);
+ if (WARN_ON(!data))
+ return -EINVAL;
+
+ phy_mode = of_get_phy_mode(np);
+ if (phy_mode < 0) {
+ dev_err(dev, "phy-mode not found\n");
+ return -EINVAL;
+ }
+ if ((!phy_interface_mode_is_rgmii(phy_mode)) &&
+ phy_mode != PHY_INTERFACE_MODE_RMII &&
+ phy_mode != PHY_INTERFACE_MODE_MII) {
+ dev_err(dev, "phy-mode is invalid\n");
+ return -EINVAL;
+ }
+
+ irq = platform_get_irq(pdev, 0);
+ if (irq < 0) {
+ dev_err(dev, "IRQ not found\n");
+ return irq;
+ }
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ base = devm_ioremap_resource(dev, res);
+ if (IS_ERR(base))
+ return PTR_ERR(base);
+
+ /* alloc netdevice */
+ ndev = alloc_etherdev(sizeof(struct ave_private));
+ if (!ndev) {
+ dev_err(dev, "can't allocate ethernet device\n");
+ return -ENOMEM;
+ }
+
+ ndev->netdev_ops = &ave_netdev_ops;
+ ndev->ethtool_ops = &ave_ethtool_ops;
+ SET_NETDEV_DEV(ndev, dev);
+
+ /* support hardware checksum */
+ ndev->features |= (NETIF_F_IP_CSUM | NETIF_F_RXCSUM);
+ ndev->hw_features |= (NETIF_F_IP_CSUM | NETIF_F_RXCSUM);
+
+ ndev->max_mtu = AVE_MAX_ETHFRAME - (ETH_HLEN + ETH_FCS_LEN);
+
+ /* get mac address */
+ mac_addr = of_get_mac_address(np);
+ if (mac_addr)
+ ether_addr_copy(ndev->dev_addr, mac_addr);
+
+ /* if the mac address is invalid, use random mac address */
+ if (!is_valid_ether_addr(ndev->dev_addr)) {
+ eth_hw_addr_random(ndev);
+ dev_warn(dev, "Using random MAC address: %pM\n",
+ ndev->dev_addr);
+ }
+
+ priv = netdev_priv(ndev);
+ priv->base = base;
+ priv->irq = irq;
+ priv->ndev = ndev;
+ priv->msg_enable = netif_msg_init(-1, AVE_DEFAULT_MSG_ENABLE);
+ priv->phy_mode = phy_mode;
+ priv->data = data;
+
+ if (IS_DESC_64BIT(priv)) {
+ priv->desc_size = AVE_DESC_SIZE_64;
+ priv->tx.daddr = AVE_TXDM_64;
+ priv->rx.daddr = AVE_RXDM_64;
+ } else {
+ priv->desc_size = AVE_DESC_SIZE_32;
+ priv->tx.daddr = AVE_TXDM_32;
+ priv->rx.daddr = AVE_RXDM_32;
+ }
+ priv->tx.ndesc = AVE_NR_TXDESC;
+ priv->rx.ndesc = AVE_NR_RXDESC;
+
+ u64_stats_init(&priv->stats_rx.syncp);
+ u64_stats_init(&priv->stats_tx.syncp);
+
+ /* get clock */
+ priv->clk = clk_get(dev, NULL);
+ if (IS_ERR(priv->clk))
+ priv->clk = NULL;
+
+ /* get reset */
+ priv->rst = reset_control_get(dev, NULL);
+ if (IS_ERR(priv->rst))
+ priv->rst = NULL;
+
+ /* create MDIO bus */
+ priv->mdio = devm_mdiobus_alloc(dev);
+ if (!priv->mdio) {
+ ret = -ENOMEM;
+ goto out_free_netdev;
+ }
+ priv->mdio->priv = ndev;
+ priv->mdio->parent = dev;
+ priv->mdio->read = ave_mdiobus_read;
+ priv->mdio->write = ave_mdiobus_write;
+ priv->mdio->name = "uniphier-mdio";
+ snprintf(priv->mdio->id, MII_BUS_ID_SIZE, "%s-%x",
+ pdev->name, pdev->id);
+
+ /* Register as a NAPI supported driver */
+ netif_napi_add(ndev, &priv->napi_rx, ave_poll_rx, priv->rx.ndesc);
+ netif_tx_napi_add(ndev, &priv->napi_tx, ave_poll_tx, priv->tx.ndesc);
+
+ ret = register_netdev(ndev);
+ if (ret) {
+ dev_err(dev, "failed to register netdevice\n");
+ goto out_del_napi;
+ }
+
+ platform_set_drvdata(pdev, ndev);
+
+ /* get ID and version */
+ ave_id = ave_r32(ndev, AVE_IDR);
+ ave_get_fwversion(ndev, buf, sizeof(buf));
+
+ dev_info(dev, "Socionext %c%c%c%c Ethernet IP %s (irq=%d, phy=%s)\n",
+ (ave_id >> 24) & 0xff, (ave_id >> 16) & 0xff,
+ (ave_id >> 8) & 0xff, (ave_id >> 0) & 0xff,
+ buf, priv->irq, phy_modes(phy_mode));
+
+ return 0;
+
+out_del_napi:
+ netif_napi_del(&priv->napi_rx);
+ netif_napi_del(&priv->napi_tx);
+out_free_netdev:
+ free_netdev(ndev);
+
+ return ret;
+}
+
+static int ave_remove(struct platform_device *pdev)
+{
+ struct net_device *ndev = platform_get_drvdata(pdev);
+ struct ave_private *priv = netdev_priv(ndev);
+
+ unregister_netdev(ndev);
+ netif_napi_del(&priv->napi_rx);
+ netif_napi_del(&priv->napi_tx);
+ mdiobus_unregister(priv->mdio);
+ free_netdev(ndev);
+
+ clk_disable_unprepare(priv->clk);
+ clk_put(priv->clk);
+
+ reset_control_assert(priv->rst);
+ reset_control_put(priv->rst);
+
+ return 0;
+}
+
+static const struct ave_soc_data ave_pro4_data = {
+ .is_desc_64bit = false,
+};
+
+static const struct ave_soc_data ave_pxs2_data = {
+ .is_desc_64bit = false,
+};
+
+static const struct ave_soc_data ave_ld20_data = {
+ .is_desc_64bit = true,
+};
+
+static const struct ave_soc_data ave_ld11_data = {
+ .is_desc_64bit = false,
+};
+
+static const struct of_device_id of_ave_match[] = {
+ {
+ .compatible = "socionext,uniphier-pro4-ave4",
+ .data = &ave_pro4_data,
+ },
+ {
+ .compatible = "socionext,uniphier-pxs2-ave4",
+ .data = &ave_pxs2_data,
+ },
+ {
+ .compatible = "socionext,uniphier-ld20-ave4",
+ .data = &ave_ld20_data,
+ },
+ {
+ .compatible = "socionext,uniphier-ld11-ave4",
+ .data = &ave_ld11_data,
+ },
+ { /* Sentinel */ }
+};
+MODULE_DEVICE_TABLE(of, of_ave_match);
+
+static struct platform_driver ave_driver = {
+ .probe = ave_probe,
+ .remove = ave_remove,
+ .driver = {
+ .name = "ave",
+ .of_match_table = of_ave_match,
+ },
+};
+module_platform_driver(ave_driver);
+
+MODULE_DESCRIPTION("Socionext UniPhier AVE ethernet driver");
+MODULE_LICENSE("GPL v2");
--
2.7.4
^ permalink raw reply related
* [next-queue PATCH v7 0/6] TSN: Add qdisc based config interface for CBS
From: Vinicius Costa Gomes @ 2017-10-13 0:39 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Vinicius Costa Gomes, jhs, xiyou.wangcong, jiri, andre.guedes,
ivan.briano, jesus.sanchez-palencia, boon.leong.ong,
richardcochran, henrik, levipearson, rodney.cummings
Hi,
Changes since v6:
- Fixed compilation for 32bit arches;
- Aligned the behaviour of .select_queue() of the mq qdisc to be the
same as mqprio;
Changes since v5:
- Fixed comments from Jiri Pirko;
Changes since v4:
- Added a software implementation of the CBS algorithm;
Changes since v3:
- None, only a clean patchset without old patches;
Changes since v2:
- squashed the patch introducing the userspace API into the patch
implementing CBS;
Changes since v1:
- Solved the mqprio dependency;
- Fixed a mqprio bug, that caused the inner qdisc to have a wrong
dev_queue associated with it;
Changes from the RFC:
- Fixed comments from Henrik Austad;
- Simplified the Qdisc, using the generic implementation of callbacks
where possible;
- Small refactor on the driver (igb) code;
This patchset is a proposal of how the Traffic Control subsystem can
be used to offload the configuration of the Credit Based Shaper
(defined in the IEEE 802.1Q-2014 Section 8.6.8.2) into supported
network devices.
As part of this work, we've assessed previous public discussions
related to TSN enabling: patches from Henrik Austad (Cisco), the
presentation from Eric Mann at Linux Plumbers 2012, patches from
Gangfeng Huang (National Instruments) and the current state of the
OpenAVNU project (https://github.com/AVnu/OpenAvnu/).
Overview
========
Time-sensitive Networking (TSN) is a set of standards that aim to
address resources availability for providing bandwidth reservation and
bounded latency on Ethernet based LANs. The proposal described here
aims to cover mainly what is needed to enable the following standards:
802.1Qat and 802.1Qav.
The initial target of this work is the Intel i210 NIC, but other
controllers' datasheet were also taken into account, like the Renesas
RZ/A1H RZ/A1M group and the Synopsis DesignWare Ethernet QoS
controller.
Proposal
========
Feature-wise, what is covered here is the configuration interfaces for
HW implementations of the Credit-Based shaper (CBS, 802.1Qav). CBS is
a per-queue shaper. Given that this feature is related to traffic
shaping, and that the traffic control subsystem already provides a
queueing discipline that offloads config into the device driver (i.e.
mqprio), designing a new qdisc for the specific purpose of offloading
the config for the CBS shaper seemed like a good fit.
For steering traffic into the correct queues, we use the socket option
SO_PRIORITY and then a mechanism to map priority to traffic classes /
Tx queues. The qdisc mqprio is currently used in our tests.
As for the CBS config interface, this patchset is proposing a new
qdisc called 'cbs'. Its 'tc' cmd line is:
$ tc qdisc add dev IFACE parent ID cbs locredit N hicredit M sendslope S \
idleslope I
Note that the parameters for this qdisc are the ones defined by the
802.1Q-2014 spec, so no hardware specific functionality is exposed here.
Per-stream shaping, as defined by IEEE 802.1Q-2014 Section 34.6.1, is
not yet covered by this proposal.
Testing this RFC
================
Attached to this cover letter are:
- calculate_cbs_params.py: A Python script to calculate the
parameters to the CBS queueing discipline;
- tsn-talker.c: A sample C implementation of the talker side of a stream;
- tsn-listener.c: A sample C implementation of the listener side of a
stream;
For testing the patches of this series, you may want to use the
attached samples to this cover letter and use the 'mqprio' qdisc to
setup the priorities to Tx queues mapping, together with the 'cbs'
qdisc to configure the HW shaper of the i210 controller:
1) Setup priorities to traffic classes to hardware queues mapping
$ tc qdisc replace dev ens4 handle 100: parent root mqprio num_tc 3 \
map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 0
For a more detailed explanation, see mqprio(8), in short, this command
will map traffic with priority 3 to the hardware queue 0, traffic with
priority 2 to hardware queue 1, and the rest will be mapped to
hardware queues 2 and 3.
2) Check scheme. You want to get the inner qdiscs ID from the bottom up
$ tc -g class show dev ens4
Ex.:
+---(100:3) mqprio
| +---(100:6) mqprio
| +---(100:7) mqprio
|
+---(100:2) mqprio
| +---(100:5) mqprio
|
+---(100:1) mqprio
+---(100:4) mqprio
* Here '100:4' is Tx Queue #0 and '100:5' is Tx Queue #1.
3) Calculate CBS parameters for classes A and B. i.e. BW for A is 20Mbps and
for B is 10Mbps:
$ calc_cbs_params.py -A 20000 -a 1500 -B 10000 -b 1500
4) Configure CBS for traffic class A (priority 3) as provided by the script:
$ tc qdisc replace dev ens4 parent 100:4 cbs locredit -1470 \
hicredit 30 sendslope -980000 idleslope 20000
5) Configure CBS for traffic class B (priority 2):
$ tc qdisc replace dev ens4 parent 100:5 cbs \
locredit -1485 hicredit 31 sendslope -990000 idleslope 10000
6) Run Listener:
$ ./tsn-listener -d 01:AA:AA:AA:AA:AA -i ens4 -s 1500
7) Run Talker for class A (prio 3 here), compiled from samples/tsn/talker.c
$ ./tsn-talker -d 01:AA:AA:AA:AA:AA -i ens4 -p 3 -s 1500
* The bandwidth displayed on the listener output at this stage should be very
close to the one configured for class A.
8) You can also run a Talker for class B (prio 2 here and using a
different address):
$ ./tsn-talker -d 01:BB:BB:BB:BB:BB -i ens4 -p 2 -s 1500
Authors
=======
- Andre Guedes <andre.guedes@intel.com>
- Ivan Briano <ivan.briano@intel.com>
- Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
- Vinicius Gomes <vinicius.gomes@intel.com>
Andre Guedes (1):
igb: Add support for CBS offload
Jesus Sanchez-Palencia (3):
net/sched: Check for null dev_queue on create flow
net/sched: Change behavior of mq select_queue()
net/sched: Add select_queue() class_ops for mqprio
Vinicius Costa Gomes (2):
net/sched: Introduce Credit Based Shaper (CBS) qdisc
net/sched: Add support for HW offloading for CBS
drivers/net/ethernet/intel/igb/e1000_defines.h | 23 ++
drivers/net/ethernet/intel/igb/e1000_regs.h | 8 +
drivers/net/ethernet/intel/igb/igb.h | 6 +
drivers/net/ethernet/intel/igb/igb_main.c | 347 ++++++++++++++++++++++
include/linux/netdevice.h | 1 +
include/net/pkt_sched.h | 9 +
include/uapi/linux/pkt_sched.h | 18 ++
net/sched/Kconfig | 11 +
net/sched/Makefile | 1 +
net/sched/sch_cbs.c | 394 +++++++++++++++++++++++++
net/sched/sch_generic.c | 8 +-
net/sched/sch_mq.c | 10 +-
net/sched/sch_mqprio.c | 7 +
13 files changed, 833 insertions(+), 10 deletions(-)
create mode 100644 net/sched/sch_cbs.c
Annex: Sample files
===================
calc_cbs_params.py
--8<---------------cut here---------------start------------->8---
#!/usr/bin/env python
#
# Copyright (c) 2017, Intel Corporation
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright notice,
# this list of conditions and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright
# notice, this list of conditions and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
# * Neither the name of Intel Corporation nor the names of its contributors
# may be used to endorse or promote products derived from this software
# without specific prior written permission.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
import argparse
import math
def print_cbs_params_for_class_a(args):
idleslope = args.idleslope_a
sendslope = idleslope - args.link_speed
# According to 802.1Q-2014 spec, Annex L, hiCredit and
# loCredit for SR class A are calculated following the
# equations L-10 and L-12, respectively.
hicredit = math.ceil(idleslope * args.frame_non_sr / args.link_speed)
locredit = math.ceil(sendslope * args.frame_a / args.link_speed)
print("tc qdisc add dev <IFNAME> parent <QDISC-ID> cbs idleslope %d sendslope %d hicredit %d locredit %d" % \
(idleslope, sendslope, hicredit, locredit))
def print_cbs_params_for_class_b(args):
idleslope = args.idleslope_b
sendslope = idleslope - args.link_speed
# Annex L doesn't present a straightforward equation to
# calculate hiCredit for Class B so we have to derive it
# based on generic equations presented in that Annex.
#
# L-3 is the primary equation to calculate hiCredit. Section
# L.2 states that the 'maxInterferenceSize' for SR class B
# is the maximum burst size for SR class A plus the
# maxInterferenceSize from SR class A (which is equal to the
# maximum frame from non-SR traffic).
#
# The maximum burst size for SR class A equation is shown in
# L-16. Merging L-16 into L-3 we get the resulting equation
# which calculates hiCredit B (refer to section L.3 in case
# you're not familiar with the legend):
#
# hiCredit B = Rb * ( Mo Ma )
# ---------- + ------
# Ro - Ra Ro
#
hicredit = math.ceil(idleslope * \
((args.frame_non_sr / (args.link_speed - args.idleslope_a)) + \
(args.frame_a / args.link_speed)))
# loCredit B is calculated following equation L-2.
locredit = math.ceil(sendslope * args.frame_b / args.link_speed)
print("tc qdisc add dev <IFNAME> parent <QDISC-ID> cbs idleslope %d sendslope %d hicredit %d locredit %d" % \
(idleslope, sendslope, hicredit, locredit))
def main():
parser = argparse.ArgumentParser()
parser.add_argument('-S', dest='link_speed', default=1000000.0, type=float,
help='Link speed in kbps')
parser.add_argument('-s', dest='frame_non_sr', default=1500.0, type=float,
help='Maximum frame size from non-SR traffic (MTU size'
'usually')
parser.add_argument('-A', dest='idleslope_a', default=0, type=float,
help='Idleslope for SR class A in kbps')
parser.add_argument('-a', dest='frame_a', default=0, type=float,
help='Maximum frame size for SR class A traffic')
parser.add_argument('-B', dest='idleslope_b', default=0, type=float,
help='Idleslope for SR class B in kbps')
parser.add_argument('-b', dest='frame_b', default=0, type=float,
help='Maximum frame size for SR class B traffic')
args = parser.parse_args()
if args.idleslope_a > 0:
print_cbs_params_for_class_a(args)
if args.idleslope_b > 0:
print_cbs_params_for_class_b(args)
if __name__ == "__main__":
main()
--8<---------------cut here---------------end--------------->8---
tsn-talker.c
--8<---------------cut here---------------start------------->8---
/*
* Copyright (c) 2017, Intel Corporation
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Intel Corporation nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <alloca.h>
#include <argp.h>
#include <arpa/inet.h>
#include <inttypes.h>
#include <linux/if.h>
#include <linux/if_ether.h>
#include <linux/if_packet.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/ioctl.h>
#include <unistd.h>
#define MAGIC 0xCC
static uint8_t ifname[IFNAMSIZ];
static uint8_t macaddr[ETH_ALEN];
static int priority = -1;
static size_t size = 1500;
static uint64_t seq;
static int delay = -1;
static struct argp_option options[] = {
{"dst-addr", 'd', "MACADDR", 0, "Stream Destination MAC address" },
{"delay", 'D', "NUM", 0, "Delay (in us) between packet transmission" },
{"ifname", 'i', "IFNAME", 0, "Network Interface" },
{"prio", 'p', "NUM", 0, "SO_PRIORITY to be set in socket" },
{"packet-size", 's', "NUM", 0, "Size of packets to be transmitted" },
{ 0 }
};
static error_t parser(int key, char *arg, struct argp_state *state)
{
int res;
switch (key) {
case 'd':
res = sscanf(arg, "%hhx:%hhx:%hhx:%hhx:%hhx:%hhx",
&macaddr[0], &macaddr[1], &macaddr[2],
&macaddr[3], &macaddr[4], &macaddr[5]);
if (res != 6) {
printf("Invalid address\n");
exit(EXIT_FAILURE);
}
break;
case 'D':
delay = atoi(arg);
break;
case 'i':
strncpy(ifname, arg, sizeof(ifname) - 1);
break;
case 'p':
priority = atoi(arg);
break;
case 's':
size = atoi(arg);
break;
}
return 0;
}
static struct argp argp = { options, parser };
int main(int argc, char *argv[])
{
int fd, res;
struct ifreq req;
uint8_t *data;
struct sockaddr_ll sk_addr = {
.sll_family = AF_PACKET,
.sll_protocol = htons(ETH_P_TSN),
.sll_halen = ETH_ALEN,
};
argp_parse(&argp, argc, argv, 0, NULL, NULL);
fd = socket(AF_PACKET, SOCK_DGRAM, htons(ETH_P_TSN));
if (fd < 0) {
perror("Couldn't open socket");
return 1;
}
strncpy(req.ifr_name, ifname, sizeof(req.ifr_name));
res = ioctl(fd, SIOCGIFINDEX, &req);
if (res < 0) {
perror("Couldn't get interface index");
goto err;
}
sk_addr.sll_ifindex = req.ifr_ifindex;
memcpy(&sk_addr.sll_addr, macaddr, ETH_ALEN);
if (priority != -1) {
res = setsockopt(fd, SOL_SOCKET, SO_PRIORITY, &priority,
sizeof(priority));
if (res < 0) {
perror("Couldn't set priority");
goto err;
}
}
data = alloca(size);
memset(data, MAGIC, size);
printf("Sending packets...\n");
while (1) {
uint64_t *seq_ptr = (uint64_t *) &data[0];
ssize_t n;
*seq_ptr = seq++;
n = sendto(fd, data, size, 0, (struct sockaddr *) &sk_addr,
sizeof(sk_addr));
if (n < 0)
perror("Failed to send data");
if (delay > 0)
usleep(delay);
}
close(fd);
return 0;
err:
close(fd);
return 1;
}
--8<---------------cut here---------------end--------------->8---
tsn-listener.c
--8<---------------cut here---------------start------------->8---
/*
* Copyright (c) 2017, Intel Corporation
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* * Redistributions of source code must retain the above copyright notice,
* this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of Intel Corporation nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
* FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
* COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
* STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
* OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <alloca.h>
#include <argp.h>
#include <arpa/inet.h>
#include <inttypes.h>
#include <linux/if.h>
#include <linux/if_ether.h>
#include <linux/if_packet.h>
#include <poll.h>
#include <stdbool.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/ioctl.h>
#include <sys/timerfd.h>
#include <unistd.h>
static uint8_t ifname[IFNAMSIZ];
static uint8_t macaddr[ETH_ALEN];
static uint64_t data_count;
static int size = 1500;
static time_t interval = 1;
static bool check_seq = false;
static uint64_t expected_seq;
static struct argp_option options[] = {
{"check-seq", 'c', NULL, 0, "Check sequence number within packet" },
{"dst-addr", 'd', "MACADDR", 0, "Stream Destination MAC address" },
{"ifname", 'i', "IFNAME", 0, "Network Interface" },
{"interval", 'I', "SEC", 0, "Interval between bandwidth reports" },
{"packet-size", 's', "NUM", 0, "Expected packet size" },
{ 0 }
};
static error_t parser(int key, char *arg, struct argp_state *state)
{
int res;
switch (key) {
case 'c':
check_seq = true;
break;
case 'd':
res = sscanf(arg, "%hhx:%hhx:%hhx:%hhx:%hhx:%hhx",
&macaddr[0], &macaddr[1], &macaddr[2],
&macaddr[3], &macaddr[4], &macaddr[5]);
if (res != 6) {
printf("Invalid address\n");
exit(EXIT_FAILURE);
}
break;
case 'i':
strncpy(ifname, arg, sizeof(ifname) - 1);
break;
case 'I':
interval = atoi(arg);
break;
case 's':
size = atoi(arg);
break;
}
return 0;
}
static struct argp argp = { options, parser };
static int setup_timer(void)
{
int fd, res;
struct itimerspec tspec = { 0 };
fd = timerfd_create(CLOCK_MONOTONIC, 0);
if (fd < 0) {
perror("Couldn't create timer");
return -1;
}
tspec.it_value.tv_sec = interval;
tspec.it_interval.tv_sec = interval;
res = timerfd_settime(fd, 0, &tspec, NULL);
if (res < 0) {
perror("Couldn't set timer");
close(fd);
return -1;
}
return fd;
}
static int setup_socket(void)
{
int fd, res;
struct sockaddr_ll sk_addr = {
.sll_family = AF_PACKET,
.sll_protocol = htons(ETH_P_TSN),
};
fd = socket(AF_PACKET, SOCK_DGRAM, htons(ETH_P_TSN));
if (fd < 0) {
perror("Couldn't open socket");
return -1;
}
/* If user provided a network interface, bind() to it. */
if (ifname[0] != '\0') {
struct ifreq req;
strncpy(req.ifr_name, ifname, sizeof(req.ifr_name));
res = ioctl(fd, SIOCGIFINDEX, &req);
if (res < 0) {
perror("Couldn't get interface index");
goto err;
}
sk_addr.sll_ifindex = req.ifr_ifindex;
res = bind(fd, (struct sockaddr *) &sk_addr, sizeof(sk_addr));
if (res < 0) {
perror("Couldn't bind() to interface");
goto err;
}
}
/* If user provided the stream destination address, set it as multicast
* address.
*/
if (macaddr[0] != '\0') {
struct packet_mreq mreq;
mreq.mr_ifindex = sk_addr.sll_ifindex;
mreq.mr_type = PACKET_MR_MULTICAST;
mreq.mr_alen = ETH_ALEN;
memcpy(&mreq.mr_address, macaddr, ETH_ALEN);
res = setsockopt(fd, SOL_PACKET, PACKET_ADD_MEMBERSHIP,
&mreq, sizeof(struct packet_mreq));
if (res < 0) {
perror("Couldn't set PACKET_ADD_MEMBERSHIP");
goto err;
}
}
return fd;
err:
close(fd);
return -1;
}
static void recv_packet(int fd)
{
uint8_t *data = alloca(size);
ssize_t n = recv(fd, data, size, 0);
if (n < 0) {
perror("Failed to receive data");
return;
}
if (n != size)
printf("Size mismatch: expected %d, got %d\n", size, n);
if (check_seq) {
uint64_t *seq = (uint64_t *) &data[0];
/* If 'expected_seq' is equal to zero, it means this is the
* first packet we received so we don't know what sequence
* number to expect.
*/
if (expected_seq == 0)
expected_seq = *seq;
if (*seq != expected_seq) {
printf("Sequence mismatch: expected %llu, got %llu\n",
expected_seq, *seq);
expected_seq = *seq;
}
expected_seq++;
}
data_count += n;
}
static void report_bw(int fd)
{
uint64_t expirations;
ssize_t n = read(fd, &expirations, sizeof(uint64_t));
if (n < 0) {
perror("Couldn't read timerfd");
return;
}
if (expirations != 1)
printf("Some went wrong with timerfd\n");
printf("Receiving data rate: %llu kbps\n", (data_count * 8) / (1000 * interval));
data_count = 0;
}
int main(int argc, char *argv[])
{
int sk_fd, timer_fd, res;
struct pollfd fds[2];
argp_parse(&argp, argc, argv, 0, NULL, NULL);
sk_fd = setup_socket();
if (sk_fd < 0)
return 1;
timer_fd = setup_timer();
if (timer_fd < 0) {
close(sk_fd);
return 1;
}
fds[0].fd = sk_fd;
fds[0].events = POLLIN;
fds[1].fd = timer_fd;
fds[1].events = POLLIN;
printf("Waiting for packets...\n");
while (1) {
res = poll(fds, 2, -1);
if (res < 0) {
perror("Error on poll()");
goto err;
}
if (fds[0].revents & POLLIN)
recv_packet(fds[0].fd);
if (fds[1].revents & POLLIN) {
report_bw(fds[1].fd);
}
}
close(timer_fd);
close(sk_fd);
return 0;
err:
close(timer_fd);
close(sk_fd);
return 1;
}
--8<---------------cut here---------------end--------------->8---
^ permalink raw reply
* [next-queue PATCH v7 2/6] net/sched: Change behavior of mq select_queue()
From: Vinicius Costa Gomes @ 2017-10-13 0:40 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Jesus Sanchez-Palencia, jhs, xiyou.wangcong, jiri, andre.guedes,
ivan.briano, boon.leong.ong, richardcochran, henrik, levipearson,
rodney.cummings
In-Reply-To: <20171013004005.17416-1-vinicius.gomes@intel.com>
From: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
Currently, the class_ops select_queue() implementation on sch_mq
returns a pointer to netdev_queue #0 when it receives and invalid
qdisc id. That can be misleading since all of mq's inner qdiscs are
attached to a valid netdev_queue.
Here we fix that by returning NULL when a qdisc id is invalid. This is
aligned with how select_queue() is implemented for sch_mqprio in the
next patch on this series, keeping a consistent behavior between these
two qdiscs.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
---
net/sched/sch_mq.c | 10 +---------
1 file changed, 1 insertion(+), 9 deletions(-)
diff --git a/net/sched/sch_mq.c b/net/sched/sch_mq.c
index f3a3e507422b..213b586a06a0 100644
--- a/net/sched/sch_mq.c
+++ b/net/sched/sch_mq.c
@@ -130,15 +130,7 @@ static struct netdev_queue *mq_queue_get(struct Qdisc *sch, unsigned long cl)
static struct netdev_queue *mq_select_queue(struct Qdisc *sch,
struct tcmsg *tcm)
{
- unsigned int ntx = TC_H_MIN(tcm->tcm_parent);
- struct netdev_queue *dev_queue = mq_queue_get(sch, ntx);
-
- if (!dev_queue) {
- struct net_device *dev = qdisc_dev(sch);
-
- return netdev_get_tx_queue(dev, 0);
- }
- return dev_queue;
+ return mq_queue_get(sch, TC_H_MIN(tcm->tcm_parent));
}
static int mq_graft(struct Qdisc *sch, unsigned long cl, struct Qdisc *new,
--
2.14.2
^ permalink raw reply related
* [next-queue PATCH v7 1/6] net/sched: Check for null dev_queue on create flow
From: Vinicius Costa Gomes @ 2017-10-13 0:40 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Jesus Sanchez-Palencia, jhs, xiyou.wangcong, jiri, andre.guedes,
ivan.briano, boon.leong.ong, richardcochran, henrik, levipearson,
rodney.cummings
In-Reply-To: <20171013004005.17416-1-vinicius.gomes@intel.com>
From: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
In qdisc_alloc() the dev_queue pointer was used without any checks
being performed. If qdisc_create() gets a null dev_queue pointer, it
just passes it along to qdisc_alloc(), leading to a crash. That
happens if a root qdisc implements select_queue() and returns a null
dev_queue pointer for an "invalid handle", for example, or if the
dev_queue associated with the parent qdisc is null.
This patch is in preparation for the next in this series, where
select_queue() is being added to mqprio and as it may return a null
dev_queue.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
---
net/sched/sch_generic.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index a0a198768aad..de2408f1ccd3 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -603,8 +603,14 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
struct Qdisc *sch;
unsigned int size = QDISC_ALIGN(sizeof(*sch)) + ops->priv_size;
int err = -ENOBUFS;
- struct net_device *dev = dev_queue->dev;
+ struct net_device *dev;
+
+ if (!dev_queue) {
+ err = -EINVAL;
+ goto errout;
+ }
+ dev = dev_queue->dev;
p = kzalloc_node(size, GFP_KERNEL,
netdev_queue_numa_node_read(dev_queue));
--
2.14.2
^ permalink raw reply related
* [next-queue PATCH v7 4/6] net/sched: Introduce Credit Based Shaper (CBS) qdisc
From: Vinicius Costa Gomes @ 2017-10-13 0:40 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Vinicius Costa Gomes, jhs, xiyou.wangcong, jiri, andre.guedes,
ivan.briano, jesus.sanchez-palencia, boon.leong.ong,
richardcochran, henrik, levipearson, rodney.cummings
In-Reply-To: <20171013004005.17416-1-vinicius.gomes@intel.com>
This queueing discipline implements the shaper algorithm defined by
the 802.1Q-2014 Section 8.6.8.2 and detailed in Annex L.
It's primary usage is to apply some bandwidth reservation to user
defined traffic classes, which are mapped to different queues via the
mqprio qdisc.
Only a simple software implementation is added for now.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
---
include/uapi/linux/pkt_sched.h | 18 +++
net/sched/Kconfig | 11 ++
net/sched/Makefile | 1 +
net/sched/sch_cbs.c | 314 +++++++++++++++++++++++++++++++++++++++++
4 files changed, 344 insertions(+)
create mode 100644 net/sched/sch_cbs.c
diff --git a/include/uapi/linux/pkt_sched.h b/include/uapi/linux/pkt_sched.h
index 099bf5528fed..41e349df4bf4 100644
--- a/include/uapi/linux/pkt_sched.h
+++ b/include/uapi/linux/pkt_sched.h
@@ -871,4 +871,22 @@ struct tc_pie_xstats {
__u32 maxq; /* maximum queue size */
__u32 ecn_mark; /* packets marked with ecn*/
};
+
+/* CBS */
+struct tc_cbs_qopt {
+ __u8 offload;
+ __s32 hicredit;
+ __s32 locredit;
+ __s32 idleslope;
+ __s32 sendslope;
+};
+
+enum {
+ TCA_CBS_UNSPEC,
+ TCA_CBS_PARMS,
+ __TCA_CBS_MAX,
+};
+
+#define TCA_CBS_MAX (__TCA_CBS_MAX - 1)
+
#endif
diff --git a/net/sched/Kconfig b/net/sched/Kconfig
index e70ed26485a2..c03d86a7775e 100644
--- a/net/sched/Kconfig
+++ b/net/sched/Kconfig
@@ -172,6 +172,17 @@ config NET_SCH_TBF
To compile this code as a module, choose M here: the
module will be called sch_tbf.
+config NET_SCH_CBS
+ tristate "Credit Based Shaper (CBS)"
+ ---help---
+ Say Y here if you want to use the Credit Based Shaper (CBS) packet
+ scheduling algorithm.
+
+ See the top of <file:net/sched/sch_cbs.c> for more details.
+
+ To compile this code as a module, choose M here: the
+ module will be called sch_cbs.
+
config NET_SCH_GRED
tristate "Generic Random Early Detection (GRED)"
---help---
diff --git a/net/sched/Makefile b/net/sched/Makefile
index 7b915d226de7..80c8f92d162d 100644
--- a/net/sched/Makefile
+++ b/net/sched/Makefile
@@ -52,6 +52,7 @@ obj-$(CONFIG_NET_SCH_FQ_CODEL) += sch_fq_codel.o
obj-$(CONFIG_NET_SCH_FQ) += sch_fq.o
obj-$(CONFIG_NET_SCH_HHF) += sch_hhf.o
obj-$(CONFIG_NET_SCH_PIE) += sch_pie.o
+obj-$(CONFIG_NET_SCH_CBS) += sch_cbs.o
obj-$(CONFIG_NET_CLS_U32) += cls_u32.o
obj-$(CONFIG_NET_CLS_ROUTE4) += cls_route.o
diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c
new file mode 100644
index 000000000000..0643587e6dc8
--- /dev/null
+++ b/net/sched/sch_cbs.c
@@ -0,0 +1,314 @@
+/*
+ * net/sched/sch_cbs.c Credit Based Shaper
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Authors: Vinicius Costa Gomes <vinicius.gomes@intel.com>
+ *
+ */
+
+/* Credit Based Shaper (CBS)
+ * =========================
+ *
+ * This is a simple rate-limiting shaper aimed at TSN applications on
+ * systems with known traffic workloads.
+ *
+ * Its algorithm is defined by the IEEE 802.1Q-2014 Specification,
+ * Section 8.6.8.2, and explained in more detail in the Annex L of the
+ * same specification.
+ *
+ * There are four tunables to be considered:
+ *
+ * 'idleslope': Idleslope is the rate of credits that is
+ * accumulated (in kilobits per second) when there is at least
+ * one packet waiting for transmission. Packets are transmitted
+ * when the current value of credits is equal or greater than
+ * zero. When there is no packet to be transmitted the amount of
+ * credits is set to zero. This is the main tunable of the CBS
+ * algorithm.
+ *
+ * 'sendslope':
+ * Sendslope is the rate of credits that is depleted (it should be a
+ * negative number of kilobits per second) when a transmission is
+ * ocurring. It can be calculated as follows, (IEEE 802.1Q-2014 Section
+ * 8.6.8.2 item g):
+ *
+ * sendslope = idleslope - port_transmit_rate
+ *
+ * 'hicredit': Hicredit defines the maximum amount of credits (in
+ * bytes) that can be accumulated. Hicredit depends on the
+ * characteristics of interfering traffic,
+ * 'max_interference_size' is the maximum size of any burst of
+ * traffic that can delay the transmission of a frame that is
+ * available for transmission for this traffic class, (IEEE
+ * 802.1Q-2014 Annex L, Equation L-3):
+ *
+ * hicredit = max_interference_size * (idleslope / port_transmit_rate)
+ *
+ * 'locredit': Locredit is the minimum amount of credits that can
+ * be reached. It is a function of the traffic flowing through
+ * this qdisc (IEEE 802.1Q-2014 Annex L, Equation L-2):
+ *
+ * locredit = max_frame_size * (sendslope / port_transmit_rate)
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/skbuff.h>
+#include <net/netlink.h>
+#include <net/sch_generic.h>
+#include <net/pkt_sched.h>
+
+#define BYTES_PER_KBIT (1000 / 8)
+
+struct cbs_sched_data {
+ s64 port_rate; /* in bytes/s */
+ s64 last; /* timestamp in ns */
+ s64 credits; /* in bytes */
+ s32 locredit; /* in bytes */
+ s32 hicredit; /* in bytes */
+ s32 sendslope; /* in kbits/s */
+ s32 idleslope; /* in kbits/s */
+ struct qdisc_watchdog watchdog;
+ int (*enqueue)(struct sk_buff *skb, struct Qdisc *sch);
+ struct sk_buff *(*dequeue)(struct Qdisc *sch);
+};
+
+static int cbs_enqueue_soft(struct sk_buff *skb, struct Qdisc *sch)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+
+ if (sch->q.qlen == 0 && q->credits > 0) {
+ /* We need to stop accumulating credits when there's
+ * no packet enqueued packets and q->credits is
+ * positive.
+ */
+ q->credits = 0;
+ q->last = ktime_get_ns();
+ }
+
+ return qdisc_enqueue_tail(skb, sch);
+}
+
+static int cbs_enqueue(struct sk_buff *skb, struct Qdisc *sch,
+ struct sk_buff **to_free)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+
+ return q->enqueue(skb, sch);
+}
+
+/* timediff is in ns, slope is in kbps */
+static s64 timediff_to_credits(s64 timediff, s32 slope)
+{
+ s64 credits = timediff * slope * BYTES_PER_KBIT;
+
+ do_div(credits, NSEC_PER_SEC);
+
+ return credits;
+}
+
+static s64 delay_from_credits(s64 credits, s32 slope)
+{
+ s64 rate = slope * BYTES_PER_KBIT;
+ s64 delay;
+
+ if (unlikely(rate == 0))
+ return S64_MAX;
+
+ delay = -credits * NSEC_PER_SEC;
+ do_div(delay, rate);
+
+ return delay;
+}
+
+static s64 credits_from_len(unsigned int len, s32 slope, s64 port_rate)
+{
+ /* As do_div() only works on unsigned quantities, convert
+ * slope to a positive number here, and credits to a negative
+ * number before returning.
+ */
+ s64 rate = -slope * BYTES_PER_KBIT;
+ s64 credits;
+
+ if (unlikely(port_rate == 0))
+ return S64_MAX;
+
+ credits = len * rate;
+ do_div(credits, port_rate);
+
+ return -credits;
+}
+
+static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+ s64 now = ktime_get_ns();
+ struct sk_buff *skb;
+ s64 credits;
+ int len;
+
+ if (q->credits < 0) {
+ credits = timediff_to_credits(now - q->last, q->idleslope);
+
+ credits = q->credits + credits;
+ q->credits = min_t(s64, credits, q->hicredit);
+
+ if (q->credits < 0) {
+ s64 delay;
+
+ delay = delay_from_credits(q->credits, q->idleslope);
+ qdisc_watchdog_schedule_ns(&q->watchdog, now + delay);
+
+ q->last = now;
+
+ return NULL;
+ }
+ }
+
+ skb = qdisc_dequeue_head(sch);
+ if (!skb)
+ return NULL;
+
+ len = qdisc_pkt_len(skb);
+
+ /* As sendslope is a negative number, this will decrease the
+ * amount of q->credits.
+ */
+ credits = credits_from_len(len, q->sendslope, q->port_rate);
+ credits += q->credits;
+
+ q->credits = max_t(s64, credits, q->locredit);
+ q->last = now;
+
+ return skb;
+}
+
+static struct sk_buff *cbs_dequeue(struct Qdisc *sch)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+
+ return q->dequeue(sch);
+}
+
+static const struct nla_policy cbs_policy[TCA_CBS_MAX + 1] = {
+ [TCA_CBS_PARMS] = { .len = sizeof(struct tc_cbs_qopt) },
+};
+
+static int cbs_change(struct Qdisc *sch, struct nlattr *opt)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+ struct net_device *dev = qdisc_dev(sch);
+ struct nlattr *tb[TCA_CBS_MAX + 1];
+ struct ethtool_link_ksettings ecmd;
+ struct tc_cbs_qopt *qopt;
+ s64 link_speed;
+ int err;
+
+ err = nla_parse_nested(tb, TCA_CBS_MAX, opt, cbs_policy, NULL);
+ if (err < 0)
+ return err;
+
+ if (!tb[TCA_CBS_PARMS])
+ return -EINVAL;
+
+ qopt = nla_data(tb[TCA_CBS_PARMS]);
+
+ if (qopt->offload)
+ return -EOPNOTSUPP;
+
+ if (!__ethtool_get_link_ksettings(dev, &ecmd))
+ link_speed = ecmd.base.speed;
+ else
+ link_speed = SPEED_1000;
+
+ q->port_rate = link_speed * 1000 * BYTES_PER_KBIT;
+
+ q->enqueue = cbs_enqueue_soft;
+ q->dequeue = cbs_dequeue_soft;
+
+ q->hicredit = qopt->hicredit;
+ q->locredit = qopt->locredit;
+ q->idleslope = qopt->idleslope;
+ q->sendslope = qopt->sendslope;
+
+ return 0;
+}
+
+static int cbs_init(struct Qdisc *sch, struct nlattr *opt)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+
+ if (!opt)
+ return -EINVAL;
+
+ qdisc_watchdog_init(&q->watchdog, sch);
+
+ return cbs_change(sch, opt);
+}
+
+static void cbs_destroy(struct Qdisc *sch)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+
+ qdisc_watchdog_cancel(&q->watchdog);
+}
+
+static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb)
+{
+ struct cbs_sched_data *q = qdisc_priv(sch);
+ struct tc_cbs_qopt opt = { };
+ struct nlattr *nest;
+
+ nest = nla_nest_start(skb, TCA_OPTIONS);
+ if (!nest)
+ goto nla_put_failure;
+
+ opt.hicredit = q->hicredit;
+ opt.locredit = q->locredit;
+ opt.sendslope = q->sendslope;
+ opt.idleslope = q->idleslope;
+ opt.offload = 0;
+
+ if (nla_put(skb, TCA_CBS_PARMS, sizeof(opt), &opt))
+ goto nla_put_failure;
+
+ return nla_nest_end(skb, nest);
+
+nla_put_failure:
+ nla_nest_cancel(skb, nest);
+ return -1;
+}
+
+static struct Qdisc_ops cbs_qdisc_ops __read_mostly = {
+ .id = "cbs",
+ .priv_size = sizeof(struct cbs_sched_data),
+ .enqueue = cbs_enqueue,
+ .dequeue = cbs_dequeue,
+ .peek = qdisc_peek_dequeued,
+ .init = cbs_init,
+ .reset = qdisc_reset_queue,
+ .destroy = cbs_destroy,
+ .change = cbs_change,
+ .dump = cbs_dump,
+ .owner = THIS_MODULE,
+};
+
+static int __init cbs_module_init(void)
+{
+ return register_qdisc(&cbs_qdisc_ops);
+}
+
+static void __exit cbs_module_exit(void)
+{
+ unregister_qdisc(&cbs_qdisc_ops);
+}
+module_init(cbs_module_init)
+module_exit(cbs_module_exit)
+MODULE_LICENSE("GPL");
--
2.14.2
^ permalink raw reply related
* [next-queue PATCH v7 3/6] net/sched: Add select_queue() class_ops for mqprio
From: Vinicius Costa Gomes @ 2017-10-13 0:40 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Jesus Sanchez-Palencia, jhs, xiyou.wangcong, jiri, andre.guedes,
ivan.briano, boon.leong.ong, richardcochran, henrik, levipearson,
rodney.cummings
In-Reply-To: <20171013004005.17416-1-vinicius.gomes@intel.com>
From: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
When replacing a child qdisc from mqprio, tc_modify_qdisc() must fetch
the netdev_queue pointer that the current child qdisc is associated
with before creating the new qdisc.
Currently, when using mqprio as root qdisc, the kernel will end up
getting the queue #0 pointer from the mqprio (root qdisc), which leaves
any new child qdisc with a possibly wrong netdev_queue pointer.
Implementing the Qdisc_class_ops select_queue() on mqprio fixes this
issue and avoid an inconsistent state when child qdiscs are replaced.
Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia@intel.com>
---
net/sched/sch_mqprio.c | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/net/sched/sch_mqprio.c b/net/sched/sch_mqprio.c
index 6bcdfe6e7b63..8c042ae323e3 100644
--- a/net/sched/sch_mqprio.c
+++ b/net/sched/sch_mqprio.c
@@ -396,6 +396,12 @@ static void mqprio_walk(struct Qdisc *sch, struct qdisc_walker *arg)
}
}
+static struct netdev_queue *mqprio_select_queue(struct Qdisc *sch,
+ struct tcmsg *tcm)
+{
+ return mqprio_queue_get(sch, TC_H_MIN(tcm->tcm_parent));
+}
+
static const struct Qdisc_class_ops mqprio_class_ops = {
.graft = mqprio_graft,
.leaf = mqprio_leaf,
@@ -403,6 +409,7 @@ static const struct Qdisc_class_ops mqprio_class_ops = {
.walk = mqprio_walk,
.dump = mqprio_dump_class,
.dump_stats = mqprio_dump_class_stats,
+ .select_queue = mqprio_select_queue,
};
static struct Qdisc_ops mqprio_qdisc_ops __read_mostly = {
--
2.14.2
^ permalink raw reply related
* [next-queue PATCH v7 6/6] igb: Add support for CBS offload
From: Vinicius Costa Gomes @ 2017-10-13 0:40 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Andre Guedes, jhs, xiyou.wangcong, jiri, ivan.briano,
jesus.sanchez-palencia, boon.leong.ong, richardcochran, henrik,
levipearson, rodney.cummings
In-Reply-To: <20171013004005.17416-1-vinicius.gomes@intel.com>
From: Andre Guedes <andre.guedes@intel.com>
This patch adds support for Credit-Based Shaper (CBS) qdisc offload
from Traffic Control system. This support enable us to leverage the
Forwarding and Queuing for Time-Sensitive Streams (FQTSS) features
from Intel i210 Ethernet Controller. FQTSS is the former 802.1Qav
standard which was merged into 802.1Q in 2014. It enables traffic
prioritization and bandwidth reservation via the Credit-Based Shaper
which is implemented in hardware by i210 controller.
The patch introduces the igb_setup_tc() function which implements the
support for CBS qdisc hardware offload in the IGB driver. CBS offload
is the only traffic control offload supported by the driver at the
moment.
FQTSS transmission mode from i210 controller is automatically enabled
by the IGB driver when the CBS is enabled for the first hardware
queue. Likewise, FQTSS mode is automatically disabled when CBS is
disabled for the last hardware queue. Changing FQTSS mode requires NIC
reset.
FQTSS feature is supported by i210 controller only.
Signed-off-by: Andre Guedes <andre.guedes@intel.com>
---
drivers/net/ethernet/intel/igb/e1000_defines.h | 23 ++
drivers/net/ethernet/intel/igb/e1000_regs.h | 8 +
drivers/net/ethernet/intel/igb/igb.h | 6 +
drivers/net/ethernet/intel/igb/igb_main.c | 347 +++++++++++++++++++++++++
4 files changed, 384 insertions(+)
diff --git a/drivers/net/ethernet/intel/igb/e1000_defines.h b/drivers/net/ethernet/intel/igb/e1000_defines.h
index 1de82f247312..83cabff1e0ab 100644
--- a/drivers/net/ethernet/intel/igb/e1000_defines.h
+++ b/drivers/net/ethernet/intel/igb/e1000_defines.h
@@ -353,7 +353,18 @@
#define E1000_RXPBS_CFG_TS_EN 0x80000000
#define I210_RXPBSIZE_DEFAULT 0x000000A2 /* RXPBSIZE default */
+#define I210_RXPBSIZE_MASK 0x0000003F
+#define I210_RXPBSIZE_PB_32KB 0x00000020
#define I210_TXPBSIZE_DEFAULT 0x04000014 /* TXPBSIZE default */
+#define I210_TXPBSIZE_MASK 0xC0FFFFFF
+#define I210_TXPBSIZE_PB0_8KB (8 << 0)
+#define I210_TXPBSIZE_PB1_8KB (8 << 6)
+#define I210_TXPBSIZE_PB2_4KB (4 << 12)
+#define I210_TXPBSIZE_PB3_4KB (4 << 18)
+
+#define I210_DTXMXPKTSZ_DEFAULT 0x00000098
+
+#define I210_SR_QUEUES_NUM 2
/* SerDes Control */
#define E1000_SCTL_DISABLE_SERDES_LOOPBACK 0x0400
@@ -1051,4 +1062,16 @@
#define E1000_VLAPQF_P_VALID(_n) (0x1 << (3 + (_n) * 4))
#define E1000_VLAPQF_QUEUE_MASK 0x03
+/* TX Qav Control fields */
+#define E1000_TQAVCTRL_XMIT_MODE BIT(0)
+#define E1000_TQAVCTRL_DATAFETCHARB BIT(4)
+#define E1000_TQAVCTRL_DATATRANARB BIT(8)
+
+/* TX Qav Credit Control fields */
+#define E1000_TQAVCC_IDLESLOPE_MASK 0xFFFF
+#define E1000_TQAVCC_QUEUEMODE BIT(31)
+
+/* Transmit Descriptor Control fields */
+#define E1000_TXDCTL_PRIORITY BIT(27)
+
#endif
diff --git a/drivers/net/ethernet/intel/igb/e1000_regs.h b/drivers/net/ethernet/intel/igb/e1000_regs.h
index 58adbf234e07..8eee081d395f 100644
--- a/drivers/net/ethernet/intel/igb/e1000_regs.h
+++ b/drivers/net/ethernet/intel/igb/e1000_regs.h
@@ -421,6 +421,14 @@ do { \
#define E1000_I210_FLA 0x1201C
+#define E1000_I210_DTXMXPKTSZ 0x355C
+
+#define E1000_I210_TXDCTL(_n) (0x0E028 + ((_n) * 0x40))
+
+#define E1000_I210_TQAVCTRL 0x3570
+#define E1000_I210_TQAVCC(_n) (0x3004 + ((_n) * 0x40))
+#define E1000_I210_TQAVHC(_n) (0x300C + ((_n) * 0x40))
+
#define E1000_INVM_DATA_REG(_n) (0x12120 + 4*(_n))
#define E1000_INVM_SIZE 64 /* Number of INVM Data Registers */
diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
index 06ffb2bc713e..92845692087a 100644
--- a/drivers/net/ethernet/intel/igb/igb.h
+++ b/drivers/net/ethernet/intel/igb/igb.h
@@ -281,6 +281,11 @@ struct igb_ring {
u16 count; /* number of desc. in the ring */
u8 queue_index; /* logical index of the ring*/
u8 reg_idx; /* physical index of the ring */
+ bool cbs_enable; /* indicates if CBS is enabled */
+ s32 idleslope; /* idleSlope in kbps */
+ s32 sendslope; /* sendSlope in kbps */
+ s32 hicredit; /* hiCredit in bytes */
+ s32 locredit; /* loCredit in bytes */
/* everything past this point are written often */
u16 next_to_clean;
@@ -621,6 +626,7 @@ struct igb_adapter {
#define IGB_FLAG_EEE BIT(14)
#define IGB_FLAG_VLAN_PROMISC BIT(15)
#define IGB_FLAG_RX_LEGACY BIT(16)
+#define IGB_FLAG_FQTSS BIT(17)
/* Media Auto Sense */
#define IGB_MAS_ENABLE_0 0X0001
diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c
index 837d9b46a390..be2cf263efa9 100644
--- a/drivers/net/ethernet/intel/igb/igb_main.c
+++ b/drivers/net/ethernet/intel/igb/igb_main.c
@@ -34,6 +34,7 @@
#include <linux/slab.h>
#include <net/checksum.h>
#include <net/ip6_checksum.h>
+#include <net/pkt_sched.h>
#include <linux/net_tstamp.h>
#include <linux/mii.h>
#include <linux/ethtool.h>
@@ -62,6 +63,17 @@
#define BUILD 0
#define DRV_VERSION __stringify(MAJ) "." __stringify(MIN) "." \
__stringify(BUILD) "-k"
+
+enum queue_mode {
+ QUEUE_MODE_STRICT_PRIORITY,
+ QUEUE_MODE_STREAM_RESERVATION,
+};
+
+enum tx_queue_prio {
+ TX_QUEUE_PRIO_HIGH,
+ TX_QUEUE_PRIO_LOW,
+};
+
char igb_driver_name[] = "igb";
char igb_driver_version[] = DRV_VERSION;
static const char igb_driver_string[] =
@@ -1271,6 +1283,12 @@ static int igb_alloc_q_vector(struct igb_adapter *adapter,
ring->count = adapter->tx_ring_count;
ring->queue_index = txr_idx;
+ ring->cbs_enable = false;
+ ring->idleslope = 0;
+ ring->sendslope = 0;
+ ring->hicredit = 0;
+ ring->locredit = 0;
+
u64_stats_init(&ring->tx_syncp);
u64_stats_init(&ring->tx_syncp2);
@@ -1598,6 +1616,284 @@ static void igb_get_hw_control(struct igb_adapter *adapter)
ctrl_ext | E1000_CTRL_EXT_DRV_LOAD);
}
+static void enable_fqtss(struct igb_adapter *adapter, bool enable)
+{
+ struct net_device *netdev = adapter->netdev;
+ struct e1000_hw *hw = &adapter->hw;
+
+ WARN_ON(hw->mac.type != e1000_i210);
+
+ if (enable)
+ adapter->flags |= IGB_FLAG_FQTSS;
+ else
+ adapter->flags &= ~IGB_FLAG_FQTSS;
+
+ if (netif_running(netdev))
+ schedule_work(&adapter->reset_task);
+}
+
+static bool is_fqtss_enabled(struct igb_adapter *adapter)
+{
+ return (adapter->flags & IGB_FLAG_FQTSS) ? true : false;
+}
+
+static void set_tx_desc_fetch_prio(struct e1000_hw *hw, int queue,
+ enum tx_queue_prio prio)
+{
+ u32 val;
+
+ WARN_ON(hw->mac.type != e1000_i210);
+ WARN_ON(queue < 0 || queue > 4);
+
+ val = rd32(E1000_I210_TXDCTL(queue));
+
+ if (prio == TX_QUEUE_PRIO_HIGH)
+ val |= E1000_TXDCTL_PRIORITY;
+ else
+ val &= ~E1000_TXDCTL_PRIORITY;
+
+ wr32(E1000_I210_TXDCTL(queue), val);
+}
+
+static void set_queue_mode(struct e1000_hw *hw, int queue, enum queue_mode mode)
+{
+ u32 val;
+
+ WARN_ON(hw->mac.type != e1000_i210);
+ WARN_ON(queue < 0 || queue > 1);
+
+ val = rd32(E1000_I210_TQAVCC(queue));
+
+ if (mode == QUEUE_MODE_STREAM_RESERVATION)
+ val |= E1000_TQAVCC_QUEUEMODE;
+ else
+ val &= ~E1000_TQAVCC_QUEUEMODE;
+
+ wr32(E1000_I210_TQAVCC(queue), val);
+}
+
+/**
+ * igb_configure_cbs - Configure Credit-Based Shaper (CBS)
+ * @adapter: pointer to adapter struct
+ * @queue: queue number
+ * @enable: true = enable CBS, false = disable CBS
+ * @idleslope: idleSlope in kbps
+ * @sendslope: sendSlope in kbps
+ * @hicredit: hiCredit in bytes
+ * @locredit: loCredit in bytes
+ *
+ * Configure CBS for a given hardware queue. When disabling, idleslope,
+ * sendslope, hicredit, locredit arguments are ignored. Returns 0 if
+ * success. Negative otherwise.
+ **/
+static void igb_configure_cbs(struct igb_adapter *adapter, int queue,
+ bool enable, int idleslope, int sendslope,
+ int hicredit, int locredit)
+{
+ struct net_device *netdev = adapter->netdev;
+ struct e1000_hw *hw = &adapter->hw;
+ u32 tqavcc;
+ u16 value;
+
+ WARN_ON(hw->mac.type != e1000_i210);
+ WARN_ON(queue < 0 || queue > 1);
+
+ if (enable) {
+ set_tx_desc_fetch_prio(hw, queue, TX_QUEUE_PRIO_HIGH);
+ set_queue_mode(hw, queue, QUEUE_MODE_STREAM_RESERVATION);
+
+ /* According to i210 datasheet section 7.2.7.7, we should set
+ * the 'idleSlope' field from TQAVCC register following the
+ * equation:
+ *
+ * For 100 Mbps link speed:
+ *
+ * value = BW * 0x7735 * 0.2 (E1)
+ *
+ * For 1000Mbps link speed:
+ *
+ * value = BW * 0x7735 * 2 (E2)
+ *
+ * E1 and E2 can be merged into one equation as shown below.
+ * Note that 'link-speed' is in Mbps.
+ *
+ * value = BW * 0x7735 * 2 * link-speed
+ * -------------- (E3)
+ * 1000
+ *
+ * 'BW' is the percentage bandwidth out of full link speed
+ * which can be found with the following equation. Note that
+ * idleSlope here is the parameter from this function which
+ * is in kbps.
+ *
+ * BW = idleSlope
+ * ----------------- (E4)
+ * link-speed * 1000
+ *
+ * That said, we can come up with a generic equation to
+ * calculate the value we should set it TQAVCC register by
+ * replacing 'BW' in E3 by E4. The resulting equation is:
+ *
+ * value = idleSlope * 0x7735 * 2 * link-speed
+ * ----------------- -------------- (E5)
+ * link-speed * 1000 1000
+ *
+ * 'link-speed' is present in both sides of the fraction so
+ * it is canceled out. The final equation is the following:
+ *
+ * value = idleSlope * 61034
+ * ----------------- (E6)
+ * 1000000
+ */
+ value = DIV_ROUND_UP_ULL(idleslope * 61034ULL, 1000000);
+
+ tqavcc = rd32(E1000_I210_TQAVCC(queue));
+ tqavcc &= ~E1000_TQAVCC_IDLESLOPE_MASK;
+ tqavcc |= value;
+ wr32(E1000_I210_TQAVCC(queue), tqavcc);
+
+ wr32(E1000_I210_TQAVHC(queue), 0x80000000 + hicredit * 0x7735);
+ } else {
+ set_tx_desc_fetch_prio(hw, queue, TX_QUEUE_PRIO_LOW);
+ set_queue_mode(hw, queue, QUEUE_MODE_STRICT_PRIORITY);
+
+ /* Set idleSlope to zero. */
+ tqavcc = rd32(E1000_I210_TQAVCC(queue));
+ tqavcc &= ~E1000_TQAVCC_IDLESLOPE_MASK;
+ wr32(E1000_I210_TQAVCC(queue), tqavcc);
+
+ /* Set hiCredit to zero. */
+ wr32(E1000_I210_TQAVHC(queue), 0);
+ }
+
+ /* XXX: In i210 controller the sendSlope and loCredit parameters from
+ * CBS are not configurable by software so we don't do any 'controller
+ * configuration' in respect to these parameters.
+ */
+
+ netdev_dbg(netdev, "CBS %s: queue %d idleslope %d sendslope %d hiCredit %d locredit %d\n",
+ (enable) ? "enabled" : "disabled", queue,
+ idleslope, sendslope, hicredit, locredit);
+}
+
+static int igb_save_cbs_params(struct igb_adapter *adapter, int queue,
+ bool enable, int idleslope, int sendslope,
+ int hicredit, int locredit)
+{
+ struct igb_ring *ring;
+
+ if (queue < 0 || queue > adapter->num_tx_queues)
+ return -EINVAL;
+
+ ring = adapter->tx_ring[queue];
+
+ ring->cbs_enable = enable;
+ ring->idleslope = idleslope;
+ ring->sendslope = sendslope;
+ ring->hicredit = hicredit;
+ ring->locredit = locredit;
+
+ return 0;
+}
+
+static bool is_any_cbs_enabled(struct igb_adapter *adapter)
+{
+ struct igb_ring *ring;
+ int i;
+
+ for (i = 0; i < adapter->num_tx_queues; i++) {
+ ring = adapter->tx_ring[i];
+
+ if (ring->cbs_enable)
+ return true;
+ }
+
+ return false;
+}
+
+static void igb_setup_tx_mode(struct igb_adapter *adapter)
+{
+ struct net_device *netdev = adapter->netdev;
+ struct e1000_hw *hw = &adapter->hw;
+ u32 val;
+
+ /* Only i210 controller supports changing the transmission mode. */
+ if (hw->mac.type != e1000_i210)
+ return;
+
+ if (is_fqtss_enabled(adapter)) {
+ int i, max_queue;
+
+ /* Configure TQAVCTRL register: set transmit mode to 'Qav',
+ * set data fetch arbitration to 'round robin' and set data
+ * transfer arbitration to 'credit shaper algorithm.
+ */
+ val = rd32(E1000_I210_TQAVCTRL);
+ val |= E1000_TQAVCTRL_XMIT_MODE | E1000_TQAVCTRL_DATATRANARB;
+ val &= ~E1000_TQAVCTRL_DATAFETCHARB;
+ wr32(E1000_I210_TQAVCTRL, val);
+
+ /* Configure Tx and Rx packet buffers sizes as described in
+ * i210 datasheet section 7.2.7.7.
+ */
+ val = rd32(E1000_TXPBS);
+ val &= ~I210_TXPBSIZE_MASK;
+ val |= I210_TXPBSIZE_PB0_8KB | I210_TXPBSIZE_PB1_8KB |
+ I210_TXPBSIZE_PB2_4KB | I210_TXPBSIZE_PB3_4KB;
+ wr32(E1000_TXPBS, val);
+
+ val = rd32(E1000_RXPBS);
+ val &= ~I210_RXPBSIZE_MASK;
+ val |= I210_RXPBSIZE_PB_32KB;
+ wr32(E1000_RXPBS, val);
+
+ /* Section 8.12.9 states that MAX_TPKT_SIZE from DTXMXPKTSZ
+ * register should not exceed the buffer size programmed in
+ * TXPBS. The smallest buffer size programmed in TXPBS is 4kB
+ * so according to the datasheet we should set MAX_TPKT_SIZE to
+ * 4kB / 64.
+ *
+ * However, when we do so, no frame from queue 2 and 3 are
+ * transmitted. It seems the MAX_TPKT_SIZE should not be great
+ * or _equal_ to the buffer size programmed in TXPBS. For this
+ * reason, we set set MAX_ TPKT_SIZE to (4kB - 1) / 64.
+ */
+ val = (4096 - 1) / 64;
+ wr32(E1000_I210_DTXMXPKTSZ, val);
+
+ /* Since FQTSS mode is enabled, apply any CBS configuration
+ * previously set. If no previous CBS configuration has been
+ * done, then the initial configuration is applied, which means
+ * CBS is disabled.
+ */
+ max_queue = (adapter->num_tx_queues < I210_SR_QUEUES_NUM) ?
+ adapter->num_tx_queues : I210_SR_QUEUES_NUM;
+
+ for (i = 0; i < max_queue; i++) {
+ struct igb_ring *ring = adapter->tx_ring[i];
+
+ igb_configure_cbs(adapter, i, ring->cbs_enable,
+ ring->idleslope, ring->sendslope,
+ ring->hicredit, ring->locredit);
+ }
+ } else {
+ wr32(E1000_RXPBS, I210_RXPBSIZE_DEFAULT);
+ wr32(E1000_TXPBS, I210_TXPBSIZE_DEFAULT);
+ wr32(E1000_I210_DTXMXPKTSZ, I210_DTXMXPKTSZ_DEFAULT);
+
+ val = rd32(E1000_I210_TQAVCTRL);
+ /* According to Section 8.12.21, the other flags we've set when
+ * enabling FQTSS are not relevant when disabling FQTSS so we
+ * don't set they here.
+ */
+ val &= ~E1000_TQAVCTRL_XMIT_MODE;
+ wr32(E1000_I210_TQAVCTRL, val);
+ }
+
+ netdev_dbg(netdev, "FQTSS %s\n", (is_fqtss_enabled(adapter)) ?
+ "enabled" : "disabled");
+}
+
/**
* igb_configure - configure the hardware for RX and TX
* @adapter: private board structure
@@ -1609,6 +1905,7 @@ static void igb_configure(struct igb_adapter *adapter)
igb_get_hw_control(adapter);
igb_set_rx_mode(netdev);
+ igb_setup_tx_mode(adapter);
igb_restore_vlan(adapter);
@@ -2150,6 +2447,55 @@ igb_features_check(struct sk_buff *skb, struct net_device *dev,
return features;
}
+static int igb_offload_cbs(struct igb_adapter *adapter,
+ struct tc_cbs_qopt_offload *qopt)
+{
+ struct e1000_hw *hw = &adapter->hw;
+ int err;
+
+ /* CBS offloading is only supported by i210 controller. */
+ if (hw->mac.type != e1000_i210)
+ return -EOPNOTSUPP;
+
+ /* CBS offloading is only supported by queue 0 and queue 1. */
+ if (qopt->queue < 0 || qopt->queue > 1)
+ return -EINVAL;
+
+ err = igb_save_cbs_params(adapter, qopt->queue, qopt->enable,
+ qopt->idleslope, qopt->sendslope,
+ qopt->hicredit, qopt->locredit);
+ if (err)
+ return err;
+
+ if (is_fqtss_enabled(adapter)) {
+ igb_configure_cbs(adapter, qopt->queue, qopt->enable,
+ qopt->idleslope, qopt->sendslope,
+ qopt->hicredit, qopt->locredit);
+
+ if (!is_any_cbs_enabled(adapter))
+ enable_fqtss(adapter, false);
+
+ } else {
+ enable_fqtss(adapter, true);
+ }
+
+ return 0;
+}
+
+static int igb_setup_tc(struct net_device *dev, enum tc_setup_type type,
+ void *type_data)
+{
+ struct igb_adapter *adapter = netdev_priv(dev);
+
+ switch (type) {
+ case TC_SETUP_CBS:
+ return igb_offload_cbs(adapter, type_data);
+
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
static const struct net_device_ops igb_netdev_ops = {
.ndo_open = igb_open,
.ndo_stop = igb_close,
@@ -2175,6 +2521,7 @@ static const struct net_device_ops igb_netdev_ops = {
.ndo_set_features = igb_set_features,
.ndo_fdb_add = igb_ndo_fdb_add,
.ndo_features_check = igb_features_check,
+ .ndo_setup_tc = igb_setup_tc,
};
/**
--
2.14.2
^ permalink raw reply related
* [next-queue PATCH v7 5/6] net/sched: Add support for HW offloading for CBS
From: Vinicius Costa Gomes @ 2017-10-13 0:40 UTC (permalink / raw)
To: netdev, intel-wired-lan
Cc: Vinicius Costa Gomes, jhs, xiyou.wangcong, jiri, andre.guedes,
ivan.briano, jesus.sanchez-palencia, boon.leong.ong,
richardcochran, henrik, levipearson, rodney.cummings
In-Reply-To: <20171013004005.17416-1-vinicius.gomes@intel.com>
This adds support for offloading the CBS algorithm to the controller,
if supported. Drivers wanting to support CBS offload must implement
the .ndo_setup_tc callback and handle the TC_SETUP_CBS (introduced
here) type.
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
---
include/linux/netdevice.h | 1 +
include/net/pkt_sched.h | 9 ++++
net/sched/sch_cbs.c | 104 ++++++++++++++++++++++++++++++++++++++++------
3 files changed, 102 insertions(+), 12 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 31bb3010c69b..1f6c44ef5b21 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -775,6 +775,7 @@ enum tc_setup_type {
TC_SETUP_CLSFLOWER,
TC_SETUP_CLSMATCHALL,
TC_SETUP_CLSBPF,
+ TC_SETUP_CBS,
};
/* These structures hold the attributes of xdp state that are being passed
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 259bc191ba59..7c597b050b36 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -146,4 +146,13 @@ static inline bool is_classid_clsact_egress(u32 classid)
TC_H_MIN(classid) == TC_H_MIN(TC_H_MIN_EGRESS);
}
+struct tc_cbs_qopt_offload {
+ u8 enable;
+ s32 queue;
+ s32 hicredit;
+ s32 locredit;
+ s32 idleslope;
+ s32 sendslope;
+};
+
#endif
diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c
index 0643587e6dc8..7d2100c5b8aa 100644
--- a/net/sched/sch_cbs.c
+++ b/net/sched/sch_cbs.c
@@ -68,6 +68,8 @@
#define BYTES_PER_KBIT (1000 / 8)
struct cbs_sched_data {
+ bool offload;
+ int queue;
s64 port_rate; /* in bytes/s */
s64 last; /* timestamp in ns */
s64 credits; /* in bytes */
@@ -80,6 +82,11 @@ struct cbs_sched_data {
struct sk_buff *(*dequeue)(struct Qdisc *sch);
};
+static int cbs_enqueue_offload(struct sk_buff *skb, struct Qdisc *sch)
+{
+ return qdisc_enqueue_tail(skb, sch);
+}
+
static int cbs_enqueue_soft(struct sk_buff *skb, struct Qdisc *sch)
{
struct cbs_sched_data *q = qdisc_priv(sch);
@@ -190,6 +197,11 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch)
return skb;
}
+static struct sk_buff *cbs_dequeue_offload(struct Qdisc *sch)
+{
+ return qdisc_dequeue_head(sch);
+}
+
static struct sk_buff *cbs_dequeue(struct Qdisc *sch)
{
struct cbs_sched_data *q = qdisc_priv(sch);
@@ -201,14 +213,66 @@ static const struct nla_policy cbs_policy[TCA_CBS_MAX + 1] = {
[TCA_CBS_PARMS] = { .len = sizeof(struct tc_cbs_qopt) },
};
+static void cbs_disable_offload(struct net_device *dev,
+ struct cbs_sched_data *q)
+{
+ struct tc_cbs_qopt_offload cbs = { };
+ const struct net_device_ops *ops;
+ int err;
+
+ if (!q->offload)
+ return;
+
+ q->enqueue = cbs_enqueue_soft;
+ q->dequeue = cbs_dequeue_soft;
+
+ ops = dev->netdev_ops;
+ if (!ops->ndo_setup_tc)
+ return;
+
+ cbs.queue = q->queue;
+ cbs.enable = 0;
+
+ err = ops->ndo_setup_tc(dev, TC_SETUP_CBS, &cbs);
+ if (err < 0)
+ pr_warn("Couldn't disable CBS offload for queue %d\n",
+ cbs.queue);
+}
+
+static int cbs_enable_offload(struct net_device *dev, struct cbs_sched_data *q,
+ const struct tc_cbs_qopt *opt)
+{
+ const struct net_device_ops *ops = dev->netdev_ops;
+ struct tc_cbs_qopt_offload cbs = { };
+ int err;
+
+ if (!ops->ndo_setup_tc)
+ return -EOPNOTSUPP;
+
+ cbs.queue = q->queue;
+
+ cbs.enable = 1;
+ cbs.hicredit = opt->hicredit;
+ cbs.locredit = opt->locredit;
+ cbs.idleslope = opt->idleslope;
+ cbs.sendslope = opt->sendslope;
+
+ err = ops->ndo_setup_tc(dev, TC_SETUP_CBS, &cbs);
+ if (err < 0)
+ return err;
+
+ q->enqueue = cbs_enqueue_offload;
+ q->dequeue = cbs_dequeue_offload;
+
+ return 0;
+}
+
static int cbs_change(struct Qdisc *sch, struct nlattr *opt)
{
struct cbs_sched_data *q = qdisc_priv(sch);
struct net_device *dev = qdisc_dev(sch);
struct nlattr *tb[TCA_CBS_MAX + 1];
- struct ethtool_link_ksettings ecmd;
struct tc_cbs_qopt *qopt;
- s64 link_speed;
int err;
err = nla_parse_nested(tb, TCA_CBS_MAX, opt, cbs_policy, NULL);
@@ -220,23 +284,30 @@ static int cbs_change(struct Qdisc *sch, struct nlattr *opt)
qopt = nla_data(tb[TCA_CBS_PARMS]);
- if (qopt->offload)
- return -EOPNOTSUPP;
+ if (!qopt->offload) {
+ struct ethtool_link_ksettings ecmd;
+ s64 link_speed;
- if (!__ethtool_get_link_ksettings(dev, &ecmd))
- link_speed = ecmd.base.speed;
- else
- link_speed = SPEED_1000;
+ if (!__ethtool_get_link_ksettings(dev, &ecmd))
+ link_speed = ecmd.base.speed;
+ else
+ link_speed = SPEED_1000;
- q->port_rate = link_speed * 1000 * BYTES_PER_KBIT;
+ q->port_rate = link_speed * 1000 * BYTES_PER_KBIT;
- q->enqueue = cbs_enqueue_soft;
- q->dequeue = cbs_dequeue_soft;
+ cbs_disable_offload(dev, q);
+ } else {
+ err = cbs_enable_offload(dev, q, qopt);
+ if (err < 0)
+ return err;
+ }
+ /* Everything went OK, save the parameters used. */
q->hicredit = qopt->hicredit;
q->locredit = qopt->locredit;
q->idleslope = qopt->idleslope;
q->sendslope = qopt->sendslope;
+ q->offload = qopt->offload;
return 0;
}
@@ -244,10 +315,16 @@ static int cbs_change(struct Qdisc *sch, struct nlattr *opt)
static int cbs_init(struct Qdisc *sch, struct nlattr *opt)
{
struct cbs_sched_data *q = qdisc_priv(sch);
+ struct net_device *dev = qdisc_dev(sch);
if (!opt)
return -EINVAL;
+ q->queue = sch->dev_queue - netdev_get_tx_queue(dev, 0);
+
+ q->enqueue = cbs_enqueue_soft;
+ q->dequeue = cbs_dequeue_soft;
+
qdisc_watchdog_init(&q->watchdog, sch);
return cbs_change(sch, opt);
@@ -256,8 +333,11 @@ static int cbs_init(struct Qdisc *sch, struct nlattr *opt)
static void cbs_destroy(struct Qdisc *sch)
{
struct cbs_sched_data *q = qdisc_priv(sch);
+ struct net_device *dev = qdisc_dev(sch);
qdisc_watchdog_cancel(&q->watchdog);
+
+ cbs_disable_offload(dev, q);
}
static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb)
@@ -274,7 +354,7 @@ static int cbs_dump(struct Qdisc *sch, struct sk_buff *skb)
opt.locredit = q->locredit;
opt.sendslope = q->sendslope;
opt.idleslope = q->idleslope;
- opt.offload = 0;
+ opt.offload = q->offload;
if (nla_put(skb, TCA_CBS_PARMS, sizeof(opt), &opt))
goto nla_put_failure;
--
2.14.2
^ permalink raw reply related
* Re: [RFC] Support for UNARP (RFC 1868)
From: Mahesh Bandewar (महेश बंडेवार) @ 2017-10-13 0:53 UTC (permalink / raw)
To: Girish Moodalbail; +Cc: Eric Dumazet, linux-netdev, David Miller, kuznet
In-Reply-To: <c75c04fe-46d6-8ec3-844d-fe379710496b@oracle.com>
On Thu, Oct 12, 2017 at 4:06 PM, Girish Moodalbail
<girish.moodalbail@oracle.com> wrote:
> Hello Eric,
>
> The basic idea is to mark the ARP entry either FAILED or STALE as soon as we
> can so that the subsequent packets that depend on that ARP entry will take
> the slow path (neigh_resolve_output()).
>
> Say, if base_reachable_time is 30 seconds, then an ARP entry will be in
> reachable state somewhere between 15 to 45 seconds. Assuming the worst case,
> the ARP entry will be in REACHABLE state for 45 seconds and the packets
> continue to traverse the network towards the source machine and gets dropped
> their since the VM has moved to destination machine.
>
> Instead, based on the received UNARP packet if we mark the ARP entry
>
> (a) FAILED
> - we move to INCOMPLETE state and start the address resolution by sending
> out ARP packets (up to allowed maximum number) until we get ARP
> response
> back at which point we move the ARP entry state to reachable.
>
> (b) STALE
> - we move to DELAY state and set the next timer to DELAY_PROBE_TIME
> (1 second) and continue to send queued packets in arp_queue.
> - After 1 sec we move to PROBE state and start the address resolution
> like
> in the case(a) above.
>
> I was leaning towards (a).
One could arbitrarily increase the stale timeout (by changing no of
probes). So sender
will continue sending traffic to something that has already gone away.
STALE doesn't
mean bad but here the sender is clearly indicating it's going away so
FAILED seems to
be the only logical option.
> Please see in-line..
>
> <snip>
>
>>
>> Hi Girish
>>
>> Your description (or patch title) is misleading. You apparently
>> implement the receive side of the RFC.
>
>
> You are right, it implements only the receive side of the RFC. If this RFC
> is accepted, then we can change arping(8) to generate UNARP requests. We
> could also add an option to ip-address(8) delete subcommand to generate
> UNARP whenever an address is deleted from the interface.
>
>> And the RFC had Proxy ARP in mind.
>>
>> What about security implications ?
>
>
> Yes, this feature will extend the attack surface for L2 networks. However,
> the attack vectors for this feature should be same as that of the gratuitous
> ARP, right? The same attack mitigation techniques for gratuitous ARPs is
> equally applicable here.
>
>> Will TCP flows be terminated, instead
>> of being smoothly migrated (TCP_REPAIR)
>
>
> The TCP flows will not be terminated. Upon receiving UNARP packet, the ARP
> entry will be marked FAILED. The subsequent TCP packets from the client
> (towards that IP) will be queued (the first 3 packets in arp_queue and then
> other packets get dropped) until the IP address is resolved again through
> the slow path neigh_resolve_output().
>
> The slow path marks the entry as INCOMPLETE and will start sending several
> ARP requests (ucast_solicit + app_solicit + mcast_solicit) to resolve the
> IP. If the resolution is successful, then the TCP packets will be sent out.
> If not, we will invalidate the ARP entry and call arp_error_report() on the
> queued packets (which will end up sending ICMP_HOST_UNREACH error). This
> behavior is same as what will occur if TCP server disappears in the middle
> of a connection.
>
>>
>> What about IPv6 ? Or maybe more abruptly, do we still need to add
>> features to IPv4 in 2017, 22 years after this RFC came ? ;)
>
>
> Legit question :). Well one thing I have seen in Networking is that an old
> idea circles back around later and turns out to be useful in new contexts
> and use cases. Like I enumerated in my initial email there are certain use
> cases in Cloud that might benefit from UNARP.
>
It doesn't make sense to have this implemented only for IPv4. At this time if
equivalent IPv6 feature is missing, I don't see it being useful / acceptable.
> regards,
> ~Girish
>
>>
>> Thanks.
>>
>>
>
^ permalink raw reply
* Re: [patch net-next 27/34] nfp: bpf: Convert ndo_setup_tc offloads to block callbacks
From: Jakub Kicinski @ 2017-10-13 1:08 UTC (permalink / raw)
To: Jiri Pirko; +Cc: netdev, Daniel Borkmann
In-Reply-To: <20171012171823.1431-28-jiri@resnulli.us>
On Thu, 12 Oct 2017 19:18:16 +0200, Jiri Pirko wrote:
> diff --git a/drivers/net/ethernet/netronome/nfp/bpf/offload.c b/drivers/net/ethernet/netronome/nfp/bpf/offload.c
> index a88bb5b..9e9af88 100644
> --- a/drivers/net/ethernet/netronome/nfp/bpf/offload.c
> +++ b/drivers/net/ethernet/netronome/nfp/bpf/offload.c
> @@ -246,6 +246,10 @@ int nfp_net_bpf_offload(struct nfp_net *nn, struct tc_cls_bpf_offload *cls_bpf)
> void *code;
> int err;
>
> + if (cls_bpf->common.protocol != htons(ETH_P_ALL) ||
> + cls_bpf->common.chain_index)
> + return -EOPNOTSUPP;
> +
> max_instr = nn_readw(nn, NFP_NET_CFG_BPF_MAX_LEN);
>
> switch (cls_bpf->command) {
It is certainly very ugly but I send a fake struct tc_cls_bpf_offload
here for XDP. Refactoring this mess is pretty high on my priority list
but one way or the other this function will be called from XDP so TC
checks must stay in the TC handler... :(
^ permalink raw reply
* Re: [PATCH 2/6] ath9k: add a quirk to set use_msi automatically
From: AceLan Kao @ 2017-10-13 1:12 UTC (permalink / raw)
To: Daniel Drake
Cc: Kalle Valo, Christoph Hellwig, QCA ath9k Development,
linux-wireless, netdev, Linux-Kernel@Vger. Kernel. Org
In-Reply-To: <CAFv23QkLJ0m112Y8iAGv+6DyFT52Hqnxq5Fy=ryWTWxu8oTXdw@mail.gmail.com>
Hi Daniel,
After applied the 2 commits you mentioned in the email, ath9k works.
https://marc.info/?l=linux-wireless&m=150631274108016&w=2
https://github.com/endlessm/linux/commit/739c7a924db8f4434a9617657
Best regards,
AceLan Kao.
2017-10-05 14:39 GMT+08:00 AceLan Kao <acelan.kao@canonical.com>:
> Hi all,
>
> Please drop my patches, Qualcomm is working internally and will submit
> the MSI patch by themselves.
> Thanks.
>
> Hi Daniel,
>
> I'll try your patches tomorrow.
>
> Best regards,
> AceLan Kao.
>
> 2017-10-02 12:21 GMT+08:00 Daniel Drake <drake@endlessm.com>:
>> Hi AceLan,
>>
>> On Thu, Sep 28, 2017 at 4:28 PM, AceLan Kao <acelan.kao@canonical.com> wrote:
>>> Hi Daniel,
>>>
>>> I've tried your patch, but it doesn't work for me.
>>> Wifi can scan AP, but can't get connected.
>>
>> Can you please clarify which patch(es) you have tried?
>>
>> This is the base patch which adds the infrastructure to request
>> specific MSI IRQ vectors:
>> https://marc.info/?l=linux-wireless&m=150631274108016&w=2
>>
>> This is the ath9k MSI patch which makes use of that:
>> https://github.com/endlessm/linux/commit/739c7a924db8f4434a9617657
>>
>> If you were already able to use ath9k MSI interrupts without specific
>> consideration for which MSI vector numbers were used, these are the
>> possible explanations that spring to mind:
>>
>> 1. You got lucky and it picked a vector number that is 4-aligned. You
>> can check this in the "lspci -vvv" output. You'll see something like:
>> Capabilities: [50] MSI: Enable+ Count=1/4 Maskable+ 64bit+
>> Address: 00000000fee0300c Data: 4142
>> The lower number is the vector number. In my example here 0x42 (66) is
>> not 4-aligned so the failure condition will be hit.
>>
>> 2. You are using interrupt remapping, which I suspect may provide a
>> high likelihood of MSI interrupt vectors being 4-aligned. See if
>> /proc/interrupts shows the IRQ type as IR-PCI-MSI
>> Unfortunately interrupt remapping is not available here,
>> https://lists.linuxfoundation.org/pipermail/iommu/2017-August/023717.html
>>
>> 3. My assumption that all ath9k hardware corrupts the MSI vector
>> number could wrong. However we've seen this on different wifi modules
>> in laptops produced by different OEMs and ODMs, so it seems to be a
>> somewhat widespread problem at least.
>>
>> 4. My assumption that ath9k hardware is corrupting the MSI vector
>> number could be wrong; maybe another component is to blame, could it
>> be a BIOS issue? Admittedly I don't really know how I can debug the
>> layers inbetween seeing the MSI Message Data value disagree with the
>> vector number being handled inside do_IRQ().
>>
>> Daniel
^ 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