Development

Ticket #11362 (Support counterfeit FTDI devices that were bricked by FTDI's Windows ...) created

Latest Bugs & Tasks - Thu, 2014-10-23 14:36

For what that means, please see: ​dissasembly of the Windows FTDI driver, ​attempts by FTDI to get Linux to do the same, and ​FTDI's response that shows it's intentional.

Looking at ​Linux's pending patch for this, it appears all we need to do is to add a PID of 0 to the list ​here.

As to why we should support these "counterfeit" devices:

  • They are extremely widespread (e.g. all Arduinos will be bricked by this)
  • Most consumers have no idea which one is in their device, we shouldn't punish consumers for something done by the manufacturer.
Categories: Development

Ticket #11361 (junk) closed

Latest Bugs & Tasks - Thu, 2014-10-23 14:04
junk
Categories: Development

Ticket #11361 (junk) created

Latest Bugs & Tasks - Thu, 2014-10-23 13:44
Categories: Development

aiff_reader: fix for hybrid build.

Source Activity - Thu, 2014-10-23 07:12
* Not really useful now that ffmpeg can decode AIFF correctly... * Not really useful now that ffmpeg can decode AIFF correctly...
Categories: Development

Remove useless file.

Source Activity - Thu, 2014-10-23 07:11
* This is a leftover of in-tree ffmpeg and was not included anywhere. * This is a leftover of in-tree ffmpeg and was not included anywhere.
Categories: Development

Remove the workaround for #11018 in two other places.

Source Activity - Thu, 2014-10-23 07:07
The bug is fixed, the workaround can be removed. The bug is fixed, the workaround can be removed.
Categories: Development

haiku_loader: Fix wrong size of gBootGDT on x86_64.

Source Activity - Wed, 2014-10-22 18:55
The BOOT_GDT_SEGMENT_COUNT was based on USER_DATA_SEGMENT on both x86 and x86_64. However, on x86_64 the order of the segments is different, leading to a too small gBootGDT array. Move the define to the arch specific headers so they can be setup correctly in either case. Also add a STATIC_ASSERT() to check that the descriptors fit into the array. Pointed out by CID 1210898. The BOOT_GDT_SEGMENT_COUNT was based on USER_DATA_SEGMENT on both x86 and x86_64. However, on x86_64 the order of the segments is different, leading to a too small gBootGDT array. Move the define to the arch specific headers so they can be setup correctly in either case. Also add a STATIC_ASSERT() to check that the descriptors fit into the array. Pointed out by CID 1210898.
Categories: Development

AreaTest: 64bit fix.

Source Activity - Wed, 2014-10-22 17:36
Categories: Development

Ticket #9065 (Audio CD does not play) closed

Latest Bugs & Tasks - Wed, 2014-10-22 17:30
invalid:

CDPlayer has been removed from the repo.

Categories: Development

Ticket #11360 (usb Device timeouts - older hardware.) created

Latest Bugs & Tasks - Wed, 2014-10-22 16:59

Connecting a older usb input device results in the following timeouts, which means
it will not apear in the listusb either.

kern: usb hub 7: port 0: new device connected
kern: usb error control pipe 44: timeout waiting for queued request to complete
kern: usb error control pipe 44: timeout waiting for queued request to complete
KERN: usb error uhci -1: error while setting device address

This is an older device (trust tb-2100 wireless tablet) but i suspect this timeout
to be more general.

Categories: Development

gcc4 fix.

Source Activity - Wed, 2014-10-22 16:58
Categories: Development

Ticket #2023 (CD Player won't stop playing audio cd) closed

Latest Bugs & Tasks - Wed, 2014-10-22 15:54
fixed:

There is no CD player anymore.

Categories: Development

Ticket #11359 (ProcessController: improve "quit an application" and kill actions) created

Latest Bugs & Tasks - Wed, 2014-10-22 15:33

In the ProcessController's "Threads and CPU usage" submenu, when you click on a process, an alert window will ask if we are sure of this action (kill). Instead, under the submenu "quit an application" an alert window is not displayed, and this could lead to troubles, if someone does the mistake to quit a system process. Eg if you kill the registrar, Haiku will become unusable, and could be the same for some other system processes (eg the input_server). So my suggestion is to exclude system processes from the "quit an application" menu (and maybe also from "Threads and CPU usage") or implement an alert window for quit system processes, (also if i don't see the need to kill/quit system apps). Any thoughts?

Categories: Development

Ticket #7047 (Noisy AIFF audio files) closed

Latest Bugs & Tasks - Wed, 2014-10-22 15:28
fixed:

Fixed in hrev48071.

Categories: Development
Syndicate content