{20} Detailed changes to bugs tagged 8.2.0 (25391 matches)

Arguments

Pull out details of the recent changes to release-critical bugs for convenient release management.

Results (1 - 100 of 25391)

1 2 3 4 5 6 7 8 9 10 11

#12: Type writing program (9 matches)

Ticket Author Date Created Field Description
#12 krstic Sep 1, 2006 3:16:30 AM priority

wishlist

#12 krstic Sep 1, 2006 3:16:30 AM type

enhancement

#12 krstic Sep 1, 2006 3:16:30 AM milestone

rev1 final

#12 cjl Apr 12, 2010 11:20:03 AM comment

I would consider Typing Turtle to be an adequate response to this ticket allowing it to be closed as fixed.

http://wiki.sugarlabs.org/go/Activities/Typing_Turtle

#12 cjl Apr 12, 2010 11:20:03 AM next_action

never set

#12 cjl Apr 12, 2010 11:20:03 AM verified

0

#12 cjb Apr 12, 2010 1:48:52 PM comment

Thanks, closing.

#12 cjb Apr 12, 2010 1:48:52 PM resolution

fixed

#12 cjb Apr 12, 2010 1:48:52 PM status

closed

#14: Time synchronization (19 matches)

Ticket Author Date Created Field Description
#14 ksankar Sep 3, 2006 12:22:42 AM comment

IMHO, the best way to achieve tight time synchronization, is to implement the IEEE 1588 algorithm for distributed precision time sync. It takes care of latency et al, would be interesting to see how it behaves over a multi-hop mesh.

I remember seeing somewhere that for NTP, there is a top level server name which when configured will do the distributed lookup. (WIll update this when I get more information)

NTP would require internet connectivity, which we might not be able to assume all the time (? or can we)

#14 ksankar Sep 3, 2006 12:23:09 AM owner

ksankar

#14 ksankar Sep 3, 2006 12:23:09 AM status

assigned

#14 krstic Sep 3, 2006 12:28:02 AM description

Clients

For multiplayer mesh networks to work at all well for sound and games, we need very tight, robust time synchronization, even more fun when you think about our suspend/resume for power management problem.

We don't yet know how accurate the TOY clocks may be on the machine, (typically not very), and the latency of a mesh network, particularly in areas requiring long hops and multiple hops, will be much more than is typically seen.

Questions:

  • Will existing NTP daemons cope with this? (if any)
  • If not, how do we get something that does?
  • If yes, which one copes best with this environment?
  • How quickly can two machines that begin communications for applications needing tight synchronization that are out of sync become synchronized? Do we need application level help?

Servers

  • Which NTP server should school servers run?
  • Autoconfiguration of the time server. We don't want manual intervention to be necessary. Anycast to find upstream servers, maybe?
  • Will schools want local time sources to use (e.g. GPS receivers).
#14 krstic Sep 3, 2006 12:28:02 AM comment

No, permanent connectivity may certainly not be assumed.

#14 krstic Sep 3, 2006 12:28:02 AM milestone

rev1 final

#14 ksankar Sep 3, 2006 1:37:29 AM comment

Replying to krstic:

No, permanent connectivity may certainly not be assumed.

  • No GPS assumed. Actually we do not need that kind of accuracy anyway
  • NTP through intermittent internet connection (or a server which is connected to the internet) is assumed, to setup the time whenever possible/needed
  • We can also ask the kid “Looks like I have lost my time, can you tell me what time it is” to get the current time.
  • Should provide a sync button in configuration settings somewhere – to sync with NTP – either with a school machine or with NTP via internet
  • Should provide distributed precision time sync API for applications that require precision time sync
  • We should not wake up just to sync and also not waste too much bandwidth, energy for time sync. So will be very conservative and if apps require time sync, they can call the API. I assume the apps that require precision time sync most probably will be using the network anyway and so a few more sync messages will not be that bad on energy consumption
  • Sync & accuracy two dimensions, we will provide both primitives
    • Sync thru IEEE 1588 and accuracy thru NTP or user input
  • Accuracy budget
    • Milliseconds – possible
    • Micro seconds – probable
    • Nano seconds – impossible
  • Need to experiment and document the accuracies and sync we can achieve, especially with the mesh network primitives
    • I will do that
  • When sync in a mesh, will assume one of the machines as master
  • Will use some algorithm to select the master clock (say the one with the highest IP) and all in that mesh will sync with that
  • Will use wisdom of crowds for sync i.e. if 5 machines show 3:00 PM and one shows 12:00 AM, we assume that one is wrong. It is possible to arrive at wrong time by this, for example kids from out of state visit a friend.
#14 jg Sep 3, 2006 10:35:21 AM comment

Andy Lippman pointed to Aggelos Bletsas' work as likely useful in the mesh environment we face.

#14 jg Mar 26, 2007 3:10:22 PM status

new

#14 jg Mar 26, 2007 3:10:22 PM milestone

CTest

#14 jg Mar 26, 2007 3:10:22 PM owner

wad

#14 jondo May 17, 2007 6:49:14 PM verified

0

#14 jondo May 17, 2007 6:49:14 PM cc

jondo

#14 kimquirk Sep 24, 2007 10:46:37 PM milestone

V1.1

#14 jg Sep 25, 2007 10:14:05 AM comment

Lots of things break without time being roughly correct. Having ntp configured will prevent a lot of support trouble.

#14 jg Sep 25, 2007 10:14:05 AM milestone

First Deployment, V1.0

#14 wad Sep 14, 2010 2:17:26 AM owner

martin.langhoff

#14 wad Sep 14, 2010 2:17:26 AM next_action

never set

#14 wad Sep 14, 2010 2:17:26 AM comment

I suspect we can declare this one closed.

I'm formally passing ownership to Martin Langhoff, who can decide.

#17: Conventional Email (12 matches)

Ticket Author Date Created Field Description
#17 jg Jul 26, 2006 3:56:22 PM component

infrastructure

#17 jg Jul 26, 2006 3:56:22 PM owner

krstic

#17 krstic Sep 1, 2006 3:20:58 AM milestone

rev1 final

#17 krstic Sep 1, 2006 3:20:58 AM priority

high

#17 krstic Oct 31, 2006 10:13:40 PM status

assigned

#17 jg Mar 26, 2007 3:13:26 PM owner

wad

#17 jg Mar 26, 2007 3:13:26 PM status

new

#17 jondo May 17, 2007 5:48:54 PM cc

jondo

#17 jondo May 17, 2007 5:48:54 PM verified

0

#17 sj Oct 10, 2007 3:38:09 PM comment

see Mark Doffman's recent work on Tinymail Python bindings for OLPC .

#17 wad Nov 25, 2009 9:58:57 PM owner

martin.langhoff

#17 wad Nov 25, 2009 9:58:57 PM next_action

never set

#18: Server IPv6 configuration (11 matches)

Ticket Author Date Created Field Description
#18 krstic Sep 1, 2006 3:22:33 AM description

The first problem is automating the allocation of the school's network from the address registrar.

Does network happen at first school server installation? Or is it assigned up front as part of the deployment process? What do we do at re-installation of the server bits? Somehow this all feels fraught with manual processes to go wrong.

Presume we've passed that hurdle.

The next one is how the school server is connected to the internet.

There are three cases and solutions here:

  1. server on a native IPv6 network: use the IPv6 allocated network for the school, and we're done
  2. server on a native IPv4 network, with routable address: Use 6to4, Use the IPv6 allocated network for the school, and we're done
  3. server on a native IPv4 network, natted, without routable address: Use tunneling: Lookup from DNS an anycast address to find the tunnel endpoint, set up tunnel, use the IPv6 allocated network for the school, and we're done.

This should be autodetected, and "just work", without having to explain to people all this stuff, or the three cases.

#18 krstic Sep 1, 2006 3:22:33 AM milestone

rev1 final

#18 ksankar Sep 4, 2006 1:21:29 PM comment
  • Shouldn’t this (as well as ticket #23 & #41) be part of a school network configuration specification.
  • Hopefully one of the artifacts we are working on is an expected/recommended school topology and services deployment pattern – detailing the connectivity, network services and interfaces/capabilities that applications can expect
#18 jg Mar 26, 2007 3:09:12 PM owner

wad

#18 jg Mar 26, 2007 3:09:12 PM milestone

CTest

#18 holger May 27, 2007 12:15:23 PM cc

holger

#18 holger May 27, 2007 12:15:23 PM verified

0

#18 kimquirk Sep 24, 2007 10:47:03 PM milestone

First Deployment, V1.0

#18 wad Sep 14, 2010 2:39:10 AM owner

martin.langhoff

#18 wad Sep 14, 2010 2:39:10 AM comment

Reassigning to its rightful owner.

#18 wad Sep 14, 2010 2:39:10 AM next_action

never set

#20: SPAM (12 matches)

Ticket Author Date Created Field Description
#20 imr@stanford.edu Jul 27, 2006 4:17:40 PM comment

I'm working on this to some extent in the email client: see wiki.laptop.org/go/Email.

#20 krstic Sep 1, 2006 3:24:54 AM milestone

rev1 final

#20 krstic Sep 1, 2006 3:24:54 AM priority

high

#20 krstic Sep 1, 2006 3:26:09 AM component

security

#20 krstic Oct 31, 2006 10:13:37 PM status

assigned

#20 jg Mar 26, 2007 3:15:25 PM status

new

#20 jg Mar 26, 2007 3:15:25 PM owner

wad

#20 jondo May 17, 2007 3:50:08 PM verified

0

#20 jondo May 17, 2007 3:50:08 PM cc

jondo

#20 wad Nov 25, 2009 9:59:52 PM priority

low

#20 wad Nov 25, 2009 9:59:52 PM next_action

never set

#20 wad Nov 25, 2009 9:59:52 PM owner

martin.langhoff

#25: I18N and Localization of OLPC applications (13 matches)

Ticket Author Date Created Field Description
#25 anonymous Jul 27, 2006 3:58:26 AM status

assigned

#25 anonymous Jul 27, 2006 3:58:26 AM owner

anonymous

#25 krstic Sep 1, 2006 3:29:17 AM status

new

#25 krstic Sep 1, 2006 3:29:17 AM component

distro

#25 krstic Sep 1, 2006 3:29:17 AM milestone

rev1 final

#25 blizzard Oct 2, 2006 6:20:24 PM owner

blizzard

#25 blizzard Oct 2, 2006 6:20:24 PM comment

My task.

#25 giangy Mar 1, 2007 5:01:04 AM comment

Replying to jg:

We need to test our I18N for the languages we support in the OLPC specific applications

And find people in country to help with the localizations.

Available to translate OLPC specific applications if Italian localization is needed/will be included.

#25 cjl Apr 12, 2010 11:24:35 AM comment

While testing is a continuing exercise, I am not sure what purpose is served by leaving this ticket open.

The localization list would be the place to raise specific questions.

http://lists.laptop.org/listinfo/localization

#25 cjl Apr 12, 2010 11:24:35 AM next_action

never set

#25 cjb Apr 12, 2010 1:49:11 PM resolution

fixed

#25 cjb Apr 12, 2010 1:49:11 PM comment

Thanks, closing.

#25 cjb Apr 12, 2010 1:49:11 PM status

closed

#60: Suspend/Resume from RAM (24 matches)

Ticket Author Date Created Field Description
#60 krstic Sep 1, 2006 3:42:41 AM priority

high

#60 krstic Sep 1, 2006 3:42:41 AM description

David Woodhouse: Do we also have some attempt at a resume-from-RAM path?

Jim Gettys: Not yet; Mitch will work on this right after Cafe' bringup.

#60 krstic Sep 1, 2006 3:42:41 AM milestone

rev1 final

#60 hai Sep 20, 2006 9:32:15 PM comment

Do we need VSA support for that?

#60 wmb@firmworks.com Sep 20, 2006 9:34:53 PM comment

We hope we won't need additional VSA support - but we will find out...

#60 cjb Oct 2, 2006 1:28:16 PM milestone

BTest-1

#60 cjb Oct 2, 2006 2:15:58 PM comment

Resolving this bug involves having enough resume code in LinuxBIOS for us to be able to start on driver suspend/resume testing; that testing isn't needed by BTest-1, though.

#60 cjb Oct 2, 2006 2:15:58 PM description

David Woodhouse: Do we also have some attempt at a resume-from-RAM path?

Jim Gettys: Not yet; Mitch will work on this right after Cafe' bringup.

#60 wmb@firmworks.com Oct 2, 2006 5:48:27 PM comment

The wakeup.patch attachment shows a proposed change to the LinuxBIOS startup that will reenter Linux after a wakeup. It compiles, but is otherwise untested; I'm posting it early in case someone wants to get started early while I'm in Taiwan.

Notes: a) This is probably not the right place to put the code. It should be in a separate file that is included in auto.inc . If someone could show me how to do it that way, I'd appreciate it.

b) For fast wakeup, you need to eliminate early startup messages. The first step is to dial down the loglevel - see *_LOGLEVEL in targets/olpc/rev_a/config.*.lb . 3 seems to be a good choice to eliminate the early messages.

c) The other message that needs to be removed is print_err("done cpuRegInit\n"); in src/mainboard/olpc/rev_a/auto.c

#60 wmb@firmworks.com Oct 2, 2006 5:48:27 PM status

assigned

#60 wmb@firmworks.com Oct 2, 2006 5:54:54 PM comment

What Linux needs to do to use the wakeup-via-LinuxBIOS patch:

During the suspend-to-RAM sequence, Linux should:

  • Save any important data at locations 0 and 4
  • Write the wakeup entry point address to location 4 (long)
  • Write 0x51ee4129 to location 0.

When LinuxBIOS comes out of reset upon waking up, it will turn on the memory controller and check for that signature at location 0. Upon seeing it, LinuxBIOS will jump to the address that is stored at location 4. LinuxBIOS doesn't set any of the other registers to specific values. The only guarantee is that the system is in 32-bit protected mode with 4GB identity-mapped code and data segments.

#60 dwmw2 Oct 31, 2006 9:09:17 AM cc

dwmw2@…

#60 jg Oct 31, 2006 9:15:53 AM milestone

BTest-2

#60 jg Dec 19, 2006 4:24:43 PM comment

We think B3 is a good target for real working resume... So putting this on its realistic milestone. Feel free to delight us, of course!

#60 jg Dec 19, 2006 4:24:43 PM milestone

BTest-3

#60 jg May 16, 2007 11:15:06 AM comment

Resume is working on GX, but we still have LX problems.

#60 jg May 16, 2007 11:15:06 AM keywords

power

#60 jg May 16, 2007 11:15:06 AM verified

0

#60 wmb@firmworks.com May 22, 2007 8:18:33 PM owner

jg

#60 wmb@firmworks.com May 22, 2007 8:18:33 PM comment

The firmware aspect of this problem is mostly under control; most of the work now is OS-related. Several individual OS engineers are involved, so it appears that the overall responsibility for this omnibus task is in the management domain. Therefore I am reassigning it back to management.

#60 wmb@firmworks.com May 22, 2007 8:18:33 PM status

new

#60 wmb@firmworks.com May 22, 2007 8:18:33 PM verified

0

#60 jg May 29, 2007 9:26:59 PM priority

blocker

#60 jg Jun 5, 2007 9:27:18 PM comment

SD resume will come in after BTest-4 with the update to Master.

1 2 3 4 5 6 7 8 9 10 11
Note: See TracReports for help on using and creating reports.