Sign Up
Log In
Log In
or
Sign Up
Places
All Projects
Status Monitor
Collapse sidebar
openSUSE:Evergreen:11.2:Test
dcbd
dcbd-0.9.2_to_0.9.2.2.patch
Overview
Repositories
Revisions
Requests
Users
Attributes
Meta
File dcbd-0.9.2_to_0.9.2.2.patch of Package dcbd
diff -pur dcbd-0.9.2/dcbd7 dcbd-0.9.2.2/dcbd7 --- dcbd-0.9.2/dcbd7 2008-12-02 17:36:19.000000000 -0800 +++ dcbd-0.9.2.2/dcbd7 2008-12-22 17:26:46.000000000 -0800 @@ -4,55 +4,80 @@ .\" .\" * Other names and brands may be claimed as the property of others. .\" -.TH dcbd 1 "September 10, 2008" +.TH dcbd 1 "December 12, 2008" .SH NAME -dcbd \-DCBX daemon +dcbd \- Data Center Bridging capabilities exchange daemon .SH SYNOPSIS .B dcbd [-h] .B [-v] -.I [-d] [-f filename] +.B [-d] +.BI "[-f" " filename" "]" .SH DESCRIPTION -Executes the DCB capabilities exchange protocol to exchange DCB -configuration with the peer device using LLDP -.BR -retrieves and stores DCB configuration to a configuration file -.BR -controls the DCB settings of the network driver based on the +Executes the Data Center Bridging (DCB) capabilities exchange protocol on DCB +capable Ethernet interfaces in order to exchange DCB configuration with the +peer device. +.PP +Major functions of +.B dcbd +include: +.br +- retrieving and storing DCB configuration to a configuration file +.br +- configuring the DCB settings of the network driver based on the operation of the DCB capabilities exchange protocol -.BR -detects and reacts to link up and link down events on network ports -.BR -supports the Priority Group, Priority Flow Control and -App TLV (FCoE, subtype 0) features. -.BR -generates client interface events when the operational configuration -or state of a feature changes. +.br +- detecting and reacting to link up and link down events on network ports +.br +- supporting the following DCB features: Priority Groups, Priority Flow Control, +FCoE and FCoE Logical Link Status. +.br +- providing an interface for client applications to query and configure DCB +features. Events are also generated on the client interface when the +operational state or configuration of a feature changes. + +.B dcbd +supports the versions of the DCB capabilities exchange protocol described here: +.TP +version 1 +<http://download.intel.com/technology/eedc/dcb_cep_spec.pdf> +.PP +and here: +.TP +version 2 +<http://www.ieee802.org/1/files/public/docs2008/az-wadekar-dcbx-capability-exchange-discovery-protocol-1108-v1.01.pdf> +.PP +See +.B dcbtool +for information on how to configure which version of the protocol +.B dcbd +executes. + .SH OPTIONS -dcbd has the following command line options: -.IP -h -show usage information -.IP -f -configfile -use the specified file as the config file instead of the default file -- /etc/sysconfig/dcbd/dcbd.conf -.IP -d -run dcbd as a daemon -.IP -v -show dcbd version -.IP -h +.B dcbd +has the following command line options: +.TP +.B \-h show usage information -.B configfile -use the specified file as the config file instead of the default file -- /etc/sysconfig/dcbd/dcbd.conf -.IP -d +.TP +.BI "-f" " filename" +use the specified file as the configuration file instead of the default file: +/etc/sysconfig/dcbd/dcbd.conf +.B dcbd +expects the directory of the configuration file to exist, but if the +configuration file does not exist, then a default configuration file will +be created. +.TP +.B \-d run dcbd as a daemon -.IP -v +.TP +.B \-v show dcbd version .SH LICENSE -dcbd and dcbtool - DCB daemon and command line utility DCB configuration +dcbd - DCB daemon +.br Copyright(c) 2007-2008 Intel Corporation. .BR -Portions of dcbd and dcbtool (basically program framework) are based on: +Portions of dcbd are based on: .IP hostapd-0.5.7 .IP Copyright (c) 2004-2008, Jouni Malinen <j@w1.fi> @@ -74,5 +99,5 @@ The full GNU General Public License is i the file called "COPYING". .SH SUPPORT Contact Information: -e1000-dcb Mailing List <e1000-dcb@lists.sourceforge.net> +e1000-eedc Mailing List <e1000-eedc@lists.sourceforge.net> Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 diff -pur dcbd-0.9.2/dcbd.spec dcbd-0.9.2.2/dcbd.spec --- dcbd-0.9.2/dcbd.spec 2008-12-02 17:36:21.000000000 -0800 +++ dcbd-0.9.2.2/dcbd.spec 2008-12-22 17:26:48.000000000 -0800 @@ -1,6 +1,6 @@ Name: dcbd Summary: Intel(R) EEDC Connection -Version: 0.9.2 +Version: 0.9.2.2 Release: 1 Source: %{name}-%{version}.tar.gz Vendor: Intel Corporation diff -pur dcbd-0.9.2/dcbtool7 dcbd-0.9.2.2/dcbtool7 --- dcbd-0.9.2/dcbtool7 2008-12-02 17:36:21.000000000 -0800 +++ dcbd-0.9.2.2/dcbtool7 2008-12-22 17:26:47.000000000 -0800 @@ -4,130 +4,337 @@ .\" .\" * Other names and brands may be claimed as the property of others. .\" -.TH dcbtool 1 "September 10, 2008" +.TH dcbtool 1 "December 11, 2008" .SH NAME -dcbtool \-display or change the DCB settings of a DCB interface +dcbtool \- manage the Data Center Bridging (DCB) settings of a DCB interface .SH SYNOPSIS .B dcbtool -h +.br .B dcbtool -v -.B dcbtool [command] [command arguments] -.B dcbtool interactive mode +.br +.B dcbtool [-rR] +.br +.BI "dcbtool [-rR] [" "command" "] [" "command arguments" "]" + .SH DESCRIPTION -dcbtool is used to query and set the DCB and DCBX settings of a -DCB capable Ethernet device. It connects to the client interface -of 'dcbd' to perform these operations. +.B dcbtool +is used to query and set the DCB settings of a +DCB capable Ethernet interface. It connects to the client interface +of +.B dcbd +to perform these operations. +.B dcbtool +will operate in interactive mode if it is executed without a +.I command. +In interactive mode, +.B dcbtool +also functions as an event listener and will print out events +received from +.B dcbd +as they arrive. .SH OPTIONS -The basic usage of dcbtool is: -.B dcbtool [options] <cmd> [cmd arguments] -If no command is specified, then dcbtool will enter an interactive mode which -allows commands to be entered interactively. When invoked in interactive -mode, dcbtool will also register with dcbd as an event monitor. -.IP -h -shows the dcbtool usage message -.IP -v -shows dcbtool version information +.TP +.B \-h +show the +.B dcbtool +usage message +.TP +.B \-v +shows +.B dcbtool +version information +.TP +.B \-r +display the raw +.B dcbd +client interface messages as well as the readable output. +.TP +.B \-R +display only the raw +.B dcbd +client interface messages .SH COMMANDS -.IP help +.TP +.B help shows the dcbtool usage message -.IP level <value> -sets the threshold dcbd uses to send event messages to this interface. -.B Valid Range: -0: MSG_DUMP - for dcbd debugging -1: MSG_DEBUG - for dcbd debugging -2: MSG_INFO - for free form dcbd informational messages -3: MSG_WARNING - for free form dcbd warning messages -4: MSG_DCB - for formated DCB event messages -.LP -.IP ping -test command. dcbd responds with "PONG" -.IP license -displays dcbtool license information -.IP quit +.TP +.B ping +test command. The +.B dcbd +daemon responds with "PONG" if the client interface is operational. +.TP +.B license +displays +.B dcbtool +license information +.TP +.B quit exit from interactive mode -.IP gc <ifname> <dcb | pg | pfc | app:<subtype>> -Get configuration command. Returns the local configuration for -the specified interface and DCB feature. Features are: -dcb - DCB state of port -pg - priority groups -pfc - priority flow control -app:<subtype> - application with subtypes: - fcoe -.IP go <ifname> <pg | pfc | app:<subtype>> -Get operational status command. Returns the current DCB -operational status and configuration for the specified -interface and DCB feature. -.IP gp <ifname> <pg | pfc | app:<subtype>> -Get peer status command. Returns the current DCB status -and configuration for the DCB peer of the specified interface -and DCB feature. -.IP sc <ifname> dcb <on | off> -Set the DCB state of the port. -.IP sc <ifname> <pg | pfc | app:<subtype>> [a:<0|1>] [e:<0|1>] [w:<0|1>] -[feature attributes] -Set configuration command. Allows setting, per DCB feature, the following: -.IP a:<0|1> - enable/disable advertising the feature via DCB -.IP e:<0|1> - enable/disable the DCB feature -.IP w:<0|1> - enable/disable willing flag for the feature - -If one of the above parameters is not specified in the set config command, -then the DCB feature attributes must be specified. - -.IP pg - priority group attributes -Priority group attributes which are not supplied will remain -unchanged. Except for the 'uppct'. If 'uppct' is not -supplied and the resulting percentages within each PGID do -not add up to 100%, then dcbd will compute new user priority -percentages for each PGID so that the sum for each PGID is 100%. -.IP up2tc:<0-7><0-7><0-7><0-7><0-7><0-7><0-7><0-7> -This is the user priority to traffic class mapping parameter. -If not specified, the default value is "01234567". -.IP pgid:<0-7F><0-7F><0-7F><0-7F><0-7F><0-7F><0-7F><0-7F> -This is the priority group ID mapping. From left to right, -each number indicates the priority group ID to which the -corresponding user priority, starting with zero, belongs. -A value of 'F' (0xF) indicates that the user priority -belongs to the link strict group. -.IP pgpct:<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>, -<0-100> -This is the priority group bandwidth percentage of link -setting. From left to right, each number indicates the -percentage of total link bandwidth which is allocated to the -corresponding priority group ID, starting with zero. -.IP uppct:<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>, -<0-100> -This is the user priority bandwidth percentage of bandwidth -group setting. From left to right, each number indicates the -percentage of total priority group bandwidth which is -allocated to the corresponding user priority, starting with -zero. -.IP strict:<0-2><0-2><0-2><0-2><0-2><0-2><0-2><0-2> -This is the strict priority settings for each user priority. -.BR -From left to right, each number indicates the strict priority -setting for each user priority, starting with zero. A value -of 0 represents no strict priority setting, 1 represents group -strict. -.B pfc - priority flow control attributes -.IP pfcup:<0|1><0|1><0|1><0|1><0|1><0|1><0|1><0|1> -This is the priority flow control enable setting per user -priority. From left to right, each number indicates the enable -setting of priority flow control for the corresponding user -priority, starting from zero. 0 means disabled, 1 means -enabled. -.IP app - application attributes -.IP appcfg:<xx> -This represents the application settings for the specified -application subtype. Assuming that the application attribute -setting is an arbitrary sequence of bytes of a given length, -the setting is supplied to dcbtool by representing each byte as -a 2 character hexadecimal string. Thus, if the application -attribute setting is the value 0x5a, the dcbtool attribute -would be "appcfg:5a" + +.PP +The following commands interact with the +.B dcbd +daemon to manage the daemon and DCB features on DCB capable interfaces. + +.B dcbd +general configuration commands: +.TP +.RB "<" "gc" "|" "go" ">" " dcbx" +get the configured or operational version of the DCB capabilities exchange +protocol. If different, the configured version will take effect (and become +the operational version) after +.B dcbd +is restarted. +.TP +.BR "sc dcbx v:" "[" "1" "|" "2" "]" +set the version of the DCB capabilities exchange protocol which will be used +the next time +.B dcbd +is started. Information about version 1 can be found at: +.br +<http://download.intel.com/technology/eedc/dcb_cep_spec.pdf> +Information about version 2 can be found at: +.br +<http://www.ieee802.org/1/files/public/docs2008/az-wadekar-dcbx-capability-exchange-discovery-protocol-1108-v1.01.pdf> +.PP +DCB per-interface commands: +.TP +\fBgc \fR<\fIifname\fR> <\fIfeature\fR> +get configuration of \fIfeature\fR on interface \fIifname\fR. +.TP +\fBgo \fR<\fIifname\fR> <\fIfeature\fR> +get operational status of \fIfeature\fR on interface \fIifname\fR. +.TP +\fBgp \fR<\fIifname\fR> <\fIfeature\fR> +get peer configuration of \fIfeature\fR on interface \fIifname\fR. +.TP +\fBsc \fR<\fIifname\fR> <\fIfeature\fR> <\fIargs\fR> +set the configuration of \fIfeature\fR on interface \fIifname\fR. +.PP +.I feature +may be one of the following: +.TP +.B dcb +DCB state of the port +.TP +.B pg +priority groups +.TP +.B pfc +priority flow control +.TP +\fBapp:\fR<\fIsubtype\fR> +application specific data +.TP +\fBll:\fR<\fIsubtype\fR> +logical link status +.TP +.B bcn +backward congestion notification +.PP +.I subtype +can be: +.TP +.BR "0" "|" "fcoe" +Fiber Channel over Ethernet (FCoE) +.PP +.I args +can include: +.TP +.BR "e:" "<" "0" "|" "1" ">" +controls feature enable +.TP +.BR "a:" "<" "0" "|" "1" ">" +controls whether the feature is advertised via DCBX to the peer +.TP +.BR "w:" "<" "0" "|" "1" ">" +controls whether the feature is willing to change its operational +configuration based on what is received from the peer +.TP +.RI "[" "feature specific args" "]" +arguments specific to a DCB feature + +.PP +Feature specific arguments for +.BR "dcb" ":" +.TP +.BR "on" "|" "off" +enable or disable +.Bdcb +for the interface. The +.B go +and +.B gp +commands are not needed for the +.B dcb +feature. Also, the +.RI "enable(" "e" "), advertise(" "a" ") and willing(" "w" ")" +arguments are not required. + +.PP +Feature specific arguments for +.BR "pg" ":" +.TP +.BI "pgid:" "xxxxxxxx" +Priority group ID for the 8 priorities. From left to right +(priorities 0-7), +.I x +is the corresponding priority group ID value, which +can be +.I 0-7 +for priority groups with bandwidth allocations or +.I f +(priority group ID 15) for the unrestricted priority group. + +.TP +.BI "pgpct:" "x" "," "x" "," "x" "," "x" "," "x" "," "x" "," "x" "," "x" +Priority group percentage of link bandwidth. From left to right +(priority groups 0-7), +.I x +is the percentage of link bandwidth allocated +to the corresponding priority group. The total bandwidth must equal 100%. + +.TP +.BI "uppct:" "x" "," "x" "," "x" "," "x" "," "x" "," "x" "," "x" "," "x" +Priority percentage of priority group bandwidth. From left to right +(priorities 0-7), +.I x +is the percentage of priority group bandwidth +allocated to the corresponding priority. The sum of percentages for +priorities which belong to the same priority group must total 100% (except +for priority group 15). + +.TP +.BI "strict:" "xxxxxxxx" +Strict priority setting. From left to right (priorities 0-7), +.I x +.RB "is " "0" " or " "1" ". " "1" +indicates that the priority may utilize all of the bandwidth allocated to +its priority group. + +.TP +.BI "up2tc:" "xxxxxxxx" +Priority to traffic class mapping. From left to right (priorities 0-7), +.I x +is the traffic class (0-7) to which the priority is mapped. + +.PP +Feature specific arguments for +.BR "pfc" ":" +.TP +.BI "pfcup:" "xxxxxxxx" +Enable/disable priority flow control. From left to right +(priorities 0-7), +.I x +.RB "is " "0" " or " "1" ". " "1" +indicates that the corresponding priority is configured to transmit +priority pause. + +.PP +Feature specific arguments for \fBapp:\fR<\fIsubtype\fR>: +.TP +.BI "appcfg:" "xx" +.I xx +is a hexadecimal value representing an 8 bit bitmap where +1 bits indicate the priorities which frames for the applications specified by +.I subtype +should use. The lowest order bit maps to priority 0. + +.PP +Feature specific arguments for \fBll:\fR<\fIsubtype\fR>: +.TP +\fBstatus:\fR[\fB0\fR|\fB1\fR] +For testing purposes, the logical link status may be set to 0 or 1. This +setting is not persisted in the configuration file. + +.PP +Feature specific arguments for +.BR "bcn" ":" + +.TP +.BI "rpadmin:" "xxxxxxxx" +Priorities for which a BCN reaction point should be enabled. From left to +right (priorities 0-7), +.I x +is 0 or 1. +A 1 indicates that the corresponding priorities is enabled. + +.TP +.BR "alpha:" "<" "0.0-1.0" ">" +alpha - floating point value between 0 and 1 + +.TP +.BR "beta:" "<" "0.0-1.0" ">" +beta -floating point value between 0 and 1 + +.TP +.BR "gd:" "<" "0.0-1.0" ">" +decrement coefficient - floating point value between 0 and 1 + +.TP +.BR "gi:" "<" "0.0-1.0" ">" +increment coefficient - floating point value between 0 and 1 + +.TP +.BR "tmax:" "<" "0-1000000" ">" +max time to backoff after BCN0 + +.TP +.BR "td:" "<" "1-10000" ">" +drift interval + +.TP +.BR "rmin:" "<" "1-1000" ">" +default rate after first BCN0 + +.TP +.BR "dw:" "<" "1-64" ">" +derived weight (1 = 0.125) + +.TP +.BR "rd:" "<" "1-100" ">" +drift factor + +.TP +.BR "ru:" "<" "1-100" ">" +rate unit + +.TP +.BR "wrtt:" "<" "0-10" ">" +RTT moving average weight + +.TP +.BR "ri:" "<" "0-10000" ">" +initial rate + +.SH EXAMPLES + +.PP +Enable DCB on interface \fIeth2\fR +.PP +.B dcbtool sc eth2 dcb on + +.PP +Assign priorites 0-3 to priority group 0, priorites 4-6 to priority group 1 and +priority 7 to the unrestricted priority. Also, allocate 25% of link +bandwidth to priority group 0 and 75% to group 1. +.PP +.B dcbtool sc eth2 pg pgid:0000111f pgpct:25,75,0,0,0,0,0,0 + +.PP +Enable transmit of Priority Flow Control for priority 3 and assign FCoE to +priority 3. +.PP +.B dcbtool sc eth2 pfc pfcup:00010000 +.br +.B dcbtool sc eth2 app:0 appcfg:08 + .SH LICENSE -dcbd and dcbtool - DCB daemon and command line utility DCB configuration +dcbtool - DCB configuration utility +.br Copyright(c) 2008 Intel Corporation. .BR -Portions of dcbd and dcbtool (basically program framework) are based on: +Portions of dcbtool are based on: .IP hostapd-0.5.7 .IP Copyright (c) 2004-2008, Jouni Malinen <j@w1.fi> @@ -149,6 +356,6 @@ The full GNU General Public License is i the file called "COPYING". .SH SUPPORT Contact Information: -e1000-dcb Mailing List <e1000-dcb@lists.sourceforge.net> +e1000-eedc Mailing List <e1000-eedc@lists.sourceforge.net> Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497 diff -pur dcbd-0.9.2/dcbtool.c dcbd-0.9.2.2/dcbtool.c --- dcbd-0.9.2/dcbtool.c 2008-12-02 17:36:20.000000000 -0800 +++ dcbd-0.9.2.2/dcbtool.c 2008-12-22 17:26:46.000000000 -0800 @@ -118,7 +118,7 @@ static const char *commands_help = " 'feature' can be:\n" " dcb DCB state of port\n" " pg priority groups\n" -" pfc priority groups\n" +" pfc priority flow control\n" " app:<subtype> application specific data\n" " bcn backwards congestion notification\n" " ll:<subtype> logical link status\n\n" @@ -143,7 +143,7 @@ static const char *commands_help = " [pgpct:x,x,x,x,x,x,x,x] priority group percent of link.\n" " From left to right (priority groups 0-7),\n" " x is the percentage of link bandwidth\n" -" assigned the total must equal 100%.\n" +" assigned. The total must equal 100%.\n" " [uppct:x,x,x,x,x,x,x,x] user priority percent of bandwidth group.\n" " From left to right (priorities 0-7),\n" " x is the percentage of priority group\n" @@ -211,13 +211,13 @@ static void usage(void) " dcbtool -h\n" " dcbtool -v\n" " dcbtool [-rR] [ command ]\n" - " dcbtool interactive mode\n" + " dcbtool interactive mode\n" "\n" "Options:\n" - " -h help (show this usage text)\n" - " -v shown version information\n" - " -r show raw messages\n" - " -R show only raw messages\n" + " -h help (show this usage text)\n" + " -v shown version information\n" + " -r show raw messages\n" + " -R show only raw messages\n" "%s", commands_help); } diff -pur dcbd-0.9.2/include/version.h dcbd-0.9.2.2/include/version.h --- dcbd-0.9.2/include/version.h 2008-12-02 17:36:21.000000000 -0800 +++ dcbd-0.9.2.2/include/version.h 2008-12-22 17:26:48.000000000 -0800 @@ -28,7 +28,7 @@ #ifndef VERSION_H #define VERSION_H -#define VERSION_STR "0.9.2" +#define VERSION_STR "0.9.2.2" #define DCBTOOL_VERSION VERSION_STR #endif /* VERSION_H */ diff -pur dcbd-0.9.2/lldp/rx.c dcbd-0.9.2.2/lldp/rx.c --- dcbd-0.9.2/lldp/rx.c 2008-12-02 17:36:21.000000000 -0800 +++ dcbd-0.9.2.2/lldp/rx.c 2008-12-22 17:26:48.000000000 -0800 @@ -1106,6 +1106,10 @@ boolean_t unpack_dcbx2_tlvs(struct port current += dcbtlv->length + sizeof(tl); offset += dcbtlv->length; switch(dcbtlv->type) { + case DCB_CONTROL_TLV: + if (port->rx.manifest->dcbx1 == NULL) + port->rx.manifest->dcbx_ctrl = dcbtlv; + break; case DCB_PRIORITY_GROUPS_TLV2: port->rx.manifest->dcbx_pg = dcbtlv; break; diff -pur dcbd-0.9.2/README dcbd-0.9.2.2/README --- dcbd-0.9.2/README 2008-12-02 17:36:20.000000000 -0800 +++ dcbd-0.9.2.2/README 2008-12-22 17:26:47.000000000 -0800 @@ -1,7 +1,7 @@ Data Center Bridging (DCB) for Intel(R) Network Connections =========================================================== -September 10, 2008 +December 19, 2008 Contents ======== @@ -27,28 +27,24 @@ was to provide QoS in software, as hardw interfaces to support it. In 2.6.23, Intel pushed the notion of multiqueue support into the qdisc layer. This provides a mechanism to map the software queues in the qdisc structure into multiple hardware queues in underlying -devices. In the case of Intel adapters, we leverage this combination to map -qdisc queues into the queues within our hardware controllers. +devices. In the case of Intel adapters, this mechanism is leveraged to map +qdisc queues onto the queues within the hardware controllers. Within the Data Center, the perception is that traditional Ethernet: a) has high latency -b) is prone to losing frames rendering it unacceptable for storage applications +b) is prone to losing frames, rendering it unacceptable for storage applications In an effort to address these issues, Intel and a host of industry leaders have been working on addressing these problems. Specifically, within the IEEE 802.1 standards body there are a number of task forces working on enhancements to -address these concerns. Listed below are the applicable standards bodies: +address these concerns. Listed below are the applicable standards: - Enhanced Transmission Selection - Priority Groups: IEEE 802.1Qaz - Lossless Traffic Class - Priority Flow Control: IEEE 802.1Qbb - Congestion Notification: IEEE 802.1Qau - Improved bi-sectional bandwidth - Shortest Path Bridging: IEEE 802.1Qaq / IETF TRILL - Dynamic Configuration - DCB Capability exchange protocol: In consideration in IEEE 802. - 1Qtbd + Enhanced Transmission Selection + Priority Groups: IEEE 802.1Qaz + Lossless Traffic Class + Priority Flow Control: IEEE 802.1Qbb + Congestion Notification: IEEE 802.1Qau + DCB Capability exchange protocol: IEEE 802.1Qaz The software solution that is being released represents Intel's implementation of these efforts. It is worth noting that many of these standards have not been @@ -59,19 +55,22 @@ solutions in development. As these solut ratified, we will work with ecosystem partners and the standards body to ensure that the Intel solution works as expected. + Requirements ============ -- Linux kernel version 2.6.27 or later. -- Linux ixgbe driver version 1.3.31.2 or newer. -- 2.6.23 or newer version of the "iproute2" package should be downloaded and - installed from http://developer.osdl.org/dev/iproute2/download in - order to obtain a multi-queue aware version of the 'tc' utility. -- Version 2.5.33 of Flex should be installed (to support 2.6.23 version of - iproute2). SLES10 is known to have an older version of Flex. The latest +- Linux kernel version 2.6.29 or later. +- Linux ixgbe driver (for Intel 82598-based adapters) from kernel 2.6.29 or + later. +- 2.6.29 or newer version of the "iproute2" package should be downloaded and + installed in order to obtain a multi-queue aware version of the 'tc' utility. + Check for new versions at: + http://www.linuxfoundation.org/en/Net:Iproute2 +- Version 2.5.33 of Flex should be installed (to support iproute2). + SLES10 is known to have an older version of Flex. The latest Flex source can be obtained from http://flex.sourceforge.net/ - An up to date netlink library needs to be installed in order to compile dcbd. -- 82598-based Intel adapter. +- An 82598-based Intel adapter. Functionality ============= @@ -79,22 +78,33 @@ Functionality dcbd - Executes the DCB capabilities exchange protocol to exchange DCB configuration with the peer device using LLDP. - - Supports DCBX versions 1 and 2 + - Supports the versions of the DCB capabilities exchange protocol + described here: + - version 1 + <http://download.intel.com/technology/eedc/dcb_cep_spec.pdf> + and here: + - version 2 + <http://www.ieee802.org/1/files/public/docs2008/az-wadekar-dcbx- + capability-exchange-discovery-protocol-1108-v1.01.pdf> - Retrieves and stores DCB configuration to a configuration file. - Controls the DCB settings of the network driver based on the - operation of the DCB capabilities exchange protocol. - - Detects and reacts to link up and link down events on network ports - - Supports the Priority Group, Priority Flow Control and App TLV (FCoE, - subtype 0) features. - - Generates client interface events when the operational configuration - or state of a feature changes. + operation of the DCB capabilities exchange protocol. Interaction with + a supporting network driver is achieved via DCB operations added to the + rtnetlink interface in kernel 2.6.29. + - Detects and reacts to link up and link down events on network ports. + - Supports the following DCB features: Priority Group, + Priority Flow Control, FCoE, and FCoE Logical Link Status. + - Provides an interface for client applications to query and configure + DCB features. Generates client interface events when the operational + configuration or state of a feature changes. dcbtool - Interacts with dcbd via the client interface. - Queries the state of the local, operational and peer configuration for - the Priority Group, Priority Flow Control and App TLV features. + the supported DCB features. + - Supports configuring the supported DCB features. - Interactive mode allows multiple commands to be entered interactively, - as well as displaying asynchronous events. + as well as displaying event messages. - Enables or disables DCB for an interface. How To Build a DCB-Capable System @@ -103,59 +113,56 @@ How To Build a DCB-Capable System Linux kernel install: --------------------- -1. Requires 2.6.27 kernel. +1. Requires 2.6.29 kernel. 2. Untar and make the kernel. Listed below are the required kernel options: Required configuration options: From make menu config -a.In the Device Drivers, Network device support menu - Select Netdevice multiple hardware queue support. -b. In the Networking, Network Options, QoS menu - Select Hardware Multiqueue-aware Multi Band Queuing (MULTIQ) - Select Multi Band Priority Queueing (PRIO) - Select Elementary classification ( BASIC ) - Select Universal 32bit comparisons w/ hashing (U32) - Select Extended Matches and make sure U32 key is selected - Select Actions -> SKB Editing - +a. In Networking support -> Networking options, enable Data Center Bridging +b. In Networking support -> Networking options -> QoS and/or fair queuing, + enable: + Hardware Multiqueue-aware Multi Band Queuing (MULTIQ) + Multi Band Priority Queueing (PRIO) + Elementary classification ( BASIC ) + Universal 32bit comparisons w/ hashing (U32) + Extended Matches and make sure U32 key is selected + Actions -> SKB Editing +c. To enable ixgbe driver support for DCB, + In Device Drivers -> Network device support -> Ethernet (10000 Mbit) + enable Intel(R) 10GbE PCI Express adapters support and the + Data Center Bridging (DCB) Support option + 3. Build the kernel. + 4. Create a link from /usr/include/linux to /usr/src/kernels/linux-2.x.xx.x/include/linux. Use the following command: ln -s /usr/src/kernels/linux-2.x.xx.x/include/linux /usr/include/linux. -ixgbe Base Driver Install -------------------------- - -1. Download the ixgbe driver from the e1000 project in Sourceforge. - Please ensure the driver version is TBD or later. -2. Build & install as outlined in the ixgbe readme. - (tar zxvf ixgbe-xxx; make; make install). dcbd Application Install ------------------------ -1. Download iproute2 from the web. Listed below is a link for iproute: - - http://devresources.linux-foundation.org/dev/iproute2/download/. - +1. Download iproute2 from the web. + See: http://www.linuxfoundation.org/en/Net:Iproute2 Please ensure that you use the version that corresponds to the kernel - verison that you are using. Follow the build/installation instructions - in the README with the tarball. Typically, the commands - ./configure;make;make install should work. + version that you are using. Follow the build/installation instructions + in the README. Typically, the commands: + ./configure; make; make install + will work. -2. Download the latest version dcbd-x.y.z tarball from the e1000 project in - sourceforge and untar it. Go into the dcbd-x.y.z directory and run +2. Download the latest version of the dcbd-x.y.z tarball from the e1000 project + in Sourceforge and untar it. Go into the dcbd-x.y.z directory and run the following commands make clean; make; make install. This will build and copy 'dcbd' and 'dcbtool' to /usr/sbin, make the - '/etc/sysconfig/dcbd'directory (default location of the dcbd.conf file) + '/etc/sysconfig/dcbd' directory (default location of the dcbd.conf file) and setup dcbd to run as a system service using the chkconfig program. - Verify that the dcbd service is working as expected with the 'service dcbd - status' command. If the service is not on, issue the command 'service dcbd - start' + Verify that the dcbd service is working as expected with the + 'service dcbd status' command. If the service is not on, issue the command + 'service dcbd start' dcbd will create the dcbd.conf file if it does not exist. @@ -182,7 +189,8 @@ SETUP: dcbd start" to start. Or, run "dcbd -d" from the command line to start. -3. Enable DCB on the selected ixgbe port: dcbtool sc ethX dcb on +3. Enable DCB on the selected ixgbe port: + dcbtool sc ethX dcb on 4. The dcbtool command can be used to query and change the DCB configuration (ie., various percentages to different queues). Use dcbtool -h to see a @@ -191,243 +199,286 @@ SETUP: Operation ========= -The 'tc' command is used to setup the qdisc and filters to cause network -traffic to be transmitted on different queues. +dcbd and dcbtool can be used to configure a DCB capable driver, such as +the ixgbe driver, which supports the rtnetlink DCB interface. Once the +DCB features are configured, the next step is to classify traffic to be +identified with an 802.1p priority and the associated DCB features. This can +be done by using the 'tc' command to setup the qdisc and filters to cause +network traffic to be transmitted on different queues. + +The skbedit action mechanism can be used in a tc filter to classify traffic +patterns to a specific queue_mapping value from 0-7. The ixgbe driver will +place traffic with a given queue_mapping value onto the corresponding hardware +queue and tag the outgoing frames with the corresponding 802.1p priority value. -To configure the MULTIQ qdisc to have 8 bands: +Set up the multi-queue qdisc for the selected interface: # tc qdisc add dev ethX root handle 1: multiq -Using the Flow id as the target in a TC filter allows you to classify a packet -into a band. Here are some examples of how to filter traffic into various -bands using the flow ids: +Setting the queue_mapping in a TC filter allows the ixgbe driver to classify a +packet into a queue. Here are some examples of how to filter traffic into +various queues using the flow ids: # tc filter add dev ethX protocol ip parent 1: u32 match ip dport 80 \ -0xffff action skbedit queue_mapping 1 +0xffff action skbedit queue_mapping 0 # tc filter add dev ethX protocol ip parent 1: u32 match ip dport 53 \ -0xffff action skbedit queue_mapping 2 +0xffff action skbedit queue_mapping 1 # tc filter add dev ethX protocol ip parent 1: u32 match ip dport 5001 \ -0xffff action skbedit queue_mapping 3 - -# tc filter add dev ethX protocol ip parent 1: u32 match ip dport 22 \ -0xffff action skbedit queue_mapping 4 - -# tc filter add dev ethX protocol ip parent 1: u32 match ip dport 137 \ -0xffff action skbedit queue_mapping 5 - -# tc filter add dev ethX protocol ip parent 1: u32 match ip dport 138 \ -0xffff action skbedit queue_mapping 5 - -# tc filter add dev ethX protocol ip parent 1: u32 match ip dport 139 \ -0xffff action skbedit queue_mapping 5 - -# tc filter add dev ethX protocol ip parent 1: u32 match ip dport 25 \ -0xffff action skbedit queue_mapping 6 - -# tc filter add dev ethX protocol ip parent 1: u32 match ip dport 21 \ -0xffff action skbedit queue_mapping 7 +0xffff action skbedit queue_mapping 2 # tc filter add dev ethX protocol ip parent 1: u32 match ip dport 20 \ -0xffff action skbedit queue_mapping 8 +0xffff action skbedit queue_mapping 7 Here is an example that sets up a filter based on EtherType. In this example, the EtherType is 0x8906. # tc filter add dev ethX protocol 802_3 parent 1: handle 0xfc0e basic match \ -'cmp(u16 at 12 layer 1 mask 0xffff eq 35078)' action skbedit queue_mapping 4 +'cmp(u16 at 12 layer 1 mask 0xffff eq 35078)' action skbedit queue_mapping 3 + Testing ======= To test in a back-to-back setup, use the following tc commands to setup the -qdisc and filters for TCP ports 5000 through 5007. +qdisc and filters for TCP ports 5000 through 5007. Then use a tool, such as +iperf, to generate UDP or TCP traffic on ports 5000-5007. + +Statistics for each queue of the ixgbe driver can be checked using the +ethtool utility: ethtool -S ethX # tc qdisc add dev ethX root handle 1: multiq # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5000 0xffff action skbedit queue_mapping 1 +u32 match ip dport 5000 0xffff action skbedit queue_mapping 0 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5000 0xffff action skbedit queue_mapping 1 +u32 match ip sport 5000 0xffff action skbedit queue_mapping 0 -# tc filter add dev ethx protocol ip parent 1: \ -u32 match ip dport 5001 0xffff action skbedit queue_mapping 2 +# tc filter add dev ethX protocol ip parent 1: \ +u32 match ip dport 5001 0xffff action skbedit queue_mapping 1 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5001 0xffff action skbedit queue_mapping 2 +u32 match ip sport 5001 0xffff action skbedit queue_mapping 1 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5002 0xffff action skbedit queue_mapping 3 +u32 match ip dport 5002 0xffff action skbedit queue_mapping 2 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5002 0xffff action skbedit queue_mapping 3 +u32 match ip sport 5002 0xffff action skbedit queue_mapping 2 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5003 0xffff action skbedit queue_mapping 4 +u32 match ip dport 5003 0xffff action skbedit queue_mapping 3 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5003 0xffff action skbedit queue_mapping 4 +u32 match ip sport 5003 0xffff action skbedit queue_mapping 3 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5004 0xffff action skbedit queue_mapping 5 +u32 match ip dport 5004 0xffff action skbedit queue_mapping 4 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5004 0xffff action skbedit queue_mapping 5 +u32 match ip sport 5004 0xffff action skbedit queue_mapping 4 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5005 0xffff action skbedit queue_mapping 6 +u32 match ip dport 5005 0xffff action skbedit queue_mapping 5 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5005 0xffff action skbedit queue_mapping 6 +u32 match ip sport 5005 0xffff action skbedit queue_mapping 5 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5006 0xffff action skbedit queue_mapping 7 +u32 match ip dport 5006 0xffff action skbedit queue_mapping 6 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5006 0xffff action skbedit queue_mapping 7 +u32 match ip sport 5006 0xffff action skbedit queue_mapping 6 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip dport 5007 0xffff action skbedit queue_mapping 8 +u32 match ip dport 5007 0xffff action skbedit queue_mapping 7 # tc filter add dev ethX protocol ip parent 1: \ -u32 match ip sport 5007 0xffff action skbedit queue_mapping 8 +u32 match ip sport 5007 0xffff action skbedit queue_mapping 7 + dcbtool Overview ================= + dcbtool is used to query and set the DCB settings of a DCB capable Ethernet + interface. It connects to the client interface of dcbd to perform these + operations. dcbtool will operate in interactive mode if it is executed + without a command. In interactive mode, dcbtool also functions as an event + listener and will print out events received from dcbd as they arrive. + + SYNOPSIS + dcbtool -h + dcbtool -v + dcbtool [-rR] + dcbtool [-rR] [command] [command arguments] + + OPTIONS + -h shows the dcbtool usage message + + -v shows dcbtool version information + + -r displays the raw dcbd client interface messages as well as the + readable output. + + -R displays only the raw dcbd client interface messages + + COMMANDS + help shows the dcbtool usage message + + ping test command. The dcbd daemon responds with "PONG" if the + client interface is operational. + + license + displays dcbtool license information + + quit exits from interactive mode + + The following commands interact with the dcbd daemon to manage the dae- + mon and DCB features on DCB capable interfaces. + + dcbd general configuration commands: + ------------------------------------ + <gc|go> dcbx + gets the configured or operational version of the DCB capabili- + ties exchange protocol. If different, the configured version + will take effect (and become the operational version) after dcbd + is restarted. + + sc dcbx v:[1|2] + sets the version of the DCB capabilities exchange protocol which + will be used the next time dcbd is started. Information about + version 1 can be found at: + <http://download.intel.com/technology/eedc/dcb_cep_spec.pdf> + Information about version 2 can be found at: + <http://www.ieee802.org/1/files/public/docs2008/az-wadekar-dcbx- + capability-exchange-discovery-protocol-1108-v1.01.pdf> + + DCB-per interface commands: + --------------------------- + gc <ifname> <feature> + gets configuration of feature on interface ifname. + + go <ifname> <feature> + gets operational status of feature on interface ifname. + + gp <ifname> <feature> + gets peer configuration of feature on interface ifname. + + sc <ifname> <feature> <args> + sets the configuration of feature on interface ifname. + + feature may be one of the following: + ------------------------------------ + dcb DCB state of the port + + pg priority groups + + pfc priority flow control + + app:<subtype> + application specific data + + ll:<subtype> + logical link status + + subtype can be: + --------------- + 0|fcoe Fiber Channel over Ethernet (FCoE) + + args can include: + ----------------- + e:<0|1> + controls feature enable + + a:<0|1> + controls whether the feature is advertised via DCBX to the peer + + w:<0|1> + controls whether the feature is willing to change its opera- + tional configuration based on what is received from the peer + + [feature specific args] + arguments specific to a DCB feature + + Feature specific arguments for dcb: + ----------------------------------- + on|off enable or disable for the interface. The go and gp commands are + not needed for the dcb feature. Also, the enable, advertise and + willing" parameters are not required. + + Feature specific arguments for pg: + ---------------------------------- + pgid:xxxxxxxx + Priority group ID for the 8 priorities. From left to right + (priorities 0-7), x is the corresponding priority group ID + value, which can be 0-7 for priority groups with bandwidth allo- + cations or f (priority group ID 15) for the unrestricted prior- + ity group. + + pgpct:x,x,x,x,x,x,x,x + Priority group percentage of link bandwidth. From left to right + (priority groups 0-7), x is the percentage of link bandwidth + allocated to the corresponding priority group. The total band- + width must equal 100%. + + uppct:x,x,x,x,x,x,x,x + Priority percentage of priority group bandwidth. From left to + right (priorities 0-7), x is the percentage of priority group + bandwidth allocated to the corresponding priority. The sum of + percentages for priorities which belong to the same priority + group must total 100% (except for priority group 15). + + strict:xxxxxxxx + Strict priority setting. From left to right (priorities 0-7), x + is 0 or 1. 1 indicates that the priority may utilize all of the + bandwidth allocated to its priority group. + + up2tc:xxxxxxxx + Priority to traffic class mapping. From left to right (priori- + ties 0-7), x is the traffic class (0-7) to which the priority is + mapped. (This setting is ignored for Intel 82598-based + adapters.) + + Feature specific arguments for pfc: + ----------------------------------- + pfcup:xxxxxxxx + Enable/disable priority flow control. From left to right (pri- + orities 0-7), x is 0 or 1. 1 indicates that the corresponding + priority is configured to transmit priority pause. + + Feature specific arguments for app:< subtype>: + ---------------------------------------------- + appcfg:xx + xx is a hexadecimal value representing an 8 bit bitmap where รข + bits indicate the priority which frames for the applications + specified by subtype should use. The lowest order bit maps to + priority 0. + + Feature specific arguments for ll:<subtype>: + -------------------------------------------- + status:[0|1] + For testing purposes, the logical link status may be set to 0 or + 1. This setting is not persisted in the configuration file. + + + EXAMPLES + Enable DCB on interface eth2 + + dcbtool sc eth2 dcb on + + Assign priorities 0-3 to priority group 0, priorities 4-6 to priority + group 1 and priority 7 to the unrestricted priority. Also, allocate + 25% of link bandwidth to priority group 0 and 75% to group 1. -dcbtool is a tool to query and set the DCB settings of an Ethernet device. + dcbtool sc eth2 pg pgid:0000111f pgpct:25,75,0,0,0,0,0,0 -Options -------- + Enable transmit of Priority Flow Control for priority 3 and assign FCoE + to priority 3. -The basic usage of dcbtool is: - dcbtool [options] <cmd> [cmd arguments] + dcbtool sc eth2 pfc pfcup:00010000 + dcbtool sc eth2 app:0 appcfg:08 -If no command is specified, then dcbtool will enter an interactive mode which -allows commands to be entered interactively. When invoked in interactive -mode, dcbtool will also register with dcbd as an event monitor. - --h shows the dcbtool usage message --v shows dcbtool version information - -Commands --------- - -help shows the dcbtool usage message - -level <value> sets the threshold dcbd uses to send event messages to this - interface. - Levels are: - 0: MSG_DUMP - for dcbd debugging - 1: MSG_DEBUG - for dcbd debugging - 2: MSG_INFO - for free form dcbd informational - messages - 3: MSG_WARNING - for free form dcbd warning - messages - 4: MSG_DCB - for formated DCB event messages - -ping test command. dcbd responds with "PONG" - -license displays dcbtool license information - -quit exit from interactive mode - -gc <ifname> <dcb | pg | pfc | app:<subtype>> - Get configuration command. Returns the local configuration for - the specified interface and DCB feature. Features are: - dcb - DCB state of port - pg - priority groups - pfc - priority flow control - app:<subtype> - application with subtypes: - fcoe - -go <ifname> <pg | pfc | app:<subtype>> - Get operational status command. Returns the current DCB - operational status and configuration for the specified - interface and DCB feature. - -gp <ifname> <pg | pfc | app:<subtype>> - Get peer status command. Returns the current DCB status - and configuration for the DCB peer of the specified interface - and DCB feature. - -sc <ifname> dcb <on | off> - Set the DCB state of the port. - -sc <ifname> <pg | pfc | app:<subtype>> [a:<0|1>] [e:<0|1>] [w:<0|1>] - [feature attributes] - Set configuration command. Allows setting, per DCB feature, - the following: - a:<0|1> - enable/disable advertising the feature via DCB - e:<0|1> - enable/disable the DCB feature - w:<0|1> - enable/disable willing flag for the feature - If one of the above parameters is not specified in the set - config command, then the DCB feature attributes must be - specified. - - pg - priority group attributes - Priority group attributes which are not supplied will remain - unchanged. Except for the 'uppct'. If 'uppct' is not - supplied and the resulting percentages within each PGID do - not add up to 100%, then dcbd will compute new user priority - percentages for each PGID so that the sum for each PGID is - 100%. - - up2tc:<0-7><0-7><0-7><0-7><0-7><0-7><0-7><0-7> - This is the user priority to traffic class mapping parameter. - If not specified, the default value is "01234567". - - pgid:<0-7F><0-7F><0-7F><0-7F><0-7F><0-7F><0-7F><0-7F> - This is the priority group ID mapping. From left to right, - each number indicates the priority group ID to which the - corresponding user priority, starting with zero, belongs. - A value of 'F' (0xF) indicates that the user priority - belongs to the link strict group. - - pgpct:<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>, - <0-100> - This is the priority group bandwidth percentage of link - setting. From left to right, each number indicates the - percentage of total link bandwidth which is allocated to the - corresponding priority group ID, starting with zero. - - uppct:<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>,<0-100>, - <0-100> - This is the user priority bandwidth percentage of bandwidth - group setting. From left to right, each number indicates the - percentage of total priority group bandwidth which is - allocated to the corresponding user priority, starting with - zero. - - strict:<0-2><0-2><0-2><0-2><0-2><0-2><0-2><0-2> - This is the strict priority settings for each user priority. - - From left to right, each number indicates the strict priority - setting for each user priority, starting with zero. A value - of 0 represents no strict priority setting, 1 represents group - strict. - - pfc - priority flow control attributes - - pfcup:<0|1><0|1><0|1><0|1><0|1><0|1><0|1><0|1> - This is the priority flow control enable setting per user - priority. From left to right, each number indicates the enable - setting of priority flow control for the corresponding user - priority, starting from zero. 0 means disabled, 1 means - enabled. - - app - application attributes - - appcfg:<xx> - This represents the application settings for the specified - application subtype. Assuming that the application attribute - setting is an arbitrary sequence of bytes of a given length, - the setting is supplied to dcbtool by representing each byte as - a 2 character hexadecimal string. Thus, if the application - attribute setting is the value 0x5a, the dcbtool attribute - would be "appcfg:5a" FAQ === @@ -444,7 +495,7 @@ FAQ Known Issues ============ -- Prior to kernel 2.6.26, tso will be disabled when the driver is put into DCB +- Prior to kernel 2.6.26, TSO will be disabled when the driver is put into DCB mode. - A TX unit hang may be observed when link strict priority is set when a large diff -pur dcbd-0.9.2/SUMS dcbd-0.9.2.2/SUMS --- dcbd-0.9.2/SUMS 2008-12-02 17:36:22.000000000 -0800 +++ dcbd-0.9.2.2/SUMS 2008-12-22 17:26:48.000000000 -0800 @@ -1,4 +1,4 @@ -53544 3 dcbd7 +14460 4 dcbd7 56524 36 clif_cmds.c 23104 2 libconfig/Makefile 61336 131 libconfig/libconfig-1.3.1.patch @@ -8,7 +8,7 @@ 28010 10 common.c 57774 9 ctrl_iface.c 21712 23 dcbtool_cmds.c -45810 19 dcbtool.c +29449 19 dcbtool.c 01403 6 clif.c 39773 19 COPYING 23984 30 nltest.c @@ -17,8 +17,8 @@ 28102 24 parse_cli.c 13564 12 dcb_rule_chk.cpp 56189 4 os_unix.c -49466 19 README -62150 7 dcbtool7 +58993 20 README +36764 9 dcbtool7 25349 21 drv_cfg.c 28671 126 dcb_protocol.cpp 62067 3 include/ctrl_iface.h @@ -33,7 +33,7 @@ 08794 7 include/clif_cmds.h 08101 7 include/clif.h 44799 2 include/config.h -03546 2 include/version.h +11404 2 include/version.h 27192 3 include/messages.h 41092 12 include/common.h 55964 2 include/dcb_osdep.h @@ -45,7 +45,7 @@ 21949 2 include/event_iface.h 50008 11 include/dcb_types.h 09063 3 include/parse_cli.h -44070 2 dcbd.spec +46738 2 dcbd.spec 04129 7 dcbd.init 52219 8 event_iface.c 62795 4 nltest.h @@ -56,7 +56,7 @@ 08229 2 lldp/agent.h 15669 2 lldp/Makefile 44452 10 lldp/ports.c -58821 46 lldp/rx.c +38179 46 lldp/rx.c 24608 5 lldp/ports.h 48526 25 lldp/tlv.c 15542 2 lldp/mibdata.h
Locations
Projects
Search
Status Monitor
Help
OpenBuildService.org
Documentation
API Documentation
Code of Conduct
Contact
Support
@OBShq
Terms
openSUSE Build Service is sponsored by
The Open Build Service is an
openSUSE project
.
Sign Up
Log In
Places
Places
All Projects
Status Monitor
