* [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
* 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] 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
* [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 23:45 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 $@
--
2.7.4
^ permalink raw reply related
* [PATCH] wanxl: use m68k-linux-gnu-as if available
From: Adam Borowski @ 2017-10-12 23:24 UTC (permalink / raw)
To: Krzysztof Halasa, netdev; +Cc: Adam Borowski
This fixes build failure on Debian based systems: GNU as is the only m68k
assembler available in the archive (package binutils-m68k-linux-gnu).
Signed-off-by: Adam Borowski <kilobyte@angband.pl>
---
I have no relevant hardware, thus I can't check whether the built firmware
actually works. Some opcodes are translated differently, thus it might be
possible that some extra options are required. Or possibly the assembler
has long since changed -- the prebuilt copy hasn't been updated anywhere
within recorded history.
In any case, I admit I don't really care about this driver -- it's merely
a notorious cause of randconfig failures.
drivers/net/wan/Makefile | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/wan/Makefile b/drivers/net/wan/Makefile
index 73c2326603fc..fa17b18765b0 100644
--- a/drivers/net/wan/Makefile
+++ b/drivers/net/wan/Makefile
@@ -38,14 +38,20 @@ obj-$(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
* 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
* 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 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] 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: [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 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: [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 3/5] net: dsa: add slave to port helper
From: Florian Fainelli @ 2017-10-12 22:59 UTC (permalink / raw)
To: Vivien Didelot, netdev; +Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <20171012225156.20758-4-vivien.didelot@savoirfairelinux.com>
On 10/12/2017 03:51 PM, Vivien Didelot wrote:
> Many portions of DSA core code require to get the dsa_port structure
> corresponding to a slave net_device. For this purpose, introduce a
> dsa_slave_to_port() helper.
>
> Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
--
Florian
^ permalink raw reply
* Re: [PATCH net-next 2/5] net: dsa: add slave notify helper
From: Florian Fainelli @ 2017-10-12 22:55 UTC (permalink / raw)
To: Vivien Didelot, netdev; +Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <20171012225156.20758-3-vivien.didelot@savoirfairelinux.com>
On 10/12/2017 03:51 PM, Vivien Didelot wrote:
> Both DSA slave create and destroy functions call call_dsa_notifiers with
> respectively DSA_PORT_REGISTER and DSA_PORT_UNREGISTER and the same
> dsa_notifier_register_info structure.
>
> Wrap this in a dsa_slave_notify helper so prevent cluttering these
> functions.
>
> Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Much nicer, thanks!
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
--
Florian
^ permalink raw reply
* [PATCH] netfilter: nf_conntrack_h323: Remove typedef struct
From: Harsha Sharma @ 2017-10-12 22:53 UTC (permalink / raw)
To: pablo, kadlec, fw, davem
Cc: netfilter-devel, coreteam, netdev, linux-kernel, outreachy-kernel,
Harsha Sharma
Remove typedef from struct as linux-kernel coding style tends to
avoid using typedefs.
Done using following coccinelle semantic patch
@r1@
type T;
@@
typedef struct { ... } T;
@script:python c1@
T2;
T << r1.T;
@@
if T[-2:] =="_t" or T[-2:] == "_T":
coccinelle.T2 = T[:-2];
else:
coccinelle.T2 = T;
print T, coccinelle.T2
@r2@
type r1.T;
identifier c1.T2;
@@
-typedef
struct
+ T2
{ ... }
-T
;
@r3@
type r1.T;
identifier c1.T2;
@@
-T
+struct T2
Signed-off-by: Harsha Sharma <harshasharmaiitr@gmail.com>
---
net/netfilter/nf_conntrack_h323_asn1.c | 80 +++++++++++++++++-----------------
1 file changed, 40 insertions(+), 40 deletions(-)
diff --git a/net/netfilter/nf_conntrack_h323_asn1.c b/net/netfilter/nf_conntrack_h323_asn1.c
index 89b2e46925c4..7831aa1effc9 100644
--- a/net/netfilter/nf_conntrack_h323_asn1.c
+++ b/net/netfilter/nf_conntrack_h323_asn1.c
@@ -91,41 +91,41 @@ typedef struct field_t {
} field_t;
/* Bit Stream */
-typedef struct {
+struct bitstr {
unsigned char *buf;
unsigned char *beg;
unsigned char *end;
unsigned char *cur;
unsigned int bit;
-} bitstr_t;
+};
/* Tool Functions */
#define INC_BIT(bs) if((++(bs)->bit)>7){(bs)->cur++;(bs)->bit=0;}
#define INC_BITS(bs,b) if(((bs)->bit+=(b))>7){(bs)->cur+=(bs)->bit>>3;(bs)->bit&=7;}
#define BYTE_ALIGN(bs) if((bs)->bit){(bs)->cur++;(bs)->bit=0;}
#define CHECK_BOUND(bs,n) if((bs)->cur+(n)>(bs)->end)return(H323_ERROR_BOUND)
-static unsigned int get_len(bitstr_t *bs);
-static unsigned int get_bit(bitstr_t *bs);
-static unsigned int get_bits(bitstr_t *bs, unsigned int b);
-static unsigned int get_bitmap(bitstr_t *bs, unsigned int b);
-static unsigned int get_uint(bitstr_t *bs, int b);
+static unsigned int get_len(struct bitstr *bs);
+static unsigned int get_bit(struct bitstr *bs);
+static unsigned int get_bits(struct bitstr *bs, unsigned int b);
+static unsigned int get_bitmap(struct bitstr *bs, unsigned int b);
+static unsigned int get_uint(struct bitstr *bs, int b);
/* Decoder Functions */
-static int decode_nul(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_bool(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_oid(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_int(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_enum(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_bitstr(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_numstr(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_octstr(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_bmpstr(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_seq(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_seqof(bitstr_t *bs, const struct field_t *f, char *base, int level);
-static int decode_choice(bitstr_t *bs, const struct field_t *f, char *base, int level);
+static int decode_nul(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_bool(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_oid(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_int(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_enum(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_bitstr(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_numstr(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_octstr(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_bmpstr(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_seq(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_seqof(struct bitstr *bs, const struct field_t *f, char *base, int level);
+static int decode_choice(struct bitstr *bs, const struct field_t *f, char *base, int level);
/* Decoder Functions Vector */
-typedef int (*decoder_t)(bitstr_t *, const struct field_t *, char *, int);
+typedef int (*decoder_t)(struct bitstr *, const struct field_t *, char *, int);
static const decoder_t Decoders[] = {
decode_nul,
decode_bool,
@@ -150,7 +150,7 @@ static const decoder_t Decoders[] = {
* Functions
****************************************************************************/
/* Assume bs is aligned && v < 16384 */
-static unsigned int get_len(bitstr_t *bs)
+static unsigned int get_len(struct bitstr *bs)
{
unsigned int v;
@@ -166,7 +166,7 @@ static unsigned int get_len(bitstr_t *bs)
}
/****************************************************************************/
-static unsigned int get_bit(bitstr_t *bs)
+static unsigned int get_bit(struct bitstr *bs)
{
unsigned int b = (*bs->cur) & (0x80 >> bs->bit);
@@ -177,7 +177,7 @@ static unsigned int get_bit(bitstr_t *bs)
/****************************************************************************/
/* Assume b <= 8 */
-static unsigned int get_bits(bitstr_t *bs, unsigned int b)
+static unsigned int get_bits(struct bitstr *bs, unsigned int b)
{
unsigned int v, l;
@@ -203,7 +203,7 @@ static unsigned int get_bits(bitstr_t *bs, unsigned int b)
/****************************************************************************/
/* Assume b <= 32 */
-static unsigned int get_bitmap(bitstr_t *bs, unsigned int b)
+static unsigned int get_bitmap(struct bitstr *bs, unsigned int b)
{
unsigned int v, l, shift, bytes;
@@ -242,7 +242,7 @@ static unsigned int get_bitmap(bitstr_t *bs, unsigned int b)
/****************************************************************************
* Assume bs is aligned and sizeof(unsigned int) == 4
****************************************************************************/
-static unsigned int get_uint(bitstr_t *bs, int b)
+static unsigned int get_uint(struct bitstr *bs, int b)
{
unsigned int v = 0;
@@ -264,7 +264,7 @@ static unsigned int get_uint(bitstr_t *bs, int b)
}
/****************************************************************************/
-static int decode_nul(bitstr_t *bs, const struct field_t *f,
+static int decode_nul(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name);
@@ -273,7 +273,7 @@ static int decode_nul(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_bool(bitstr_t *bs, const struct field_t *f,
+static int decode_bool(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name);
@@ -285,7 +285,7 @@ static int decode_bool(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_oid(bitstr_t *bs, const struct field_t *f,
+static int decode_oid(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
int len;
@@ -302,7 +302,7 @@ static int decode_oid(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_int(bitstr_t *bs, const struct field_t *f,
+static int decode_int(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int len;
@@ -346,7 +346,7 @@ static int decode_int(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_enum(bitstr_t *bs, const struct field_t *f,
+static int decode_enum(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
PRINT("%*.s%s\n", level * TAB_SIZE, " ", f->name);
@@ -362,7 +362,7 @@ static int decode_enum(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_bitstr(bitstr_t *bs, const struct field_t *f,
+static int decode_bitstr(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int len;
@@ -396,7 +396,7 @@ static int decode_bitstr(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_numstr(bitstr_t *bs, const struct field_t *f,
+static int decode_numstr(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int len;
@@ -414,7 +414,7 @@ static int decode_numstr(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_octstr(bitstr_t *bs, const struct field_t *f,
+static int decode_octstr(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int len;
@@ -463,7 +463,7 @@ static int decode_octstr(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_bmpstr(bitstr_t *bs, const struct field_t *f,
+static int decode_bmpstr(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int len;
@@ -489,7 +489,7 @@ static int decode_bmpstr(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_seq(bitstr_t *bs, const struct field_t *f,
+static int decode_seq(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int ext, bmp, i, opt, len = 0, bmp2, bmp2_len;
@@ -606,7 +606,7 @@ static int decode_seq(bitstr_t *bs, const struct field_t *f,
}
/****************************************************************************/
-static int decode_seqof(bitstr_t *bs, const struct field_t *f,
+static int decode_seqof(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int count, effective_count = 0, i, len = 0;
@@ -696,7 +696,7 @@ static int decode_seqof(bitstr_t *bs, const struct field_t *f,
/****************************************************************************/
-static int decode_choice(bitstr_t *bs, const struct field_t *f,
+static int decode_choice(struct bitstr *bs, const struct field_t *f,
char *base, int level)
{
unsigned int type, ext, len = 0;
@@ -772,7 +772,7 @@ int DecodeRasMessage(unsigned char *buf, size_t sz, RasMessage *ras)
FNAME("RasMessage") CHOICE, 5, 24, 32, DECODE | EXT,
0, _RasMessage
};
- bitstr_t bs;
+ struct bitstr bs;
bs.buf = bs.beg = bs.cur = buf;
bs.end = buf + sz;
@@ -789,7 +789,7 @@ static int DecodeH323_UserInformation(unsigned char *buf, unsigned char *beg,
FNAME("H323-UserInformation") SEQ, 1, 2, 2, DECODE | EXT,
0, _H323_UserInformation
};
- bitstr_t bs;
+ struct bitstr bs;
bs.buf = buf;
bs.beg = bs.cur = beg;
@@ -808,7 +808,7 @@ int DecodeMultimediaSystemControlMessage(unsigned char *buf, size_t sz,
FNAME("MultimediaSystemControlMessage") CHOICE, 2, 4, 4,
DECODE | EXT, 0, _MultimediaSystemControlMessage
};
- bitstr_t bs;
+ struct bitstr bs;
bs.buf = bs.beg = bs.cur = buf;
bs.end = buf + sz;
--
2.11.0
^ permalink raw reply related
* Re: [PATCH net-next 1/5] net: dsa: use port's cpu_dp when creating a slave
From: Florian Fainelli @ 2017-10-12 22:53 UTC (permalink / raw)
To: Vivien Didelot, netdev; +Cc: linux-kernel, kernel, David S. Miller, Andrew Lunn
In-Reply-To: <20171012225156.20758-2-vivien.didelot@savoirfairelinux.com>
On 10/12/2017 03:51 PM, Vivien Didelot wrote:
> When dsa_slave_create is called, the related port already has a CPU port
> assigned to it, available in its cpu_dp member. Use it instead of the
> unique tree cpu_dp.
>
> Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
--
Florian
^ permalink raw reply
* [PATCH net-next 5/5] net: dsa: split dsa_port's netdev member
From: Vivien Didelot @ 2017-10-12 22:51 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, kernel, David S. Miller, Florian Fainelli,
Andrew Lunn, Vivien Didelot
In-Reply-To: <20171012225156.20758-1-vivien.didelot@savoirfairelinux.com>
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.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
---
drivers/net/dsa/bcm_sf2.c | 6 +++---
drivers/net/dsa/mt7530.c | 2 +-
drivers/net/dsa/mv88e6xxx/chip.c | 2 +-
include/net/dsa.h | 7 ++++++-
net/dsa/dsa.c | 6 +++---
net/dsa/dsa2.c | 22 +++++++++++-----------
net/dsa/dsa_priv.h | 4 ++--
net/dsa/legacy.c | 14 +++++++-------
net/dsa/slave.c | 6 +++---
9 files changed, 37 insertions(+), 32 deletions(-)
diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c
index 32025b990437..b43c063b9634 100644
--- a/drivers/net/dsa/bcm_sf2.c
+++ b/drivers/net/dsa/bcm_sf2.c
@@ -601,7 +601,7 @@ static void bcm_sf2_sw_fixed_link_update(struct dsa_switch *ds, int port,
* state machine and make it go in PHY_FORCING state instead.
*/
if (!status->link)
- netif_carrier_off(ds->ports[port].netdev);
+ netif_carrier_off(ds->ports[port].slave);
status->duplex = 1;
} else {
status->link = 1;
@@ -690,7 +690,7 @@ static int bcm_sf2_sw_resume(struct dsa_switch *ds)
static void bcm_sf2_sw_get_wol(struct dsa_switch *ds, int port,
struct ethtool_wolinfo *wol)
{
- struct net_device *p = ds->ports[port].cpu_dp->netdev;
+ struct net_device *p = ds->ports[port].cpu_dp->master;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
struct ethtool_wolinfo pwol;
@@ -713,7 +713,7 @@ static void bcm_sf2_sw_get_wol(struct dsa_switch *ds, int port,
static int bcm_sf2_sw_set_wol(struct dsa_switch *ds, int port,
struct ethtool_wolinfo *wol)
{
- struct net_device *p = ds->ports[port].cpu_dp->netdev;
+ struct net_device *p = ds->ports[port].cpu_dp->master;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
s8 cpu_port = ds->ports[port].cpu_dp->index;
struct ethtool_wolinfo pwol;
diff --git a/drivers/net/dsa/mt7530.c b/drivers/net/dsa/mt7530.c
index 034241696ce2..fea2e665d0cb 100644
--- a/drivers/net/dsa/mt7530.c
+++ b/drivers/net/dsa/mt7530.c
@@ -933,7 +933,7 @@ mt7530_setup(struct dsa_switch *ds)
* controller also is the container for two GMACs nodes representing
* as two netdev instances.
*/
- dn = ds->ports[MT7530_CPU_PORT].netdev->dev.of_node->parent;
+ dn = ds->ports[MT7530_CPU_PORT].master->dev.of_node->parent;
priv->ethernet = syscon_node_to_regmap(dn);
if (IS_ERR(priv->ethernet))
return PTR_ERR(priv->ethernet);
diff --git a/drivers/net/dsa/mv88e6xxx/chip.c b/drivers/net/dsa/mv88e6xxx/chip.c
index d74c7335c512..955f4e214191 100644
--- a/drivers/net/dsa/mv88e6xxx/chip.c
+++ b/drivers/net/dsa/mv88e6xxx/chip.c
@@ -1124,7 +1124,7 @@ static int mv88e6xxx_port_check_hw_vlan(struct dsa_switch *ds, int port,
if (dsa_is_dsa_port(ds, i) || dsa_is_cpu_port(ds, i))
continue;
- if (!ds->ports[port].netdev)
+ if (!ds->ports[port].slave)
continue;
if (vlan.member[i] ==
diff --git a/include/net/dsa.h b/include/net/dsa.h
index ce1d622734d7..4c769bc8a8b5 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -164,6 +164,12 @@ struct dsa_mall_tc_entry {
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;
+
/* CPU port tagging operations used by master or slave devices */
const struct dsa_device_ops *tag_ops;
@@ -176,7 +182,6 @@ struct dsa_port {
unsigned int index;
const char *name;
struct dsa_port *cpu_dp;
- struct net_device *netdev;
struct device_node *dn;
unsigned int ageing_time;
u8 stp_state;
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 832c659ff993..a3abf7a7b9a2 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -201,7 +201,7 @@ static int dsa_switch_rcv(struct sk_buff *skb, struct net_device *dev,
#ifdef CONFIG_PM_SLEEP
static bool dsa_is_port_initialized(struct dsa_switch *ds, int p)
{
- return ds->enabled_port_mask & (1 << p) && ds->ports[p].netdev;
+ return ds->enabled_port_mask & (1 << p) && ds->ports[p].slave;
}
int dsa_switch_suspend(struct dsa_switch *ds)
@@ -213,7 +213,7 @@ int dsa_switch_suspend(struct dsa_switch *ds)
if (!dsa_is_port_initialized(ds, i))
continue;
- ret = dsa_slave_suspend(ds->ports[i].netdev);
+ ret = dsa_slave_suspend(ds->ports[i].slave);
if (ret)
return ret;
}
@@ -240,7 +240,7 @@ int dsa_switch_resume(struct dsa_switch *ds)
if (!dsa_is_port_initialized(ds, i))
continue;
- ret = dsa_slave_resume(ds->ports[i].netdev);
+ ret = dsa_slave_resume(ds->ports[i].slave);
if (ret)
return ret;
}
diff --git a/net/dsa/dsa2.c b/net/dsa/dsa2.c
index 54ed054777bd..5ac78edd756f 100644
--- a/net/dsa/dsa2.c
+++ b/net/dsa/dsa2.c
@@ -279,7 +279,7 @@ static int dsa_user_port_apply(struct dsa_port *port)
if (err) {
dev_warn(ds->dev, "Failed to create slave %d: %d\n",
port->index, err);
- port->netdev = NULL;
+ port->slave = NULL;
return err;
}
@@ -289,7 +289,7 @@ static int dsa_user_port_apply(struct dsa_port *port)
if (err)
return err;
- devlink_port_type_eth_set(&port->devlink_port, port->netdev);
+ devlink_port_type_eth_set(&port->devlink_port, port->slave);
return 0;
}
@@ -297,9 +297,9 @@ static int dsa_user_port_apply(struct dsa_port *port)
static void dsa_user_port_unapply(struct dsa_port *port)
{
devlink_port_unregister(&port->devlink_port);
- if (port->netdev) {
- dsa_slave_destroy(port->netdev);
- port->netdev = NULL;
+ if (port->slave) {
+ dsa_slave_destroy(port->slave);
+ port->slave = NULL;
port->ds->enabled_port_mask &= ~(1 << port->index);
}
}
@@ -337,7 +337,7 @@ static int dsa_ds_apply(struct dsa_switch_tree *dst, struct dsa_switch *ds)
return err;
if (ds->ops->set_addr) {
- err = ds->ops->set_addr(ds, dst->cpu_dp->netdev->dev_addr);
+ err = ds->ops->set_addr(ds, dst->cpu_dp->master->dev_addr);
if (err < 0)
return err;
}
@@ -438,9 +438,9 @@ static int dsa_dst_apply(struct dsa_switch_tree *dst)
* sent to the tag format's receive function.
*/
wmb();
- dst->cpu_dp->netdev->dsa_ptr = dst->cpu_dp;
+ dst->cpu_dp->master->dsa_ptr = dst->cpu_dp;
- err = dsa_master_ethtool_setup(dst->cpu_dp->netdev);
+ err = dsa_master_ethtool_setup(dst->cpu_dp->master);
if (err)
return err;
@@ -457,9 +457,9 @@ static void dsa_dst_unapply(struct dsa_switch_tree *dst)
if (!dst->applied)
return;
- dsa_master_ethtool_restore(dst->cpu_dp->netdev);
+ dsa_master_ethtool_restore(dst->cpu_dp->master);
- dst->cpu_dp->netdev->dsa_ptr = NULL;
+ dst->cpu_dp->master->dsa_ptr = NULL;
/* If we used a tagging format that doesn't have an ethertype
* field, make sure that all packets from this point get sent
@@ -505,7 +505,7 @@ static int dsa_cpu_parse(struct dsa_port *port, u32 index,
if (!dst->cpu_dp) {
dst->cpu_dp = port;
- dst->cpu_dp->netdev = ethernet_dev;
+ dst->cpu_dp->master = ethernet_dev;
}
/* Initialize cpu_port_mask now for drv->setup()
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index f1dc5a856fda..b47c46fba376 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -130,7 +130,7 @@ static inline struct net_device *dsa_master_get_slave(struct net_device *dev,
if (port < 0 || port >= ds->num_ports)
return NULL;
- return ds->ports[port].netdev;
+ return ds->ports[port].slave;
}
/* port.c */
@@ -181,7 +181,7 @@ dsa_slave_get_master(const struct net_device *dev)
{
struct dsa_port *dp = dsa_slave_to_port(dev);
- return dp->cpu_dp->netdev;
+ return dp->cpu_dp->master;
}
/* switch.c */
diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c
index 6f2254753859..ce6bc80911d0 100644
--- a/net/dsa/legacy.c
+++ b/net/dsa/legacy.c
@@ -120,7 +120,7 @@ static int dsa_switch_setup_one(struct dsa_switch *ds,
return -EINVAL;
}
dst->cpu_dp = &ds->ports[i];
- dst->cpu_dp->netdev = master;
+ dst->cpu_dp->master = master;
ds->cpu_port_mask |= 1 << i;
} else if (!strcmp(name, "dsa")) {
ds->dsa_port_mask |= 1 << i;
@@ -267,10 +267,10 @@ static void dsa_switch_destroy(struct dsa_switch *ds)
if (!(ds->enabled_port_mask & (1 << port)))
continue;
- if (!ds->ports[port].netdev)
+ if (!ds->ports[port].slave)
continue;
- dsa_slave_destroy(ds->ports[port].netdev);
+ dsa_slave_destroy(ds->ports[port].slave);
}
/* Disable configuration of the CPU and DSA ports */
@@ -607,7 +607,7 @@ static int dsa_setup_dst(struct dsa_switch_tree *dst, struct net_device *dev,
wmb();
dev->dsa_ptr = dst->cpu_dp;
- return dsa_master_ethtool_setup(dst->cpu_dp->netdev);
+ return dsa_master_ethtool_setup(dst->cpu_dp->master);
}
static int dsa_probe(struct platform_device *pdev)
@@ -672,9 +672,9 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst)
{
int i;
- dsa_master_ethtool_restore(dst->cpu_dp->netdev);
+ dsa_master_ethtool_restore(dst->cpu_dp->master);
- dst->cpu_dp->netdev->dsa_ptr = NULL;
+ dst->cpu_dp->master->dsa_ptr = NULL;
/* If we used a tagging format that doesn't have an ethertype
* field, make sure that all packets from this point get sent
@@ -689,7 +689,7 @@ static void dsa_remove_dst(struct dsa_switch_tree *dst)
dsa_switch_destroy(ds);
}
- dev_put(dst->cpu_dp->netdev);
+ dev_put(dst->cpu_dp->master);
}
static int dsa_remove(struct platform_device *pdev)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index d2c780f13d78..672977689e1b 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1123,7 +1123,7 @@ static void dsa_slave_notify(struct net_device *dev, unsigned long val)
int dsa_slave_create(struct dsa_port *port, const char *name)
{
struct dsa_port *cpu_dp = port->cpu_dp;
- struct net_device *master = cpu_dp->netdev;
+ struct net_device *master = cpu_dp->master;
struct dsa_switch *ds = port->ds;
struct net_device *slave_dev;
struct dsa_slave_priv *p;
@@ -1170,7 +1170,7 @@ int dsa_slave_create(struct dsa_port *port, const char *name)
p->old_link = -1;
p->old_duplex = -1;
- port->netdev = slave_dev;
+ port->slave = slave_dev;
netif_carrier_off(slave_dev);
@@ -1198,7 +1198,7 @@ int dsa_slave_create(struct dsa_port *port, const char *name)
out_free:
free_percpu(p->stats64);
free_netdev(slave_dev);
- port->netdev = NULL;
+ port->slave = NULL;
return ret;
}
--
2.14.2
^ permalink raw reply related
* [PATCH net-next 4/5] net: dsa: add slave get master helper
From: Vivien Didelot @ 2017-10-12 22:51 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, kernel, David S. Miller, Florian Fainelli,
Andrew Lunn, Vivien Didelot
In-Reply-To: <20171012225156.20758-1-vivien.didelot@savoirfairelinux.com>
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>
---
net/dsa/dsa_priv.h | 13 ++++++++-----
net/dsa/slave.c | 26 +++++++++-----------------
2 files changed, 17 insertions(+), 22 deletions(-)
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 569a4929b4c9..f1dc5a856fda 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -176,6 +176,14 @@ static inline struct dsa_port *dsa_slave_to_port(const struct net_device *dev)
return p->dp;
}
+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;
+}
+
/* switch.c */
int dsa_switch_register_notifier(struct dsa_switch *ds);
void dsa_switch_unregister_notifier(struct dsa_switch *ds);
@@ -204,9 +212,4 @@ extern const struct dsa_device_ops qca_netdev_ops;
/* tag_trailer.c */
extern const struct dsa_device_ops trailer_netdev_ops;
-static inline struct net_device *dsa_master_netdev(struct dsa_slave_priv *p)
-{
- return p->dp->cpu_dp->netdev;
-}
-
#endif
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 894602c88b09..d2c780f13d78 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -64,15 +64,12 @@ void dsa_slave_mii_bus_init(struct dsa_switch *ds)
/* slave device handling ****************************************************/
static int dsa_slave_get_iflink(const struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
-
- return dsa_master_netdev(p)->ifindex;
+ return dsa_slave_get_master(dev)->ifindex;
}
static int dsa_slave_open(struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
+ struct net_device *master = dsa_slave_get_master(dev);
struct dsa_port *dp = dsa_slave_to_port(dev);
int err;
@@ -120,8 +117,7 @@ static int dsa_slave_open(struct net_device *dev)
static int dsa_slave_close(struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
+ struct net_device *master = dsa_slave_get_master(dev);
struct dsa_port *dp = dsa_slave_to_port(dev);
if (dev->phydev)
@@ -144,8 +140,7 @@ static int dsa_slave_close(struct net_device *dev)
static void dsa_slave_change_rx_flags(struct net_device *dev, int change)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
+ struct net_device *master = dsa_slave_get_master(dev);
if (change & IFF_ALLMULTI)
dev_set_allmulti(master, dev->flags & IFF_ALLMULTI ? 1 : -1);
@@ -155,8 +150,7 @@ static void dsa_slave_change_rx_flags(struct net_device *dev, int change)
static void dsa_slave_set_rx_mode(struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
+ struct net_device *master = dsa_slave_get_master(dev);
dev_mc_sync(master, dev);
dev_uc_sync(master, dev);
@@ -164,8 +158,7 @@ static void dsa_slave_set_rx_mode(struct net_device *dev)
static int dsa_slave_set_mac_address(struct net_device *dev, void *a)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
+ struct net_device *master = dsa_slave_get_master(dev);
struct sockaddr *addr = a;
int err;
@@ -409,7 +402,7 @@ static netdev_tx_t dsa_slave_xmit(struct sk_buff *skb, struct net_device *dev)
/* Queue the SKB for transmission on the parent interface, but
* do not modify its EtherType
*/
- nskb->dev = dsa_master_netdev(p);
+ nskb->dev = dsa_slave_get_master(dev);
dev_queue_xmit(nskb);
return NETDEV_TX_OK;
@@ -632,8 +625,8 @@ static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e)
static int dsa_slave_netpoll_setup(struct net_device *dev,
struct netpoll_info *ni)
{
+ struct net_device *master = dsa_slave_get_master(dev);
struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
struct netpoll *netpoll;
int err = 0;
@@ -1115,8 +1108,7 @@ int dsa_slave_resume(struct net_device *slave_dev)
static void dsa_slave_notify(struct net_device *dev, unsigned long val)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct net_device *master = dsa_master_netdev(p);
+ struct net_device *master = dsa_slave_get_master(dev);
struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_notifier_register_info rinfo = {
.switch_number = dp->ds->index,
--
2.14.2
^ permalink raw reply related
* [PATCH net-next 3/5] net: dsa: add slave to port helper
From: Vivien Didelot @ 2017-10-12 22:51 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, kernel, David S. Miller, Florian Fainelli,
Andrew Lunn, Vivien Didelot
In-Reply-To: <20171012225156.20758-1-vivien.didelot@savoirfairelinux.com>
Many portions of DSA core code require to get the dsa_port structure
corresponding to a slave net_device. For this purpose, introduce a
dsa_slave_to_port() helper.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
---
net/dsa/dsa_priv.h | 7 +++
net/dsa/legacy.c | 6 +-
net/dsa/slave.c | 167 +++++++++++++++++++++++++-------------------------
net/dsa/tag_brcm.c | 9 ++-
net/dsa/tag_dsa.c | 10 +--
net/dsa/tag_edsa.c | 10 +--
net/dsa/tag_ksz.c | 4 +-
net/dsa/tag_lan9303.c | 4 +-
net/dsa/tag_mtk.c | 4 +-
net/dsa/tag_qca.c | 5 +-
net/dsa/tag_trailer.c | 4 +-
11 files changed, 115 insertions(+), 115 deletions(-)
diff --git a/net/dsa/dsa_priv.h b/net/dsa/dsa_priv.h
index 2850077cc9cc..569a4929b4c9 100644
--- a/net/dsa/dsa_priv.h
+++ b/net/dsa/dsa_priv.h
@@ -169,6 +169,13 @@ int dsa_slave_resume(struct net_device *slave_dev);
int dsa_slave_register_notifier(void);
void dsa_slave_unregister_notifier(void);
+static inline struct dsa_port *dsa_slave_to_port(const struct net_device *dev)
+{
+ struct dsa_slave_priv *p = netdev_priv(dev);
+
+ return p->dp;
+}
+
/* switch.c */
int dsa_switch_register_notifier(struct dsa_switch *ds);
void dsa_switch_unregister_notifier(struct dsa_switch *ds);
diff --git a/net/dsa/legacy.c b/net/dsa/legacy.c
index 19ff6e0a21dc..6f2254753859 100644
--- a/net/dsa/legacy.c
+++ b/net/dsa/legacy.c
@@ -746,8 +746,7 @@ int dsa_legacy_fdb_add(struct ndmsg *ndm, struct nlattr *tb[],
const unsigned char *addr, u16 vid,
u16 flags)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
return dsa_port_fdb_add(dp, addr, vid);
}
@@ -756,8 +755,7 @@ int dsa_legacy_fdb_del(struct ndmsg *ndm, struct nlattr *tb[],
struct net_device *dev,
const unsigned char *addr, u16 vid)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
return dsa_port_fdb_del(dp, addr, vid);
}
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index f31737256f69..894602c88b09 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -72,8 +72,8 @@ static int dsa_slave_get_iflink(const struct net_device *dev)
static int dsa_slave_open(struct net_device *dev)
{
struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
struct net_device *master = dsa_master_netdev(p);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
int err;
if (!(master->flags & IFF_UP))
@@ -122,7 +122,7 @@ static int dsa_slave_close(struct net_device *dev)
{
struct dsa_slave_priv *p = netdev_priv(dev);
struct net_device *master = dsa_master_netdev(p);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
if (dev->phydev)
phy_stop(dev->phydev);
@@ -246,14 +246,13 @@ dsa_slave_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb,
struct net_device *dev, struct net_device *filter_dev,
int *idx)
{
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_slave_dump_ctx dump = {
.dev = dev,
.skb = skb,
.cb = cb,
.idx = *idx,
};
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
int err;
err = dsa_port_fdb_dump(dp, dsa_slave_port_fdb_do_dump, &dump);
@@ -274,8 +273,7 @@ static int dsa_slave_port_attr_set(struct net_device *dev,
const struct switchdev_attr *attr,
struct switchdev_trans *trans)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
int ret;
switch (attr->id) {
@@ -301,8 +299,7 @@ static int dsa_slave_port_obj_add(struct net_device *dev,
const struct switchdev_obj *obj,
struct switchdev_trans *trans)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
int err;
/* For the prepare phase, ensure the full set of changes is feasable in
@@ -329,8 +326,7 @@ static int dsa_slave_port_obj_add(struct net_device *dev,
static int dsa_slave_port_obj_del(struct net_device *dev,
const struct switchdev_obj *obj)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
int err;
switch (obj->id) {
@@ -351,8 +347,8 @@ static int dsa_slave_port_obj_del(struct net_device *dev,
static int dsa_slave_port_attr_get(struct net_device *dev,
struct switchdev_attr *attr)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
switch (attr->id) {
case SWITCHDEV_ATTR_ID_PORT_PARENT_ID:
@@ -431,11 +427,11 @@ static void dsa_slave_get_drvinfo(struct net_device *dev,
static int dsa_slave_get_regs_len(struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (ds->ops->get_regs_len)
- return ds->ops->get_regs_len(ds, p->dp->index);
+ return ds->ops->get_regs_len(ds, dp->index);
return -EOPNOTSUPP;
}
@@ -443,11 +439,11 @@ static int dsa_slave_get_regs_len(struct net_device *dev)
static void
dsa_slave_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *_p)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (ds->ops->get_regs)
- ds->ops->get_regs(ds, p->dp->index, regs, _p);
+ ds->ops->get_regs(ds, dp->index, regs, _p);
}
static u32 dsa_slave_get_link(struct net_device *dev)
@@ -462,8 +458,8 @@ static u32 dsa_slave_get_link(struct net_device *dev)
static int dsa_slave_get_eeprom_len(struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (ds->cd && ds->cd->eeprom_len)
return ds->cd->eeprom_len;
@@ -477,8 +473,8 @@ static int dsa_slave_get_eeprom_len(struct net_device *dev)
static int dsa_slave_get_eeprom(struct net_device *dev,
struct ethtool_eeprom *eeprom, u8 *data)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (ds->ops->get_eeprom)
return ds->ops->get_eeprom(ds, eeprom, data);
@@ -489,8 +485,8 @@ static int dsa_slave_get_eeprom(struct net_device *dev,
static int dsa_slave_set_eeprom(struct net_device *dev,
struct ethtool_eeprom *eeprom, u8 *data)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (ds->ops->set_eeprom)
return ds->ops->set_eeprom(ds, eeprom, data);
@@ -501,8 +497,8 @@ static int dsa_slave_set_eeprom(struct net_device *dev,
static void dsa_slave_get_strings(struct net_device *dev,
uint32_t stringset, uint8_t *data)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (stringset == ETH_SS_STATS) {
int len = ETH_GSTRING_LEN;
@@ -512,7 +508,7 @@ static void dsa_slave_get_strings(struct net_device *dev,
strncpy(data + 2 * len, "rx_packets", len);
strncpy(data + 3 * len, "rx_bytes", len);
if (ds->ops->get_strings)
- ds->ops->get_strings(ds, p->dp->index, data + 4 * len);
+ ds->ops->get_strings(ds, dp->index, data + 4 * len);
}
}
@@ -520,8 +516,9 @@ static void dsa_slave_get_ethtool_stats(struct net_device *dev,
struct ethtool_stats *stats,
uint64_t *data)
{
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_switch *ds = dp->ds;
struct pcpu_sw_netstats *s;
unsigned int start;
int i;
@@ -543,13 +540,13 @@ static void dsa_slave_get_ethtool_stats(struct net_device *dev,
data[3] += rx_bytes;
}
if (ds->ops->get_ethtool_stats)
- ds->ops->get_ethtool_stats(ds, p->dp->index, data + 4);
+ ds->ops->get_ethtool_stats(ds, dp->index, data + 4);
}
static int dsa_slave_get_sset_count(struct net_device *dev, int sset)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (sset == ETH_SS_STATS) {
int count;
@@ -566,29 +563,29 @@ static int dsa_slave_get_sset_count(struct net_device *dev, int sset)
static void dsa_slave_get_wol(struct net_device *dev, struct ethtool_wolinfo *w)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (ds->ops->get_wol)
- ds->ops->get_wol(ds, p->dp->index, w);
+ ds->ops->get_wol(ds, dp->index, w);
}
static int dsa_slave_set_wol(struct net_device *dev, struct ethtool_wolinfo *w)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
int ret = -EOPNOTSUPP;
if (ds->ops->set_wol)
- ret = ds->ops->set_wol(ds, p->dp->index, w);
+ ret = ds->ops->set_wol(ds, dp->index, w);
return ret;
}
static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
int ret;
/* Port's PHY and MAC both need to be EEE capable */
@@ -598,7 +595,7 @@ static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e)
if (!ds->ops->set_mac_eee)
return -EOPNOTSUPP;
- ret = ds->ops->set_mac_eee(ds, p->dp->index, e);
+ ret = ds->ops->set_mac_eee(ds, dp->index, e);
if (ret)
return ret;
@@ -613,8 +610,8 @@ static int dsa_slave_set_eee(struct net_device *dev, struct ethtool_eee *e)
static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
int ret;
/* Port's PHY and MAC both need to be EEE capable */
@@ -624,7 +621,7 @@ static int dsa_slave_get_eee(struct net_device *dev, struct ethtool_eee *e)
if (!ds->ops->get_mac_eee)
return -EOPNOTSUPP;
- ret = ds->ops->get_mac_eee(ds, p->dp->index, e);
+ ret = ds->ops->get_mac_eee(ds, dp->index, e);
if (ret)
return ret;
@@ -676,9 +673,9 @@ static void dsa_slave_poll_controller(struct net_device *dev)
static int dsa_slave_get_phys_port_name(struct net_device *dev,
char *name, size_t len)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
- if (snprintf(name, len, "p%d", p->dp->index) >= len)
+ if (snprintf(name, len, "p%d", dp->index) >= len)
return -EINVAL;
return 0;
@@ -701,14 +698,15 @@ static int dsa_slave_add_cls_matchall(struct net_device *dev,
struct tc_cls_matchall_offload *cls,
bool ingress)
{
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_slave_priv *p = netdev_priv(dev);
struct dsa_mall_tc_entry *mall_tc_entry;
__be16 protocol = cls->common.protocol;
- struct dsa_switch *ds = p->dp->ds;
struct net *net = dev_net(dev);
- struct dsa_slave_priv *to_p;
+ struct dsa_switch *ds = dp->ds;
struct net_device *to_dev;
const struct tc_action *a;
+ struct dsa_port *to_dp;
int err = -EOPNOTSUPP;
LIST_HEAD(actions);
int ifindex;
@@ -741,13 +739,12 @@ static int dsa_slave_add_cls_matchall(struct net_device *dev,
mall_tc_entry->type = DSA_PORT_MALL_MIRROR;
mirror = &mall_tc_entry->mirror;
- to_p = netdev_priv(to_dev);
+ to_dp = dsa_slave_to_port(to_dev);
- mirror->to_local_port = to_p->dp->index;
+ mirror->to_local_port = to_dp->index;
mirror->ingress = ingress;
- err = ds->ops->port_mirror_add(ds, p->dp->index, mirror,
- ingress);
+ err = ds->ops->port_mirror_add(ds, dp->index, mirror, ingress);
if (err) {
kfree(mall_tc_entry);
return err;
@@ -762,9 +759,9 @@ static int dsa_slave_add_cls_matchall(struct net_device *dev,
static void dsa_slave_del_cls_matchall(struct net_device *dev,
struct tc_cls_matchall_offload *cls)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_mall_tc_entry *mall_tc_entry;
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_switch *ds = dp->ds;
if (!ds->ops->port_mirror_del)
return;
@@ -777,8 +774,7 @@ static void dsa_slave_del_cls_matchall(struct net_device *dev,
switch (mall_tc_entry->type) {
case DSA_PORT_MALL_MIRROR:
- ds->ops->port_mirror_del(ds, p->dp->index,
- &mall_tc_entry->mirror);
+ ds->ops->port_mirror_del(ds, dp->index, &mall_tc_entry->mirror);
break;
default:
WARN_ON(1);
@@ -855,25 +851,25 @@ static void dsa_slave_get_stats64(struct net_device *dev,
static int dsa_slave_get_rxnfc(struct net_device *dev,
struct ethtool_rxnfc *nfc, u32 *rule_locs)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (!ds->ops->get_rxnfc)
return -EOPNOTSUPP;
- return ds->ops->get_rxnfc(ds, p->dp->index, nfc, rule_locs);
+ return ds->ops->get_rxnfc(ds, dp->index, nfc, rule_locs);
}
static int dsa_slave_set_rxnfc(struct net_device *dev,
struct ethtool_rxnfc *nfc)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
+ struct dsa_switch *ds = dp->ds;
if (!ds->ops->set_rxnfc)
return -EOPNOTSUPP;
- return ds->ops->set_rxnfc(ds, p->dp->index, nfc);
+ return ds->ops->set_rxnfc(ds, dp->index, nfc);
}
static const struct ethtool_ops dsa_slave_ethtool_ops = {
@@ -933,8 +929,9 @@ static struct device_type dsa_type = {
static void dsa_slave_adjust_link(struct net_device *dev)
{
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_switch *ds = dp->ds;
unsigned int status_changed = 0;
if (p->old_link != dev->phydev->link) {
@@ -953,7 +950,7 @@ static void dsa_slave_adjust_link(struct net_device *dev)
}
if (ds->ops->adjust_link && status_changed)
- ds->ops->adjust_link(ds, p->dp->index, dev->phydev);
+ ds->ops->adjust_link(ds, dp->index, dev->phydev);
if (status_changed)
phy_print_status(dev->phydev);
@@ -962,14 +959,14 @@ static void dsa_slave_adjust_link(struct net_device *dev)
static int dsa_slave_fixed_link_update(struct net_device *dev,
struct fixed_phy_status *status)
{
- struct dsa_slave_priv *p;
struct dsa_switch *ds;
+ struct dsa_port *dp;
if (dev) {
- p = netdev_priv(dev);
- ds = p->dp->ds;
+ dp = dsa_slave_to_port(dev);
+ ds = dp->ds;
if (ds->ops->fixed_link_update)
- ds->ops->fixed_link_update(ds, p->dp->index, status);
+ ds->ops->fixed_link_update(ds, dp->index, status);
}
return 0;
@@ -978,8 +975,9 @@ static int dsa_slave_fixed_link_update(struct net_device *dev,
/* slave device setup *******************************************************/
static int dsa_slave_phy_connect(struct net_device *slave_dev, int addr)
{
+ struct dsa_port *dp = dsa_slave_to_port(slave_dev);
struct dsa_slave_priv *p = netdev_priv(slave_dev);
- struct dsa_switch *ds = p->dp->ds;
+ struct dsa_switch *ds = dp->ds;
slave_dev->phydev = mdiobus_get_phy(ds->slave_mii_bus, addr);
if (!slave_dev->phydev) {
@@ -997,14 +995,15 @@ static int dsa_slave_phy_connect(struct net_device *slave_dev, int addr)
static int dsa_slave_phy_setup(struct net_device *slave_dev)
{
+ struct dsa_port *dp = dsa_slave_to_port(slave_dev);
struct dsa_slave_priv *p = netdev_priv(slave_dev);
- struct dsa_switch *ds = p->dp->ds;
- struct device_node *phy_dn, *port_dn;
+ struct device_node *port_dn = dp->dn;
+ struct dsa_switch *ds = dp->ds;
+ struct device_node *phy_dn;
bool phy_is_fixed = false;
u32 phy_flags = 0;
int mode, ret;
- port_dn = p->dp->dn;
mode = of_get_phy_mode(port_dn);
if (mode < 0)
mode = PHY_INTERFACE_MODE_NA;
@@ -1025,7 +1024,7 @@ static int dsa_slave_phy_setup(struct net_device *slave_dev)
}
if (ds->ops->get_phy_flags)
- phy_flags = ds->ops->get_phy_flags(ds, p->dp->index);
+ phy_flags = ds->ops->get_phy_flags(ds, dp->index);
if (phy_dn) {
int phy_id = of_mdio_parse_addr(&slave_dev->dev, phy_dn);
@@ -1061,10 +1060,10 @@ static int dsa_slave_phy_setup(struct net_device *slave_dev)
* MDIO bus instead
*/
if (!slave_dev->phydev) {
- ret = dsa_slave_phy_connect(slave_dev, p->dp->index);
+ ret = dsa_slave_phy_connect(slave_dev, dp->index);
if (ret) {
netdev_err(slave_dev, "failed to connect to port %d: %d\n",
- p->dp->index, ret);
+ dp->index, ret);
if (phy_is_fixed)
of_phy_deregister_fixed_link(port_dn);
return ret;
@@ -1118,7 +1117,7 @@ static void dsa_slave_notify(struct net_device *dev, unsigned long val)
{
struct dsa_slave_priv *p = netdev_priv(dev);
struct net_device *master = dsa_master_netdev(p);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct dsa_notifier_register_info rinfo = {
.switch_number = dp->ds->index,
.port_number = dp->index,
@@ -1202,8 +1201,8 @@ int dsa_slave_create(struct dsa_port *port, const char *name)
out_phy:
phy_disconnect(slave_dev->phydev);
- if (of_phy_is_fixed_link(p->dp->dn))
- of_phy_deregister_fixed_link(p->dp->dn);
+ if (of_phy_is_fixed_link(port->dn))
+ of_phy_deregister_fixed_link(port->dn);
out_free:
free_percpu(p->stats64);
free_netdev(slave_dev);
@@ -1213,10 +1212,9 @@ int dsa_slave_create(struct dsa_port *port, const char *name)
void dsa_slave_destroy(struct net_device *slave_dev)
{
+ struct dsa_port *dp = dsa_slave_to_port(slave_dev);
struct dsa_slave_priv *p = netdev_priv(slave_dev);
- struct device_node *port_dn;
-
- port_dn = p->dp->dn;
+ struct device_node *port_dn = dp->dn;
netif_carrier_off(slave_dev);
if (slave_dev->phydev) {
@@ -1239,8 +1237,7 @@ static bool dsa_slave_dev_check(struct net_device *dev)
static int dsa_slave_changeupper(struct net_device *dev,
struct netdev_notifier_changeupper_info *info)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
- struct dsa_port *dp = p->dp;
+ struct dsa_port *dp = dsa_slave_to_port(dev);
int err = NOTIFY_DONE;
if (netif_is_bridge_master(info->upper_dev)) {
@@ -1283,14 +1280,14 @@ static void dsa_slave_switchdev_event_work(struct work_struct *work)
container_of(work, struct dsa_switchdev_event_work, work);
struct net_device *dev = switchdev_work->dev;
struct switchdev_notifier_fdb_info *fdb_info;
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
int err;
rtnl_lock();
switch (switchdev_work->event) {
case SWITCHDEV_FDB_ADD_TO_DEVICE:
fdb_info = &switchdev_work->fdb_info;
- err = dsa_port_fdb_add(p->dp, fdb_info->addr, fdb_info->vid);
+ err = dsa_port_fdb_add(dp, fdb_info->addr, fdb_info->vid);
if (err) {
netdev_dbg(dev, "fdb add failed err=%d\n", err);
break;
@@ -1301,7 +1298,7 @@ static void dsa_slave_switchdev_event_work(struct work_struct *work)
case SWITCHDEV_FDB_DEL_TO_DEVICE:
fdb_info = &switchdev_work->fdb_info;
- err = dsa_port_fdb_del(p->dp, fdb_info->addr, fdb_info->vid);
+ err = dsa_port_fdb_del(dp, fdb_info->addr, fdb_info->vid);
if (err) {
netdev_dbg(dev, "fdb del failed err=%d\n", err);
dev_close(dev);
diff --git a/net/dsa/tag_brcm.c b/net/dsa/tag_brcm.c
index cc4f472fbd77..2a0fa9f9f7ca 100644
--- a/net/dsa/tag_brcm.c
+++ b/net/dsa/tag_brcm.c
@@ -61,7 +61,7 @@
static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
u16 queue = skb_get_queue_mapping(skb);
u8 *brcm_tag;
@@ -82,15 +82,14 @@ static struct sk_buff *brcm_tag_xmit(struct sk_buff *skb, struct net_device *dev
((queue & BRCM_IG_TC_MASK) << BRCM_IG_TC_SHIFT);
brcm_tag[1] = 0;
brcm_tag[2] = 0;
- if (p->dp->index == 8)
+ if (dp->index == 8)
brcm_tag[2] = BRCM_IG_DSTMAP2_MASK;
- brcm_tag[3] = (1 << p->dp->index) & BRCM_IG_DSTMAP1_MASK;
+ brcm_tag[3] = (1 << dp->index) & BRCM_IG_DSTMAP1_MASK;
/* Now tell the master network device about the desired output queue
* as well
*/
- skb_set_queue_mapping(skb, BRCM_TAG_SET_PORT_QUEUE(p->dp->index,
- queue));
+ skb_set_queue_mapping(skb, BRCM_TAG_SET_PORT_QUEUE(dp->index, queue));
return skb;
}
diff --git a/net/dsa/tag_dsa.c b/net/dsa/tag_dsa.c
index c77218f173d1..fe3c9700a8c8 100644
--- a/net/dsa/tag_dsa.c
+++ b/net/dsa/tag_dsa.c
@@ -18,7 +18,7 @@
static struct sk_buff *dsa_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
u8 *dsa_header;
/*
@@ -34,8 +34,8 @@ static struct sk_buff *dsa_xmit(struct sk_buff *skb, struct net_device *dev)
* Construct tagged FROM_CPU DSA tag from 802.1q tag.
*/
dsa_header = skb->data + 2 * ETH_ALEN;
- dsa_header[0] = 0x60 | p->dp->ds->index;
- dsa_header[1] = p->dp->index << 3;
+ dsa_header[0] = 0x60 | dp->ds->index;
+ dsa_header[1] = dp->index << 3;
/*
* Move CFI field from byte 2 to byte 1.
@@ -55,8 +55,8 @@ static struct sk_buff *dsa_xmit(struct sk_buff *skb, struct net_device *dev)
* Construct untagged FROM_CPU DSA tag.
*/
dsa_header = skb->data + 2 * ETH_ALEN;
- dsa_header[0] = 0x40 | p->dp->ds->index;
- dsa_header[1] = p->dp->index << 3;
+ dsa_header[0] = 0x40 | dp->ds->index;
+ dsa_header[1] = dp->index << 3;
dsa_header[2] = 0x00;
dsa_header[3] = 0x00;
}
diff --git a/net/dsa/tag_edsa.c b/net/dsa/tag_edsa.c
index 0b83cbe0c9e8..c683e240bafc 100644
--- a/net/dsa/tag_edsa.c
+++ b/net/dsa/tag_edsa.c
@@ -19,7 +19,7 @@
static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
u8 *edsa_header;
/*
@@ -43,8 +43,8 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
edsa_header[1] = ETH_P_EDSA & 0xff;
edsa_header[2] = 0x00;
edsa_header[3] = 0x00;
- edsa_header[4] = 0x60 | p->dp->ds->index;
- edsa_header[5] = p->dp->index << 3;
+ edsa_header[4] = 0x60 | dp->ds->index;
+ edsa_header[5] = dp->index << 3;
/*
* Move CFI field from byte 6 to byte 5.
@@ -68,8 +68,8 @@ static struct sk_buff *edsa_xmit(struct sk_buff *skb, struct net_device *dev)
edsa_header[1] = ETH_P_EDSA & 0xff;
edsa_header[2] = 0x00;
edsa_header[3] = 0x00;
- edsa_header[4] = 0x40 | p->dp->ds->index;
- edsa_header[5] = p->dp->index << 3;
+ edsa_header[4] = 0x40 | dp->ds->index;
+ edsa_header[5] = dp->index << 3;
edsa_header[6] = 0x00;
edsa_header[7] = 0x00;
}
diff --git a/net/dsa/tag_ksz.c b/net/dsa/tag_ksz.c
index b241c990cde0..66e443fd7675 100644
--- a/net/dsa/tag_ksz.c
+++ b/net/dsa/tag_ksz.c
@@ -34,7 +34,7 @@
static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct sk_buff *nskb;
int padlen;
u8 *tag;
@@ -72,7 +72,7 @@ static struct sk_buff *ksz_xmit(struct sk_buff *skb, struct net_device *dev)
tag = skb_put(nskb, KSZ_INGRESS_TAG_LEN);
tag[0] = 0;
- tag[1] = 1 << p->dp->index; /* destination port */
+ tag[1] = 1 << dp->index; /* destination port */
return nskb;
}
diff --git a/net/dsa/tag_lan9303.c b/net/dsa/tag_lan9303.c
index 4f211e56c81f..c709adb5efd9 100644
--- a/net/dsa/tag_lan9303.c
+++ b/net/dsa/tag_lan9303.c
@@ -42,7 +42,7 @@
static struct sk_buff *lan9303_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
u16 *lan9303_tag;
/* insert a special VLAN tag between the MAC addresses
@@ -62,7 +62,7 @@ static struct sk_buff *lan9303_xmit(struct sk_buff *skb, struct net_device *dev)
lan9303_tag = (u16 *)(skb->data + 2 * ETH_ALEN);
lan9303_tag[0] = htons(ETH_P_8021Q);
- lan9303_tag[1] = htons(p->dp->index | BIT(4));
+ lan9303_tag[1] = htons(dp->index | BIT(4));
return skb;
}
diff --git a/net/dsa/tag_mtk.c b/net/dsa/tag_mtk.c
index 968586c5d40e..5abebbfd7060 100644
--- a/net/dsa/tag_mtk.c
+++ b/net/dsa/tag_mtk.c
@@ -23,7 +23,7 @@
static struct sk_buff *mtk_tag_xmit(struct sk_buff *skb,
struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
u8 *mtk_tag;
if (skb_cow_head(skb, MTK_HDR_LEN) < 0)
@@ -36,7 +36,7 @@ static struct sk_buff *mtk_tag_xmit(struct sk_buff *skb,
/* Build the tag after the MAC Source Address */
mtk_tag = skb->data + 2 * ETH_ALEN;
mtk_tag[0] = 0;
- mtk_tag[1] = (1 << p->dp->index) & MTK_HDR_XMIT_DP_BIT_MASK;
+ mtk_tag[1] = (1 << dp->index) & MTK_HDR_XMIT_DP_BIT_MASK;
mtk_tag[2] = 0;
mtk_tag[3] = 0;
diff --git a/net/dsa/tag_qca.c b/net/dsa/tag_qca.c
index 8d33d9ebf910..b0274b6781cb 100644
--- a/net/dsa/tag_qca.c
+++ b/net/dsa/tag_qca.c
@@ -38,7 +38,7 @@
static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
u16 *phdr, hdr;
dev->stats.tx_packets++;
@@ -54,8 +54,7 @@ static struct sk_buff *qca_tag_xmit(struct sk_buff *skb, struct net_device *dev)
/* Set the version field, and set destination port information */
hdr = QCA_HDR_VERSION << QCA_HDR_XMIT_VERSION_S |
- QCA_HDR_XMIT_FROM_CPU |
- BIT(p->dp->index);
+ QCA_HDR_XMIT_FROM_CPU | BIT(dp->index);
*phdr = htons(hdr);
diff --git a/net/dsa/tag_trailer.c b/net/dsa/tag_trailer.c
index 61668be267f5..8b205bb5f521 100644
--- a/net/dsa/tag_trailer.c
+++ b/net/dsa/tag_trailer.c
@@ -16,7 +16,7 @@
static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev)
{
- struct dsa_slave_priv *p = netdev_priv(dev);
+ struct dsa_port *dp = dsa_slave_to_port(dev);
struct sk_buff *nskb;
int padlen;
u8 *trailer;
@@ -48,7 +48,7 @@ static struct sk_buff *trailer_xmit(struct sk_buff *skb, struct net_device *dev)
trailer = skb_put(nskb, 4);
trailer[0] = 0x80;
- trailer[1] = 1 << p->dp->index;
+ trailer[1] = 1 << dp->index;
trailer[2] = 0x10;
trailer[3] = 0x00;
--
2.14.2
^ permalink raw reply related
* [PATCH net-next 2/5] net: dsa: add slave notify helper
From: Vivien Didelot @ 2017-10-12 22:51 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, kernel, David S. Miller, Florian Fainelli,
Andrew Lunn, Vivien Didelot
In-Reply-To: <20171012225156.20758-1-vivien.didelot@savoirfairelinux.com>
Both DSA slave create and destroy functions call call_dsa_notifiers with
respectively DSA_PORT_REGISTER and DSA_PORT_UNREGISTER and the same
dsa_notifier_register_info structure.
Wrap this in a dsa_slave_notify helper so prevent cluttering these
functions.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
---
net/dsa/slave.c | 29 +++++++++++++++++------------
1 file changed, 17 insertions(+), 12 deletions(-)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index c6f4829645bf..f31737256f69 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1114,9 +1114,23 @@ int dsa_slave_resume(struct net_device *slave_dev)
return 0;
}
+static void dsa_slave_notify(struct net_device *dev, unsigned long val)
+{
+ struct dsa_slave_priv *p = netdev_priv(dev);
+ struct net_device *master = dsa_master_netdev(p);
+ struct dsa_port *dp = p->dp;
+ struct dsa_notifier_register_info rinfo = {
+ .switch_number = dp->ds->index,
+ .port_number = dp->index,
+ .master = master,
+ .info.dev = dev,
+ };
+
+ call_dsa_notifiers(val, dev, &rinfo.info);
+}
+
int dsa_slave_create(struct dsa_port *port, const char *name)
{
- struct dsa_notifier_register_info rinfo = { };
struct dsa_port *cpu_dp = port->cpu_dp;
struct net_device *master = cpu_dp->netdev;
struct dsa_switch *ds = port->ds;
@@ -1175,11 +1189,7 @@ int dsa_slave_create(struct dsa_port *port, const char *name)
goto out_free;
}
- rinfo.info.dev = slave_dev;
- rinfo.master = master;
- rinfo.port_number = p->dp->index;
- rinfo.switch_number = p->dp->ds->index;
- call_dsa_notifiers(DSA_PORT_REGISTER, slave_dev, &rinfo.info);
+ dsa_slave_notify(slave_dev, DSA_PORT_REGISTER);
ret = register_netdev(slave_dev);
if (ret) {
@@ -1204,7 +1214,6 @@ int dsa_slave_create(struct dsa_port *port, const char *name)
void dsa_slave_destroy(struct net_device *slave_dev)
{
struct dsa_slave_priv *p = netdev_priv(slave_dev);
- struct dsa_notifier_register_info rinfo = { };
struct device_node *port_dn;
port_dn = p->dp->dn;
@@ -1216,11 +1225,7 @@ void dsa_slave_destroy(struct net_device *slave_dev)
if (of_phy_is_fixed_link(port_dn))
of_phy_deregister_fixed_link(port_dn);
}
- rinfo.info.dev = slave_dev;
- rinfo.master = p->dp->cpu_dp->netdev;
- rinfo.port_number = p->dp->index;
- rinfo.switch_number = p->dp->ds->index;
- call_dsa_notifiers(DSA_PORT_UNREGISTER, slave_dev, &rinfo.info);
+ dsa_slave_notify(slave_dev, DSA_PORT_UNREGISTER);
unregister_netdev(slave_dev);
free_percpu(p->stats64);
free_netdev(slave_dev);
--
2.14.2
^ permalink raw reply related
* [PATCH net-next 1/5] net: dsa: use port's cpu_dp when creating a slave
From: Vivien Didelot @ 2017-10-12 22:51 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, kernel, David S. Miller, Florian Fainelli,
Andrew Lunn, Vivien Didelot
In-Reply-To: <20171012225156.20758-1-vivien.didelot@savoirfairelinux.com>
When dsa_slave_create is called, the related port already has a CPU port
assigned to it, available in its cpu_dp member. Use it instead of the
unique tree cpu_dp.
Signed-off-by: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
---
net/dsa/slave.c | 7 ++-----
1 file changed, 2 insertions(+), 5 deletions(-)
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 45f4ea845c07..c6f4829645bf 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1117,16 +1117,13 @@ int dsa_slave_resume(struct net_device *slave_dev)
int dsa_slave_create(struct dsa_port *port, const char *name)
{
struct dsa_notifier_register_info rinfo = { };
+ struct dsa_port *cpu_dp = port->cpu_dp;
+ struct net_device *master = cpu_dp->netdev;
struct dsa_switch *ds = port->ds;
- struct net_device *master;
struct net_device *slave_dev;
struct dsa_slave_priv *p;
- struct dsa_port *cpu_dp;
int ret;
- cpu_dp = ds->dst->cpu_dp;
- master = cpu_dp->netdev;
-
if (!ds->num_tx_queues)
ds->num_tx_queues = 1;
--
2.14.2
^ permalink raw reply related
* [PATCH net-next 0/5] net: dsa: master and slave helpers
From: Vivien Didelot @ 2017-10-12 22:51 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, kernel, David S. Miller, Florian Fainelli,
Andrew Lunn, Vivien Didelot
This patch series adds a few helpers to DSA core for clarity and
readability but brings no functional changes.
A dsa_slave_notify helper calls the DSA notifiers when (un)registering a
slave device.
Most of the DSA slave code only needs to access the dsa_port structure,
not the dsa_slave_priv (which only contains a few PHY-specific members).
Thus a dsa_slave_to_port helper returns a dsa_port structure of a slave
device.
A dsa_slave_get_master returns the master device of a slave device.
After that the netdev member of the dsa_port structure is split into two
explicit master and slave members to avoid confusion, even though it is
not planned to create a slave for DSA or CPU ports for the moment.
Vivien Didelot (5):
net: dsa: use port's cpu_dp when creating a slave
net: dsa: add slave notify helper
net: dsa: add slave to port helper
net: dsa: add slave get master helper
net: dsa: split dsa_port's netdev member
drivers/net/dsa/bcm_sf2.c | 6 +-
drivers/net/dsa/mt7530.c | 2 +-
drivers/net/dsa/mv88e6xxx/chip.c | 2 +-
include/net/dsa.h | 7 +-
net/dsa/dsa.c | 6 +-
net/dsa/dsa2.c | 22 ++--
net/dsa/dsa_priv.h | 22 ++--
net/dsa/legacy.c | 20 ++--
net/dsa/slave.c | 227 +++++++++++++++++++--------------------
net/dsa/tag_brcm.c | 9 +-
net/dsa/tag_dsa.c | 10 +-
net/dsa/tag_edsa.c | 10 +-
net/dsa/tag_ksz.c | 4 +-
net/dsa/tag_lan9303.c | 4 +-
net/dsa/tag_mtk.c | 4 +-
net/dsa/tag_qca.c | 5 +-
net/dsa/tag_trailer.c | 4 +-
17 files changed, 183 insertions(+), 181 deletions(-)
--
2.14.2
^ permalink raw reply
* [Patch net-next v2] tcp: add a tracepoint for tcp_retransmit_skb()
From: Cong Wang @ 2017-10-12 22:48 UTC (permalink / raw)
To: netdev
Cc: Cong Wang, Eric Dumazet, Alexei Starovoitov, Hannes Frederic Sowa,
Brendan Gregg, Neal Cardwell
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)
+ ),
+
+ TP_fast_assign(
+ struct ipv6_pinfo *np = inet6_sk(sk);
+ struct inet_sock *inet = inet_sk(sk);
+ struct in6_addr *pin6;
+ __be32 *p32;
+
+ __entry->skbaddr = skb;
+ __entry->skaddr = sk;
+
+ __entry->sport = ntohs(inet->inet_sport);
+ __entry->dport = ntohs(inet->inet_dport);
+
+ p32 = (__be32 *) __entry->saddr;
+ *p32 = inet->inet_saddr;
+
+ p32 = (__be32 *) __entry->daddr;
+ *p32 = inet->inet_daddr;
+
+ if (np) {
+ pin6 = (struct in6_addr *)__entry->saddr_v6;
+ *pin6 = np->saddr;
+ pin6 = (struct in6_addr *)__entry->daddr_v6;
+ *pin6 = *(np->daddr_cache);
+ } else {
+ pin6 = (struct in6_addr *)__entry->saddr_v6;
+ ipv6_addr_set_v4mapped(inet->inet_saddr, pin6);
+ pin6 = (struct in6_addr *)__entry->daddr_v6;
+ ipv6_addr_set_v4mapped(inet->inet_daddr, pin6);
+ }
+ ),
+
+ TP_printk("sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 saddrv6=%pI6 daddrv6=%pI6",
+ __entry->sport, __entry->dport, __entry->saddr, __entry->daddr,
+ __entry->saddr_v6, __entry->daddr_v6)
+);
+
+#endif /* _TRACE_TCP_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/net/core/net-traces.c b/net/core/net-traces.c
index 1132820c8e62..f4e4fa2db505 100644
--- a/net/core/net-traces.c
+++ b/net/core/net-traces.c
@@ -31,6 +31,7 @@
#include <trace/events/napi.h>
#include <trace/events/sock.h>
#include <trace/events/udp.h>
+#include <trace/events/tcp.h>
#include <trace/events/fib.h>
#include <trace/events/qdisc.h>
#if IS_ENABLED(CONFIG_IPV6)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 696b0a168f16..e1e7410a5b60 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -42,6 +42,8 @@
#include <linux/gfp.h>
#include <linux/module.h>
+#include <trace/events/tcp.h>
+
/* People can turn this off for buggy TCP's found in printers etc. */
int sysctl_tcp_retrans_collapse __read_mostly = 1;
@@ -2875,6 +2877,7 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs)
if (likely(!err)) {
TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS;
+ trace_tcp_retransmit_skb(sk, skb, segs);
} else if (err != -EBUSY) {
NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRETRANSFAIL);
}
--
2.13.0
^ permalink raw reply related
* Re: [PATCH] Add -target to clang switches while cross compiling.
From: Abhijit Ayarekar @ 2017-10-12 22:43 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: Abhijit Ayarekar, ast, daniel, netdev, linux-kernel, yhs
In-Reply-To: <20171012222302.3zlndmcm7ixywbs6@ast-mbp>
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.
> > 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
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