Page 3 of 15
Posted: Tue 12 Feb 2013, 13:46
by Ray MK
Manual frugal install of 5.5alpha1(5.4.90) with k3.2.29 to an ntfs partition and booting via grub4dos from an SDcard. Emachines 732 laptop.
So far all seems good to this average user - everything seems to work as expected.
Very low on resource usage and temperature. Nice.
Report for your info.
Posted: Tue 12 Feb 2013, 15:13
by Hogweed
Installed the non PAE build and everything seems fine. In fact more than fine. There's a phenomenal improvement in Flash Video that I can't explain.
I am using Seamonkey 2.15.2 and Flash 11.2.202.262 in both RetroPrecise 5.4.3 and 5.4.91 Alpha builds. glxgears reports 60 fps in both builds.
Decent resolution BBC iplayer Flash videos normally play ok for me in all versions of Puppy at their standard resolution window but normally are very jumpy or totally unwatchable in full screen mode (on this old laptop that is). However in the 5.4.91 alpha build the videos play fine in full screen mode. Flash hardware acceleration is enabled in both but Flash reports it is using software decoding and rendering in both cases.
I can also play higher resolution Youtube videos in full screen which I couldn't play before. The difference is like night and day.
I have never seen Flash perform as well in any previous build of Puppy Linux I've tried. Yes they all have used the Intel graphics driver. Anyone any ideas why the massive performance jump?
Computer
Processor Intel(R) Celeron(R) M processor 1.40GHz
Memory 741MB (309MB used)
Operating System Unknown distribution
User Name root (root)
Date/Time Tue 12 Feb 2013 15:00:32 GMT
Display
Resolution 1280x800 pixels
OpenGL Renderer Mesa DRI Intel(R) 852GM/855GM x86/MMX/SSE2
X11 Vendor The X.Org Foundation
Version
Kernel Linux 3.8.0-rc7 (i686)
Compiled #1 SMP Mon Feb 11 13:55:51 GMT-8 2013
C Library GNU C Library version 2.15 (stable)
Default C Compiler Unknown
Distribution Unknown distribution
Posted: Tue 12 Feb 2013, 15:51
by rerwin
BarryK wrote:I don't yet know whether to include Frisbee in 5.5beta (the next build).
I will rely on feedback, how well it works in 5.5alpha1, and whether it stomps on the toes of the other connection systems or not.
Barry,
In testing the frisbee-1.0-20120209 package on my wifi-less PC, I am seeing problems. It aborts silently (actually hangs) when it accesses an invalid interface, "none". I am digging into it. Please consider this when deciding whether to include it. Maybe with a warning to not try it if no wifi hardware is installed. Or let users get the latest from the "frisbee beta" thread, where I work with others to perfect it.
BTW, I have already upgraded network_tray-2.6 to correct and improve handling of dialup connections. I will post it in the frisbee-beta thread.
Richard
UPDATE: I have fixed the above problem and uploaded newer versions of frisbee and the network_tray in the Frisbee beta thread:
frisbee:
http://www.murga-linux.com/puppy/viewto ... 092#684092
network_tray:
http://www.murga-linux.com/puppy/viewto ... 694#684694
Barring any surprises, they should be ready for the 5.5 beta.
Richard
Re: Problem with report-video
Posted: Tue 12 Feb 2013, 16:34
by ETP
BarryK wrote:ETP wrote:Currently picks up my pci DVB-T card details rather than my graphics card.
ETP,
Would you mind running these two, and post them here (this forum thread):
# scanpci | gzip - > scanpci.log.gz
# lspci | gzip - > lspci.log.gz
Barry,
Logs are attached from 5.4.91 which has the same issue.
Code: Select all
# report-video
VIDEO REPORT: Precise Puppy, version 5.4.91
Chip description:
Philips Semiconductors SAA7131/SAA7133/SAA7135 Video Broadcast Decoder
Requested by /etc/X11/xorg.conf:
Probing Xorg startup log file (/var/log/Xorg.0.log):
Loaded modules: dbe dri dri2 exa extmod fb glx kbd mouse record shadowfb
Actual rendering on monitor:
Resolution: 1920x1080 pixels (507x285 millimeters)
Depth: 24 planes
...the above also recorded in /tmp/report-video
Code: Select all
video-info-glx 1.5.1 Tue 12 Feb 2013 on Precise Puppy 5.4.91 Linux 3.8.0-rc7 i686
0.0 VGA compatible controller: NVIDIA Corporation GT218 [GeForce 210] (rev a2)
oem: NVIDIA
product: GT218 Board - 08740000 Chip Rev
X Server: Xorg Driver: nouveau
X.Org version: 1.11.3
dimensions: 1920x1080 pixels (507x285 millimeters)
depth of root window: 24 planes
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.4
OpenGL vendor string: nouveau
OpenGL renderer string: Gallium 0.4 on NVA8
OpenGL version string: 2.1 Mesa 8.0.4
Intel(R) Pentium(R) D CPU 3.40GHz
Core 0: @3400 1: @3400 MHz
...the above also recorded at '/tmp/root/video-info-glx'.
mount.cifs broken
Posted: Tue 12 Feb 2013, 17:21
by jrb
Hi Barry,
Thanks for another great puppy.
Prescise 5491
I couldn't get Pnethood or YASSM to connect to my Samba shares. I tried connecting on command line using the command I normally use in other Puppies:
Code: Select all
# mount.cifs //192.168.1.120/usr=root_passwd=woofwoof /root/YASSM/usr=root_passwd=woofwoof@LENOVO-120 -o username=root,password=woofwoof
mount error 22 = Invalid argument
Refer to the mount.cifs(8) manual page (e.g.man mount.cifs)
Further investigation reveals there are two different mount.cifs, /sbin/mount.cifs (18k, 11feb13) and /usr/sbin/mount.cifs (23k, 7jul2010). Deleting either one doesn't seem to help.
problems with ACER aspire 7715z
Posted: Tue 12 Feb 2013, 17:36
by gilloraymondo
Hi,
I tried the 2 versions : 5.4.91 and 5.4.90 and I have the same thing than with all others precise versions :
The last thing I see on the screen is "loading kernel modules".
But looking very carefuly, it can (hardly) see that the precise desktop is there.
If I link my computer to another screen with a cable, it's works on the other screen......
Please excuse my bad english
Is there anything to do with this problem ?
On Internet, I have seen that everybody has this problem with this computer and this new kernel (not only with puppy, but also with ubuntu)
Friendly yours
Posted: Tue 12 Feb 2013, 17:54
by rcrsn51
@jrb: Both YASSM and Pnethood are working fine for me. But I don't have any very long share names like you do.
This was testing 5490. Downloading 5491 now...
Posted: Tue 12 Feb 2013, 17:55
by vicmz
BarryK wrote:vicmz wrote: I've seen this in Precise (the alpha version too), both in JWM and Openbox: when I right-click a file from ROX and select either Copy, Rename, or Move (several times, the one I use more frequently is Copy), the screen turns black, showing only the taskbar, then if I try to use other programs the taskbar disappears and all is finally black. I restart X and keep doing things as usual, then after many times copying files from ROX right-click menu I get the black screen again.
vicmz,
With ROX, I just drag a file, then when I release the left-button a menu pops up asking whether want to copy, move or link.
Does this way work OK for you?
Yes, it does. I usually copy that way, and use the right-click menu when it's a single file. Renaming a file, however, doesn't have a second option (not one that doesn't imply a second application, at least), yet the bug is so weird you can't tell when the screen will turn black after clicking the Rename option. I think I'll post again about this when I get some specific information, maybe it's something I installed.
Posted: Tue 12 Feb 2013, 18:15
by jrb
rcrsn51 wrote:@jrb: Both YASSM and Pnethood are working fine for me. But I don't have any very long share names like you do.
This was testing 5490. Downloading 5491 now...
I just downloaded 5490 and both Pnethood and YASSM work fine. It also has two different mount.cifs. One in /sbin and one in /usr/sbin.
Posted: Tue 12 Feb 2013, 18:31
by rcrsn51
Yes, there is definitely something wrong with mount.cifs in 5491. I tried a bunch of alternate syntaxes, but nothing worked.
Posted: Tue 12 Feb 2013, 18:59
by Ray MK
Manual frugal install of 5.5alpha1(5.4.91) with k3.8.rc7 to an ntfs partition and booting via grub4dos from an SDcard. Emachines (Acer) 732 laptop.
Boot and setup ok but - no wifi.
Guessing that the needed broadcom bits are not in new kernel.
Or maybe not activated?
Is there a pet or something I can try - or is it more complex than that?
Hardinfo report if it helps for your info.
JFYI - Puppy 5.4.90 works fine on the same laptop with identical setup.
(posting from 5.4.90 now)
Posted: Tue 12 Feb 2013, 19:24
by pemasu
Ray MK. Yeah. Your broadcom BCM43225 would need wl or brcmsmac driver.
brcmsmac relies nowadays in bcma. So...when there is this:
CONFIG_BCMA is not set
and not this:
CONFIG_BCMA=m
CONFIG_BCMA_BLOCKIO=y
CONFIG_BCMA_HOST_PCI_POSSIBLE=y
CONFIG_BCMA_HOST_PCI=y
you wont see this selection as tickable option:
CONFIG_BRCMSMAC=m
and so the kernel wont have brcmsmac driver as module included.
So the 3.8 rc kernel does not have your needed brcmsmac driver and it has not the proprietary wl module either.
Posted: Tue 12 Feb 2013, 20:24
by peebee
rcrsn51 wrote:Yes, there is definitely something wrong with mount.cifs in 5491. I tried a bunch of alternate syntaxes, but nothing worked.
Both YASSM and PNetHood work fine for me in 5491.....my samba share is running on Slacko 5.4 on my desktop
Posted: Tue 12 Feb 2013, 21:38
by rcrsn51
@peebee: I'm guessing that your shares do not require authentication?
Because some new kernels have a bug in how they parse the line of options.
@jrb: Try this:
Code: Select all
mount.cifs //host/share mount-point -o user=xxx
At the prompt, type in the password.
precise-5.4.91-k3.8rc7-nopae
Posted: Tue 12 Feb 2013, 21:42
by Keef
I've reported before that having a CF card in PCMCIA adapter will stall booting on precise pups (the exception being Pemasu's Upup Precise 3.7.2). Removing the adapter will allow booting to continue, but the USB socket will no longer function. Previously I could insert the card after boot up and there would be no problem, but now the system will either freeze completely or disable the USB port again.
This is from /var/log/messages, and shows what is happening when I re-insert the card:
Code: Select all
19:49:16 puppypc32605 user.notice kernel: [ 548.140049] pcmcia_socket pcmcia_socket0: pccard: PCMCIA card inserted into slot 0
Feb 13 19:49:16 puppypc32605 user.info kernel: [ 548.140216] pcmcia_socket pcmcia_socket0: cs: memory probe 0x0d4000-0x0effff:
Feb 13 19:49:16 puppypc32605 user.warn kernel: [ 548.146597] clean.
Feb 13 19:49:16 puppypc32605 user.info kernel: [ 548.146611] pcmcia_socket pcmcia_socket0: cs: memory probe 0x60000000-0x60ffffff:
Feb 13 19:49:16 puppypc32605 user.warn kernel: [ 548.153887] clean.
Feb 13 19:49:16 puppypc32605 user.info kernel: [ 548.153899] pcmcia_socket pcmcia_socket0: cs: memory probe 0xa0000000-0xa0ffffff:
Feb 13 19:49:16 puppypc32605 user.warn kernel: [ 548.161099] clean.
Feb 13 19:49:16 puppypc32605 user.notice kernel: [ 548.161352] pcmcia 0.0: pcmcia: registering new device pcmcia0.0 (IRQ: 11)
Feb 13 19:49:16 puppypc32605 user.emerg kernel: [ 548.204570] NMI: PCI system error (SERR) for reason a9 on CPU 0.
Feb 13 19:49:16 puppypc32605 user.emerg kernel: [ 548.204583] Dazed and confused, but trying to continue
Feb 13 19:49:16 puppypc32605 user.info kernel: [ 548.205603] scsi2 : pata_pcmcia
Feb 13 19:49:16 puppypc32605 user.info kernel: [ 548.206065] ata3: PATA max PIO0 cmd 0x100 ctl 0x10e irq 11
Feb 13 19:49:57 puppypc32605 user.notice kernel: [ 589.004867] pcmcia_socket pcmcia_socket0: pccard: card ejected from slot 0
Feb 13 19:50:02 puppypc32605 user.notice kernel: [ 594.263376] pcmcia_socket pcmcia_socket0: pccard: PCMCIA card inserted into slot 0
Feb 13 19:50:02 puppypc32605 user.notice kernel: [ 594.263763] pcmcia 0.0: pcmcia: registering new device pcmcia0.0 (IRQ: 11)
Feb 13 19:50:03 puppypc32605 user.info kernel: [ 594.307132] scsi3 : pata_pcmcia
Feb 13 19:50:03 puppypc32605 user.info kernel: [ 594.307411] ata4: PATA max PIO0 cmd 0x100 ctl 0x10e irq 11
Feb 13 19:50:03 puppypc32605 user.err kernel: [ 594.387799] uhci_hcd 0000:00:07.2: host controller process error, something bad happened!
Feb 13 19:50:03 puppypc32605 user.err kernel: [ 594.387810] uhci_hcd 0000:00:07.2: host controller halted, very bad!
Feb 13 19:50:03 puppypc32605 user.err kernel: [ 594.387834] uhci_hcd 0000:00:07.2: HC died; cleaning up
Compaq Armada M700
Processor : Pentium III (Coppermine)
Memory : 578MB (212MB used)
Operating System : Unknown distribution
User Name : root (root)
Date/Time : Tue 12 Feb 2013 21:34:52 GMT
-Display-
Resolution : 1024x768 pixels
OpenGL Renderer : Gallium 0.4 on softpipe
X11 Vendor : The X.Org Foundation
-Multimedia-
Audio Adapter : ES1978 - ESS ES1978 (Maestro 2E)
-Input Devices-
AT Translated Set 2 keyboard
HID 413c:3010
TPPS/2 IBM TrackPoint
Sleep Button
Lid Switch
Power Button
PC Speaker
ES1978
-Printers (CUPS)-
CUPS-PDF : <i>Default</i>
-SCSI Disks-
ATA TOSHIBA MK4032GA
COMPAQ CD-224E
Other than that it is running quite well - possibly a little snappier than previous precise versions.
Posted: Tue 12 Feb 2013, 22:15
by peebee
rcrsn51 wrote:@peebee: I'm guessing that your shares do not require authentication?
Because some new kernels have a bug in how they parse the line of options.
@jrb: Try this:
Code: Select all
mount.cifs //host/share mount-point -o user=xxx
At the prompt, type in the password.
I do enter a password for root and I did setup samba as instructed by setting a new password for both linux and samba (same).
start in powersave mode
Posted: Tue 12 Feb 2013, 22:53
by anikin
Hi all,
I would like to apply elroy's tip, that starts the machine in powersave mode.
Code: Select all
#!/bin/bash
killall acpid
if [ -f /root/.acpid_on ] ; then
exec acpid &
acpitool -a |grep off-line
if [[ $? == 0 ]] ; then
/etc/acpi/powersave.sh
else
/etc/acpi/performance.sh
fi
exec ppower &
fi
/etc/acpi/powersave.sh
However, I can't find the 0acpid script in root/Startup. Every pup that I know has it in that directory, but not this one. Can I just inject my own file there?
This build (5.4.91 Linux 3.8.0-rc7) runs
extremely cool on my Intel atom n270/945GME eeepc. The above trick, as I hope will make it run even cooler.
Thank you in advance.
Posted: Tue 12 Feb 2013, 22:57
by jrb
rcrsn51 wrote:@peebee: I'm guessing that your shares do not require authentication?
Because some new kernels have a bug in how they parse the line of options.
@jrb: Try this:
Code: Select all
mount.cifs //host/share mount-point -o user=xxx
At the prompt, type in the password.
Code: Select all
# mount.cifs //192.168.1.120/usr=root_passwd=woofwoof /root/YASSM/usr=root_passwd=woofwoof@LENOVO-120 -o user=root
Password:
mount error 22 = Invalid argument
Refer to the mount.cifs(8) manual page (e.g.man mount.cifs)
Posted: Tue 12 Feb 2013, 23:42
by rcrsn51
My mistake. My Win7 shares work fine in YASSM, as does my NAS box. the problem is specific to my Samba-TNG server.
Maybe it's an smb1 vs smb2 issue.
Posted: Tue 12 Feb 2013, 23:57
by jrb
rcrsn51 wrote:My mistake. My Win7 shares work fine in YASSM, as does my NAS box. the problem is specific to my Samba-TNG server.
Maybe it's an smb1 vs smb2 issue.
I never thought to try connecting to a windows machine. Works OK.
