Helium Sources (kernel patchset)

Message

hirakendu · Post by **hirakendu** » Tue Jun 05, 2007 9:20 pm

(Updated 20110403, latest is 2.6.38-helium0. )

Yet another linux kernel patchset, helium sources:

http://hirakendu.mooo.com/helium-sources/

README for patchset (somewhat obsolete):

http://hirakendu.mooo.com/helium-source ... ources.txt

README for precompiled sources (somewhat obsolete):

http://hirakendu.mooo.com/helium-source ... mpiled.txt

For details of a patch, see broken-out/ folder of corresponding patchset. (And see LATEST_IS* file for latest version. Currently 2.6.38-helium0.)

Short description : As close to vanilla kernel as it gets with goodies like reiser4, aufs, (squashfs), tuxonice, fbcondecor(gensplash). The ebuild supports each of these feature patches as USE flags, e.g.,

Code: Select all

[ebuild  N    ] sys-kernel/helium-sources-2.6.38-r2  USE="aufs config fbcondecor reiser4 tuxonice -build -deblob -symlink" 74,204 kB [1]

.

vipernicus · Post by **vipernicus** » Tue Jun 05, 2007 10:45 pm

hirakendu wrote:Here is yet another linux kernel patchset (as if there is a dearth of them ), helium sources :
http://hirakendu.mooo.com/helium-source ... .patch.bz2 (the patch)
http://hirakendu.mooo.com/helium-sources/ (etc)
Start with vanilla 2.6.21.

Summary : 2.6.22-rc4, suspend2, reiser4, genpatches (includes squashfs and fbsplash), unionfs, and powertop. Thats it! Nothing fancy . Thanks to beyond-sources (no more ?), kamikaze (klight?) sources, skunk-sources, viper + nicus sources, mm-patchset and ...

Would be more appropriate if you renamed it to 2.6.22-rc4-helium1.

hirakendu · Post by **hirakendu** » Tue Jun 05, 2007 11:20 pm

Actually, yes. Just kept it so to keep the name short

. Of course, if you meant to reflect the RC nature of the kernel, it should indeed be there. In fact, I forgot to mention like all other patchsets that it could be pretty unstable. As an aside, last few months have been a rough time for me for playing with kernels, since say december - perhaps since 2.6.18 or 2.6.19. Too many changes and new features, but overall the kernel releases have been unstable. Although there is hope. 2.6.21 has been somewhat usable on my athlon 64x2 machine (its a via k8t890 mobo) with a ton of bootloader options and workarounds. 2.6.22 looks pretty promising. Yesterday I was in a mood of euphoria with the powertop patches - for some reason my machine seems to be flying (perhaps could be just a placebo effect - but its only through powertop I came to know how much irqbalance and nolapic_timer were pulling it down - as of now interrupts per second are down to < 100

). So much joy that I thought to cook up a patchset :p. But missing the good old days of nitro-sources and beyond-sources.

I have been so irked over the last few months to see the oops messages, that today morning when I ssh'ed into this machine to see this on my newly compiled kernel (the usual suspect - nvidia-drivers 9755)...
WARNING: at mm/slab.c:777 __find_general_cachep()
[<c0188aef>] __kmalloc+0x9f/0xd0
[<f9970a4c>] os_alloc_mem+0x5e/0xa3 [nvidia]
[<f9650bfa>] _nv003401rm+0x16/0x28 [nvidia]
[<f9636546>] _nv002573rm+0x202/0x384 [nvidia]
[<f9659796>] rm_set_interrupts+0x142/0x15c [nvidia]
[<f962f5dd>] _nv004362rm+0x81/0xb8 [nvidia]
[<f965be1d>] _nv002557rm+0x3d/0x770 [nvidia]
[<f965c123>] _nv002557rm+0x343/0x770 [nvidia]
[<c016cdd0>] find_get_page+0x20/0x60
[<c016f7a8>] filemap_nopage+0x238/0x340
[<c018be57>] do_sync_read+0xc7/0x110
[<f9659a7c>] rm_ioctl+0x1c/0x24 [nvidia]
[<f996e3e9>] nv_kern_ioctl+0x2f0/0x361 [nvidia]
[<c0384641>] pci_create_sysfs_dev_files+0x141/0x310
[<f996e48f>] nv_kern_unlocked_ioctl+0x18/0x1d [nvidia]
[<f996e477>] nv_kern_unlocked_ioctl+0x0/0x1d [nvidia]
[<c0197afb>] do_ioctl+0x2b/0xc0
[<c0121116>] do_page_fault+0x346/0x640
[<c0197bec>] vfs_ioctl+0x5c/0x290
[<c0197e5d>] sys_ioctl+0x3d/0x70
[<c01040c2>] sysenter_past_esp+0x5f/0x85
[<c0384641>] pci_create_sysfs_dev_files+0x141/0x310
=======================
... I thought, its yet another oops (oh no, not again!). Fortunately, it was just a warning. The pleasant surprise - when I came to my office - X was still fine and machine didnt lock up (those are the most frequent things over last few months

). And so we live dangerously - but enjoying the flight

.

deno · Post by **deno** » Wed Jun 06, 2007 1:17 am

Hi! Cannot compile vmware-modules here with this kernel. 32 bit system.

Code: Select all

>>> Compiling source in /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work ...
 * Preparing vmmon module
Using 2.6.x kernel build system.
make -C /usr/src/linux/include/.. M=$PWD SRCROOT=$PWD/. modules
make[1]: Entering directory `/usr/src/linux-2.6.22-helium1'
make[1]: warning: jobserver unavailable: using -j1.  Add `+' to parent make rule.
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/linux/driver.o
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/linux/hostif.o
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/common/cpuid.o
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/common/hash.o
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/common/memtrack.o
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/common/phystrack.o
  CC [M]  /var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/common/task.o
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for C/ObjC but not for C++
cc1plus: warning: command line option "-Wdeclaration-after-statement" is valid for C/ObjC but not for C++
cc1plus: warning: command line option "-Wno-pointer-sign" is valid for C/ObjC but not for C++
cc1plus: warning: command line option "-Wstrict-prototypes" is valid for C/ObjC but not for C++
cc1plus: warning: command line option "-ffreestanding" is valid for C/ObjC but not for C++
include/asm/page.h: In function 'pte_t native_make_pte(long unsigned int)':
include/asm/page.h:112: error: expected primary-expression before ')' token
include/asm/page.h:112: error: expected ';' before '{' token
include/asm/page.h:112: error: expected primary-expression before '.' token
include/asm/page.h:112: error: expected `;' before '}' token
make[3]: *** [/var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only/common/task.o] Error 1
make[2]: *** [_module_/var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/work/vmmon-only] Error 2
make[1]: *** [modules] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.22-helium1'
make: *** [vmmon.ko] Error 2

!!! ERROR: app-emulation/vmware-modules-1.0.0.15-r1 failed.
Call stack:
  ebuild.sh, line 1615:   Called dyn_compile
  ebuild.sh, line 972:   Called qa_call 'src_compile'
  ebuild.sh, line 44:   Called src_compile
  ebuild.sh, line 1305:   Called linux-mod_src_compile
  linux-mod.eclass, line 516:   Called die

!!! Unable to make   auto-build VMWARE_VER=VME_S1B1 KERNEL_DIR=/usr/src/linux KBUILD_OUTPUT=/lib/modules/2.6.22-helium1/build.
!!! If you need support, post the topmost build error, and the call stack if relevant.
!!! A complete build log is located at '/var/tmp/portage/app-emulation/vmware-modules-1.0.0.15-r1/temp/build.log'.

vipernicus · Post by **vipernicus** » Wed Jun 06, 2007 3:08 am

deno wrote:Hi! Cannot compile vmware-modules here with this kernel. 32 bit system.

http://www.vmware.com/community/thread. ... 0&tstart=0

Personally I've gotten used to Virtualbox as an alternative.

hirakendu · Post by **hirakendu** » Wed Jun 06, 2007 4:40 am

Thanks for the update

. (Sorry for the delay. Had a 3 hr lecture to attend in between.) I have the same problem. But from the thread, it seems to be pretty broken. Aside, /usr/src/linux/include/asm/page.h seems fine (the lines around 110 to 113) and seems to be a syntax confusion between c and c++ (?). Using virtualbox for the moment

. (I mostly use vmware, just recently tried out virtualbox - and found it fine - happy with either of them

.)

chittoor · Post by **chittoor** » Wed Jun 06, 2007 6:42 am

Hi Das!! (Hirakendu) nice to see you here on Gentoo Forums. Will give your Helium a try

Indeed the 22 series will be most usable kernel in the libata era

. Hoping that the stable release comes out as promised

Phlogiston · Post by **Phlogiston** » Wed Jun 06, 2007 7:09 am

hirakendu wrote:Here is yet another linux kernel patchset (as if there is a dearth of them ), helium sources :
http://hirakendu.mooo.com/helium-source ... .patch.bz2 (the patch)
http://hirakendu.mooo.com/helium-sources/ (etc)
Start with vanilla 2.6.21.

Summary : 2.6.22-rc4, suspend2, reiser4, genpatches (includes squashfs and fbsplash), unionfs, and powertop. Thats it! Nothing fancy . Thanks to beyond-sources (no more ?), kamikaze (klight?) sources, skunk-sources, viper + nicus sources, mm-patchset and ...

I would give it a try if you include thinkpad stuff

hirakendu · Post by **hirakendu** » Wed Jun 06, 2007 9:19 am

Ok, added thinkpad stuff also (from kamikaze's patches

). Actually there was some conflict with the hdaps from powertop yesterday, so I whimsically ignored thinkpad patches yesterday at 3 am

. Anyway, here they are. I believe its 'just' the tp-smapi and some ipw, bluetooth, hdaps, ec stuff or is it more than that ?
Anyway, it doesnt matter to me as I dont have hdaps :p.

So as of now, from what I have checked, almost all other additional drivers from portage work, except vmware-modules. Also included ipw3945 (from kamikaze) in the kernel itself, as the one from portage wants things the old way as in 2.6.21.

So all is fine on my desktop (athlon64x2, via k8t890, nv gf 7300gs) and also on my notebook hp dv5000t (core duo t2050, 945 chipset, nv go force 7400, ipw3945 and a stupid intel hda with conexant codecs :p).

Phlogiston · Post by **Phlogiston** » Wed Jun 06, 2007 9:44 am

hirakendu wrote:Ok, added thinkpad stuff also (from kamikaze's patches ). Actually there was some conflict with the hdaps from powertop yesterday, so I whimsically ignored thinkpad patches yesterday at 3 am . Anyway, here they are. I believe its 'just' the tp-smapi and some ipw, bluetooth, hdaps, ec stuff or is it more than that ?
Anyway, it doesnt matter to me as I dont have hdaps :p.

So as of now, from what I have checked, almost all other additional drivers from portage work, except vmware-modules. Also included ipw3945 (from kamikaze) in the kernel itself, as the one from portage wants things the old way as in 2.6.21.

So all is fine on my desktop (athlon64x2, via k8t890, nv gf 7300gs) and also on my notebook hp dv5000t (core duo t2050, 945 chipset, nv go force 7400, ipw3945 and a stupid intel hda with conexant codecs :p).

Sounds good. And yes in my case its actually tp_smapi and this one updates hdaps too...

Phlogiston · Post by **Phlogiston** » Wed Jun 06, 2007 9:45 am

hirakendu wrote:Ok, added thinkpad stuff also (from kamikaze's patches ).

But where can I get it? =)

hirakendu · Post by **hirakendu** » Wed Jun 06, 2007 9:58 am

. forgot to mention, its the same name :
http://hirakendu.mooo.com/helium-source ... .patch.bz2

Phlogiston · Post by **Phlogiston** » Wed Jun 06, 2007 11:23 am

hirakendu wrote:. forgot to mention, its the same name :
http://hirakendu.mooo.com/helium-source ... .patch.bz2

and applies on 2.6.21

Phlogiston · Post by **Phlogiston** » Thu Jun 07, 2007 10:52 am

Hmm did you include linux phc? Becuase undervolting does not work here. And squashfs seems to not be there too:

Code: Select all

cat .config | grep SQU

gives nothing.

But at least it boots

Phlogiston · Post by **Phlogiston** » Thu Jun 07, 2007 11:00 am

Oh wait my fault

Phlogiston · Post by **Phlogiston** » Thu Jun 07, 2007 12:13 pm

Its me again

I can't boot but its seems to be a .22 issue. I heard that there are a lot of sata changes. Maybe thats why my kernel does not find my harddrive. Futhermore at boot it says that the option:

Code: Select all

libata.atapi_enabled=1

is ignored.

I checked the configuration and the same drivers under ata are active than with older kernels. Here is my lspci. (its an ibm thinkpad T43)

Code: Select all

00:00.0 Host bridge: Intel Corporation Mobile 915GM/PM/GMS/910GML Express Processor to DRAM Controller (rev 03)
00:01.0 PCI bridge: Intel Corporation Mobile 915GM/PM Express PCI Express Root Port (rev 03)
00:1c.0 PCI bridge: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) PCI Express Port 1 (rev 03)
00:1c.2 PCI bridge: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) PCI Express Port 3 (rev 03)
00:1d.0 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #1 (rev 03)
00:1d.1 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #2 (rev 03)
00:1d.2 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #3 (rev 03)
00:1d.3 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB UHCI #4 (rev 03)
00:1d.7 USB Controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) USB2 EHCI Controller (rev 03)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev d3)
00:1e.2 Multimedia audio controller: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Audio Controller (rev 03)
00:1e.3 Modem: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) AC'97 Modem Controller (rev 03)
00:1f.0 ISA bridge: Intel Corporation 82801FBM (ICH6M) LPC Interface Bridge (rev 03)
00:1f.2 IDE interface: Intel Corporation 82801FBM (ICH6M) SATA Controller (rev 03)
00:1f.3 SMBus: Intel Corporation 82801FB/FBM/FR/FW/FRW (ICH6 Family) SMBus Controller (rev 03)
01:00.0 VGA compatible controller: ATI Technologies Inc M22 [Mobility Radeon X300]
02:00.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5751M Gigabit Ethernet PCI Express (rev 11)
04:00.0 CardBus bridge: Ricoh Co Ltd RL5c476 II (rev 8d)
04:02.0 Ethernet controller: Atheros Communications, Inc. AR5212 802.11abg NIC (rev 01)

Futhermore the kernel config here: http://phpfi.com/240027

hirakendu · Post by **hirakendu** » Thu Jun 07, 2007 9:16 pm

oopsie. looks that i accidentally unsubscribed from the thread. (i have bookmarked this thread for now

.)

About the hard-drive not recognized, perhaps you are a victim of hda changing to sda for newer kernel. Btw, in my default .config, I have no ide support (i.e., the old ata/atapi/mfm/rll), and its only through sata and pata drivers. So, two options :
1) deselect pata drivers and choose the old ide support
or
2) change all references of /dev/hd* to /dev/sd* - both in bootloader and fstab. Keep a livecd like slax or dsl handy

.

And although not formally said, I have put up a 2.6.22-helium2 patch which is simply kamikaze-rc4 patch with 4k-stacks disabled by default and a patch for creating /dev/loop* entries from mm-patches. And so it also inherits phc patches

. Links :

http://137.110.115.18/helium-sources/2. ... .patch.bz2 (the patch : applies on 2.6.21 vanilla)
http://137.110.115.18/helium-sources/2.6.22-helium2/ (whatever)

PS : If anyone is using virtualbox, I suggest to upgrade to 1.4.0. On 1.3.8, my livecd with this same config just stops loading arbirtarily. (No messages, apparently just dozes off :p.) I dont know whether its a issue with the guest or host with this kernel. Anyway, on 1.4.0, its all fine.

hirakendu · Post by **hirakendu** » Thu Jun 07, 2007 9:34 pm

Some more updates on that :

1) Your kernel config looks fine to me. Possibly you can add "Generic PCI IDE Chipset Support" (under ata/atapi/mfm/rll support). Actually, so far as I know, there is no harm in enabling both the ide driver stack and pata driver stack in the kernel. In fact, can you just quickly see through the kernel boot messages scrolling by (do disable the 'quiet' option in bootloader) as to whether your drives are detected as sda (or hda) or something?

2) The libata.atapi_enabled=1 doesn't seem to bother me. The output of my :

$ dmesg | grep libata

is :

Kernel command line: root=/dev/sda5 vga=794 resume2=swap:/dev/sda6 libata.atapi_enabled=1 quiet noirqbalance acpi_sleep=s3_mode,s3_bios
libata version 2.21 loaded.

Some history about that option. Apparently its required only if you are using the pata drivers, which you have anyway disabled. Aside, it was required on thinkpads on earlier kernels for enabling dma on dvd drives. Incidentally, its not required on most systems. On my hp notebook, if and only if I disable ahci sata in bios, its required to enable dma on my dvd drive. (Ps : if I disable ahci sata, there is no dma on my dvd drive on windows

. And there was a long workaround to get ahci sata working on my windows when i reinstalled it

. Incidentally, windows xp install disks dont even recognize this intel ahci sata controller

.)

Phlogiston · Post by **Phlogiston** » Fri Jun 08, 2007 8:04 am

Ok first of all thanks for your kind help. I was always using sata drivers on this laptop (by the way it is a thinkpad and I exactly had those dma problems with my drive

), so I'm using /dev/sda.

So hmm I really don't know what to change to get it working I think. Furhtermore I have no options about sata/ide in my bios.

hirakendu · Post by **hirakendu** » Fri Jun 08, 2007 8:34 am

Hmm, seems funny because you have a pretty generic (usual intel ich6) chipset. May I suggest something - to deselect the ide drivers and go for the pata drivers. On the safe side, I am pretty sure my generic config will surely work

. (yes, it will take some time to compile because of great many drivers, but there will not be any performance gap.)

PS : Your kernel config is pretty clean - but you sure wanted smp disabled

? Local apic is also disabled. Btw, you also have pci express support disabled - do your graphics card and ethernet card still work ? (You also have disabled wireless). Oh ya, most importantly, I also see reiserfs disabled. Hope you are just using ext3.

Phlogiston · Post by **Phlogiston** » Fri Jun 08, 2007 9:06 am

hirakendu wrote:Hmm, seems funny because you have a pretty generic (usual intel ich6) chipset. May I suggest something - to deselect the ide drivers and go for the pata drivers. On the safe side, I am pretty sure my generic config will surely work . (yes, it will take some time to compile because of great many drivers, but there will not be any performance gap.)

PS : Your kernel config is pretty clean - but you sure wanted smp disabled ? Local apic is also disabled. Btw, you also have pci express support disabled - do your graphics card and ethernet card still work ? (You also have disabled wireless). Oh ya, most importantly, I also see reiserfs disabled. Hope you are just using ext3.

Hey man. I will try that with all ide stuff disabled.

to the PS: Yes I know. I selected everything by myself and only stuff I need

And I think there is no PCI-E in my laptop. Or hmm wait... maybe. But I never needed that one. The 3d support works with intel agp. Futhermore I don't like reiserfs, it screwed up a whole partition once and yes thats it.

hirakendu · Post by **hirakendu** » Fri Jun 08, 2007 3:39 pm

The pci-e stuff sounds interesting. Especially because your network card (in lspci) says its pci-e. Meanwhile, let me try out without pci-e on my notebook

. (the amusing thing at my machine is that nvidia-settings show its a 16x pci-e graphics card. wonder why intel_agp is also loaded as a module. perhaps still required.)

Reiserfs - its good. What more can I say

. I think I have been using it ever since I came to know what a filesystem is :p. Anyway, whatever works for you

. (PS : Never put the spanner on a working system, no matter how broken it is - as long as it works

.)

hirakendu · Post by **hirakendu** » Fri Jun 08, 2007 5:51 pm

Also added some precompiled sources. For a try, may use :
http://hirakendu.mooo.com/helium-source ... ed.tar.bz2

Phlogiston · Post by **Phlogiston** » Mon Jun 11, 2007 7:28 am

hey hirakendu

I disabled all the ide stuff and the kernel booted

So far so good, but the following did not work:

linux phc (undervolting) and thinkpad fanctontrol. Until that kernel adding the option "ibm_acpi.experimental=1" was sufficent to be able to control the fan speed. Do you know if there were any changes?

hirakendu · Post by **hirakendu** » Mon Jun 11, 2007 8:55 am

Nope, frankly I have no clue about undervolting and fan control. But tell me something - so without this thinkpad fan control, does your notebook cpu fan keep whining at full speed all the time? Because on my hp dv5000, I dont have any fancy things in the kernel (except the usual acpi fan support), and things seem fine - its quiet usually and whines only while doing intensive tasks.

Helium Sources (kernel patchset)

Helium Sources (kernel patchset)

Re: Helium Sources : 2.6.22-helium1

Re: Helium Sources : 2.6.22-helium1 (rc4)