fobbit-developer Mailing List for Fobbit
Brought to you by:
johngeeaswell,
mchang
Menu
▾
▴
fobbit-developer — Development discussion.
You can subscribe to this list here.
| 2002 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(1) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(5) |
Dec
(1) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2003 |
Jan
|
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Mark L. C. <mar...@ac...> - 2003-02-16 02:24:55
|
I integrated these patches into the cvs tree and into the 0.60rc1
release candidate.
I was able to successfully compile this on a 2.4.20 kernel machine, but
can't do it on my own 2.4.18 machine due to some weirdness with the
kernel source I have (RH 7.3) ... I'll see if I can do anything about that.
In the future, if you could create patches against the cvs tree, that'd
be great.
Mark
David R Bacon wrote:
> Hello All-
>
> I have made 3 separate patches to the linux driver:
> 1/3 update urb_t to struct urb (**needed to compile**)
> 2/3 update __FUNCTION__ to __LINE__ (fixes deprecation warnings)
> 3/3 add devfs support
>
> They apply in any order against v0.56, and probably v0.57, since it is
> largely unchanged.
>
> I'm not sure the best way to include 3 patches, so I attached them.
>
> Ciao,
> Dave
>
>
>
> ------------------------------------------------------------------------
>
> diff -Naur pristine/v0.56/linux_drv/usbvb.c voipblaster/v0.56/linux_drv/usbvb.c
> --- pristine/v0.56/linux_drv/usbvb.c 2002-04-04 21:47:14.000000000 -0600
> +++ voipblaster/v0.56/linux_drv/usbvb.c 2003-02-01 18:49:30.000000000 -0600
> @@ -117,6 +117,8 @@
> struct usb_device *udev; // save off the usb device pointer
> int minor; // starting minor number for this device
>
> + devfs_handle_t devfs; // devfs device node
> +
> u_char out_buffer[VOUT_SIZE]; // voice/command output buffer for urb
> u_char inp_buffer[VINP_SIZE]; // voice/status input buffer for urb
> #ifdef BG_READ
> @@ -140,6 +142,10 @@
> struct semaphore inp_sem; // locks this structure
> };
>
> +// the global devfs handle
> +extern devfs_handle_t usb_devfs_handle;
> +static struct file_operations vb_fops; // forward decl of fops for devfs use
> +
> // array of pointers to our devices that are currently connected
> static struct usb_vb *minor_table[MAX_DEVICES];
>
> @@ -974,6 +980,7 @@
> struct usb_vb *c_dev = NULL;
> struct usb_vb *v_dev = NULL;
> int minor;
> + char name[10];
>
> // See if the device offered us matches what we can accept
> if ((udev->descriptor.idVendor != USB_VB_VENDOR_ID ) ||
> @@ -1012,6 +1019,29 @@
> goto done;
> }
>
> + // initialize the devfs node for voice and control device and register them
> + sprintf(name, "vbc%d", c_dev->minor);
> +
> + c_dev->devfs = devfs_register(usb_devfs_handle, name,
> + DEVFS_FL_DEFAULT,
> + USB_MAJOR,
> + USB_VB_MINOR_BASE + c_dev->minor,
> + S_IFCHR | S_IRUSR | S_IWUSR |
> + S_IRGRP | S_IWGRP |
> + S_IROTH,
> + &vb_fops, NULL);
> +
> + sprintf(name, "vbv%d", v_dev->minor - (MAX_DEVICES/2));
> +
> + v_dev->devfs = devfs_register(usb_devfs_handle, name,
> + DEVFS_FL_DEFAULT,
> + USB_MAJOR,
> + USB_VB_MINOR_BASE + v_dev->minor,
> + S_IFCHR | S_IRUSR | S_IWUSR |
> + S_IRGRP | S_IWGRP |
> + S_IROTH,
> + &vb_fops, NULL);
> +
> // let the user know what node this device is now attached to
> info("USB Voip-Blaster (%d) connected, ifnum(%d)", minor, ifnum);
>
> @@ -1041,6 +1071,9 @@
>
> minor = c_dev->minor; // get before do any cleanup
>
> + // remove our control devfs node
> + devfs_unregister(c_dev->devfs);
> +
> // if the device is not opened, then we clean up right now
> if (!c_dev->open_count) {
> up (&c_dev->inp_sem);
> @@ -1059,6 +1092,9 @@
> down (&v_dev->out_sem);
> down (&v_dev->inp_sem);
>
> + // remove our voice devfs node
> + devfs_unregister(v_dev->devfs);
> +
> // if the device is not opened, then we clean up right now
> if (!v_dev->open_count)
> {
>
>
> ------------------------------------------------------------------------
>
> diff -Naur pristine/v0.56/linux_drv/usbvb.c voipblaster/v0.56/linux_drv/usbvb.c
> --- pristine/v0.56/linux_drv/usbvb.c 2002-04-04 21:47:14.000000000 -0600
> +++ voipblaster/v0.56/linux_drv/usbvb.c 2003-02-01 16:21:37.000000000 -0600
> @@ -65,12 +65,12 @@
> // Use our own dbg() macro
> #undef dbg
> #define dbg(format, arg...) if (debug) \
> - printk(KERN_DEBUG __FILE__ "," __FUNCTION__ ": " format "\n" , ## arg); \
> + printk(KERN_DEBUG __FILE__ ", %d: " format "\n" , __LINE__, ## arg); \
>
> // Use our own err() macro
> #undef err
> #define err(format, arg...) \
> - printk(KERN_ERR __FILE__ "," __FUNCTION__ ": " format "\n" , ## arg); \
> + printk(KERN_ERR __FILE__ ", %d : " format "\n" , __LINE__, ## arg); \
>
> // Version Information
> #define DRIVER_VERSION "v0.20"
> @@ -553,7 +553,7 @@
> // usb_vb_debug_data
> //
> static inline void
> -usb_vb_debug_data (const char *function, int size, const unsigned char *data)
> +usb_vb_debug_data (int lineno, int size, const unsigned char *data)
> {
> int i;
>
> @@ -562,8 +562,8 @@
> return;
> }
>
> - printk (KERN_DEBUG __FILE__": %s - length = %d, data = ",
> - function, size);
> + printk (KERN_DEBUG __FILE__": line %d - length = %d, data = ",
> + lineno, size);
> for (i = 0; i < size; ++i)
> {
> printk ("%.2x ", data[i]);
> @@ -668,11 +668,11 @@
> dev = (struct usb_vb *)file->private_data;
> if (dev == NULL)
> {
> - dbg (__FUNCTION__ " - object is NULL");
> + dbg ("%d - object is NULL", __LINE__);
> return -ENODEV;
> }
>
> - dbg(__FUNCTION__ " - minor %d", dev->minor);
> + dbg("%d - minor %d", __LINE__, dev->minor);
>
> // lock our minor table
> down (&minor_table_mutex);
> @@ -683,7 +683,7 @@
>
> if (dev->open_count <= 0)
> {
> - dbg (__FUNCTION__ " - device not opened");
> + dbg ("%d - device not opened", __LINE__);
> retval = -ENODEV;
> goto exit_not_opened;
> }
> @@ -710,7 +710,7 @@
>
> if (dev->udev == NULL)
> {
> - dbg (__FUNCTION__ " - device unplugged");
> + dbg ("%d - device unplugged", __LINE__);
> // the device was unplugged before the file was released
> // so held off rest of 'vb_disconnect' till now
> up (&dev->inp_sem);
> @@ -864,7 +864,7 @@
> goto exit;
> }
>
> - usb_vb_debug_data (__FUNCTION__, write_count, dev->out_buffer);
> + usb_vb_debug_data (__LINE__, write_count, dev->out_buffer);
>
> // send the data to the device
> retval = vb_bulk_msg (dev->udev,
>
>
> ------------------------------------------------------------------------
>
> diff -Naur pristine/v0.56/linux_drv/usbvb.c voipblaster/v0.56/linux_drv/usbvb.c
> --- pristine/v0.56/linux_drv/usbvb.c 2002-04-04 21:47:14.000000000 -0600
> +++ voipblaster/v0.56/linux_drv/usbvb.c 2003-01-31 21:52:44.000000000 -0600
> @@ -126,7 +126,7 @@
> int rdr_status; // interface to read system call
> wait_queue_head_t rdr_wqh ; // Processes waiting
> wait_queue_head_t rdr_pwait ; // Processes waiting
> - urb_t *have_urb ;
> + struct urb *have_urb ;
> #endif
>
> int out_size; // max size for output pipe
> @@ -160,7 +160,7 @@
> #define VB_WRITE_TIMEOUT (0)
>
> static void
> -vb_blocking_completion(urb_t *urb)
> +vb_blocking_completion(struct urb *urb)
> {
> api_wrapper_data *awd = (api_wrapper_data *)urb->context;
>
> @@ -185,7 +185,7 @@
> DECLARE_WAIT_QUEUE_HEAD(wqh);
> api_wrapper_data awd;
> int status, was_sig;
> - urb_t *urb;
> + struct urb *urb;
>
> if (len < 0)
> {
> @@ -275,7 +275,7 @@
> #define VB_WRITE_TIMEOUT (HZ*10)
>
> static void
> -vb_blocking_completion(urb_t *urb)
> +vb_blocking_completion(struct urb *urb)
> {
> struct usb_api_data *awd = (struct usb_api_data *)urb->context;
>
> @@ -291,7 +291,7 @@
> DECLARE_WAITQUEUE(wait, current);
> struct usb_api_data awd;
> int status, was_sig;
> - urb_t *urb;
> + struct urb *urb;
>
> if (len < 0)
> {
> @@ -404,7 +404,7 @@
>
> // gets called when read-urb completes
> static void
> -vb_read_restart(urb_t *urb)
> +vb_read_restart(struct urb *urb)
> {
> int status;
> int next_head;
> @@ -512,7 +512,7 @@
> vb_read_start(struct usb_vb *dev)
> {
> int status;
> - urb_t *urb;
> + struct urb *urb;
>
> dbg("dev(%d) entry", dev->minor);
>
|
|
From: David R B. <db...@ci...> - 2003-02-02 01:15:03
|
Hello All- I have made 3 separate patches to the linux driver: 1/3 update urb_t to struct urb (**needed to compile**) 2/3 update __FUNCTION__ to __LINE__ (fixes deprecation warnings) 3/3 add devfs support They apply in any order against v0.56, and probably v0.57, since it is largely unchanged. I'm not sure the best way to include 3 patches, so I attached them. Ciao, Dave |
|
From: Anup M. <an...@bu...> - 2002-12-08 20:11:10
|
In an effort to reconcile the mixed (Linux & windows) behavior of the OS X port, I'd like to propose three potentially significant changes for discussion. First, I'd like to re-organize the file tree a bit. I'm suggesting moving the various readme's & platform notes to the root level of the tree and leaving only source files in src/. The sounds directory wouldn't change. While less critical, I'd also propose creating platform specific sub-dirs within src for platform specific files. Any thoughts? Second, I'd like to introduce autoconf. This isn't a huge issue but it would be nice to have rather than introducing more macro symbols. Finally, I'd still like to break up vblast.h into smaller pieces. Also not a huge problem but it would make it easier to read. I'm testing the OS X merge now and hope to submit something for check-in over the weekend. anup |
|
From: Anup M. <an...@bu...> - 2002-11-20 21:17:24
|
Excellent! I'll do the checkout and start my merge. anup At 10:50 PM 11/19/2002 -0800, you wrote: >Got note from M. Hulden about a merge with the encryption code. Everyone >around, check it out and give it a whirl! Exciting news, no doubt. How >about M. Hulden slap up some news from the Admin interface, eh? :) > >I will get around to peeking at this and the Mac OS X port after >Wednesday..... BUSY. > >Mark > > > >------------------------------------------------------- >This sf.net email is sponsored by: To learn the basics of securing your >web site with SSL, click here to get a FREE TRIAL of a Thawte Server >Certificate: http://www.gothawte.com/rd524.html >_______________________________________________ >Fobbit-developer mailing list >Fob...@li... >https://lists.sourceforge.net/lists/listinfo/fobbit-developer |
|
From: Mark L. C. <mar...@ac...> - 2002-11-20 06:50:47
|
Got note from M. Hulden about a merge with the encryption code. Everyone around, check it out and give it a whirl! Exciting news, no doubt. How about M. Hulden slap up some news from the Admin interface, eh? :) I will get around to peeking at this and the Mac OS X port after Wednesday..... BUSY. Mark |
|
From: Anup M. <sk...@us...> - 2002-11-15 00:03:06
|
Mark & fobbit team, As I've commented in the public forums, I have a working OS X port with one known bug. In order to exit the app, you have to unplug the usb device. This is because I can't figure out how to best force my launched pthreads to quit. I'm close to getting it working but I don't know enough about pthreads so my progress is a bit slow. Barak, how are your efforts coming along? I'd love to collaborate on the OS X port. I'd also like to understand what clean-ups people have done in order to best figure out what to submit of my changes. I've listed some areas that I'd propose changing: 1. Platform macros This is the most critical as the Darwin (the kernel in MacOS X) port uses most of the behavior from the unix port but has some elements from the windows port (threads to read from the USB pipes and dump that data into network buffers). Just using the So the usb interface is almost windows like while the rest of the app is BSD like. Setting VB_IS_UNIX doesn't work for Darwin. I have some thoughts about reworking the platform macros but want to properly coordinate that with your efforts. 2. Build process I'd like to have the unix flavors using the configure/automake functionality. I've not started this effort yet but I'm willing to take it on if there is interest in switching. Not a high priority in my mind but it would make it more compatible with other porting/packaging efforts. 3. vblast.h Maybe it's just me but I'd love to find a way to break up this file into more manageable chunks. If we all start contributing to the project, it might help not to have constant merges on one all-encompassing header file. 4. exit() This is the usual path if any error occurs. I think the behavior is fine but it makes error reporting/recovery/cleanup more difficult. Maybe this could change. I'd love to hear what others are preparing... anup |
|
From: Anup M. <sk...@us...> - 2002-11-15 00:03:06
|
Mark & fobbit team, As I've commented in the public forums, I have a working OS X port with one known bug. In order to exit the app, you have to unplug the usb device. This is because I can't figure out how to best force my launched pthreads to quit. I'm close to getting it working but I don't know enough about pthreads so my progress is a bit slow. Barak, how are your efforts coming along? I'd love to collaborate on the OS X port. I'd also like to understand what clean-ups people have done in order to best figure out what to submit of my changes. I've listed some areas that I'd propose changing: 1. Platform macros This is the most critical as the Darwin (the kernel in MacOS X) port uses most of the behavior from the unix port but has some elements from the windows port (threads to read from the USB pipes and dump that data into network buffers). Just using the So the usb interface is almost windows like while the rest of the app is BSD like. Setting VB_IS_UNIX doesn't work for Darwin. I have some thoughts about reworking the platform macros but want to properly coordinate that with your efforts. 2. Build process I'd like to have the unix flavors using the configure/automake functionality. I've not started this effort yet but I'm willing to take it on if there is interest in switching. Not a high priority in my mind but it would make it more compatible with other porting/packaging efforts. 3. vblast.h Maybe it's just me but I'd love to find a way to break up this file into more manageable chunks. If we all start contributing to the project, it might help not to have constant merges on one all-encompassing header file. 4. exit() This is the usual path if any error occurs. I think the behavior is fine but it makes error reporting/recovery/cleanup more difficult. Maybe this could change. I'd love to hear what others are preparing... anup |
|
From: Mark L. C. <mar...@ac...> - 2002-11-14 06:33:59
|
Currently there are three "developers" on this project. Me, Barak and M. Hulden. Barak is working on the Mac OSX port and M. Hulden, among other things, is working on encryption of the network traffic. I propse that we use this list (for the first time, mind you) to get an open dialog going about merging source trees. I made the very first merge and release (0.57) of M. Hulden's code to minimize to the Windows system tray a few days ago, and it went well. Both of you have access to the CVS tree, and we should start trying to see how difficult it will be to merge the two sets of changes. I know that there has been talk between us about code cleanup. I'll take a peek at the code this weekend and form some thoughts on that. Please share what you guys think. Thanks, Mark |
|
From: Mark C. <mar...@ho...> - 2002-05-23 17:27:46
|
Greetings. I duplicated the current (v0.56) fobbit source in the sourceforge cvs and released a package. It should be the same as the ones available at fobbit.com. The goal here is to expose this project to more developers as well as provide forums and mailing lists to improve discussion. Hopefully, this will catch on soon. Mark _________________________________________________________________ MSN Photos is the easiest way to share and print your photos: http://photos.msn.com/support/worldwide.aspx |
