Hello,
I have my 4GB SanDisk with the Fedora live, I've copied the developer key to /security/develop.sig.
Upon booting the XO, I get
Type the Esc key to interrupt automatic startup SDHCI: Card didn't power up after 1 second Boot device: /sd/disk:\boot\olpc.fth Arguments: Boot device: /sd/disk:\syslinux\vmlinuz0 Arguments: root=UUID=6462-3436 riitfstype=vfat rw liveimg overlay=UUID=6462-3436 quiet rhgb persistenthome=mtd0 reset_overlay SDHCI: Card didn't power up after 1 second buffer@ffb676d0:33: Can't open boot device
Then I do
dir sd:\boot
on the ok prompt, the first attempts returns
SDHCI: Card didn't power up after 1 second Can't open directory
but when I rerun the dir command, I get the /boot listing (showing OLPC.FTH there). So the XO seems to be able to talk to the SD card, at least somehow.
When I try
boot sd:\boot\olpc.fth
I get
Boot device: /sd/disk:\boot\olpc.fth Arguments: root=UUID=6462-3436 riitfstype=vfat rw liveimg overlay=UUID=6462-3436 quiet rhgb persistenthome=mtd0 reset_overlay SDHCI: Card didn't power up after 1 second
Can't open boot device
and when I do it again I get
Boot device: /sd/disk:\boot\olpc.fth Arguments: root=UUID=6462-3436 riitfstype=vfat rw liveimg overlay=UUID=6462-3436 quiet rhgb persistenthome=mtd0 reset_overlay buffer@ffb651d8:2: Can't find word to replace
but when I do just
boot
Fedora starts to boot.
Is the
SDHCI: Card didn't power up after 1 second
message an expected thing?
On Sun, 2008-10-26 at 20:54 +0100, Jan Pazdziora wrote:
Upon booting the XO, I get
Type the Esc key to interrupt automatic startup SDHCI: Card didn't power up after 1 second
What version of the firmware do you have on the XO? It should say on bootup. If it's not Q2E18, you'll want to upgrade
Jeremy
On Sun, Oct 26, 2008 at 05:53:26PM -0400, Jeremy Katz wrote:
On Sun, 2008-10-26 at 20:54 +0100, Jan Pazdziora wrote:
Upon booting the XO, I get
Type the Esc key to interrupt automatic startup SDHCI: Card didn't power up after 1 second
What version of the firmware do you have on the XO? It should say on bootup. If it's not Q2E18, you'll want to upgrade
I've checked that I have Q2E18.
On Mon, Oct 27, 2008 at 12:37:49PM +0100, Jan Pazdziora wrote:
On Sun, Oct 26, 2008 at 05:53:26PM -0400, Jeremy Katz wrote:
On Sun, 2008-10-26 at 20:54 +0100, Jan Pazdziora wrote:
Upon booting the XO, I get
Type the Esc key to interrupt automatic startup SDHCI: Card didn't power up after 1 second
What version of the firmware do you have on the XO? It should say on bootup. If it's not Q2E18, you'll want to upgrade
I've checked that I have Q2E18.
I've run badblocks on the SD card and it seems to be completely fine.
When I type
boot
Fedora starts booting, there seem to be some error messages under the bar at the bottom of the screen (they are overwritten quite fast but I can take a picture if it seems important) and when the bar at the bottom of the screen is all white, the screen clears and I get
-------------------------------------- WARNING: Cannot find root file system! --------------------------------------
Create symlink /dev/root and then exit this shell to continue the boot sequence.
bash: no job control in this shell bash-3.2#
I'm affraid I'm facing some strange nondeterministic problem, provided other people seem to be able to boot the Snap3 just fine, and for me it seems to give error message but then starts to boot, and when part of the boot process finishes, it breaks again.
On Mon, Oct 27, 2008 at 7:12 PM, Jan Pazdziora jpazdziora@redhat.com wrote:
I'm affraid I'm facing some strange nondeterministic problem, provided other people seem to be able to boot the Snap3 just fine, and for me it seems to give error message but then starts to boot, and when part of the boot process finishes, it breaks again.
Interestingly, I'm hitting the exact same problem with the Xfce LiveCD of rawhide 20081111. Firmware Q2E18, fresh SD card (just got my XO).
Is there something more we can do to diagnose this?
On Wed, Nov 12, 2008 at 12:26 PM, Jon Stanley jonstanley@gmail.com wrote:
On Mon, Oct 27, 2008 at 7:12 PM, Jan Pazdziora jpazdziora@redhat.com wrote:
I'm affraid I'm facing some strange nondeterministic problem, provided other people seem to be able to boot the Snap3 just fine, and for me it seems to give error message but then starts to boot, and when part of the boot process finishes, it breaks again.
Interestingly, I'm hitting the exact same problem with the Xfce LiveCD of rawhide 20081111. Firmware Q2E18, fresh SD card (just got my XO).
I had the same issue, but it went away after upgrading to the latest firmware. Now I just need to fix the borken keyboard.
On Tue, Nov 11, 2008 at 8:30 PM, Rob K robk@ningaui.net wrote:
I had the same issue, but it went away after upgrading to the latest firmware. Now I just need to fix the borken keyboard.
Later than Q2E18? Where would one get such a thing?
On Wed, Nov 12, 2008 at 12:33 PM, Jon Stanley jonstanley@gmail.com wrote:
On Tue, Nov 11, 2008 at 8:30 PM, Rob K robk@ningaui.net wrote:
I had the same issue, but it went away after upgrading to the latest firmware. Now I just need to fix the borken keyboard.
Later than Q2E18? Where would one get such a thing?
Apologies for the confusion, I meant latest firmware == Q2E18.
On Tue, Nov 11, 2008 at 8:38 PM, Rob K robk@ningaui.net wrote:
On Wed, Nov 12, 2008 at 12:33 PM, Jon Stanley jonstanley@gmail.com wrote:
On Tue, Nov 11, 2008 at 8:30 PM, Rob K robk@ningaui.net wrote:
I had the same issue, but it went away after upgrading to the latest firmware. Now I just need to fix the borken keyboard.
Later than Q2E18? Where would one get such a thing?
Apologies for the confusion, I meant latest firmware == Q2E18.
-- Rob K http://ningaui.net I swear, if I collected all seven dragonballs, I'd bring back Jon Postel. - Raph
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
Jon,
I believe this is an issue with the cards themselves, I haven't yet proved that theory but inode0 and myself are both getting the exact same problems with the 8gig cards we just bought, but I have no problem at all with my 4 gig card and if I recall you are trying to use a 16gig card.
The XO can see the larger cards just fine if you boot straight into sugar but if you try to boot to the cards, we either get the message you are or unable to locate boot device.
We even went so far as to split the 8 gig card in half and only use 4 gigs with the rest unpartitioned but that was a fail as well.
On Tue, Nov 11, 2008 at 8:58 PM, Brian Powell bpowell01@gmail.com wrote:
I believe this is an issue with the cards themselves, I haven't yet proved that theory but inode0 and myself are both getting the exact same problems with the 8gig cards we just bought, but I have no problem at all with my 4 gig card and if I recall you are trying to use a 16gig card.
Trying to use both a 4 and 16gig card gives identical results, which are quite interesting:
initial boot fails, with $SUBJECT error dir sd: works try dir sd: again, fails with $SUBJECT try again, it works try again, fail.
Exactly every other time, it doesn't work. Oddness.
On Tue, Nov 11, 2008 at 9:39 PM, Jon Stanley jonstanley@gmail.com wrote:
Exactly every other time, it doesn't work. Oddness.
Update:
Typing 'test pci/sd' at the ok prompt until it succeeds every time will lead to a successful boot and so far uneventful desktop. James noted that the Xfce team needed a leader, so I tried with an Xfce livecd that I composed today :) More stats on that to come in a separate thread. :)
On Tue, Nov 11, 2008 at 09:39:26PM -0500, Jon Stanley wrote:
On Tue, Nov 11, 2008 at 8:58 PM, Brian Powell bpowell01@gmail.com wrote:
I believe this is an issue with the cards themselves, I haven't yet proved that theory but inode0 and myself are both getting the exact same problems with the 8gig cards we just bought, but I have no problem at all with my 4 gig card and if I recall you are trying to use a 16gig card.
Trying to use both a 4 and 16gig card gives identical results, which are quite interesting:
initial boot fails, with $SUBJECT error dir sd: works try dir sd: again, fails with $SUBJECT try again, it works try again, fail.
Exactly every other time, it doesn't work. Oddness.
The same results here -- I have a 4 GB card, and the flipping success / failure result is the same here. Which means that if I manage to boot (success), it then fails with that Cannot find root file system message (failure).
Boy, I'm so relived that I'm not the only one with this problem -- I just thought I was crazy with everyone else booting just fine.
On Wed, Nov 12, 2008 at 2:47 AM, Jan Pazdziora jpazdziora@redhat.com wrote:
The same results here -- I have a 4 GB card, and the flipping success / failure result is the same here. Which means that if I manage to boot (success), it then fails with that Cannot find root file system message (failure).
Does 'test /pci/sd' several times work for you as well? I do it til it succeeds about 10 times in a row, then type boot, and then it works fine. Question is, I'm not sure how/where to report these results, it seems as though this is a definite firmware issue (possibly with a subset of cards) and not any of the cards in question (since the same results now are happening for four people - me, you, John Rose ,and Brian Powell. Interestingly, John and Brian work fine with 4GB cards, just when they went to 8GB cards they ran into this. You and I see it with 4GB cards too).
On Wed, Nov 12, 2008 at 11:02:00AM -0500, Jon Stanley wrote:
On Wed, Nov 12, 2008 at 2:47 AM, Jan Pazdziora jpazdziora@redhat.com wrote:
The same results here -- I have a 4 GB card, and the flipping success / failure result is the same here. Which means that if I manage to boot (success), it then fails with that Cannot find root file system message (failure).
Does 'test /pci/sd' several times work for you as well? I do it til it succeeds about 10 times in a row, then type boot, and then it works fine. Question is, I'm not sure how/where to report these results, it
Well, it does not help here.
I get (when I remove that quiet rhgb from boot-file)
sdhci-pci: ... Invalid iomem size ... ... Starting udev: ... and here some errors about paths missing (they scrolled away to fast)
or with another boot, I get up to
Adding live user
and then a kernel stack dump ...
Trying to think about causes:
When I was flashing the firmware to the latest version, there were blocks shown on the screen, and some of them were red. I assume that means that some memory blocks are bad. I assumed that the XO can disable them but maybe this is what ruins my day?
Jan Pazdziora píše v St 12. 11. 2008 v 22:33 +0100:
On Wed, Nov 12, 2008 at 11:02:00AM -0500, Jon Stanley wrote:
On Wed, Nov 12, 2008 at 2:47 AM, Jan Pazdziora jpazdziora@redhat.com wrote:
The same results here -- I have a 4 GB card, and the flipping success / failure result is the same here. Which means that if I manage to boot (success), it then fails with that Cannot find root file system message (failure).
Does 'test /pci/sd' several times work for you as well? I do it til it succeeds about 10 times in a row, then type boot, and then it works fine. Question is, I'm not sure how/where to report these results, it
Well, it does not help here.
I get (when I remove that quiet rhgb from boot-file)
sdhci-pci: ... Invalid iomem size ...
this is probably a "standard" message, because I see it too though I can boot from the SD card (the 20MB/s type) normally
relevant part of dmesg: sdhci: Secure Digital Host Controller Interface driver sdhci: Copyright(c) Pierre Ossman sdhci-pci 0000:00:0c.1: SDHCI controller found [11ab:4101] (rev 10) sdhci-pci 0000:00:0c.1: enabling device (0000 -> 0002) sdhci-pci 0000:00:0c.1: Invalid iomem size. You may experience problems. Registered led device: mmc0 mmc0: SDHCI controller on PCI [0000:00:0c.1] using DMA mmc0: new high speed SDHC card at address 9a88 mmcblk0: mmc0:9a88 SD04G 3979776KiB mmcblk0: p1
kernel version 2.6.27.4-68.fc10.i686
Dan
On Thu, Nov 13, 2008 at 11:57 AM, Dan Horák dan@danny.cz wrote:
Jan Pazdziora píše v St 12. 11. 2008 v 22:33 +0100:
On Wed, Nov 12, 2008 at 11:02:00AM -0500, Jon Stanley wrote:
On Wed, Nov 12, 2008 at 2:47 AM, Jan Pazdziora jpazdziora@redhat.com
wrote:
The same results here -- I have a 4 GB card, and the flipping success / failure result is the same here. Which means that if I manage to boot (success), it then fails with that Cannot find root file system message (failure).
Does 'test /pci/sd' several times work for you as well? I do it til it succeeds about 10 times in a row, then type boot, and then it works fine. Question is, I'm not sure how/where to report these results, it
Well, it does not help here.
I get (when I remove that quiet rhgb from boot-file)
sdhci-pci: ... Invalid iomem size ...
this is probably a "standard" message, because I see it too though I can boot from the SD card (the 20MB/s type) normally
relevant part of dmesg: sdhci: Secure Digital Host Controller Interface driver sdhci: Copyright(c) Pierre Ossman sdhci-pci 0000:00:0c.1: SDHCI controller found [11ab:4101] (rev 10) sdhci-pci 0000:00:0c.1: enabling device (0000 -> 0002) sdhci-pci 0000:00:0c.1: Invalid iomem size. You may experience problems. Registered led device: mmc0 mmc0: SDHCI controller on PCI [0000:00:0c.1] using DMA mmc0: new high speed SDHC card at address 9a88 mmcblk0: mmc0:9a88 SD04G 3979776KiB mmcblk0: p1
kernel version 2.6.27.4-68.fc10.i686
Dan
-- fedora-test-list mailing list fedora-test-list@redhat.com To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-test-list
Methinks it's a kernel problem. Have you tried with a vanilla kernel or something?
On Wed, Nov 12, 2008 at 11:02 AM, Jon Stanley jonstanley@gmail.com wrote:
Does 'test /pci/sd' several times work for you as well? I do it til it succeeds about 10 times in a row, then type boot, and then it works fine. Question is, I'm not sure how/where to report these results, it seems as though this is a definite firmware issue (possibly with a subset of cards) and not any of the cards in question (since the same results now are happening for four people - me, you, John Rose ,and Brian Powell. Interestingly, John and Brian work fine with 4GB cards, just when they went to 8GB cards they ran into this. You and I see it with 4GB cards too).
And another update. I reported this to the OLPC folks as http://dev.laptop.org/ticket/8987. After a marathon 5 hour debug session with Mitch Bradley, learning more about the Forth debugger than I ever wanted to know, we have a fix that will show up in firmware Q2E23.
On Fri, Nov 14, 2008 at 10:11:29PM -0500, Jon Stanley wrote:
On Wed, Nov 12, 2008 at 11:02 AM, Jon Stanley jonstanley@gmail.com wrote:
And another update. I reported this to the OLPC folks as http://dev.laptop.org/ticket/8987. After a marathon 5 hour debug session with Mitch Bradley, learning more about the Forth debugger than I ever wanted to know, we have a fix that will show up in firmware Q2E23.
So what did you find is that special difference in our cards / XOs / / whatever that makes it fail for us while it works OK for the rest of the guys?
On Tue, Nov 18, 2008 at 2:48 AM, Jan Pazdziora jpazdziora@redhat.com wrote:
So what did you find is that special difference in our cards / XOs / / whatever that makes it fail for us while it works OK for the rest of the guys?
That we don't know. There was some Marvel specific quirk that wasn't enabled.that was causing the issue for me at least. We found that if you read from the card and power it down, that was enough to make it unhappy, but if you didn't read from it, everything was fine.
Interestingly enough, the test firmware has only completely fixed the problem with my 16GB card. My 4GB still fails the first time, but any subsequent time works (so all that I have to do is type 'boot' and it boots right up). Stepping through the init with the debugger failed to find anything, the card powered up properly the first time.
On Fri, Nov 14, 2008 at 10:11:29PM -0500, Jon Stanley wrote:
On Wed, Nov 12, 2008 at 11:02 AM, Jon Stanley jonstanley@gmail.com wrote:
Does 'test /pci/sd' several times work for you as well? I do it til it succeeds about 10 times in a row, then type boot, and then it works fine. Question is, I'm not sure how/where to report these results, it seems as though this is a definite firmware issue (possibly with a subset of cards) and not any of the cards in question (since the same results now are happening for four people - me, you, John Rose ,and Brian Powell. Interestingly, John and Brian work fine with 4GB cards, just when they went to 8GB cards they ran into this. You and I see it with 4GB cards too).
And another update. I reported this to the OLPC folks as http://dev.laptop.org/ticket/8987. After a marathon 5 hour debug session with Mitch Bradley, learning more about the Forth debugger than I ever wanted to know, we have a fix that will show up in firmware Q2E23.
Hmm, I've been eagerly waiting for the q2e23 so that I could continue with the Fedora work. There is now new firmware listed on
http://wiki.laptop.org/go/Firmware
but it is
q2f01 - 2008-12-11 OLPC Firmware q2f01 First release of refactored EC code
Do you know if it is equivalent to that expected q2e23 and safe to use?
On Thu, Dec 11, 2008 at 05:25:03PM +0100, Jan Pazdziora wrote:
And another update. I reported this to the OLPC folks as http://dev.laptop.org/ticket/8987. After a marathon 5 hour debug session with Mitch Bradley, learning more about the Forth debugger than I ever wanted to know, we have a fix that will show up in firmware Q2E23.
Hmm, I've been eagerly waiting for the q2e23 so that I could continue with the Fedora work. There is now new firmware listed on
http://wiki.laptop.org/go/Firmware
but it is
q2f01 - 2008-12-11 OLPC Firmware q2f01 First release of refactored EC code
Do you know if it is equivalent to that expected q2e23 and safe to use?
I've upgraded the firmware to q2e24 and now Fedora 10 boots fine. I'll try to catchup with what you guys have been doing with your XOs in the meantime.
To do the upgrade, I had to have the firmware on USB flashdisk as having it on OLPC's nand leads to firmware image has bad internal crc error message.