[12:32] <niktaris_lap> hi, can anyone please tell me where the patches for the ubuntu kernel are located ?
[12:34] <crimsun> see ubuntu-dapper.git and ubuntu-2.6.git. There're broken-out patches that I know of for Dapper and Edgy.
[12:34] <crimsun> there're _no_
[12:46] <niktaris_lap> crimsun, they are not in the kernel source .deb. are they?
[12:47] <mjg59> No
[12:47] <niktaris_lap> mjg59, didn't think so :)
[12:48] <sn9> you could also apt-get source
[12:50] <niktaris_lap> I am trying to figure out why a pcmcia card works with the standard ubuntu kernel but not with a vanilla kernel
[12:51] <sn9> wifi?
[12:51] <niktaris_lap> a vodafone 3g card
[12:52] <niktaris_lap> the brand is option
[12:52] <sn9> when it does work, what /dev node does it use?
[12:52] <niktaris_lap> sn9, I will have to reboot to find that out
[12:52] <sn9> no need
[12:53] <sn9> you use ppp with it, right?
[12:53] <niktaris_lap> sn9, I don't actually _use_ it
[12:54] <sn9> you don't use the card? then how did you know it worked?
[12:54] <niktaris_lap> because when I use the ubuntu kernel the system sees it as usb3 modem
[12:54] <niktaris_lap> and with vanilla not
[12:55] <sn9> that also answers my question
[12:55] <niktaris_lap> using kppp to query the modem
[12:56] <niktaris_lap> using a standard vanilla kernel I only get " CardBus card inserted into slot 0"
[12:56] <niktaris_lap> while with the ubuntu kernel the system goes on and configures it
[12:56] <sn9> and you load the same module on the vanilla kernel?
[12:57] <niktaris_lap> no I don't do anything at all
[12:57] <niktaris_lap> the ubuntu kernel does everything while vanilla does nothing
[12:57] <sn9> did you even compile the module for the vanilla kernel?
[12:58] <niktaris_lap> /lib/modules/2.6.17.11/kernel/drivers/usb/serial/option.ko
[12:59] <niktaris_lap> I even used the config from the ubuntu kernel
[12:59] <sn9> ok, what happens if you try to load that module manually on the vanilla kernel?
[01:01] <niktaris_lap> nothing
[01:01] <niktaris_lap> well the module loads
[01:01] <sn9> nothing in dmesg?
[01:01] <niktaris_lap> but besides that nothing
[01:02] <niktaris_lap> only [17185386.460000]  pccard: CardBus card inserted into slot 0
[01:04] <sn9> ok, so it's not associating the card with the driver. do all other cardbus cards work?
[01:05] <niktaris_lap> it's the only one I  got 
[01:06] <sn9> it could be pcmcia/cardbus support didn't get compiled with patches to use config files in /etc, or the driver itself may be a different version
[01:07] <sn9> you should modinfo both the vanilla version of the driver and the stock version and diff the output
[01:07] <niktaris_lap> option driver version v0.4
[01:08] <niktaris_lap> in vanilla
[01:08] <niktaris_lap> don't know what is in ubuntu atm
[01:09] <niktaris_lap> besides the giving the same config from the ubuntu kernel I can only image a patch is applied someplace 
[01:09] <niktaris_lap> that's why I asked 
[01:10] <sn9> run modinfo on both files and use the diff command
[01:10] <sn9> look for changes in lines with a bunch of numbers and codes
[01:10] <niktaris_lap> sn9, I uninstalled the ubuntu one. :(
[01:11] <niktaris_lap> use the diff command where?
[01:11] <sn9> but you can apt-get it again, right?
[01:11] <sn9> in a shell
[01:11] <niktaris_lap> sn9, yes. but later as I am apt-getting something else now
[01:13] <sn9> if you still have it in /var/cache/apt/archive, you might not need to wait
[01:13] <niktaris_lap> yes. wait...
[01:17] <niktaris_lap> < alias:          usb:v16D5p6501d*dc*dsc*dp*ic*isc*ip*
[01:17] <niktaris_lap> < srcversion:     84340500E79BAB5E3CB7869
[01:17] <niktaris_lap> ---
[01:17] <niktaris_lap> > srcversion:     E986026D4704EC0498CCF92
[01:17] <niktaris_lap> sn9, is this it ?
[01:18] <niktaris_lap> < aliias : 
[01:18] <niktaris_lap> is ubuntu
[01:20] <sn9> yes, the alias lines were what i meant
[01:20] <niktaris_lap> so what should I do ?
[01:20] <sn9> but the one you show is for usb, and you said this was a cardbus device
[01:21] <niktaris_lap> the option card is recognised as a usb device
[01:22] <sn9> so this is usb over cardbus... does lsusb show it with vanilla?
[01:22] <niktaris_lap> all 5 devices are listed as Device 001: ID 0000:0000
[01:23] <niktaris_lap> wrong! wait
[01:24] <niktaris_lap> no right ( I had the card unplugged but the output is the same in any case)
[01:24] <niktaris_lap> all 5 devices are listed as Device 001: ID 0000:0000
[01:24] <sn9> and there are no other usb devices?
[01:25] <niktaris_lap> connected? no
[01:25] <sn9> ok, it seems there are two things that are not happening
[01:26] <sn9> 1) the cardbus is not being seen as usb
[01:26] <sn9> 2) even if usb were seen, the option driver wouldn't recognize it
[01:27] <niktaris_lap> when connecting a usb stick I get Bus 001 Device 002: ID 0930:651c Toshiba Corp
[01:27] <niktaris_lap> on a usb port 
[01:27] <sn9> we have not even confirmed that #1 is supposed to happen at all
[01:28] <sn9> looks like you'll have to boot the ubuntu kernel at least once more (to see whether the card is in lsusb there)
[01:28] <niktaris_lap> ok give me 1 min
[01:39] <niktaris_lap> sn9, Bus 006 Device 002: ID 0af0:5000 Option UMTS Card
[01:39] <niktaris_lap> Bus 006 Device 001: ID 0000:0000
[01:39] <niktaris_lap> those are new
[01:40] <sn9> ok, do an lsmod and save that to a file
[01:41] <sn9> the option.ko is clearly not the only module involved
[01:41] <niktaris_lap> done
[01:44] <sn9> look through dmesg for the last mentions of usb
[01:45] <niktaris_lap> sn9, where can I paste ?
[01:46] <niktaris_lap> it's at least 20 lines
[01:46] <sn9> just the last mentions of it
[01:46] <niktaris_lap> last line : [17179703.044000]  ohci_hcd: 2005 April 22 USB 1.1 'Open' Host Controller (OHCI) Driver (PCI)
[01:46] <niktaris_lap> [17179703.048000]  PCI: Enabling device 0000:07:00.0 (0000 -> 0002)
[01:46] <niktaris_lap> [17179703.048000]  ACPI: PCI Interrupt 0000:07:00.0[A]  -> Link [LNKC]  -> GSI 11 (level, low) -> IRQ 11
[01:46] <niktaris_lap> [17179703.048000]  PCI: Setting latency timer of device 0000:07:00.0 to 64
[01:46] <niktaris_lap> [17179703.048000]  ohci_hcd 0000:07:00.0: OHCI Host Controller
[01:46] <niktaris_lap> [17179703.048000]  ohci_hcd 0000:07:00.0: new USB bus registered, assigned bus number 6
[01:46] <sn9> at least one of them should mention a module other than option.ko
[01:46] <niktaris_lap> [17179703.048000]  ohci_hcd 0000:07:00.0: irq 11, io mem 0xb2000000
[01:46] <niktaris_lap> [17179703.104000]  usb usb6: configuration #1 chosen from 1 choice
[01:47] <niktaris_lap> [17179703.104000]  hub 6-0:1.0: USB hub found
[01:47] <niktaris_lap> [17179703.104000]  hub 6-0:1.0: 2 ports detected
[01:47] <niktaris_lap> [17179712.260000]  ohci_hcd 0000:07:00.0: wakeup
[01:47] <niktaris_lap> [17179712.644000]  usb 6-1: new full speed USB device using ohci_hcd and address 2
[01:47] <niktaris_lap> [17179712.904000]  spurious 8259A interrupt: IRQ7.
[01:47] <niktaris_lap> [17179712.920000]  usb 6-1: configuration #1 chosen from 1 choice
[01:47] <niktaris_lap> [17179713.072000]  usbcore: registered new driver usbserial
[01:47] <niktaris_lap> [17179713.072000]  drivers/usb/serial/usb-serial.c: USB Serial support registered for generic
[01:47] <niktaris_lap> [17179713.072000]  usbcore: registered new driver usbserial_generic
[01:47] <niktaris_lap> [17179713.072000]  drivers/usb/serial/usb-serial.c: USB Serial Driver core
[01:47] <niktaris_lap> [17179713.104000]  drivers/usb/serial/usb-serial.c: USB Serial support registered for Option 3G data card
[01:47] <niktaris_lap> [17179713.104000]  option 6-1:1.0: Option 3G data card converter detected
[01:47] <niktaris_lap> [17179713.108000]  usb 6-1: Option 3G data card converter now attached to ttyUSB0
[01:47] <niktaris_lap> [17179713.108000]  option 6-1:1.1: Option 3G data card converter detected
[01:47] <niktaris_lap> [17179713.108000]  usb 6-1: Option 3G data card converter now attached to ttyUSB1
[01:47] <niktaris_lap> [17179713.108000]  option 6-1:1.2: Option 3G data card converter detected
[01:47] <niktaris_lap> [17179713.108000]  usb 6-1: Option 3G data card converter now attached to ttyUSB2
[01:47] <niktaris_lap> [17179713.108000]  usbcore: registered new driver option
[01:47] <niktaris_lap> [17179713.108000]  drivers/usb/serial/option.c: Option Card (PC-Card to) USB to Serial Driver: v0.4
[01:47] <niktaris_lap> shit!!!!
[01:47] <niktaris_lap> sorry!!!!
[01:47] <niktaris_lap> sn9, at least you see them all :S
[01:49] <sn9> does the laptop have any built-in usb1.1 ports, or are they all usb2?
[01:50] <niktaris_lap> all usb2 
[01:51] <sn9> then the other module involved is ohci_hcd.ko
[01:52] <sn9> that's in your lsmod, right?
[02:04] <niktaris_lap> sn9, so if I reboot and load ohci_hcd.ko, usbserial and option modules it should work ?
[02:05] <sn9> before you do that,
[02:06] <sn9> see which of those is not auto-loaded
[02:06] <sn9> diff the modinfo's for that one
[02:06] <niktaris_lap> for ohci_hdc.ko ?
[02:06] <sn9> yes, if that does not auto-load on vanilla
[02:07] <niktaris_lap> ok. rebooting
[02:13] <niktaris_lap> sn9, nothing changed
[02:14] <niktaris_lap> ohci-hdc does not auto load
[02:14] <niktaris_lap> moprode ohci-hdc, usbserial , option
[02:14] <sn9> diff the modinfo's for ohci_hcd
[02:14] <niktaris_lap> still nothing
[02:15] <sn9> look for changes in the alias lines
[02:17] <niktaris_lap> alias is the same
[02:18] <sn9> no change?
[02:18] <niktaris_lap> nope
[02:18] <sn9> in other words, alias does not show up in the diff?
[02:18] <niktaris_lap> yes
[02:19] <niktaris_lap> only filename and vermagic
[02:19] <sn9> if alias shows up in modinfo but not in the diff, and you modprobed it, look at dmesg
[02:20] <niktaris_lap> alias in modinfo shows. dmesg: [17179906.956000]  pccard: CardBus card inserted into slot 0
[02:21] <sn9> no mention of ohci in dmesg?
[02:21] <niktaris_lap> no
[02:21] <sn9> even though you modprobed ohci_hcd?
[02:22] <niktaris_lap> oh sorry 
[02:22] <niktaris_lap> yes
[02:22] <niktaris_lap> [17179708.164000]  ohci_hcd: 2005 April 22 USB 1.1 'Open' Host Controller (OHCI) Driver (PCI)
[02:22] <niktaris_lap> when modprobed
[02:22] <sn9> but no other mentions of ohci?
[02:22] <niktaris_lap> nothing more
[02:23] <sn9> i think you have an acpi or irq problem, then
[02:25] <sn9> save the entire dmesg to a file from both kernels, diff them, and look for the first mentions of acpi, apic, or pci in the diff
[02:26] <niktaris_lap> sn9, diff-ing the /boot/config files of both kernel I see some diff in acpi
[02:27] <niktaris_lap> so there should be a apci patch be applied to the ubuntu kernel
[02:28] <sn9> either that or kernel args
[02:28] <niktaris_lap> eg SCSI_SATA_ACPI exists in ubuntu kernel but not in vanilla
[02:29] <sn9> unless you actually have sata in your laptop, that is not the cause
[02:32] <niktaris_lap> diff are: CONFIG_ACPI_SBS=m  CONFIG_ACPI_PCC=m CONFIG_ACPI_SONY=m CONFIG_ACPI_DEV=m CONFIG_ACPI_TC1100=m CONFIG_ACPI_INITRD=y CONFIG_ACPI_DEV=m
[02:32] <niktaris_lap> those do not exist in vanilla
[02:33] <sn9> but that's in the .config
[02:33] <niktaris_lap> yes but doesn't that mean that the modules do not exist ?
[02:33] <sn9> are there acpi/apic/pci diffs in the dmesg?
[02:34] <niktaris_lap> I remember some acpi events when pluging in the card
[02:34] <niktaris_lap> have to reboot to see though
[02:34] <sn9> yes, it means the modules do not exist, but if they were never in lsmod, what does it matter?
[02:35] <sn9> i said to save the entire dmesg to a file under both kernels
[02:35] <niktaris_lap> ok I ll do that
[02:36] <sn9> diff those, looking for the first occurrences of acpi or apic or pci in the diff
[02:36] <niktaris_lap> ok
[02:36] <niktaris_lap> rebooting
[02:43] <niktaris_lap> sn9, fist thing that happens after card is inserted is ACPI: PCI Interrupt 0000:07:00.0[A]  -> Link [LNKC]  -> GSI 11 (level, low) -> IRQ 11
[02:44] <niktaris_lap> after PCI: Enabling device 0000:07:00.0 (0000 -> 0002
[02:44] <sn9> no diffs before it's inserted?
[02:46] <niktaris_lap> [17179569.184000]  Local APIC disabled by BIOS (or by default) -- you can enable it with "lapic"
[02:46] <niktaris_lap> in ubuntu 
[02:46] <niktaris_lap> [17179569.184000]  Local APIC disabled by BIOS -- you can enable it with "lapic"
[02:46] <niktaris_lap> in vanilla
[02:47] <sn9> that's the only diff before insertion?
[02:47] <niktaris_lap> from what I can see
[02:47] <niktaris_lap> let me check again
[02:51] <niktaris_lap> one diff is ACPI: Looking for DSDT ... not found!
[02:52] <sn9> on vanilla?
[02:54] <niktaris_lap> on ubuntu
[02:55] <sn9> and that line is just skipped on vanilla?
[02:56] <niktaris_lap> yes
[02:57] <mjg59> That code isn't in vanilla, hence the difference
[02:57] <sn9> but other than those two lines, the diffs only start after insertion, right?
[02:59] <niktaris_lap> after the insetion I get ACPI: PCI Interrupt 0000:07:00.0[A]  -> Link [LNKC]  -> GSI 11 (level, low) -> IRQ 11
[02:59] <sn9> i'll take that as a yes
[03:01] <niktaris_lap> there is http://gaugusch.at/acpi-dsdt-initrd-patches/acpi-dsdt-initrd-v0.8.2-2.6.17.patch
[03:01] <niktaris_lap> maybe this is included in ubuntu
[03:04] <mjg59> It is
[03:07] <niktaris_lap> mjg59, thanks 
[03:07] <niktaris_lap> let me apply this patch first and test again
[03:08] <niktaris_lap> thanks a million sn9 
[03:08] <niktaris_lap> thanks all
[03:08] <sn9> niktaris_lap: why did you want a vanilla kernel, anyway?
[03:08] <mjg59> That patch isn't going to change anything
[03:09] <niktaris_lap> mjg59, why?
[03:09] <mjg59> Because it's got nothing to do with the problem you have
[03:09] <niktaris_lap> sn9, I like vanilla kernels 
[03:09] <mjg59> Only three lines of it are actually running on your machine, and the only thing they do is print the message you're seeing
[03:10] <mjg59> niktaris_lap: Shame that they don't seem to work in this case...
[03:11] <niktaris_lap> mjg59, well something is different in ubuntu kernel . Since I use the same config I guess some patch does the diff. 
[03:11] <niktaris_lap> at least I found one :)
[03:13] <mjg59> There's several hundred patches in the Ubuntu kernel
[03:13] <mjg59> One of them causes the difference. I don't know which one, but it isn't that one.
[03:14] <niktaris_lap> mjg59, it would be nice if they where located someplace...
[03:16] <sn9> they are
[03:16] <sn9> if you apt-get source, you get them all
[03:22] <mjg59> They're in the git repository
[03:22] <mjg59> We don't keep them as independent patches
[03:23] <zul> we use to but its more harder to maintain if we didnt use a git repository
[03:26] <niktaris_lap> have to go
[03:26] <niktaris_lap> thanks again. 
[05:54] <niktaris_lap> hi, I downloaded the git tree, how do I see the ubuntu specific changes?
[05:54] <Keybuk> bisect it with other trees?
[05:55] <niktaris_lap> ubuntu vs vanilla
[06:15] <niktaris_lap> Keybuk, do I have to download the vanilla tree also ?
[06:17] <Keybuk> niktaris_lap: I'm afraid I don't know git
[06:18] <niktaris_lap> Keybuk, me neither :)
[06:20] <Keybuk> tbh, I've never understood why the kernel developers like it so much
[06:20] <Keybuk> to me it's always seemed complicated and obtuse
[06:22] <niktaris_lap> I am just trying to find the ubuntu specific changes (aka patches
[06:23] <niktaris_lap> I hope someone is kind enough to walk me through
[07:07] <Keybuk> that's difficult to judge
[07:07] <Keybuk> e.g. is ubuntu pulling the acpi tip an ubuntu specific change?
[07:09] <mjg59> Keybuk: Because its merging stuff is magic
[07:11] <Mithrandir> zul: what's the state of xen now?  You said some days ago you were going to package up some xen-restricted-modules?
[07:11] <mjg59> Keybuk: I merged acpi tip with our own acpi crack in under an hour, despite there being  158 of them
[07:11] <mjg59> Hm. That's not a good start.
[07:12] <mjg59> "This computer does not have enough space for temporary files. Windows installation needs at least 306 megabytes of space on any partitoin for temporary files"
[07:12] <mjg59> There's 12GB free!
[09:18] <niktaris_lap> hi again, so there is no easy way of finding out what patches are applied in the ubuntu kernel ?
[09:20] <Mithrandir> niktaris_lap: patches compared with what?
[09:20] <Mithrandir> zul: do you have your xen repo online somewhere?
[09:20] <niktaris_lap> Mithrandir, to vanilla
[09:21] <niktaris_lap> Debian would work also I guess
[09:21] <Mithrandir> niktaris_lap: download the kernel.org tree, download our tree, run diff -Nru
[09:21] <Mithrandir> or look at the git commit log and diff that with Linus's git tree
[09:21] <mjg59> But no, there's no trivial way to determine whether a given patch came from upstream or us
[09:22] <niktaris_lap> Mithrandir, I downloaded ubuntu tree. Don't know how to download the vanilla git tree. where a the git commit logs located?
[09:22] <Mithrandir> niktaris_lap: "git log" gives you the commit logs, iirc
[09:23] <Mithrandir> zul: it looks like the xen kernel could benefit from having the bd_claim patch applied.
[09:24] <niktaris_lap> Mithrandir, can I see the changes based on the commit number eg. commit cda2a116920d69ff82689b4f92e98e32d346c632
[10:41] <Mithrandir> zul: also, where's the domU?