Tiny Core 3.0 Alpha 4 Testing

[curaga]

System ram and swap sizes? Could you also post /proc/meminfo from around the time that happens?

Also, how many extensions are you loading?

Final edit, any relevant dmesg output?

[sandras]

Does anybody see (very rare) mount problems, like one extension out of many that does not mount on /tce/tcloop the first time?

Repeating the mount or tce-load command solves the problem.

I see these rare mount problems from time to time in the Alpha releases.  The mount fails with a "Cannot allocate memory" message, even when there is ample memory available.  I didn't see this behaviour in the 2.x releases.

I can not reliably reproduce this problem.  It happens randomly.

happens to me, when using Appbrowser. Also, once I had to drop some things off of onboot.lst, because TC showed the same error.

[u54749]

>>> System ram and swap sizes? 
512 M system RAM with 512 M swap size

>>> Could you also post /proc/meminfo from around the time that happens?
you will have to wait until the problem manifests itself.  I have seen it happening five times or so since 3.0 Alpha is available.

>>> Also, how many extensions are you loading?
I have a quite heavily tuned setup:  I load 96 extensions in one single mount.  This mount I never have seen fail until now.  But I load Openoffice and the compiler/development stuff in the classic TC way with tce-load -i xxx.tcz and that's when I see the problem happening.

>>> Final edit, any relevant dmesg output?
I see nothing special.  The kernel sees all the RAM and sees the swap space.  Memory consumption of my environment is moderate (I think):  with Xorg, jwm, and the (ROX)-desktop loaded, dbus and CUPS running and the 96 extensions mounted  and available I consume 43,5 Megabytes so there is absolutely no RAM scarcity.

[u54749]

I had the mount problem happening for the first time today

exact message that was issued
mount: mounting /dev/loop5 on /tmp/tcloop/sed failed: Cannot allocate memory

this is the result of the free command just after it happened
                    total         used            free       shared      buffers
Mem:        514192       502596        11596            0       124088
Swap:       511992        10500       501492
Total:      1026184       513096       513088

Here is /proc/meminfo
MemTotal:         514192 kB
MemFree:           11720 kB
Buffers:          123940 kB
Cached:           199760 kB
SwapCached:        10004 kB
Active:           196632 kB
Inactive:         272268 kB
Active(anon):      65328 kB
Inactive(anon):   104736 kB
Active(file):     131304 kB
Inactive(file):   167532 kB
Unevictable:           0 kB
Mlocked:               0 kB
HighTotal:             0 kB
HighFree:              0 kB
LowTotal:         514192 kB
LowFree:           11720 kB
SwapTotal:        511992 kB
SwapFree:         501456 kB
Dirty:               720 kB
Writeback:             0 kB
AnonPages:        136580 kB
Mapped:            42216 kB
Shmem:             24848 kB
Slab:              27676 kB
SReclaimable:      12572 kB
SUnreclaim:        15104 kB
KernelStack:         936 kB
PageTables:         1364 kB
NFS_Unstable:          0 kB
Bounce:                0 kB
WritebackTmp:          0 kB
CommitLimit:      769088 kB
Committed_AS:     386536 kB
VmallocTotal:     508476 kB
VmallocUsed:        1148 kB
VmallocChunk:     504600 kB
DirectMap4k:       11708 kB
DirectMap4M:      512000 kB

There is interesting stuff in dmesg:  here are the relevant lines:

mount: page allocation failure. order:4, mode:0xd0
Pid: 3203, comm: mount Not tainted 2.6.33.3-tinycore #2012
Call Trace:
 [<c015a343>] ? 0xc015a343
 [<c01742a8>] ? 0xc01742a8
 [<c0226ecb>] ? 0xc0226ecb
 [<c01744c5>] ? 0xc01744c5
 [<e07af6ff>] ? 0xe07af6ff
 [<c022b8d8>] ? 0xc022b8d8
 [<c0195070>] ? 0xc0195070
 [<c0179a36>] ? 0xc0179a36
 [<e07af5e6>] ? 0xe07af5e6
 [<e07af6b1>] ? 0xe07af6b1
 [<c0179604>] ? 0xc0179604
 [<c01797e0>] ? 0xc01797e0
 [<c0189011>] ? 0xc0189011
 [<c011a080>] ? 0xc011a080
 [<c0187c5b>] ? 0xc0187c5b
 [<c01890c4>] ? 0xc01890c4
 [<c03bcc35>] ? 0xc03bcc35
Mem-Info:
DMA per-cpu:
CPU    0: hi:    0, btch:   1 usd:   0
Normal per-cpu:
CPU    0: hi:  186, btch:  31 usd:   0
active_anon:15742 inactive_anon:24482 isolated_anon:0
 active_file:36247 inactive_file:42147 isolated_file:0
 unevictable:0 dirty:18 writeback:0 unstable:0
 free:3647 slab_reclaimable:2650 slab_unreclaimable:2346
 mapped:10159 shmem:6072 pagetables:330 bounce:0
DMA free:2056kB min:88kB low:108kB high:132kB active_anon:36kB inactive_anon:216kB active_file:1404kB inactive_file:7748kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:15872kB mlocked:0kB dirty:0kB writeback:0kB mapped:48kB shmem:36kB slab_reclaimable:88kB slab_unreclaimable:16kB kernel_stack:8kB pagetables:0kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 491 491 491
Normal free:12532kB min:2788kB low:3484kB high:4180kB active_anon:62932kB inactive_anon:97712kB active_file:143584kB inactive_file:160840kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:503360kB mlocked:0kB dirty:72kB writeback:0kB mapped:40588kB shmem:24252kB slab_reclaimable:10512kB slab_unreclaimable:9368kB kernel_stack:768kB pagetables:1320kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
DMA: 0*4kB 1*8kB 0*16kB 0*32kB 2*64kB 1*128kB 1*256kB 1*512kB 1*1024kB 0*2048kB 0*4096kB = 2056kB
Normal: 2135*4kB 299*8kB 78*16kB 7*32kB 2*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 12532kB
84724 total pagecache pages
259 pages in swap cache
Swap cache stats: add 562, delete 303, find 118/133
Free swap  = 510568kB
Total swap = 511992kB
130927 pages RAM
0 pages HighMem
2411 pages reserved
65787 pages shared
80734 pages non-shared
SQUASHFS error: Failed to allocate zlib workspace

sed is the first mount happening in my development tools install script, which consists of nothing more than a sequence of tce-load -i  commands.
the mounts that came immediately after went all through without problems
running the script a second time mounted sed without problems

[sandras]

About my earlier posts about local=* and copy2fs:
Tried that setup again and it seems everything works fine. Seems like it was a mistake on my part, so sorry to waste everybody's time. Still, I don't remember and/or understand what did i do wrong the first time.

[althalus]

Found another problem - I've built 2 compositing managers (xcompmgr and cairo-compmgr) which I've been testing before I submit to the repo.

If i install the graphics-KERNEL extension for TC3a4, both compositing managers breaks down (blank screen with occasional flicker of open windows).

Xcompmgr has been tested and is working in TC2.11 with the appropriate graphics-KERNEL for 2.x installed.

I can make the xcompmgr available for others to test with, but it's not really ready to submit as a proper extension yet (no info file, for starters)

EDIT: Probably important to mention
lspci

Code: [Select]

01:05.0 VGA compatible controller: ATI Technologies Inc RS482 [Radeon Xpress 200M]

[bigpcman]

I created a bootable flash card using the "usb install" tool then tried booting from it on a eeepc 900a.

Using boot codes "waitusb=15 base norestore pause tce=UUID="3EA1-E917" " everything comes up fine.
Note the "tce=UUID="3EA1-E917" " is inserted by the "install usb" tool.

Using boot codes "waitusb=15  norestore pause tce=UUID="3EA1-E917" " I get a stream of "bus error" messages right after the "loading Tiny Core Applications Extensions..." message. There are no extensions installed at this point.

After I install the 915resolution.tcz and reboot I get the same results - "bus errors".

However, if I add the "showapps" boot code then I get the extension loaded messages "including the Patch mode 50 to resolution 1024x600 complete" and the "bus error" messages go away.

Any ideas as to what is going on?

edit: tried the same above procedure using a usb stick instead of a flash card and the stream of "bus error" messages is reduced to one message when no extensions are installed. After I install 915resolution I get a message after the "loading Tiny core applications extensions" as follows: udev[96]: can not read '/etc/udev/rules.d/75-cd-dvd.rules'

edit2: I just tried all of the above with tc2.11.1 and everything works perfect! Great release!

[curaga]

bigpcman, could you post your dmesg? Also, on a boot you see the udev error, the output of "ls -lh /etc/udev/rules.d".
Udev outputs that error when it can't read the file, or the file is empty.

[bigpcman]

bigpcman, could you post your dmesg? Also, on a boot you see the udev error, the output of "ls -lh /etc/udev/rules.d".
Udev outputs that error when it can't read the file, or the file is empty.

I just noticed that the eeepc_laptop was not loaded at boot. Here is the 2.11.1 lsmod report:

Code: [Select]

Module                  Size  Used by    Not tainted
vfat                    5652  1
fat                    29692  1 vfat
squashfs               11732  1
scsi_wait_scan           260  0
eeepc_laptop            5264  0
hwmon                    640  1 eeepc_laptop
atl1e                  19596  0
backlight               1404  1 eeepc_laptop
battery                 5976  0
rfkill                  4012  2 eeepc_laptop
ac                      1732  0

and here is tc3alpha4 lsmod report

Code: [Select]

Module                  Size  Used by    Not tainted
vfat                    5596  1
fat                    30220  1 vfat
squashfs               14884  1
ramzswap               10240  1
loop                    8068  2
scsi_wait_scan           276  0
sparse_keymap           1168  0
battery                 6028  0
video                  12712  0
backlight               1632  1 video
ac                      1696  0
output                   724  1 video
atl1e                  19316  0

Does this explain the problems?

I tried to do a "modprobe eeepc_laptop" but it failed with the message "unknown symbol in module or unknown parameter"

See attached dmesg report for the usb stick boot failure. It seems to be the worst case. I reran the tests with a different usb stick so the "/tce" name is different than in my previous post.

edit: The following is reported in the syslog report:

Code: [Select]

May 25 16:39:29 (none) user.warn kernel: eeepc_laptop: Unknown symbol hwmon_device_register
May 25 16:39:29 (none) user.warn kernel: eeepc_laptop: Unknown symbol pci_hp_deregister
May 25 16:39:29 (none) user.warn kernel: eeepc_laptop: Unknown symbol __pci_hp_register
May 25 16:39:29 (none) user.warn kernel: eeepc_laptop: Unknown

See attached syslog:

[maro]

I was wondering whether it would be desirable to use "KERNEL" within an extension name used as parameter of 'tce-load'. E.g.
    tce-load -wi filesystems-KERNEL
instead of
    tce-load -wi filesystems-$( uname -r)

AFAIK the "-KERNEL" placeholder is only used for dependency files but not available from the CLI.

[curaga]

@bigpcman: there were no errors in the dmesg posted. I wonder if syslog would catch them better.

I'm a bit torn about the eeepc module - all other platform modules are in the base, and work fine, but eeepc now requires both hwmon and pci-hotplug. Separating it on its own wouldn't improve much, and separating all platform drivers wouldn't make sense either.

[bigpcman]

@bigpcman: there were no errors in the dmesg posted. I wonder if syslog would catch them better.

I'm a bit torn about the eeepc module - all other platform modules are in the base, and work fine, but eeepc now requires both hwmon and pci-hotplug. Separating it on its own wouldn't improve much, and separating all platform drivers wouldn't make sense either.

I have updated my previous post with the syslog output. It does indeed show eeepc_laptop does not load.
Also, the ath9k module for wireless does not load on boot. Although unlike eeepc_laptop it does load using "sudo modprobe ath9k" and seems to work fine with wpa_supplicant.

The error messages are a pain because they overrun the boot screen wiping away all the other info.

edit: BTW the eeepc_laptop module has worked going all the way back to version tc2.2. It would be a shame to not continue the previous automatic operation.

[jur]

I am having no problems on my eeepc 1000he. I am using tc3.0a4 as my daily OS at home.

[curaga]

@bigpcman: Nothing in messages.txt either. If you haven't already, please check the md5sums of all extensions, the base image, and do a fsck. Also, is your kernel current? There was an update after the first alphas.

On eeepc, try loading the hwmon and pci-hotplug extensions.

[althalus]

Just thought I'd point out again that there seems to be a problem with the graphics extension, on radeon cards, at least.

Games that need graphics-2.6.29.1-tinycore.tcz (Games from the Humble Indie Bundle) to run at full speed in 2.11 are acting as if graphics-2.6.33.33-tinycore.tcz hasn't been installed when run under TC3.