Changelog
- January 2009: Initial version
- 17 July 2009:
- eSATA port is also an USB port, as reported by Lars Wilhelmsen.
- E-port is a perfect companion for the E4300: some notes about it.
- 18 July 2009:
- Flashing the BIOS of your E4300 on Debian
- Suspend to RAM works fine
- 22 July 2009:
- Broadcom USH firmware upgrade
- 17 August 2009:
- Broadcom USH now working w/ Cryptoflex E-Gate 32K
- 28 October 2010:
- Updated URL for Dell BIOS images (Thanks to Jesper Henriksen)
Overview
This page contains some notes on configuring/supporting the various hardware elements available in the Dell Latitude E4300. Each section is dedicated to a specific piece of hardware (smartcard reader, bluetooth module, ...) or a specific Linux feature (suspend to RAM, X server, ...)
The distribution considered here is Debian (32 bit). I expect the setup to be fairly similar if you use Ubuntu but you'll have to investigate required changes when needed. If you use another Linux distribution (redhat?), there is no reason you will not be able to do the same things, sometimes in a fairly different way but you might still find this notes useful.
Because you might have selected a different hardware configuration for your E4300, here is the output of the lspci and lsusb under a 2.6.28 kernel on mine:
arno@small:~$ lspci
00:00.0 Host bridge: Intel Corporation Mobile 4 Series Chipset Memory Controller Hub
00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated
Graphics Controller
00:02.1 Display controller: Intel Corporation Mobile 4 Series Chipset Integrated
Graphics Controller (rev 07)
00:19.0 Ethernet controller: Intel Corporation 82567LM Gigabit Network Connection
00:1a.0 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #4
00:1a.1 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #5
00:1a.2 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #6
00:1a.7 USB Controller: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #2
00:1b.0 Audio device: Intel Corporation 82801I (ICH9 Family) HD Audio Controller
00:1c.0 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 1 (rev 03)
00:1c.1 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 2 (rev 03)
00:1c.3 PCI bridge: Intel Corporation 82801I (ICH9 Family) PCI Express Port 4 (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #1
00:1d.1 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #2
00:1d.2 USB Controller: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #3
00:1d.7 USB Controller: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #1
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev 93)
00:1f.0 ISA bridge: Intel Corporation ICH9M-E LPC Interface Controller
00:1f.2 RAID bus controller: Intel Corporation Mobile 82801 SATA RAID Controller
00:1f.3 SMBus: Intel Corporation 82801I (ICH9 Family) SMBus Controller (rev 03)
02:01.0 FireWire (IEEE 1394): Ricoh Co Ltd R5C832 IEEE 1394 Controller (rev 05)
02:01.1 SD Host controller: Ricoh Co Ltd R5C822 SD/SDIO/MMC/MS/MSPro Host Adapter
02:01.2 System peripheral: Ricoh Co Ltd R5C843 MMC Host Controller (rev ff)
0c:00.0 Network controller: Intel Corporation PRO/Wireless 5300 AGN [Shiloh]
Network Connection
arno@small:~$ lsusb
Bus 008 Device 002: ID 413c:8147 Dell Computer Corp.
Bus 008 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 007 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 002: ID 0a5c:5801 Broadcom Corp.
Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 002 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 001 Device 004: ID 413c:8162 Dell Computer Corp.
Bus 001 Device 003: ID 413c:8161 Dell Computer Corp.
Bus 001 Device 002: ID 0a5c:4500 Broadcom Corp.
Bus 001 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Some additional notes:
- I did not select the camera (the option was not compatible with the HSDPA card, i.e. the WWAN option), so you will not find any info on it on this page.
- I opted for a 250 GB 5400 RPM SATA disk, not a SSD one. I might change my mind at some point when bigger models (more than 200 GB) will be available at reasonable prices.
- I ordered the Advanced E-Port station (for the 2 DVI outputs and additional USB plugs)
- I selected the backlight keyboard option (nice and useful)
- I did not ordered a DVD drive (additional weight for a useless component)
Don't hesitate to send me a mail if you have questions or corrections, or if you want me to extend a specific section with your inputs.
Keep in mind this is a work in progress so be gentle.
Smartcard Reader
The E4300 comes with a smartcard reader. A card can be inserted on the left side of the laptop located at the same level as the touchpad. The reader is a Broadcom BCM5880, internally connected to the system via USB (Note that this chip is also the one supporting the UPEK fingerprint reader).
Before reading further, if you really intend to use the smartcard reader of your E4300, you should definitely consider reading next section and spend the time needed to update the firmware of the USH to the latest available version. Old version have serious bug which prevent the use of the reader with at some cards (e.g. Cryptoflex E-Gate 32K in my specific case).
Once you have done that, come back here for more fun (additional patches required to have the beast work).
Below is an lsusb output for the device:
Bus 003 Device 002: ID 0a5c:5801 Broadcom Corp.
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 1.10
bDeviceClass 0 (Defined at Interface level)
bDeviceSubClass 0
bDeviceProtocol 0
bMaxPacketSize0 64
idVendor 0x0a5c Broadcom Corp.
idProduct 0x5801
bcdDevice 1.01
iManufacturer 1 Broadcom Corp
iProduct 2 5880
iSerial 3 0123456789ABCD
bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 139
bNumInterfaces 2
bConfigurationValue 0
iConfiguration 0
bmAttributes 0xc0
Self Powered
MaxPower 100mA
...
< snip: removed Broadcom USH w/swipe sensor Interface Descriptor <
...
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 1
bAlternateSetting 0
bNumEndpoints 3
bInterfaceClass 11 Chip/SmartCard
bInterfaceSubClass 0
bInterfaceProtocol 0
iInterface 5 Contacted SmartCard
ChipCard Interface Descriptor:
bLength 54
bDescriptorType 33
bcdCCID 1.00
nMaxSlotIndex 0
bVoltageSupport 7 5.0V 3.0V 1.8V
dwProtocols 3 T=0 T=1
dwDefaultClock 4000
dwMaxiumumClock 4000
bNumClockSupported 0
dwDataRate 9600 bps
dwMaxDataRate 250000 bps
bNumDataRatesSupp. 0
dwMaxIFSD 247
dwSyncProtocols 00000000
dwMechanical 00000000
dwFeatures 000102BA
Auto configuration based on ATR
Auto voltage selection
Auto clock change
Auto baud rate change
Auto PPS made by CCID
NAD value other than 0x00 accpeted
TPDU level exchange
dwMaxCCIDMsgLen 271
bClassGetResponse 00
bClassEnvelope 00
wlcdLayout none
bPINSupport 0
bMaxCCIDBusySlots 1
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x82 EP 2 IN
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0040 1x 64 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x02 EP 2 OUT
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0040 1x 64 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x86 EP 6 IN
bmAttributes 3
Transfer Type Interrupt
Synch Type None
Usage Type Data
wMaxPacketSize 0x0010 1x 16 bytes
bInterval 1
Device Status: 0x0000
(Bus Powered)
As you can read above, the reader is CCID compliant. It supports both ISO/IEC 7816 T=0 and T=1 protocols. I use it in combination with a Cryptoflex E-Gate 32K Card.
At the time of writing, after having installed opensc and openct packages (and their dependencies), some small configuration steps are still needed to be able to interact with the card. The broadcom reader being quite recent, it is not referenced in openct's udev script (/etc/udev/rules.d/z60_openct.rules) and a line must be added manually:
...
SUBSYSTEM=="usb", ACTION=="add", ENV{DEVTYPE}=="usb_device", \
NAME="bus/usb/$env{BUSNUM}/$env{DEVNUM}", MODE="0644"
# Broadcom 5880 <= ADDED
ENV{MODALIAS}=="usb:v0A5Cp5801*", RUN+="/lib/udev/openct_usb" <= ADDED
# egate
ENV{MODALIAS}=="usb:v0973p0001*", RUN+="/lib/udev/openct_usb"
...
This line will have udev call openct-control to attach the USB device and make the reader available (this is mostly what /lib/udev/openct_usb is used for).
On openct side, some configuration is also required to make it aware that this specific USB reader is supported via one of the internal reader driver, more specifically the ccid driver. This is done by adding the information in /etc/openct.conf file, at the end of the ccid driver ids list:
driver ccid {
ids = {
usb:03f0/1024, # HP Keyboard with CCID reader
usb:046a/0010, # Cherry smartboard G83-6744
usb:04e6/5115,
usb:04e6/5116,
usb:04e6/5117, # SCM Micro token size reader
usb:04e6/511d, # SCM Micro SCR3311
usb:04e6/E001,
usb:04e6/E003,
usb:073d/0c00, # Eutron SimPocket (doesn't work yet)
usb:076b/1021, # OmniKey CardMan 1021
usb:076b/3021,
usb:076b/5121,
usb:076b/6622, # OmniKey CardMan 6121
usb:0783/0003,
usb:08e6/3437, # Gemplus
usb:08e6/3438, # Gemplus GemPC Key SmartCard Reader
usb:08e6/4433, # Gemplus
usb:0b97/7762, # O2 Micro, Inc. Oz776 SmartCard Reader
usb:0b97/7772, # O2 Micro, Inc. Oz776 SmartCard Reader
usb:0bf8/1006, # fujitsu siemens 3.5 drive size reader
usb:0a5c/5801, # Broadcom 5880 Smartcard/UPEK Reader <= ADDED
};
};
After you have restarted openct (and possibly udev), you should be able to see your card (additional installation and configuration steps may be required if you use a different card, not directly supported by opensc):
arno@small:~$ openct-tool atr Detected CCID Compatible Card present, status changed ATR: 3b 95 18 40 ff 62 01 02 01 04
But if you try and list the elements on the card using pkcs15-tool (or do other operations using pkcs15-* utilities), you get an error:
arno@small:~$ pkcs15-tool -D PKCS#15 initialization failed: Wrong length
After some debugging at opensc level, it seems that the reader returns (via openct) some APDU with leading NULL bytes (more precisely, 252 of those NULL bytes). I wrote a small temporary fix (below) for opensc package but the problem should probably be corrected at a lower level level.
Index: opensc-0.11.4/src/libopensc/apdu.c
===================================================================
--- opensc-0.11.4.orig/src/libopensc/apdu.c 2008-12-26 09:37:55.000000000 +0100
+++ opensc-0.11.4/src/libopensc/apdu.c 2008-12-26 09:41:36.000000000 +0100
@@ -455,6 +455,25 @@
return SC_ERROR_NOT_SUPPORTED;
}
+#define BROADCOM_E4300_BUG_APDU_DROP 252
+ if (apdu->resplen >= BROADCOM_E4300_BUG_APDU_DROP) {
+ int i, allzero=1;
+ u8 *tmp = apdu->resp;
+
+ for (i=0; i < BROADCOM_E4300_BUG_APDU_DROP; i++) {
+ if (tmp[i] != 0) {
+ allzero = 0;
+ break;
+ }
+ }
+
+ if (allzero) {
+ memmove(tmp, tmp + BROADCOM_E4300_BUG_APDU_DROP,
+ apdu->resplen - BROADCOM_E4300_BUG_APDU_DROP);
+ apdu->resplen -= BROADCOM_E4300_BUG_APDU_DROP;
+ }
+ }
+
/* if the command already returned some data
* append the new data to the end of the buffer
*/
For those who are familiar with cryptoflex cards and previously used them via their native USB interface on previous version of Dell laptops (once plugged, the card was reported as a new smarcard reader), the Broadcom hardware does not seem to support that mode. The card is only accessible via the common CCID interface of the broadcom reader.
Being familiar with the Cryptoflex E-Gate, I know it is quite fast. Used in combination with a Dell SmartCard Reader Keyboard (USB ID 413c:2101) for authentication with login, sudo, screensaver unlocking ... the delay associated with the transaction has never been an issue. But when used in combination with the SmartCard Reader of the BCM5880, the delay can make it impossible to use for common tasks using only the small previous patch.
Simply put, for things to work smoothly, you need:
- at least version 11 of the USH firmware
- previous patch for opensc
- a patched version of openct
Below are some very simple statistics presenting the delay when using the Cryptoflex E-Gate with a Dell SmartCard Reader Keyboard ...
$ sudo -k # $ time sudo date Please insert your smart card or enter your username. Smart card inserted. Welcome OpenSC Card (Arno)! Smart card password: Sat Jul 25 22:49:29 CEST 2009 real 0m3.999s user 0m0.080s sys 0m0.020s
... or the BCM5880 (w/o version 11 of the USH firmware and w/o openct patch):
$ sudo -k $ time sudo date Please insert your smart card or enter your username. Smart card inserted. Welcome OpenSC Card (Arno)! Smart card password: Sat Jul 25 13:03:29 CEST 2009 real 0m14.108s user 0m0.044s sys 0m0.020s
Being presented with the password prompt for the PIN (smartcard's discovery and reset process), typing the smartcard's pin and waiting for the authentication process via the smartcard to occur before the 'date' command is executed take 14 seconds with the BCM5880. With the Dell SmartCard Reader Keyboard, the process takes only 4 seconds.
More precisely, with the BCM5880, the password prompt for the PIN is presented only after 8 seconds or so (1 second or so with the SmartCard Reader Keyboard). Then, after the PIN has been typed, the date is presented only after 5 seconds ot so (2 seconds or so with the SmartCard Reader Keyboard).
Considering the cryptographic processing occuring on the device is expected to be the one causing the delay, I started investigating why such a difference between both readers. A quick look at the lsusb output between the two devices provides the following major differences:
BCM5880: Dell SmartCard Keyboard:
bcdUSB 1.10 bcdUSB 2.00
bMaxPacketSize0 64 bMaxPacketSize0 8
idVendor 0x0a5c Broadcom Corp. idVendor 0x413c Dell Computer Corp.
idProduct 0x5801 idProduct 0x2101 SmartCard Reader Keyboard
bcdDevice 1.01 bcdDevice 1.00
iManufacturer 1 Broadcom Corp iManufacturer 1 Dell
iProduct 2 5880 iProduct 2 Dell Smart Card Reader Keyboard
iSerial 3 0123456789ABCD iSerial 0
iInterface 5 Contacted iInterface 0
SmartCard
dwDataRate 9600 bps dwDataRate 10752 bps
dwMaxDataRate 250000 bps dwMaxDataRate 344086 bps
dwMaxIFSD 247 dwMaxIFSD 254
dwFeatures 000102BA dwFeatures 00010230
Auto configuration based on ATR
Auto voltage selection
Auto clock change Auto clock change
Auto baud rate change Auto baud rate change
Auto PPS made by CCID
NAD value other than 0x00 accpeted NAD value other than 0x00 accpeted
TPDU level exchange TPDU level exchange
bPINSupport 0 bPINSupport 1 verification
bEndpointAddress 0x86 EP 6 IN bEndpointAddress 0x83 EP 3 IN
bmAttributes 3 bmAttributes 3
Transfer Type Interrupt Transfer Type Interrupt
Synch Type None Synch Type None
Usage Type Data Usage Type Data
wMaxPacketSize 0x0010 1x 16 bytes wMaxPacketSize 0x0008 1x 8 bytes
bInterval 1 bInterval 24
So, the main differences between those USB CCID devices are:
- The Dell SmartCard Reader Keyboard is a USB 2.0 device while the BCM5880 is "only" USB1.1 compliant. This may imply less bandwidth for the BCM 5880 but considering the amount of data exchanged, this should not be significant.
- the default ICC I/O data rate (in bps) of both readers differ are basically the same (10752 vs 9600 bps) but the maximum data rate for the BCM5880 is expected to be basically 38% lower than the one of the Dell SmartCard Reader Keyboard.
- The Dell SmartCard Reader Keyboard advertises basic features while the BCM5880 advertises additional ones.
Considering there was nothing to do for the two first points, I decided to take a look at the initialization steps performed by openct for the additional features of the BCM5880 (Debian package source, version 0.6.16, src/ifd/ifd-ccid.c mainly). After some time modifying the code to make BCM5880 dumber (i.e. make it advertise less features), I noticed that without passing via the code associated supporting the "Auto configuration based on ATR" made the initial latency disappear, allowing to get the following statistic:
$ sudo -k $ time sudo date Please insert your smart card or enter your username. Smart card inserted. Welcome OpenSC Card (Arno)! Smart card password: Sat Jul 25 22:49:53 CEST 2009 real 0m6.056s user 0m0.064s sys 0m0.036s
More precisely, it now takes only around one second for the paswword prompt for the PIN to be presented, just like for the Dell SmartCard Reader Keyboard. Nonetheless, the rest of the process is still taking 2 additional seconds.
The patch is here:
Index: openct-0.6.16/src/ifd/ifd-ccid.c
===================================================================
--- openct-0.6.16.orig/src/ifd/ifd-ccid.c 2009-08-03 08:29:50.000000000 +0200
+++ openct-0.6.16/src/ifd/ifd-ccid.c 2009-08-03 08:34:47.000000000 +0200
@@ -742,7 +742,9 @@
} else if (ccid.dwFeatures & 0x60000) {
st->reader_type = TYPE_APDU;
}
- if (ccid.dwFeatures & 0x2)
+ /* AUTO_ATRPARSE seems broken on BCM 5880 aka Broadcom USH */
+ if ((ccid.dwFeatures & 0x2) &&
+ !(de.idVendor == 0x0a5c && de.idProduct == 0x5801))
st->flags |= FLAG_AUTO_ATRPARSE;
if (ccid.dwFeatures & 0x4)
st->flags |= FLAG_AUTO_ACTIVATE;
As a matter of fact, the 2 seconds delay is due to some bugs in older version of the USH firmware. Upgrading the USH firmware as described in following section makes the additional 2 seconds latency disappear completely.
Updating Broadcom USH firmware
In order to upgrade BCM5880 firmware, I first took a look at Broadcom's web site to see if there was something available. There is only a single single page about the chip and no associated support/download for that low-level product.
So I decided to go for a search on Dell's website. There, looking for Broadcom Unified Securiy Hub, USH, or BCM5880 did not helped: the chip is part of something bigger called "Dell ControlPoint Security Manager". Using that designation and the Service Tag of the device (8L5024J) I ended up on that page which seemed to match my search. The package (auto-extractible exe) contains all the drivers for the various hardware parts that the ControPoint Security Manager can use on existing DELL devices:
- Broadcom Trusted Platform Module
- Atmel Trusted Platform Module
- STMicroelectronics Trusted Platform Module
- Winbond Trusted Platform Module
- O2Micro Smart Card Reader
- Dell Smart Card Keyboard
- UPEK TouchChip Fingerprint Reader
- Authentec Fingerprint Reader
- Broadcom Unified Security Hub
- Vista Storage Driver Update (for ATAPI.SYS)
This probably explains the size of the package: 72MB. The .exe is of no use under Linux (well, in fact, a patched version of unshield, and additional tools will allow you to extract the various layers contained in the archive but this is not reliable enough to be discussed in more length here). So, after spending more time on Google, I ended up via external links on Dell's FTP server "Security" repository: http://ftp.us.dell.com/Security/
There, I downloaded Broadcom_Unified-Security-Hu_A07_R210234.exe, which wine managed to handle correctly. After extraction:
$ ls -l total 20 drwxr-xr-x 2 arno arno 4096 2009-07-20 10:13 DOS drwxr-xr-x 2 arno arno 4096 2009-07-20 10:13 firmware drwxr-xr-x 2 arno arno 4096 2009-07-20 10:13 ODM -rw-r--r-- 1 arno arno 31 2008-11-11 13:58 UPDATE.BAT -rw-r--r-- 1 arno arno 781 2009-02-06 09:53 Version.txt
Update: don't bother downloading that version. As pointer to me by Klaus Meyners, there is a more recent version of broadcom USH firmware (version A11) is available elsewhere on Dell's servers, more precisely here. It fixes the latency associated with the reader (the problem with the automatic ATR parsing feature announced by the reader is still there and requires openct fix described above).
So I rebooted on an USB key running a DOS system and performed the upgrade.
In order for the firmware upgrade to start and complete, I had to deactivate the TPM in the BIOS (update utility tells you that) and switch off the wifi (update utility does not tell you that).
In the end, with this version 11, the opensc patch and the openct patch, the Cryptoflex E-gate works perfectly with the smartcard reader of the E4300, providing the exact same latency as the Dell SmartCard keyboard.
Sound
Even when the volume is set at the highest possible level, the sound is still quite low.
Disk
Not much to say:
arno@small:~$ sudo hdparm -tT /dev/sda /dev/sda: Timing cached reads: 11264 MB in 1.99 seconds = 5652.75 MB/sec Timing buffered disk reads: 164 MB in 3.04 seconds = 54.03 MB/sec
RAM
Considering RAM is quite cheap and required when using virtual machines, I ordered the laptop with 4Go (4096MB (2x2048) 800MHz DDR3 Dual Channel).
RAID Controller
What's the expected purpose of that builtin RAID controller? to be used in conjunction with an external esata disk?
00:1f.2 RAID bus controller: Intel Corporation Mobile 82801 SATA RAID Controller (rev 03)
Battery
The 6-cell battery extends the size of the laptop of about 1 inch. For that reason, I selected a primary 3-cell battery ("Primary 3-cell 30W/HR LI-ION", lasts less than 2 hours) and additional slice batteries ("Additional Slice Battery 48W/HR LI-ION")
Touchpad
Keyboard
As stated before, I asked for the backlight version, which is very useful in the dark. The backlight is control via hardware keys, i.e. no specific binding configuration is required on the system (I speak here as an old Linux PowerBook user).
Wifi
I selected an Intel 5300 AGN Card, which is supported by recent kernels (support was introduced in 2.6.26 via iwalgn if I am not mistaken).
The only required additional step (after having activated the associated option in your kernel configuration) is make the firmware available so that it can be loaded when the device is detected. At the time of writing, you have to do that manually on Debian.
Just go to "Intel Wireless Wifi Link Linux drivers for Linux" download page to grab the correct microcode tarball (the 5000 images, in our specific case). Just untar the archive somewhere and copy the iwlwifi-5000-1.ucode file to /lib/firmware/ where the kernel will look for it
Note that you have additional configuration options in the kernel configuration file to embed such firmwares in your kernel image, but I do not discuss this kind of configuration here.
Ethernet
The ethernet controller on the E4300 is an Intel 82567LM Gigabit one. It is supported by the e1000e driver.
Bluetooth
Here is what you get by default:
arno@small:~$ hcitool dev Devices:
Now, just consider the following:
arno@small~$ sudo lsusb -t
...
Bus# 1
`-Dev# 1 Vendor 0x1d6b Product 0x0001
`-Dev# 10 Vendor 0x0a5c Product 0x4500
|-Dev# 11 Vendor 0x413c Product 0x8161
`-Dev# 12 Vendor 0x413c Product 0x8162
...
The broadcom device 0a5c:4500 is a BCM 2046, which is our missing bluetooth EDR module, or at least its tail. It is seen as a hub, with 2 attached devices: 413c:8161 and 413c:8162 (usbview provides a better tree view). lsusb may let you think that those devices are respectively a keyboard and a mouse. Those are in fact some kinds of fake devices from the bluetooth adapter which can be used at boot time without the any prior OS knowledge. If you are interested, there are explanations here. You should definitely take the time and read it.
The trick to make it appear is to use hid2hci to "switch the HID proxy Bluetooth dongle into HCI mode". The thing is that latest current Debian version of bluez-utils (3.36-2) does not have information about 413c:8162 (this is the one that should be switched). Latest version of bluez-utils (4.25) has it. Patching the tool with the following does the trick:
--- hid2hci.c.orig 2009-01-05 23:13:07.000000000 +0100
+++ hid2hci.c 2008-10-26 07:40:34.000000000 +0100
@@ -264,6 +264,7 @@
{ HCI, 0x046d, 0xc71c, switch_logitech }, /* Logitech diNovo Edge */
{ HCI, 0x413c, 0x8154, switch_dell }, /* Dell Wireless 410 */
{ HCI, 0x413c, 0x8158, switch_dell }, /* Dell Wireless 370 */
+ { HCI, 0x413c, 0x8162, switch_dell }, /* Dell Wireless 365 */
{ -1 }
};
After recompiling the package/tool, switching the mode can be done the
following way:
arno@small:/tmp/bluez-utils-3.36/tools$ sudo ./hid2hci Switching device 413c:8162 to HCI mode was successful
After that, an additional devices shows up (413c:8160):
arno@small:~$ sudo lsusb -t
...
Bus# 1
`-Dev# 1 Vendor 0x1d6b Product 0x0001
`-Dev# 10 Vendor 0x0a5c Product 0x4500
|-Dev# 11 Vendor 0x413c Product 0x8161
|-Dev# 12 Vendor 0x413c Product 0x8162
`-Dev# 13 Vendor 0x413c Product 0x8160
...
which is recognized as our missing Dell Wireless 365 Bluetooth Module:
arno@small:~$ sudo lsusb -v -d 413c:8160
Bus 001 Device 013: ID 413c:8160 Dell Computer Corp.
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 2.00
bDeviceClass 224 Wireless
bDeviceSubClass 1 Radio Frequency
bDeviceProtocol 1 Bluetooth
bMaxPacketSize0 64
idVendor 0x413c Dell Computer Corp.
idProduct 0x8160
bcdDevice 1.73
iManufacturer 1 Dell Computer Corp
iProduct 2 Dell Wireless 365 Bluetooth Module
iSerial 0
bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 216
bNumInterfaces 4
bConfigurationValue 1
iConfiguration 0
bmAttributes 0xe0
Self Powered
Remote Wakeup
MaxPower 100mA
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 0
bAlternateSetting 0
bNumEndpoints 3
bInterfaceClass 224 Wireless
bInterfaceSubClass 1 Radio Frequency
bInterfaceProtocol 1 Bluetooth
...
And finally:
arno@small:~$ sudo hcitool dev
Devices:
hci0 00:22:69:C6:F2:EE
Let's reset the device to be able to use it and do a simple scan:
arno@small:~$ sudo hciconfig hci0 reset
arno@small:~$ sudo hcitool scan
Scanning ...
00:0E:07:B7:0D:93 MyDummyPhone
If you want things to happen automatically (the hid2hci call at startup and the device reset) you need two additional changes:
- Set HID2HCI_ENABLED=1 in /etc/default/bluetooth to have hid2hci called at startup.
- Have the newly added device reseted automatically: this will be the default behavior in next stable kernel (2.6.29) but current version (2.6.28) and previous ones need a fix for the HCI reset quirk. The simple patch below does the job (it is against 2.6.28).
Index: linux-2.6.28/drivers/bluetooth/btusb.c
===================================================================
--- linux-2.6.28.orig/drivers/bluetooth/btusb.c 2009-01-18 01:14:41.000000000 +0100
+++ linux-2.6.28/drivers/bluetooth/btusb.c 2009-01-18 01:16:00.000000000 +0100
@@ -105,6 +105,9 @@
{ USB_DEVICE(0x0a5c, 0x2146), .driver_info = BTUSB_RESET },
{ USB_DEVICE(0x0a5c, 0x2151), .driver_info = BTUSB_RESET },
+ /* Dell Wireless 365 */
+ { USB_DEVICE(0x413c, 0x8160), .driver_info = BTUSB_RESET },
+
/* Apple MacBook Pro with Broadcom chip */
{ USB_DEVICE(0x05ac, 0x820f), .driver_info = BTUSB_RESET },
Index: linux-2.6.28/drivers/bluetooth/hci_usb.c
===================================================================
--- linux-2.6.28.orig/drivers/bluetooth/hci_usb.c 2009-01-18 01:14:57.000000000 +0100
+++ linux-2.6.28/drivers/bluetooth/hci_usb.c 2009-01-18 01:16:21.000000000 +0100
@@ -141,6 +141,9 @@
/* Broadcom 2046 */
{ USB_DEVICE(0x0a5c, 0x2151), .driver_info = HCI_RESET },
+ /* Dell Wireless 365 */
+ { USB_DEVICE(0x413c, 0x8160), .driver_info = HCI_RESET },
+
/* Microsoft Wireless Transceiver for Bluetooth 2.0 */
{ USB_DEVICE(0x045e, 0x009c), .driver_info = HCI_RESET },
HSDPA card
The Dell Wireless 5530 HSDPA card is basically a rebranded Ericsson product (Ericsson Mobile Broadband Module F3507g). It provides both HSDPA modem (7.2Mbps DL / 2.0Mbps UL) and GPS functionality. Under Linux, both functions are controlled and used via serial interfaces (TTYs). In practice, the 3 serial interfaces are resepectively used to:
- Control the device: start some functions, get info, activates some modes. A common set of AT commands is used for that purpose for both the GPS and the HSDPA part.
- Exchange data traffic with the network: the consumer for that interface is usually pppd (if you use pppd and not the mbm solution described below).
- Receive the GPS NMEA message flow: the consumer for that interface is gpsd.
In that section, I only cover the initial setup of the device (valid kernel options) and the configuration of the HSDPA part. A specific section is dedicated to the GPS functionality.
You may already have noticed from a lsusb output that the Dell Wireless 5530 advertises a lot of Interface Descriptors for various functions. From a kernel standpoint, the good way to support the device is by using the cdc_acm and cdc_wdm (well, at the moment, I don't know yet how to use the /dev/cdc-wdm0 and /dev/cdc-wdm1 devices created when the latter is loaded):
- cdc_acm: this driver will create the /dev/ttyACM0,
/dev/ttyACM1 and /dev/ttyACM2 TTYs. Their use is explained later in
this section. Here is the location of the kernel option:
Symbol: USB_ACM [=m] Prompt: USB Modem (CDC ACM) support Defined at drivers/usb/class/Kconfig:7 Depends on: USB_SUPPORT && USB Location: -> Device Drivers -> USB support (USB_SUPPORT [=y]) - cdc_wdm: this driver will create the /dev/cdc-wdm0 and
/dev/cdc-wdm1 devices:
Symbol: USB_WDM [=m] Prompt: USB Wireless Device Management support Defined at drivers/usb/class/Kconfig:32 Depends on: USB_SUPPORT && USB Location: -> Device Drivers -> USB support (USB_SUPPORT [=y])
Note that if you also have selected the option driver below, it may be loaded before the cdc_acm and cdc_wdm drivers and will create /dev/ttyUSB0 to /dev/ttyUSB10. Basically, /dev/ttyUSB2, /dev/ttyUSB4 and /dev/ttyUSB10 will operate in the same way as /dev/ttyACM0, 1 and 2. Considering the ongoing merge request of the MBM driver by Ericsson, cdc_acm and cdc_wdm are the right drivers to use.
Symbol: USB_SERIAL_OPTION [=m]
Prompt: USB driver for GSM and CDMA modems
Defined at drivers/usb/serial/Kconfig:542
Depends on: USB_SUPPORT && USB_SERIAL
Location:
-> Device Drivers
-> USB support (USB_SUPPORT [=y])
-> USB Serial Converter support (USB_SERIAL [=m])
Simply put, the HSDPA card can be fully controlled using AT commands passed via a TTY interface (I use /dev/ttyACM0 for that purpose). Then, /dev/ttyACM1 can be used by pppd to exchange data with the network. The third TTY (/dev/ttyACM2) may be used, as described in the section dedicated to the GPS, to receive NMEA message flow.
To sum up, you can use:
- /dev/ttyACM0: as the control interface for both the modem and the GPS
- /dev/ttyACM1: as the serial line for ppp connection (if you use pppd)
- /dev/ttyACM2: as the interface on which NMEA messages flow are received
The modem supports an extensive list of AT commands (AT* can be used to get that list): many of them are officialy documented (i.e. by Sony/Ericsson). Many others are not. I have created a page dedicated to the AT commands supported by the Dell Wireless 5530, their interest an how to use them.
Typically, common interactions with the modem require knowledge of a very limited set of AT commands:
- AT+CFUN: used to set the phone functionality, i.e. activate/deactivate the GSM/WCDMA radio
- AT+CPIN: enter the PIN to unlock the SIM card
- AT+CSQ: get the quality of the signal
Now, there is a better alternative than pppd. But it requires some kernel support which is not yet available upstream (at least not in 2.6.28). But the associated patch is pretty small (less than 500 lines) and has been written by developers from Ericsson and submitted for review on netdev.
Anyway, the idea is to provide an Ethernet emulation for the device, i.e. get an ethernet interface on the system that can be used as usual (i.e. with dhclient). The information provided below on how to set things up are based~:
- A post on netdev by Carl Nordbeck for the submission of the patch providing "Ericsson Mobile Broadband Module" option (i.e. USB_NET_MBM)
- Some additional information on AT commands behaviors provided by Bjørn Mork and Carl Nordbeck (Thanks!)
- A post by Per Hallsmark on Network Manager mailing list
- the patch itself
First, you have to get the patch and apply it to your kernel tree (AFAICT, it applies fine on 2.6.27, 2.6.28 and 2.6.29-rc3). Then, you need to select the newly created option:
Symbol: USB_NET_MBM [=m]
Prompt: Ericsson Mobile Broadband Module
Defined at drivers/net/usb/Kconfig:348
Depends on: NETDEVICES && USB && NET && USB_USBNET
Location:
-> Device Drivers
-> Network device support (NETDEVICES [=y])
-> USB Network Adapters
Selects: USB_NET_CDCETHER
Note that it automatically selects USB_NET_CDCETHER. Also note that to you still need to select the USB_ACM and USB_WDM discussed previously to get the TTY lines to control the modem via AT commands (and be able to use the GPS).
When loading the mbm module, you should see sth like the following in dmesg output:
$ sudo dmesg ... usbcore: registered new interface driver cdc_ether mb0: register 'mbm' at usb-0000:00:1d.7-6, Mobile Broadband Network Device, 02:80:37:*:*:* usbcore: registered new interface driver mbm
Your lsmod output should contain the following
$ sudo lsmod Module Size Used by mbm 4864 0 cdc_ether 4736 1 mbm usbnet 14472 2 mbm,cdc_ether mii 4736 1 usbnet ...
But the best part is that a new ethernet interface (mb0) should have been created~:
$ sudo ip link
...
11: mb0: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000
link/ether 02:80:37:**:**:** brd ff:ff:ff:ff:ff:ff
if you still use ifconfig, don't forget to call it with "-a" option as the interface is down by default. For an unknow reason (at least to me), the OUI in the ethernet address is not a registered one (among others, "Ericsson Group" has 00-80-37, coincidentally).
Now, we need to issue some AT commands in order to configure the modem (unlock the SIM, activate the antenna, set account parameters, ask for Packet Switched mode, ...). The final goal is to get the mb0 device report an usable link (i.e. UP state). I describe required steps below with as many details as possible in order for the order and interactions to be explicit.
The first thing to do is to unlock your SIM card (i.e. pass it its PIN)if it is not already unlocked:
$ screen /dev/ttyACM0 AT+CPIN? << PIN needed? +CPIN: SIM PIN << Yes, PIN needed OK AT+CPIN="1234" OK AT+CPIN? << Card is unlocked +CPIN: READY OK
At some point, you need to activate the RF circuits (and optionnally select the mode in which you want it to operate). Use AT+CFUN command for that purpose. Possible values are documented on the AT commands page:
AT+CFUN=1 OK
Note that you can activate the RF circuits before or after unlocking the SIM. If the PIN has already been passed before activating the RF circuit, you should may get something like the following (depending on your mobile operator):
AT+CPIN="1234" OK *ESTKSMENU: Services SFR,0,6 2,SFR & Moi 3,Perso du mobile 4,Messages 5,Texto Chat 6,Infos & Loisirs 16,Ripertoire SIM *EMWI: 1,0 +PACSP0
At that point, we need to select the Internet Account to be used for the connection. This means creating one with specific parameters (connection type, APN name, ...) or use an existing one. The AT*EIAC can be used to get access to the list of already defined accounts (and also create one if needed). Reading/Writing the parameters of an existing account is done respectively using the AT*EIAPSR/AT*EIAPSW command. Flushing a given account or all existing accounts can be done using AT*EIAD:
AT*EIAC? << List existing accounts *EIAC: 1,1,"PS Account 1" *EIAC: 2,1,"" *EIAC: 3,1,"" *EIAC: 4,1,"PS Account 4" *EIAC: 9,1,"PS Account 9" OK AT*EIAPSR=1 << Read account w/ index 1 *EIAPSR: 1,1,"websfr",4,0,0 OK AT*EIAPSR=9 << Read account w/ index 9 *EIAPSR: 9,1,"websfr",4,0,0 OK AT*EIAD=9,1 << Flush PS IA #9 OK AT*EIAC? << List existing accounts *EIAC: 1,1,"PS Account 1" *EIAC: 2,1,"" *EIAC: 3,1,"" *EIAC: 4,1,"PS Account 4" OK AT*EIAD=0 << Flush all IA OK AT*EIAC? << List existing accounts OK << None
Now, if you need to create an account, AT*EIAC can be used to create a dummy one (i.e. get assigned an unused index, basically) for a given bearer type. Below, passing 1 to the command creates an account with bearer type "PS connection over UMTS/GPRS network". Optionally, a friendly name can be given for the account ("SFR Account" below). In the example below, created account is assigned index value 5. Parameters can then be modified using AT*EIAPSW command. Below, created account is modified to set the APN to "websfr"; other parameters being left to their default value (The value we set below are for the french operator SFR). Then, if you operator requires some credentials (login/password), those can be configured using the AT*EIAAUW command.
AT*EIAC=1,"SFR Account" << Create a dummy account for PS connections over
*EIAC: 5,"SFR Account" UMTS/GPRS network (Account name will be SFR
Account). Selected index is 5 (not chosen).
OK
AT*EIAPSR=5 << Read current parameters
*EIAPSR: 5,1,"SFR Account",4,0,0
OK
AT*EIAPSW=5,1,"websfr",4,0,0 << Set the APN to "websfr". Other params
left to their default values
OK
AT*EIAPSR=5
*EIAPSR: 5,1,"websfr",4,0,0 << Check our modification
OK
AT*EIAAUW=5,1,"mylogin","mypwd",00111,0 << optional step depending on your operator
<< (for SFR, it is pointless): credentials
AT*EIAAUR=5,1 << Check our modification
*EIAAUR: 5,1,"mylogin","mypwd",00111,0 <<
OK
Now, the AT*ENAP can be used to enable the PS bearer emulation for a given connection (after getting a PDP context from your operator's GGSN). In the example below, we use the fifth connection (hence the value 5 for the second parameter of the command).
AT*ENAP=1,5 OK
After issuing that command, the link should be ready. This can be check using the AT*ENAP command in following way:
AT*ENAP? *ENAP: 1 OK
If the value returned by the command is 0, the issue may not be an account problem but bad coverage. If it happens to you while testing, try and force GPRS for the connection by issuing an AT+CFUN=5 command.
In the end, the following line should also appear in your kernel logs:
$ sudo dmesg ... Jan 31 16:51:12 small kernel: usb 8-6: link speeds: 7200 kbps RX, 2000 kbps TX
and your mb0 interface should now be up and running:
$ sudo ip link
...
11: mb0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 02:80:37:**:**:** brd ff:ff:ff:ff:ff:ff
The link can later be set down using AT*ENAP=0.
Now that we have the ethernet emulation ok and the interface up and running, we can use dhcpcd (or any other DHCP client, like dhclient) on the interface to get an IP, a route and DNS parameters (i.e. get the information from the PDP context).
$ sudo dhcpcd mb0 dhcpcd.sh: interface mb0 has been configured with new IP=10.132.52.53 $ ping www.google.com PING www.l.google.com (74.125.39.103) 56(84) bytes of data. 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=2 ttl=241 time=1527 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=3 ttl=241 time=1074 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=4 ttl=241 time=84.2 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=5 ttl=241 time=73.7 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=6 ttl=241 time=73.8 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=7 ttl=241 time=73.6 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=8 ttl=241 time=72.9 ms 64 bytes from fx-in-f103.google.com (74.125.39.103): icmp_seq=9 ttl=241 time=72.0 ms ^C --- www.l.google.com ping statistics --- 9 packets transmitted, 8 received, 11% packet loss, time 8002ms rtt min/avg/max/mdev = 72.072/381.600/1527.329/542.826 ms, pipe 2
Contactless (RFID) reader
The RFID reader is not supported. It is part of the broadcom USH
TPM
The Broadcom TPM security chip on the E4300 is compliant with the TCG TIS 1.2 TPM specification: it is supported by the tpm_tis kernel driver:
Symbol: TCG_TPM [=y]
Prompt: TPM Hardware Support
Defined at drivers/char/tpm/Kconfig:5
Depends on: HAS_IOMEM && EXPERIMENTAL
Location:
-> Device Drivers
-> Character devices
Selects: SECURITYFS
If your kernel has been compiled with the support and the TPM is enabled in the BIOS, you should found a /dev/tpm0 on your system:
$ ls -l /dev/tpm0 crw------- 1 tss tss 10, 224 2009-01-18 01:58 /dev/tpm0
Without some userland support the device is pretty useless. You should install:
- trousers: provides an open-source TCG Software Stack (TSS). In practice, this provides tcsd, a daemon listening on 30003/tcp, which is used by userland apps to access the TPM.
- tpm-tools: provides a set of userland apps dedicated to
the user interactions with the TPM. As a small panorama of the
capabilities, here are quick descriptions of the tools in the
package:
- tpmtoken_import: import an X.509 certificate or and RSA key pair into a user's TPM PKCS#11 data store.
- tpmtoken_init: initialize the user's TPM PKCS#11 data store.
- tpmtoken_protect: encrypt or decrypt data using a symmetric key stored in the user's TPM PKCS#11 data store.
- tpmtoken_objects: display the objects in the user's TPM PKCS#11 data store
- tpmtoken_setpasswd: change the password(s) associated with the user's TPM PKCS#11 data store.
- tpm_sealdata: seal input data to the system's TPM
- tpm_takeownership: setup an owner on the TPM
- tpm_selftest: request TPM perform selftest and report
- tpm_restrictsrk: XXX
- tpm_changeownerauth: change the authorization data associated with the owner or SRK
- tpm_createek: create and Endorsement Key Pair on the TPM.
- tpm_restrictpubek: restrict the ability to display the public portion of the Endorsement Key to the owner
- tpm_revokeek: revokes the Endorsement Key Pair of the TPM
- tpm_clear: return the TPM to the default state (unowned, disabled, inactive)
- tpm_resetdalock: reset the dictionary attack lock for the user (require owner authentication)
- tpm_setenable: change TPM enable states
- tpm_setpresence: change TPM physical presence states or settings
- tpm_setoperatorauth: sets the operator authorization value in the TPM.
- tpm_setclearable: disable TPM clear operations
- tpm_version: report TPM version and manufacturer information
- tpm_getpubek: restrict the ability to display the public portion of the Endorsement Key to the owner
- tpm_setownable: change whether the TPM allows tpm_takeownership opera‐ tions
- tpm_setactive: change TPM active states
After having installed previous packages, you can get some information on the TPM:
$ sudo tpm_version TPM 1.2 Version Info: Chip Version: 1.2.6.0 Spec Level: 2 Errata Revision: 1 TPM Vendor ID: BRCM TPM Version: 01010000 Manufacturer Info: 4252434dI have found no time yet to have more fun with the TPM. I just generated an EK (i.e. the device works).
USB
The laptop comes with an USB2 port on the right side ... but it also comes with an eSATA port on the left side which acts as a combined USB/eSATA port.
I initially missed the fact that the eSATA port had that capability. Thanks to Lars Wilhelmsen for pointing that to me (and shame on me for the time it took me to update the page). As he wrote in his email, this basically increases the number of available USB ports by 100% compared to my initial review.
I still wonder why Dell does not offer a bay (to be plugged in replacement of the DVD drive) with additional USB ports.
SD Card Reader
02:01.1 SD Host controller: Ricoh Co Ltd R5C822 SD/SDIO/MMC/MS/MSPro Host Adapter
(rev 22) (prog-if 01)
Subsystem: Dell Device 024d
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping-
SERR+ FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 64, Cache Line Size: 64 bytes
Interrupt: pin B routed to IRQ 18
Region 0: Memory at f65ff600 (32-bit, non-prefetchable) [size=256]
Capabilities: [80] Power Management version 2
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
Status: D0 PME-Enable- DSel=0 DScale=2 PME-
Kernel driver in use: sdhci-pci
02:01.2 System peripheral: Ricoh Co Ltd R5C843 MMC Host Controller (rev ff) (prog-if ff)
!!! Unknown header type 7f
Kernel driver in use: ricoh-mmc
The laptop comes with a SD card reader located on the side, just below the touchpad. It is recognized and works just fine.
Fingerprint reader
The UPEK fingerprint reader is located on the far left of the touchpad, below the keyboard.
Bus 003 Device 002: ID 0a5c:5801 Broadcom Corp.
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 1.10
bDeviceClass 0 (Defined at Interface level)
bDeviceSubClass 0
bDeviceProtocol 0
bMaxPacketSize0 64
idVendor 0x0a5c Broadcom Corp.
idProduct 0x5801
bcdDevice 1.01
iManufacturer 1 Broadcom Corp
iProduct 2 5880
iSerial 3 0123456789ABCD
bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 139
bNumInterfaces 2
bConfigurationValue 0
iConfiguration 0
bmAttributes 0xc0
Self Powered
MaxPower 100mA
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 0
bAlternateSetting 0
bNumEndpoints 3
bInterfaceClass 254 Application Specific Interface
bInterfaceSubClass 0
bInterfaceProtocol 0
iInterface 4 Broadcom USH w/swipe sensor
** UNRECOGNIZED: 10 25 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x81 EP 1 IN
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0040 1x 64 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x01 EP 1 OUT
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0040 1x 64 bytes
bInterval 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x85 EP 5 IN
bmAttributes 3
Transfer Type Interrupt
Synch Type None
Usage Type Data
wMaxPacketSize 0x0010 1x 16 bytes
bInterval 1
...
< snip: removed Contacted SmartCard Interface Descriptor >
...
There is currently no support for the device (neither via fprint, thinkfinger, or any other project). Considering that Broadcom does not provide the specs, this will probably require some reverse engineering.
BIOS
The laptop was delivered with a BIOS in version A03. At the time of writing, latest version available on Dell's web site is A07. After having taken a look at which improvements the various BIOS releases brought, I decided to update my BIOS.
Some time ago, the procedure would have required the system to run some mouse oriented OS but, thanks to some efforts from Dell engineers, the procedure can now be performed from Linux too. The tool they have developed and maintain are even available as standard Debian packages.
Note that you need your kernel to be compiled with CONFIG_DELL_RBU option, either statically or as a module. If you decided to go for a module, load it now.
$ sudo modprobe dell_rbu
Then, you need to install the following packages
$ sudo apt-get update $ sudo apt-get install libsmbios-bin libsmbios2
libsmbios-bin provides various Dell utilities:
$ dpkg -L libsmbios-bin /. /usr /usr/share /usr/share/doc /usr/share/doc/libsmbios-bin /usr/share/doc/libsmbios-bin/copyright /usr/share/doc/libsmbios-bin/changelog.Debian.gz /usr/sbin /usr/sbin/mkbiospkg.sh /usr/sbin/assetTag /usr/sbin/getSystemId /usr/sbin/propertyTag /usr/sbin/serviceTag /usr/sbin/verifySmiPassword /usr/sbin/wakeupCtl /usr/sbin/dellBiosUpdate /usr/sbin/dellLcdBrightness /usr/sbin/dumpSmbios /usr/sbin/dellWirelessCtl /usr/sbin/dellMediaDirectCtl /usr/sbin/activateCmosToken /usr/sbin/createUnitTestFiles /usr/sbin/disable_console_redir /usr/sbin/dumpCmos /usr/sbin/isCmosTokenActive /usr/sbin/smitest /usr/sbin/stateByteCtl /usr/sbin/upBootCtl /usr/sbin/getPasswordFormat /usr/sbin/ascii2enUS_scancode /usr/sbin/probes /usr/sbin/dellLEDCtl
getSystemId provides various information on the system:
$ sudo getSystemId Libsmbios: 2.0.3 System ID: 0x024D Service Tag: 8L5024J Express Service Code: 18692449219 Product Name: Latitude E4300 BIOS Version: A03 Vendor: Dell Inc. Is Dell: 1
Among others, the System ID (0x024D for the E4300), which we then use to get the latest BIOS from Dell's firmware repository: http://linux.dell.com/repo/firmware/bios-hdrs/.
$ w3m http://linux.dell.com/repo/software/bios-hdrs/ ... [DIR] system_bios_ven_0x1028_dev_0x024a_version_a08/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024d_version_a00/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024d_version_a02/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024d_version_a03/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024d_version_a04/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024d_version_a05/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024d_version_a06/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024e_version_a00/ 17-Jul-2009 06:26 - [DIR] system_bios_ven_0x1028_dev_0x024e_version_a02/ 17-Jul-2009 06:26 - ... ... ... Index of /repo/software/bios-hdrs/system_bios_ven_0x1028_dev_0x024d_version_a06 Icon Name Last modified Size Description[DIR] Parent Directory - [ ] bios.hdr 24-Feb-2009 01:24 3.5M [TXT] dell-std-license.txt 17-Jul-2009 06:24 7.0K [TXT] extract.log 17-Jul-2009 01:56 3.6K [ ] package.ini 24-Feb-2009 01:24 242
At the time of writing, the latest version available on the server is A06. In fact, A07 is the latest BIOS version, but is available only as a .exe which cannot directly be used with linux tools. An extraction process is needed to create a bios.hdr file. The files available at previous URL have already been extracted and can be used directly with linux tools.
Once you have grabbed the bios.hdr file for the BIOS version you need, you can use dellBiosUpdate to perform to test if the version is correct and to perform the update. The update will only be performed during the next reboot:
$ sudo dellBiosUpdate -t -f bios.hdr BIOS file matches this system and is newer. $ sudo dellBiosUpdate -i -f bios.hdr HeaderId : $RBU Header Length: 84 Header Major Ver: 1 Header Minor Ver: 0 Num Systems: 1 Version: A06 Quick Check: Copyright 2009 Dell Inc. System ID List: 0x024d $ sudo dellBiosUpdate -u -f bios.hdr Supported RBU type for this system: (MONOLITHIC, PACKET) Using RBU v2 driver. Initializing Driver. Setting RBU type in v2 driver to: PACKET writing (4096) to file: /sys/devices/platform/dell_rbu/packet_size Writing RBU data (4096bytes/dot): ....................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................................... ......................................................... Done writing packet data. Activate CMOS bit to notify BIOS that update is ready on next boot. Update staged sucessfully. BIOS update will occur on next reboot. $ sudo reboot
After the reboot and the BIOS update, getSystemId should reflect the update:
$ sudo getSystemId Libsmbios: 2.0.3 System ID: 0x024D Service Tag: 8L5024J Express Service Code: 18692449219 Product Name: Latitude E4300 BIOS Version: A06 Vendor: Dell Inc. Is Dell: 1
If the BIOS version you want to install is not available on Dell website as a .hdr file, you can extract the .hdr file yourself from the .exe. You need wine for that purpose:
$ wine E4300A07.EXE -writehdrfile -nopause $ ls E4300A07.EXE E4300A07.hdr $ sudo dellBiosUpdate -t -f E4300A07.hdr BIOS file matches this system and is newer. $ sudo dellBiosUpdate -i -f E4300A07.hdr HeaderId : $RBU Header Length: 84 Header Major Ver: 1 Header Minor Ver: 0 Num Systems: 1 Version: A07 Quick Check: Copyright 2009 Dell Inc. System ID List: 0x024d $ sudo dellBiosUpdate -u -f E4300A07.hdr $ sudo reboot
eSATA port
The port is located on the left side of the laptop. I have no eSATA peripheral to plug in so I have not tried it yet in that mode. But the port is in fact a combined eSATA/USB port: it can be used to plug USB devices.
Express card slot
Not tried: like for eSATA, i have no such peripheral yet.
Firewire
02:01.0 FireWire (IEEE 1394): Ricoh Co Ltd R5C832 IEEE 1394 Controller (rev 05)
(prog-if 10 [OHCI])
Subsystem: Dell Device 024d
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping-
SERR+ FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort-
<MAbort- >SERR- <PERR- INTx-
Latency: 64 (500ns min, 1000ns max), Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 19
Region 0: Memory at f65ff800 (32-bit, non-prefetchable) [size=2K]
Capabilities: [dc] Power Management version 2
Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
Status: D0 PME-Enable- DSel=0 DScale=2 PME+
Kernel driver in use: ohci1394
Kernel modules: ohci1394
The laptop comes with a small (6 pins) IEEE 1394 interface. It is recognised and works fine (tested with a firewire drive using a simple 6 pin to 9 pin converter).
The annoying thing with 6 pins IEEE 1394 interfaces is that they do not provide power to the device: when plugging a disk, you need to use an external power source for the drive. If you have a Firewire iSight, do not expect you will be able to use it with an adapter.
GPS
The GPS function is provided by the Dell Wireless 5530 card (rebranded Ericsson Mobile Broadband Module F3507g). Using the cdc_acm driver, 3 serial interfaces (/dev/ttyACM{0,1,2}) are provided to control and use the various functions of the device. At least, the HSDPA and GPS functions works well.
Simply put, using simple AT commands, just like for the HSDPA part, the GPS can be activated (in either GPS or DGPS mode) to get an NMEA 0183 messages flow on /dev/ttyACM2. This flow is then perfectly handled by gpsd which then makes it available to its users (gpsdrive, tangogps, ...) on 2947/tcp.
In the following, I use:
- /dev/ttyACM0: the control interface for both the modem and the GPS
- /dev/ttyACM1: the modem interface (used by pppd)
- /dev/ttyACM2: the interface on which NMEA messages flow are received
Let's first start with the AT commands of interest for the control of the GPS. The list below is just a quick summary, extracted from the exhaustive list of commands supported by the Dell Wireless 5350 I maintain here:
- AT+CFUN: Activate/Deactivate the RF circuit (also used to switch between various modes). After enabling the RF circuit, the blue led near the bluetooth one at the top of the keyboard should get on. Activating the RF circuit is required in order to be able to use the GPS (following command will return an ERROR otherwise).
- AT*E2GPSCTL: used to start/stop the NMEA message flow on the specific TTY (selected using AT*E2GPSNPD), and configure associated parameters (GPS or DGPS, interval between the messages).
- AT*E2GPSNPD: when issued on a TTY, this specific TTY is selected to receive the NMEA message flows. The specific form (GPS, DGPS) and interval between the messages is configured using the AT*E2GPSCTL command on another TTY.
- AT*E2GPSCLM: not documented
- AT*E2GPSEPH: not documented
- AT*E2GPSSTAT: not documented
- AT*E2GPSTIME: not documented
Let's know see a concrete use of those commands (everything done by hand but this can easily be automated using via a simple script).
First, the RF circuit need to be activated by issuing AT+CFUN=1 (AT+CFUN=5 or 6 will also do the job) on the control TTY (/dev/ttyACM0). Unlocking the SIM is also required. Don't ask! I have no clue why this is needed but GPS-related AT commands will fail otherwise.
Then, AT*E2GPSCTL command is used to start receiving the NMEA message flow (first argument set to 1) on the /dev/ttyACM2 (previously selected for that purpose). The second argument is used to control the frequency: here, one message every four seconds. The last argument being set to 1 asks for DGPS message format for NMEA messages (0 is for common GPS format).
$ screen /dev/ttyACM0 AT+CFUN=1 << enable the RF circuit OK AT+CPIN="1234" << unlock the SIM OK *ESTKSMENU: Services SFR,0,6 2,SFR & Moi 3,Perso du mobile 4,Messages 5,Texto Chat 6,Infos & Loisirs 16,Ripertoire SIM *EMWI: 1,0 +PACSP0 AT*E2GPSCTL=1,4,1 << Starts NMEA flow, 4 second interval, DGPS
Then, to receive the NMEA messages flow on /dev/ttyACM2 we need to issue the AT*E2GPSNPD command on that specific TTY. After that step, your output should look like the one below. Switching between GPS mode (DGPS vs GPS) is done by changing the last value of AT*E2GSPCTL commad (using respectively 1 and 0). Before doing something else, you first need to terminate screen before starting any application that will use this TTY (like GPSD, as covered later).
$ screen /dev/ttyACM2 AT*E2GPSNPD << Leave this screen using Ctrl-a k $ screen /dev/ttyACM2 $GPGGA,,,,,,0,00,10.0,,M,,,,*34 << common GPS flow i.e. $GPRMC,134525.41,V,,,,,,,030136,,,N*7B AT*E2GPSCTL=1,X,0 has $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 been issued on control TTY $GPGGA,,,,,,0,00,10.0,,M,,,,*34 $GPRMC,134530.41,V,,,,,,,030136,,,N*7F $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 $GPGGA,,,,,,0,00,10.0,,M,,,,*34 $GPRMC,134535.41,V,,,,,,,030136,,,N*7A $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 $GPGGA,,,,,,0,00,10.0,,M,-0.323049,M,0.0030199,0130*6B << switching to DGPS by $GPRMC,000024.40,V,,,,,,,060180,,,N*70 issuing AT*E2GPSCTL=1,X,1 $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 on control TTY $GPGGA,,,,,,0,00,10.0,,M,-0.323049,M,0.0030199,0130*6B $GPRMC,000029.40,V,,,,,,,060180,,,N*7D $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 $GPGGA,,,,,,0,00,10.0,,M,-0.323049,M,0.0030199,0130*6B $GPRMC,000034.40,V,,,,,,,060180,,,N*71 $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 $GPGGA,,,,,,0,00,10.0,,M,-0.323049,M,0.0030199,0130*6B $GPRMC,000039.40,V,,,,,,,060180,,,N*7C $GPGSA,A,1,,,,,,,,,,,,,22.3,10.0,20.0*00 [screen is terminating] << Leave screen using Ctrl-a k
Now, in order for apps to be able to use the GPS, gpsd should be started (again, terminate your screen session on /dev/ttyACM2). Under Debian, the configuration of dgps is performed in /etc/default/gpsd. Mine looks like the following:
$ cat /etc/default/gpsd START_DAEMON="true" DAEMON_OPTS="" DEVICES="/dev/ttyACM2" USBAUTO="true"
You can then start the daemon and then use xgps to check everything is working as expected (you will have to go outside and wait until you get a fix):
$ sudo /etc/init.d/gpsd restart Stopping GPS (Global Positioning System) daemon: gpsd. Starting GPS (Global Positioning System) daemon: gpsd. $ xgps &
Processor
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 23
model name : Intel(R) Core(TM)2 Duo CPU P9400 @ 2.40GHz
stepping : 6
cpu MHz : 800.000
cache size : 6144 KB
physical id : 0
siblings : 2
core id : 0
cpu cores : 2
apicid : 0
initial apicid : 0
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 10
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr
pge mca cmov pat pse36 clflush dts acpi mmx fxsr
sse sse2 ss ht tm pbe nx lm constant_tsc
arch_perfmon pebs bts pni dtes64 monitor ds_cpl
vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1
lahf_lm tpr_shadow vnmi flexpriority
bogomips : 4790.75
clflush size : 64
power management:
Advanced E-Port replicator
With the laptop, I ordered the advanced E-port replicator. It has 5 USB ports, an eSATA/USB port, a serial port, a VGA output, 2 DVI ports, 2 HDMI ports, an ethernet port, mouse and keyboard ports, a parallel port, audio line and mic ports.
The idea was to be able to have external displays, a mouse and my Dell Smartcard Keyboard connected to the E-port on my desk.
With a recent kernel (currently a 2.6.30 but previous version may also work fine) and a recent version of Xorg (dpkg reports 1:7.4+1 for xserver-xorg and 2:2.7.0-1 for xserver-xorg-video-intel), I have my two Dell 2407WFP plugged on the two DVI ports, each at a 1920x1200@60Hz resolution. It just works nicely, providing a desktop with a 3840x1200 resolution.
In fact, the E-port is so useful when you are at your desk that I have one at work and another at home, both serving the same LCD monitors. The combination of such LCD displays, E-port and E4300 is simply perfect: it provides a big desktop surface in a limited amount of space and great performances. In fact, I don't even use my old desktop box anymore at work and have dropped the other boxes I had at home.
Screen/output
The screen is nice. xrandr reports the following:
Screen 0: minimum 320 x 200, current 1280 x 800, maximum 4096 x 4096 VGA disconnected (normal left inverted right x axis y axis) LVDS connected 1280x800+0+0 (normal left inverted right x axis y axis) 287mm x 180mm 1280x800 60.0*+ 40.0 1024x768 60.0 800x600 60.3 56.2 640x480 59.9 HDMI-1 disconnected (normal left inverted right x axis y axis) HDMI-2 disconnected (normal left inverted right x axis y axis) TV disconnected (normal left inverted right x axis y axis)
The laptop also has an external output (VGA) which can be used simultaneously, either as a clone or to extend your desktop.
If you need more, see previous section on the advanced E-port replicator.
Suspend
At the time of writing, suspend to RAM works just fine. I simply use s2ram with -f flag. The E4300 is not known to work yet I suppose; it may be when you'll try with yours. Under Debian, s2ram is provided by uswsusp packages:
$ sudo apt-get install uswsusp $ sudo s2ram -f