FreeCalypso > hg > freecalypso-tools
comparison doc/RVTMUX @ 0:e7502631a0f9
initial import from freecalypso-sw rev 1033:5ab737ac3ad7
| author | Mychaela Falconia <falcon@freecalypso.org> |
|---|---|
| date | Sat, 11 Jun 2016 00:13:35 +0000 |
| parents | |
| children | d9307880f59f |
comparison
equal
deleted
inserted
replaced
| -1:000000000000 | 0:e7502631a0f9 |
|---|---|
| 1 TI's Calypso GSM/GPRS baseband processor chip has not one but two UART serial | |
| 2 ports, called "MODEM" and "IrDA" in the hardware documentation. In hardware | |
| 3 terms, both support basic data-leads-only UART operation at a fixed baud rate, | |
| 4 but their extended capabilities differ: the IrDA UART adds IrDA capability (no | |
| 5 surprise), whereas the MODEM UART adds hardware flow control and autobaud. If | |
| 6 one is not implementing an actual IrDA interface, then the so-called "IrDA" | |
| 7 UART becomes a strict subset of the MODEM one in terms of hw capabilities - | |
| 8 just an extra UART, but a somewhat less capable one. | |
| 9 | |
| 10 In a classic modem design such as that present in the GTA0x smartphones made by | |
| 11 FIC/Openmoko, the Calypso presents a standard AT command interface on its MODEM | |
| 12 UART port. (In the case of GTA0x phones this serial channel is wired to the | |
| 13 phone's application processor; in a standalone modem it would be wired to a | |
| 14 USB-serial chip or even to a classic RS-232 DB25 port.) However, what is less | |
| 15 known is that the standard firmware for such modems simultaneously presents an | |
| 16 entirely different interface on the IrDA UART - an interface intended for | |
| 17 development, debugging and factory production testing (which includes RF | |
| 18 calibration and IMEI etc programming), rather than for "normal" end users. | |
| 19 | |
| 20 Normally this debug/development serial interface (called RVTMUX as will be | |
| 21 explained momentarily) is hidden from "ordinary" users - for example, on FIC | |
| 22 GTA0x phones it is wired to the analog headset jack through a hardware switch | |
| 23 which needs to be enabled through a GPIO signal from the AP. On Mot C139 and | |
| 24 its siblings the situation is similar: one needs to key in the secret magic | |
| 25 sequence **16379#, and then the firmware presents a hidden menu for switching | |
| 26 the analog headset jack between its "normal" function and the UART carrying | |
| 27 RVTMUX. | |
| 28 | |
| 29 But there also exist some oddball devices on which the RVTMUX interface is | |
| 30 presented "in your face". The Pirelli DP-L10 phone has a USB charging port | |
| 31 which is also wired (through a CP2102 USB-serial chip) to the IrDA UART on the | |
| 32 Calypso - that's right, IrDA, not MODEM - a design decision with which this | |
| 33 hacker strongly disagrees. (It'll definitely be wired to the MODEM UART | |
| 34 instead on our own semi-clone of this phone, but I digress.) Apparently Foxconn | |
| 35 (the designers of this phone) had no desire to provide a standard AT command | |
| 36 interface, and instead the only "official" way to use the "data" function of | |
| 37 their USB port (rather than the charging function) is for their "PC sync" | |
| 38 feature, i.e., their proprietary Weendoze software. And guess what, their | |
| 39 proprietary "PC sync" feature works over TI's RVTMUX interface, as that is | |
| 40 what's presented on Calypso's IrDA UART behind the CP2102! | |
| 41 | |
| 42 OK, so what is this RVTMUX? RV stands for RiViera, an application framework | |
| 43 which TI added to their GSM firmware suite in the early 2000s, T stands for | |
| 44 trace, and MUX stands for multiplexor. It's a binary packet interface, although | |
| 45 many of these packets contain ASCII debug messages inside. The framing format | |
| 46 is the same in both directions: each packet begins and ends with an STX (0x02) | |
| 47 byte, all payload bytes except 0x02 and 0x10 are sent literally, and there is a | |
| 48 DLE (0x10) byte prepended before any 0x02 or 0x10 in the payload. It's the same | |
| 49 general principle as asynchronous HDLC (RFC 1662): packets can contain any | |
| 50 binary data, and the framing provides packet boundaries - although TI's version | |
| 51 is a little less robust than async-HDLC when it comes to recovering after lost | |
| 52 synchronization. | |
| 53 | |
| 54 The firmware suite component responsible for actually sending and receiving | |
| 55 these packets over the assigned UART port (usually IrDA, but can be MODEM too, | |
| 56 as on Compal phones for example) is called RVT (RiViera Trace), and it | |
| 57 implements a MUX function. There are several logical channels multiplexed over | |
| 58 one physical serial port, and the first byte of every packet indicates which | |
| 59 logical channel it belongs to. Any component within the GSM firmware suite can | |
| 60 send packets to RVT for transmission on this serial interface, and can also | |
| 61 register to receive packets beginning with a particular type ID byte. | |
| 62 | |
| 63 Debug trace output | |
| 64 ================== | |
| 65 | |
| 66 All GSM device firmwares that are based on TI's Calypso chipset reference fw | |
| 67 continuously emit quite voluminous debug trace output on their RVTMUX serial | |
| 68 port, whether it is hidden or exposed on a given device. Like all RVTMUX | |
| 69 traffic, this debug trace output takes the form of binary packets as explained | |
| 70 above, but the content of these packets is mostly ASCII with some binary header | |
| 71 bytes prepended. FreeCalypso host utility rvtdump captures all serial output | |
| 72 from a GSM device's RVTMUX port, parses the packet structure and displays this | |
| 73 output in line-oriented pure ASCII with all binary parts decoded. | |
| 74 | |
| 75 Test Mode commands | |
| 76 ================== | |
| 77 | |
| 78 The other major use of the RVTMUX interface is sending so-called Test Mode | |
| 79 commands from an external host to a running GSM device. Depending on the | |
| 80 firmware version, a GSM device can be commanded to do any of the following | |
| 81 things through this mechanism: | |
| 82 | |
| 83 * Exercise RF test modes, e.g., transmit continuously at a set frequency and | |
| 84 power level; | |
| 85 * Read and write arbitrary memory locations in the Calypso ARM7 address space; | |
| 86 * Read and write ABB chip registers; | |
| 87 * Reboot or power off; | |
| 88 * Access and manipulate the device's flash file system (FFS). | |
| 89 | |
| 90 In the segment of history of interest to us TI has produced two different | |
| 91 target firmware components that can receive, interpret and act upon Test Mode | |
| 92 command packets: | |
| 93 | |
| 94 * The original Test Mode component of Layer 1, called L1TM or TML1: this | |
| 95 component handles all RF test modes (needed for RF calibration on device | |
| 96 production lines), and originally it also implemented memory and ABB register | |
| 97 read and write commands, and provided access to TMFFS1 (see below). In the | |
| 98 original implementation this component registered itself as the handler for | |
| 99 the "TM" RVTMUX channel (RVT packet type 0x14), so it would receive all TM | |
| 100 packets sent to the device. | |
| 101 | |
| 102 * Enhanced Test Mode (ETM) is a later invention. It registers itself (instead | |
| 103 of the old TM in L1) with RVT as the handler for the "TM" RVTMUX channel, and | |
| 104 then provides a registration service of its own, such that various components | |
| 105 in the fw suite can register to receive external command packets passing | |
| 106 first through RVT, then through ETM, and can send responses passing through | |
| 107 ETM, then through RVT back to the external host. If a given fw version | |
| 108 contains both ETM and L1TM, then L1TM registers itself with ETM; an external | |
| 109 host would send exactly the same binary command packets to exercise RF test | |
| 110 modes, but inside the firmware they now pass through ETM on their way to L1TM. | |
| 111 | |
| 112 The ETM_CORE module contained within ETM itself provides some low-level debug | |
| 113 commands: by sending the right binary command packets to the GSM device via the | |
| 114 RVTMUX serial channel, an external host can examine or modify any memory | |
| 115 location and any hardware register, cause the device to reset, etc. Prior to | |
| 116 ETM some of these functions (but not all) could be exercised through older TM3 | |
| 117 commands, but in FreeCalypso we became familiar with the ETM versions of these | |
| 118 commands long before the older ones because we got the ETM component in full | |
| 119 source form, whereas our copy of TCS211 (TI's reference fw) has L1TM in a | |
| 120 binary library. | |
| 121 | |
| 122 Our TCS211/leo2moko reference fw has both ETM and L1TM, thus it accepts both | |
| 123 ETM and TM3 command packets. ETM commands (including TMFFS2, see below) work | |
| 124 on Pirelli's fw, but Mot/Compal's original fw for the C139 has only the | |
| 125 original non-enhanced Test Mode, not ETM. | |
| 126 | |
| 127 FFS access via TM/ETM | |
| 128 ===================== | |
| 129 | |
| 130 One of the essential facilities provided in one form or another in all known | |
| 131 incarnations of the Test Mode mechanism is the ability to access and manipulate | |
| 132 the GSM device's flash file system (FFS). See TIFFS-Overview for a description | |
| 133 of this file system. TI's TMFFS1 and TMFFS2 protocols provide a command and | |
| 134 response packet interface to the FFS API functions inside the fw, and enable an | |
| 135 external host connected to the GSM device via the RVTMUX channel to perform | |
| 136 arbitrary read and write operations on the device file system. | |
| 137 | |
| 138 In the segment of history of interest to us TI has produced two different | |
| 139 and entirely incompatible versions of the TMFFS protocol: TMFFS1 and TMFFS2. | |
| 140 Or rather, what is now called TMFFS1 was originally just TMFFS, and then came | |
| 141 TMFFS2. TMFFS2 works only through ETM, whereas TMFFS1 predates ETM: in the | |
| 142 original implementation the tm_ffs() function in the FFS code was called from | |
| 143 L1TM code. | |
| 144 | |
| 145 Our copy of TCS211 reference fw includes the source for both TMFFS1 and TMFFS2; | |
| 146 it is theoretically possible to build a firmware image that includes both TMFFS | |
| 147 versions (they won't conflict because they respond to different command | |
| 148 packets), but it is pretty clear that TI never intended to have both enabled | |
| 149 at the same time. Our copy of TCS211 came with TMFFS1 enabled and we didn't | |
| 150 change it when we made the moko12 (leo2moko-r1) fw release for the Openmoko | |
| 151 community (the previous proprietary mokoN firmwares also implement TMFFS1), | |
| 152 but we have subsequently switched to TMFFS2 for our current TCS211-based work. | |
| 153 | |
| 154 Pirelli's fw implements TMFFS2: we don't have any source for this fw, but our | |
| 155 FreeCalypso host utilities written to talk the TMFFS2 protocol based on our | |
| 156 available TCS211 source work beautifully when run against Pirelli's fw. | |
| 157 | |
| 158 Use in FreeCalypso | |
| 159 ================== | |
| 160 | |
| 161 The FreeCalypso project has adopted the same general firmware architecture as | |
| 162 that exhibited by TI's standard firmwares from the Moko/Pirelli time frame. We | |
| 163 use TI's RiViera framework lifted directly out of the TCS211 reference fw, and | |
| 164 that includes the RVT module and the RVTMUX interface it presents. Our GSM fw | |
| 165 emits the same 3 kinds of debug traces (RV, L1 and GPF) as the pre-existing | |
| 166 firmwares with which we are seeking functional parity, and for Test Mode | |
| 167 functionality we have the option of including ETM, TMFFS1 and/or TMFFS2 in our | |
| 168 firmware builds. (Both TMFFS versions require ETM in our implementation, and | |
| 169 it is possible to build a firmware image with both included.) | |
| 170 | |
| 171 We have adopted ETM and TMFFS2 as the standard combination for FreeCalypso, | |
| 172 i.e., ETM_CORE for memory and ABB register reads and writes and TMFFS2 for | |
| 173 external FFS access. We needed to develop our own host tools for operating on | |
| 174 GSM device FFS via one of the two TMFFS protocols, and after studying the fw | |
| 175 source implementing both, I (Space Falcon) came to the conclusion that TMFFS2 | |
| 176 is both more capable and more reliable; my guess is that TMFFS1 was likely kept | |
| 177 around only because some of TI's crappy Weendoze host software depended on it. | |
| 178 (See gsm-fw/services/ffs/tmffs.c if you would like to judge for yourself.) | |
| 179 | |
| 180 We have the following host tools for communicating with TI-based GSM firmwares | |
| 181 (both our own and some of the existing proprietary ones): | |
| 182 | |
| 183 rvtdump This tool produces a human-readable dump of all output emitted | |
| 184 by a TI-based GSM fw in the form of RVTMUX binary packets. It | |
| 185 can also log this dump to a file. | |
| 186 | |
| 187 rvinterf This tool is a superset of rvtdump: it not only dumps and/or | |
| 188 logs all output from the GSM fw, but also provides a mechanism | |
| 189 for sending command packets to it. | |
| 190 | |
| 191 Rvinterf is the engine behind the following host tools that send Test Mode | |
| 192 commands to a target: | |
| 193 | |
| 194 fc-tmsh This is our basic tool for sending Test Mode commands to a | |
| 195 running GSM fw. It is strictly asynchronous in that commands | |
| 196 entered by the operator get sent to the target, and any response | |
| 197 packets received from the target are displayed as they come in. | |
| 198 The tool has no knowledge of any correspondence between commands | |
| 199 being sent and whatever responses they should elicit, i.e., it | |
| 200 is perfectly suited for experimental discovery of firmware | |
| 201 behaviour in response to Test Mode commands. | |
| 202 | |
| 203 This tool was written before we realized that there was/is an | |
| 204 older, more basic Test Mode predating ETM, hence in many place | |
| 205 we say "ETM" when we really should have said "TM". Oh well... | |
| 206 | |
| 207 fc-fsio This tool speaks the TMFFS2 protocol and allows a user or | |
| 208 developer to perform a wide range of operations on the file | |
| 209 system of a GSM device. It operates synchronously, i.e., it | |
| 210 sends ETM/TMFFS2 commands and expects responses in strict | |
| 211 lock-step; a single user command may translate into a large | |
| 212 number of ETM/TMFFS2 command packet exchanges. | |
| 213 | |
| 214 AT commands over RVTMUX | |
| 215 ======================= | |
| 216 | |
| 217 There is one more use to which we put the RVTMUX debug serial interface that is | |
| 218 an original FreeCalypso invention: communicating with the AT command interpreter | |
| 219 (ATI). TI's original architecture assumes that if a product is to offer a | |
| 220 standard AT command interface (the product is either a GSM/GPRS modem for which | |
| 221 this AT command interface is the sole mode of usage or a feature phone that | |
| 222 offers a data port as one of its features), then it will be presented on a | |
| 223 dedicated UART separate from RVTMUX. | |
| 224 | |
| 225 However, many of our target devices have only one UART practically accessible, | |
| 226 and even when we use Openmoko's modem as our development platform, the RVTMUX | |
| 227 interface is more convenient because it connects externally, whereas the MODEM | |
| 228 UART is connected to the application processor of the smartphone. Therefore, | |
| 229 we developed a way to pass AT commands over RVTMUX. We created a new RVTMUX | |
| 230 channel for this interface and assigned it RVT packet type 0x1A. Packets sent | |
| 231 from an external host to the GSM device carry AT commands and SMS string input, | |
| 232 whereas packets flowing the other way carry ATI's responses to commands and | |
| 233 asynchronous notifications such as incoming calls. | |
| 234 | |
| 235 The host utility for talking AT commands to a FreeCalypso GSM device via RVTMUX | |
| 236 is fc-shell; it works via rvinterf just like fc-fsio and fc-tmsh. |
