Latest Bugs & Tasks

Syndicate content Haiku
Trac Timeline
Updated: 23 min 27 sec ago

Ticket #10737 ([AHCI error] InterruptErrorHandler) reopened

42 min 18 sec ago

Unfortunately, it's not fixed. Updated to latest packages (hrev49607) today and now I have this problem too, it almost always fails to boot due to AHCI. Sometimes (rarely) it succeeds booting.

Attaching excerpt from serial debug log for a failed boot (shortened, it always outputs just the same AHCI error messages before finally giving up and KDLing).

Categories: Development

Ticket #12355 (One CPU core 100% at idle) created

17 hours 37 min ago

hrev49606
Was not running any applications, but noticed after a couple of minutes of uptime that one CPU core was pegged at 100%. ProcessController indicated input_server/PathMonitor looper problem (see attached screenshot1). When I attempted to attach the debugger to the input_server the system completely locked up for about 30 seconds. I was able to save a report (attached) before the system crashed and the PC rebooted.

Categories: Development

Ticket #12353 ([Debugger] cache function list for visited libs in Images tab) created

Mon, 2015-08-31 13:42

hrev49606.

building up the list view with function list for images which have tons of symbols takes a lot of time and freezes Debugger for a few seconds. It would be nice to to make that smoother.
Debugger could report what it's doing in its status bar as well as cache previously loaded functions.

Might need to change BCLV to make it faster.

Categories: Development

Ticket #12258 ([launch_daemon] increased boot time) closed

Mon, 2015-08-31 06:20
fixed:

Replying to pulkomandy:

So I think we can close this ticket now and try to optimize the early parts of the boot process to get it back to alpha4 speed. According to vidrep's benchmarks the launch daemon only causes a 200ms increase in boot time, and I guess this is with only one CPU?

Yes, let's close this.

Note that the alpha4 had serial debugging disabled by default, so if these tests were all with default settings, it's fully expected to have a boot time difference of several seconds. Local tests showed that boot time dropped from the above mentioned 10 to somewhat over 6 seconds when serial debugging was disabled, so it can be quite dramatic (because serial output is really slow and there's a lot of it, sometimes more and sometimes less depending on the drivers involved).

So for the sake of comparison it'd be interesting to have the boot times of a post launch_daemon build with serial debugging disabled in /boot/home/config/settings/kernel/drivers/kernel.

Categories: Development

Ticket #12352 ([WebPositive] reloading a page opens about:blank) created

Mon, 2015-08-31 05:44

hrev49605 / haikuwebkit_x86-1.4.12-1-x86_gcc2

Go to: ​​https://easyweb.td.com/
Notice that address bar is filled with url very late.
Reload the page using Alt+r
WebPositive goes to about:blank

Categories: Development

Ticket #12351 ([WebPositive] crash in ...) created

Mon, 2015-08-31 05:38

hrev49605 / haikuwebkit_x86-1.4.12-1-x86_gcc2

Tried to reproduce #12349 with:
LD_PRELOAD=/system/lib/x86/libroot_debug.so WebPositive https://easyweb.td.com

and got yet another backtrace.

Categories: Development

Ticket #12350 ([WebPositive] crash in WebCore::BUrlProtocolHandler::HeadersReceived) created

Mon, 2015-08-31 05:36

hrev49605 / haikuwebkit_x86-1.4.12-1-x86_gcc2

Tried to reproduce #12349 with:

LD_PRELOAD=/system/lib/x86/libroot_debug.so WebPositive https://easyweb.td.com

and got a different backtrace.

Categories: Development

Ticket #12349 (WebPositive crash on secure website) created

Mon, 2015-08-31 00:29

hrev49606 x86_gcc2
Go to: ​https://easyweb.td.com/
Webpositive crashes
Debug report attached

Categories: Development

Ticket #12338 (Haiku unbootable in VirtualBox since hrev49590) closed

Sun, 2015-08-30 23:11
fixed:

Fixed indeed, thank you.

Categories: Development

Ticket #12348 (Boot Failure Lenovo Ideapad z580) created

Sun, 2015-08-30 17:37

When booting latest nightly (rev 49600) I get:

PANIC: did not find any boot partitions!

I am booting from a CD. I also get the same thing if I create a USB and try to boot. I have the latest BIOS available for this laptop. I have also tried "safe mode" booting and get the same problem. Someone suggested turning off USB 3.0 but I do not have that option. I can only turn on/off legacy support for USB. I have tried both.

the specs are listed here: ​http://shop.lenovo.com/us/en/laptops/ideapad/z-series/z580/#tab-tech_specs

Categories: Development

Ticket #12347 ([regression] What happened to the mailbox deskbar replicant?) created

Sun, 2015-08-30 17:25

E-mail preflet used to offer (until January this year, I think) an option to activate a deskbar replicant that informed you about new incoming mail (the mailbox icon changed from empty to full) and gave you a kind of shortcut to your mail folders, Mail application and queries (see ​https://www.haiku-os.org/docs/userguide/en/workshop-email.html). As long as this replicant was already sitting in your deskbar before the changes from January (or you copied pre-January mail settings with an activated replicant to a new Haiku installation) you wouldn't notice a difference after updating with pkgman because the replicant would work just as before.
However, after a fresh install (with new mail settings) or after deleting the replicant, there seems no obvious way to activate the e-mail replicant again. The option seems to be gone from the E-mail preflet. The option to change the menu links (which used to be displayed by the e-mail replicant) is still there.

Categories: Development

Ticket #12345 ([Tracker] crashes on volume replicant removal) closed

Sun, 2015-08-30 14:56
duplicate:

Indeed.

Categories: Development

Ticket #12346 ([HaikuDepot] show requires/provides packages) created

Sun, 2015-08-30 11:27

It would be nice to be able to see requires/provides packages in HD.

Categories: Development

Ticket #12295 (Enhancement to the ahci driver) closed

Sun, 2015-08-30 10:44
fixed:

Applied in hrev49590. Unfortunately, the scontrol structure changed the access to a 32 bit field to 8 bits which broke some HW, and I didn't notice that.

Apart from that, I cleaned it up a little, and added a check for the ST flag in the command register as the specification suggests.

Categories: Development

Ticket #12345 ([Tracker] crashes on volume replicant removal) created

Sun, 2015-08-30 08:41

hrev49601.

To reproduce:

  • Drop volume replicant from Deskbar to Desktop.
  • Remove it.
Categories: Development

Ticket #11197 (Deskbar Volume replicant kills deskbar) closed

Sun, 2015-08-30 08:29
fixed:

Thanks for the update!

Categories: Development

Ticket #12344 ([Tracker] debug build crashes in BTrashWatcher::UpdateTrashIcons(void)) created

Sun, 2015-08-30 08:26

hrev49601.

  • Add SetConfigVar DEBUG : HAIKU_TOP src kits tracker : 1 : global ;

to build/jam/UserBuildConfig

  • Run Tracker
Categories: Development