網路晶片商的解決方案
Survey and Research:
1. Broadcom方案
(1) LDK:
Linux Development Kit (LDK) for the Broadcom iProc (integrated Processor) family of processors.
The iProcLDK currently supports all the production iProc SOCs (Northstar/Northstar+/Northstar2/ … etc). It provides as buildroot package for toolchains/compiler, U-Boot loader, Linux kernel 3.6.5 BSP and applications. The LDK now has SAMBA optimized porting for NAS application.
(2) ESDK:
WLAN Enterprise SDK (or Enterprise Wireless SDK) which support to the Broadcom reference platforms. It provides bootloader (CFE/U-boot), Linux kernel 2.6.36 BSP, WiFi driver and application (with Enterprise features).
(3) FASTPATH-UAP:
You may visit our company website with following link for product brief.
For customer doing Enterprise AP product design, they will choose ESDK or FastpathUAP (if they want turnkey solution).
For customer doing NAS product or application other than Enterprise AP they can use LDK.
The initial batch of 802.11ac routers shipped with the dated and inefficient BCM4706 (based on a 600 MHz MIPS32 CPU). Broadcom made some amends last year with the BCM4708x / BCM5301x targeting the SOHO and SMB space respectively. Today, Broadcom is introducing the StrataGX BCM5862x to satisfy the market demand for SMB storage routers.
Broadcom has made some evolutionary updates to the BCM5301x in order to increase performance and simplify design.
The BCM5862x supports both 16 and 32b DDR3 (compared to the 16b-only BCM5301x). The Cortex-A9 cores gain an extra 200 MHz to run at 1.2 GHz. L2 cache is doubled from 256KB to 512KB. The SoC integrates two SATA ports (to accommodate such a design, the BCM5301x had to use an external PCIe switch and a PCIe SATA bridge). The Cortex-A9 cluster in the BCM5862x integrates NEON and VFP, as these features are important for NAS platforms. A 5-port GbE switch is also integrated. The FlexSPARX engine in the StrataGX BCM5862x provides packet and storage acceleration that offloads CPU for other tasks and a crypto engine provides support for IPsec / SSL and VPNs. These features make it attractive for integrated NAS / branch routers in SMB environments.
Broadcom Hardware Acceleration
https://www.snbforums.com/threads/broadcoms-hardware-acceleration.18144/
CTF is Broadcom's closed-source, proprietary "secret sauce" that allows routers based on their hardware to achieve near gigabit performance. It does so through various methods which are not publicly known (even manufacturers don't get access to the ctf.ko source code AFAIK). One of it involves bypassing parts of Linux's Netfilter (the FORWARD chain is the most known one).
So as you can already see, "hardware acceleration" isn't an entirely accurate name. At least one portion of that acceleration is really a software trick (bypassing part of Linux's stack).
Due to these bypass, it prevents various firmware-level features from working. Anything that relies on the FORWARD chain for instance. The solution used by router manufacturers usually work on two different levels:
1) Some manufacturers like Asus and Netgear (if I remember correctly) will allow port-forwarding to work by modifying the Linux kernel so that any packet that gets marked will be flagged to bypass the CTF code. At the iptables level, any port-forwarded packet gets marked with a value. This way, you can have HW acceleration enabled and still use port forwards. The obvious consequence of this is that any traffic going through a port forward will not be "hardware-accelerated". So if you were to push a lot of traffic over a forwarded port, that traffic would probably not be able to reach near gigabit performance.
(caveat: I never actually tested this. I assume that CTF bypass is applied to every single packets that gets marked, not just on part of it)
2) When certain incompatible features are enabled, then the router is rebooted with CTF disabled. In this mode, the processing is then entirely done by Linux. It allows you to do anything you'd want (as a firmware developer), but performance is seriously impacted. A typical 600 MHz MIPS device (such as the RT-N66U) will reach a WAN to LAN limit of around 150-200 Mbps (less if you start heavily processing traffic through QoS, parental control, custom firewall rules, etc...). Unfortunately, it's not always clear to the end user when HW acceleration is automatically disabled by such a thing. If your router has telnet access, you could see if the ctf.ko kernel module is loaded or not, using the "lsmod" command.
CTF is what explains why most third party firmwares (such as DD-WRT) tend to have lower throughput than manufacturer stock firmwares. For people with average (North American levels there) WAN rates of 10-100 Mbps, this is not an issue. Any additional feature will come at no real cost on maximum throughput. But for our more fortunate oversea friends who get 100-1000 Mbits link speeds, CTF is virtually essential.
Due to the nature of its closed-sourceness, and also the fact that many advanced features do not work with CTF enabled, this is why most third party firmwares such as DD-WRT or OpenWRT don't support CTF.
Now, another recent topic: the different levels of hardware acceleration. Recent Broadcom chips support a new technology they call "Flow Acceleration", or "FA" for short. Broadcom's demonstration can be seen in this video:
https://www.youtube.com/watch?v=vwRmQkkZ71E
In home routers that have hardware supporting this, it gets handled by the same ctf.ko module, in addition to support being implemented at the Ethernet driver level. Unfortunately I don't know which specific Broadcom chips support this, or which specific routers support it. I know that neither the RT-AC56U or RT-AC68U (as of this date) support this at the hardware level. No idea about Netgear or Linksys's recent products.
In Asus's particular case (since it's the one I'm most familiar with - someone else could fill us up on the other manufacturers), they are handling this as a "Hardware acceleration level". Level 1 is just traditional CTF. Level 2 is traditional CTF + FA. One coming product that does support both levels will have to downgrade from Level 2 to Level 1 when one of the new features they are adding will be enabled.
One thing I do not know however is what kind of performance impact FA has on a router. Traditional CTF was already able to push things fairly close to gigabit speed with a minimal CPU impact.
(disclaimer: most of this is based on my own experience over the years. Due to the blackbox nature of CTF, I might not be 100% correct on all of this, so if anyone has any additional detail or corrections, feel free to share)
CTF is Broadcom's closed-source, proprietary "secret sauce" that allows routers based on their hardware to achieve near gigabit performance. It does so through various methods which are not publicly known (even manufacturers don't get access to the ctf.ko source code AFAIK). One of it involves bypassing parts of Linux's Netfilter (the FORWARD chain is the most known one).
So as you can already see, "hardware acceleration" isn't an entirely accurate name. At least one portion of that acceleration is really a software trick (bypassing part of Linux's stack).
Due to these bypass, it prevents various firmware-level features from working. Anything that relies on the FORWARD chain for instance. The solution used by router manufacturers usually work on two different levels:
1) Some manufacturers like Asus and Netgear (if I remember correctly) will allow port-forwarding to work by modifying the Linux kernel so that any packet that gets marked will be flagged to bypass the CTF code. At the iptables level, any port-forwarded packet gets marked with a value. This way, you can have HW acceleration enabled and still use port forwards. The obvious consequence of this is that any traffic going through a port forward will not be "hardware-accelerated". So if you were to push a lot of traffic over a forwarded port, that traffic would probably not be able to reach near gigabit performance.
(caveat: I never actually tested this. I assume that CTF bypass is applied to every single packets that gets marked, not just on part of it)
2) When certain incompatible features are enabled, then the router is rebooted with CTF disabled. In this mode, the processing is then entirely done by Linux. It allows you to do anything you'd want (as a firmware developer), but performance is seriously impacted. A typical 600 MHz MIPS device (such as the RT-N66U) will reach a WAN to LAN limit of around 150-200 Mbps (less if you start heavily processing traffic through QoS, parental control, custom firewall rules, etc...). Unfortunately, it's not always clear to the end user when HW acceleration is automatically disabled by such a thing. If your router has telnet access, you could see if the ctf.ko kernel module is loaded or not, using the "lsmod" command.
CTF is what explains why most third party firmwares (such as DD-WRT) tend to have lower throughput than manufacturer stock firmwares. For people with average (North American levels there) WAN rates of 10-100 Mbps, this is not an issue. Any additional feature will come at no real cost on maximum throughput. But for our more fortunate oversea friends who get 100-1000 Mbits link speeds, CTF is virtually essential.
Due to the nature of its closed-sourceness, and also the fact that many advanced features do not work with CTF enabled, this is why most third party firmwares such as DD-WRT or OpenWRT don't support CTF.
Now, another recent topic: the different levels of hardware acceleration. Recent Broadcom chips support a new technology they call "Flow Acceleration", or "FA" for short. Broadcom's demonstration can be seen in this video:
https://www.youtube.com/watch?v=vwRmQkkZ71E
In home routers that have hardware supporting this, it gets handled by the same ctf.ko module, in addition to support being implemented at the Ethernet driver level. Unfortunately I don't know which specific Broadcom chips support this, or which specific routers support it. I know that neither the RT-AC56U or RT-AC68U (as of this date) support this at the hardware level. No idea about Netgear or Linksys's recent products.
In Asus's particular case (since it's the one I'm most familiar with - someone else could fill us up on the other manufacturers), they are handling this as a "Hardware acceleration level". Level 1 is just traditional CTF. Level 2 is traditional CTF + FA. One coming product that does support both levels will have to downgrade from Level 2 to Level 1 when one of the new features they are adding will be enabled.
One thing I do not know however is what kind of performance impact FA has on a router. Traditional CTF was already able to push things fairly close to gigabit speed with a minimal CPU impact.
(disclaimer: most of this is based on my own experience over the years. Due to the blackbox nature of CTF, I might not be 100% correct on all of this, so if anyone has any additional detail or corrections, feel free to share)
2. Marvell LinkStreet
The Marvell Link Street family of integrated networking devices are specifically designed for the Small Office/Home Office (SOHO) and Small to Medium Business (SMB) markets. Link Street Gigabit and Fast Ethernet switches offer ideal switch configurations and functionality for products where cost, ease-of-use and flexibility are of paramount importance.
Link Street Gigabit Switches integrate a high-performance non-blocking gigabit switch fabric with four priority queues, a high-speed lookup engine, multiple MGMII Gigabit Ethernet interface ports, and 1Mb of packet memory. Other advanced features include IEEE 802.1p/IPv4/IPv6 traffic classification and port-based VLANs.
The Link Street Fast Ethernet switches integrate multiple independent Fast Ethernet Media Access Controllers (MACs), multiple 10BASE-T/100BASE-TX transceivers (PHYs), a high-speed non-blocking switch fabric, a high-performance address lookup engine, and an embedded frame memory. These devices enable PCs and IP phones to connect through a broadband Internet access pipeline (DSL modem, cable modem, or fiber Wide Area Network [WAN]) to the Internet, with all ports switched at 100 Megabits per second (Mbps) bandwidth Ethernet.
The Marvell SOHO switches with integrated PHYs build upon proven Marvell physical layer (PHY) technology, incorporate the advanced features such as smart power management, which dramatically reduces power consumption by more than 50% and Virtual Cable Tester (VCT™) technology, which performs cable diagnostics to reduce overall network support costs. These enhanced features make these switches ideal for applications such as standalone switches, media converters, IP phones, firewall appliances, wireless and wired gateway routers, and wireless access points.
For applications that require extended temperature range support, such as aggregators, gateways and network terminals that are placed outdoors and are subjected to harsh environments, Marvell also offers a range of Industrial Grade switches that can operate between -40°C and + 85°C .
Part Number
|
Description
|
I-Temp
|
Documentation
|
88E6020
|
4-Port Fast Ethernet Switch with EEE
|
Yes
| |
88E6031
|
3-Port Fast Ethernet Switch
|
No
| |
88E6035
|
Advanced feature 3-Port Fast Ethernet Switch
|
No
| |
88E6060
|
6-Port Fast Ethernet Switch
|
Yes
| |
88E6061
|
6-Port Fast Ethernet Switch w/ QoS, 802.1Q
|
Yes
| |
88E6061B
|
6-Port Fast Ethernet Switch w/ QoS, 802.1Q, and Switch Bypass
|
Yes
| |
88E6063
|
7-Port Fast Ethernet Switch w/ QoS, SNMP, 802.1Q
|
Yes
| |
88E6065
|
Advanced Feature 6-Port Fast Ethernet Managed Switch
|
Yes
| |
88E6065B
|
Advanced Feature 6-Port Fast Ethernet Managed Switch with Switch Bypass
|
Yes
| |
88E6070
|
5-Port Fast Ethernet Switch with EEE
|
No
| |
88E6071
|
7-Port Fast Ethernet Switch with EEE
|
Yes
| |
88E6085
|
Advanced Feature 10-Port Fast Ethernet Managed Switch
|
Yes
| |
88E6220
|
4-Port Fast Ethernet AVB Switch with EEE
|
Yes
| |
88E6250
|
7-Port Fast Ethernet AVB Switch with EEE
|
Yes
|
Part Number
|
Description
|
I-Temp
|
Documentation
|
88E6046
|
Advanced Feature 4-Port FE + 2-Port GE Ethernet Switch
|
Yes
| |
88E 6046F
|
Advanced Feature 4-Port FE + 2-Port GE Ethernet Switch
|
Yes
| |
88E6096
|
8-Port FE + 3-Port GE Managed Switch
|
No
| |
88E6097
|
Advanced Feature 8-Port FE + 3-Port GE Managed Switch
|
Yes
| |
88E 6097F
|
Advanced Feature 8-Port FE + 3-Port GE Managed Switch
|
Yes
| |
88E6240
|
4-port FE + 3-Port GE AVB Switch with TCAM and EEE
|
Yes
|
Part Number
|
Description
|
I-Temp
|
Documentation
|
88E6218
|
Gateway SoC with 150MHz ARM CPU with 7-port FE switch
|
No
| |
88E7221
|
AVB Gateway SoC with 400MHz ARM CPU,4-port FE switch,USB, SDIO, AVB TDM/I2S ports
|
Yes
| |
88E7251
|
AVB Gateway SoC with 400MHz ARM CPU,7-port FE switch,USB, SDIO, AVB TDM/I2S ports
|
TBD
| |
88E 7251F
|
AVB Gateway SoC with 400MHz ARM CPU,7-port FE switch,USB, SDIO, AVB TDM/I2S ports
|
No
|
Part Number
|
Description
|
I-Temp
|
Documentation
|
88E6122
|
6 Port GE Switch w/ 2 integrated GE PHYs, 3 SGMII, GMII/MII
|
Yes
| |
88E6131
|
8 Port GE Switch w/ 3 integrated GE PHYs, 4 SGMII, GMII/MII
|
Yes
| |
88E6152
|
6-Port GE Smart/Unmanaged Switch
|
No
| |
88E6155
|
6-Port GE Managed Switch
|
No
| |
88E6161
|
6-Port GE Switch w/ 5 integrated PHYs
|
Yes
| |
88E6165
|
Advanced Feature 6-Port GE Switch w/ 5 integrated PHYs
|
Yes
| |
88E6171/88E6171R
|
7-Port GE Switch w/ 5 integrated PHYs + 2GMII/RGMII
|
No
| |
88E6172
|
7-Port GE Switch w/ 5 integrated PHYs w/ 1GMII, RGMII, and EEE
|
No
| |
88E6175/88E6175R
|
7-Port GE Switch w/ 5 integrated PHYs + 2GMII/RGMII
|
No
| |
88E6176
|
7-Port GE Switch w/ 5 integrated PHYs w/ 1GMII, RGMII, 1 Serdes, and EEE
|
Yes
| |
88E6182
|
10-Port GE Smart/Unmanaged Switch
|
No
| |
88E6185
|
10-Port GE Stackable Managed Switch
|
Yes
| |
88E6350R
|
7 Port AVB Gigabit Ethernet Switch with 5 Integrated 10/100/1000 PHYs
|
Yes
| |
88E6351
|
7 Port AVB Gigabit Ethernet Switch with 5 Integrated PHYs, Synchronous Ethernet
|
Yes
| |
88E6352
|
7 Port AVB Gigabit Ethernet Switch with 5 Integrated PHYs, Synchronous Ethernet, and EEE
|
Yes
| |
88E6390X
|
11 Port AVB Ethernet Switch with Eight Integrated PHYs and Two 10Gbps Interfaces
|
TBD
|
留言
張貼留言