View previous topic :: View next topic |
Author |
Message |
buddabrod Apprentice
Joined: 15 Oct 2006 Posts: 241 Location: Germany
|
Posted: Tue Mar 06, 2007 4:42 pm Post subject: |
|
|
Are you using nicksched?
Maybe it's that rsdl works fine.
Besides i can't compile saa7134 support into kernel, but just as module.. not that it matters, just a remark. |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Tue Mar 06, 2007 4:46 pm Post subject: |
|
|
buddabrod wrote: | Are you using nicksched?
Maybe it's that rsdl works fine.
Besides i can't compile saa7134 support into kernel, but just as module.. not that it matters, just a remark. |
now how could i use nicksched if plugsched isnt even in the kernel i will try and put plugsched back in one I get 2.6.21-rc2-skunk2.0 out.... gonna reboot now and take it for a test drive
...saa7134 is what type of driver im not familiar with it? _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
buddabrod Apprentice
Joined: 15 Oct 2006 Posts: 241 Location: Germany
|
Posted: Tue Mar 06, 2007 4:51 pm Post subject: |
|
|
Gnarr i meant ingosched
saa7134 is a driver for my tv-card
Device-drivers -> Multimedia Devices -> Video Capture Adapters |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Wed Mar 07, 2007 12:51 am Post subject: |
|
|
Bump... 2.6.21-rc2-skunk2.0 is out.... cpu scheduler slowdowns should be taken care of in -mm2.... also I reverted to Jake latest original genetics patches (the ones with out kobj) and the hardlocks seem to be gone....
Have fun.... _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Wed Mar 07, 2007 6:17 am Post subject: |
|
|
Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
bollucks l33t
Joined: 27 Oct 2004 Posts: 606
|
Posted: Wed Mar 07, 2007 8:36 am Post subject: |
|
|
rmh3093 wrote: | Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome |
WTF is a "genetic RSDL" ... ck asks... |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Wed Mar 07, 2007 1:45 pm Post subject: |
|
|
bollucks wrote: | rmh3093 wrote: | Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome |
WTF is a "genetic RSDL" ... ck asks... |
Genetic RSDL is my quick hack at applying the genetic lib to tune the very few tunables for the RSDL cpu scheduler...... just like its applied to ingosched cpuscheduler. _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Wed Mar 07, 2007 4:23 pm Post subject: |
|
|
rmh3093 wrote: | bollucks wrote: | rmh3093 wrote: | Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome |
WTF is a "genetic RSDL" ... ck asks... |
Genetic RSDL is my quick hack at applying the genetic lib to tune the very few tunables for the RSDL cpu scheduler...... just like its applied to ingosched cpuscheduler. |
RSDL is supposed to use nice levels to determine your timeslice. How this will work if someone nices processes? _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Wed Mar 07, 2007 4:57 pm Post subject: |
|
|
vipernicus wrote: | rmh3093 wrote: | bollucks wrote: | rmh3093 wrote: | Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome |
WTF is a "genetic RSDL" ... ck asks... |
Genetic RSDL is my quick hack at applying the genetic lib to tune the very few tunables for the RSDL cpu scheduler...... just like its applied to ingosched cpuscheduler. |
RSDL is supposed to use nice levels to determine your timeslice. How this will work if someone nices processes? |
well since rr_interval is based of the nice value and rr_interval is what gets mutated then any change in nice level by the userland should be reflected in the mutated rr_interval, so you should still be able to set default nice levels to apps.... there is a decay rate for "top_performing_genes" so crappy genes dont accidentally get promoted and bog down your system.... _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Wed Mar 07, 2007 5:06 pm Post subject: |
|
|
rmh3093 wrote: | vipernicus wrote: | rmh3093 wrote: | bollucks wrote: | rmh3093 wrote: | Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome |
WTF is a "genetic RSDL" ... ck asks... |
Genetic RSDL is my quick hack at applying the genetic lib to tune the very few tunables for the RSDL cpu scheduler...... just like its applied to ingosched cpuscheduler. |
RSDL is supposed to use nice levels to determine your timeslice. How this will work if someone nices processes? |
well since rr_interval is based of the nice value and rr_interval is what gets mutated then any change in nice level by the userland should be reflected in the mutated rr_interval, so you should still be able to set default nice levels to apps.... there is a decay rate for "top_performing_genes" so crappy genes dont accidentally get promoted and bog down your system.... |
How does this handle things when tasks expire? _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
tranquilcool Veteran
Joined: 25 Mar 2005 Posts: 1245
|
Posted: Wed Mar 07, 2007 5:13 pm Post subject: |
|
|
CC kernel/genetic-rsdl.o
kernel/genetic-rsdl.c: In function âgenetic_cpu_sched_initâ:
kernel/genetic-rsdl.c:420: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c: In function âcpu_take_stats_snapshotâ:
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:464: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c:464: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:465: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c:465: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:466: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c:466: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c: In function âcpu_throughput_calc_fitnessâ:
kernel/genetic-rsdl.c:528: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:531: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c: In function âcpu_latency_calc_fitnessâ:
kernel/genetic-rsdl.c:543: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:546: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c: In function âcpu_context_switch_calc_fitnessâ:
kernel/genetic-rsdl.c:559: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:562: error: dereferencing pointer to incomplete type
make[1]: *** [kernel/genetic-rsdl.o] Error 1
make: *** [kernel] Error 2
genetic-rsdl doesn't compile. above are the errors. i applied both patches. _________________ this is a strange strange world.
Last edited by tranquilcool on Wed Mar 07, 2007 5:19 pm; edited 1 time in total |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Wed Mar 07, 2007 5:17 pm Post subject: |
|
|
vipernicus wrote: | rmh3093 wrote: | vipernicus wrote: | rmh3093 wrote: | bollucks wrote: | rmh3093 wrote: | Bump... I added Genetic RSDL to 2.6.21-rc2-skunk2.1....... Now please keep in mind that this has not been benchmarked... and probably could be tuned
...testing and feedback welcome |
WTF is a "genetic RSDL" ... ck asks... |
Genetic RSDL is my quick hack at applying the genetic lib to tune the very few tunables for the RSDL cpu scheduler...... just like its applied to ingosched cpuscheduler. |
RSDL is supposed to use nice levels to determine your timeslice. How this will work if someone nices processes? |
well since rr_interval is based of the nice value and rr_interval is what gets mutated then any change in nice level by the userland should be reflected in the mutated rr_interval, so you should still be able to set default nice levels to apps.... there is a decay rate for "top_performing_genes" so crappy genes dont accidentally get promoted and bog down your system.... |
How does this handle things when tasks expire? |
well the decay rates and mutation rates are constantly manipulating/mutating/changing the values of the genes so when a task expires the will be a workload change, the workload change would be reflected in schedstats and more than likely a new gene will get promoted, if that gene sucks it gets fired and a new gene gets promoted and this process keeps going until a good candidate arises _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
vipernicus Veteran
Joined: 17 Jan 2005 Posts: 1462 Location: Your College IT Dept.
|
Posted: Wed Mar 07, 2007 7:43 pm Post subject: |
|
|
rmh3093 wrote: | vipernicus wrote: | How does this handle things when tasks expire? |
well the decay rates and mutation rates are constantly manipulating/mutating/changing the values of the genes so when a task expires the will be a workload change, the workload change would be reflected in schedstats and more than likely a new gene will get promoted, if that gene sucks it gets fired and a new gene gets promoted and this process keeps going until a good candidate arises |
So if tasks with varying nice levels constantly expire and new tasks with varying nice levels are spawned, then how can a good candidate ever be found with RSDL? _________________ Viper-Sources Maintainer || nesl247 Projects || vipernicus.org blog |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Wed Mar 07, 2007 8:28 pm Post subject: |
|
|
vipernicus wrote: | rmh3093 wrote: | vipernicus wrote: | How does this handle things when tasks expire? |
well the decay rates and mutation rates are constantly manipulating/mutating/changing the values of the genes so when a task expires the will be a workload change, the workload change would be reflected in schedstats and more than likely a new gene will get promoted, if that gene sucks it gets fired and a new gene gets promoted and this process keeps going until a good candidate arises |
So if tasks with varying nice levels constantly expire and new tasks with varying nice levels are spawned, then how can a good candidate ever be found with RSDL? |
well the average user doesnt renice every program they start so how could an algorithm that does that based on scheduler statistics & i/o workload stats not work? _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
deno Guru
Joined: 13 Sep 2006 Posts: 411
|
Posted: Thu Mar 08, 2007 5:25 pm Post subject: |
|
|
Quote: | and like always, if there is a driver/patch you want included just ask
|
How about suspend2? |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Sat Mar 10, 2007 2:56 pm Post subject: Re: 2.6.21-rc2-skunk2.2 "Hindu Kush" :: Genetic RS |
|
|
bump... 2.6.21-rc2-skunk2.2 includes new version of RSDL ( v0.28 ) also contains some a few -mm hot-fixes and the genetic hot-fixes _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
HecHacker1 Apprentice
Joined: 26 Jun 2003 Posts: 213 Location: UCSD
|
Posted: Sun Mar 11, 2007 1:03 am Post subject: |
|
|
Second request for suspend2. Laptop user here.
Right now I'm on viper4, and RSDL is working great for my system, but suspend2 crashes on boot up. I am very interested in Genetic RSDL too, i wonder if it will provide even better interactivity. |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Sun Mar 11, 2007 3:35 am Post subject: |
|
|
HecHacker1 wrote: | Second request for suspend2. Laptop user here.
Right now I'm on viper4, and RSDL is working great for my system, but suspend2 crashes on boot up. I am very interested in Genetic RSDL too, i wonder if it will provide even better interactivity. |
I have never really had great success suspending my laptops so I dont have a working system to test things on... Also -mm introduces a bunch of changes to the crypto api and my last 2 attempts to port suspend2 to -mm resulted in a kernel that crashes very ugly when trying to suspend and another version that freezes with no output... I will to bring in suspend2 once a new version is released.... or if I can learn how to use git i might be able to update the -mm branch of suspend2 (im not sure how old that is)
...with respects to Genetic RSDL, or genetics in general.... I think the genetic aio scheduler is awsome... my laptop runs way quieter (fans on less often) and I dont notice any performance loss, im using the genetic rsdl cpu scheduler also and I can emerge my world and build a kernel and the 3D beryl effects like leaf spread are still fairly snappy... interactivity while cd burning with genetics is really nice also _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
HecHacker1 Apprentice
Joined: 26 Jun 2003 Posts: 213 Location: UCSD
|
Posted: Sun Mar 11, 2007 4:14 am Post subject: |
|
|
rmh3093 wrote: | HecHacker1 wrote: | Second request for suspend2. Laptop user here.
Right now I'm on viper4, and RSDL is working great for my system, but suspend2 crashes on boot up. I am very interested in Genetic RSDL too, i wonder if it will provide even better interactivity. |
I have never really had great success suspending my laptops so I dont have a working system to test things on... Also -mm introduces a bunch of changes to the crypto api and my last 2 attempts to port suspend2 to -mm resulted in a kernel that crashes very ugly when trying to suspend and another version that freezes with no output... I will to bring in suspend2 once a new version is released.... or if I can learn how to use git i might be able to update the -mm branch of suspend2 (im not sure how old that is)
...with respects to Genetic RSDL, or genetics in general.... I think the genetic aio scheduler is awsome... my laptop runs way quieter (fans on less often) and I dont notice any performance loss, im using the genetic rsdl cpu scheduler also and I can emerge my world and build a kernel and the 3D beryl effects like leaf spread are still fairly snappy... interactivity while cd burning with genetics is really nice also |
yeah, i know suspend2 causes many problems for people trying to port it. I'll await your release, but in the mean time it looks like viper is updating his kernel. Maybe the suspend issues will be fixed. When suspend works right, it works great. I can press my power button, and have it trigger hibernate on my beryl/xfce4 desktop, and upon resume everything is ready to go. Much faster than booting! Of course, this is only when it works right (and lately it hasnt).
as for genetic, I started using it because it was the only thing that would take away pauses in my music back in the days of 2.6.12. It actually works... despite the cynicism many people have for genetic. |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Sun Mar 11, 2007 2:46 pm Post subject: |
|
|
tranquilcool wrote: | CC kernel/genetic-rsdl.o
kernel/genetic-rsdl.c: In function âgenetic_cpu_sched_initâ:
kernel/genetic-rsdl.c:420: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c: In function âcpu_take_stats_snapshotâ:
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:459: error: invalid application of âsizeofâ to incomplete type âstruct sched_infoâ
kernel/genetic-rsdl.c:464: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c:464: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:465: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c:465: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:466: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c:466: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c: In function âcpu_throughput_calc_fitnessâ:
kernel/genetic-rsdl.c:528: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:531: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c: In function âcpu_latency_calc_fitnessâ:
kernel/genetic-rsdl.c:543: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:546: error: dereferencing pointer to incomplete type
kernel/genetic-rsdl.c: In function âcpu_context_switch_calc_fitnessâ:
kernel/genetic-rsdl.c:559: error: âstruct rqâ has no member named ârq_sched_infoâ
kernel/genetic-rsdl.c:562: error: dereferencing pointer to incomplete type
make[1]: *** [kernel/genetic-rsdl.o] Error 1
make: *** [kernel] Error 2
genetic-rsdl doesn't compile. above are the errors. i applied both patches. |
I didnt even see this post sneak in there.... sorry, do you get this 2.2 also??? I think the issue has to do with kernel/Kconfig.cpusched I forget to make Genetic RSDL select schedstats and geneitclib so make sure you have: GENETIC_LIB (in the libraries section) and SCHEDSTATS (in the kernel hacking section) _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
tranquilcool Veteran
Joined: 25 Mar 2005 Posts: 1245
|
Posted: Sun Mar 11, 2007 4:32 pm Post subject: |
|
|
no SCHEDSTATS in kernel hacking section if genetic rsdl is selected. _________________ this is a strange strange world. |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Sun Mar 11, 2007 4:36 pm Post subject: |
|
|
tranquilcool wrote: | no SCHEDSTATS in kernel hacking section if genetic rsdl is selected. |
"Collect scheduler statistics" you dont see that in the kernel hacking section?? you need to enable "Kernel debugging" also _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
rmh3093 Advocate
Joined: 06 Aug 2003 Posts: 2138 Location: Albany, NY
|
Posted: Mon Mar 12, 2007 3:00 am Post subject: |
|
|
version 2.6.21-rc2-skunk2.3 is now out. i fixed the kconfig issues so genetic-rsdl should now select schedstats and geneticlib.... also rsdl has been updated to v0.30..... I tried to get suspend2 to work with this release and its not happening, i tried both latest and stable versions and the snapshot from suspend2-mm.git... there are too many changes for 2.6.21 I dont want to spend much more time on it until a new version of suspend2 is out _________________ Do not meddle in the affairs of wizards, for they are subtle and quick to anger. |
|
Back to top |
|
|
Donman Tux's lil' helper
Joined: 05 Sep 2005 Posts: 117
|
Posted: Mon Mar 12, 2007 3:53 am Post subject: |
|
|
Unfortunately Ndiswrapper does not compile against this kernel for me. Here is the error I receive:
Code: |
* Preparing ndiswrapper module
make -C /usr/src/linux M=/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver
make[1]: Entering directory `/usr/src/linux-2.6.21-rc2-skunk2.3'
LD /var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/built-in.o
CC [M] /var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/crt.o
CC [M] /var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/hal.o
CC [M] /var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/iw_ndis.o
CC [M] /var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/loader.o
CC [M] /var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.o
/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.c:40:47: error: macro "INIT_WORK" passed 3 arguments, but takes just 2
/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.c: In function 'ndis_init':
/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.c:40: error: 'INIT_WORK' undeclared (first use in this function)
/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.c:40: error: (Each undeclared identifier is reported only once
/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.c:40: error: for each function it appears in.)
make[2]: *** [/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver/ndis.o] Error 1
make[1]: *** [_module_/var/tmp/portage/net-wireless/ndiswrapper-1.38/work/ndiswrapper-1.38/driver] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.21-rc2-skunk2.3'
make: *** [default] Error 2
!!! ERROR: net-wireless/ndiswrapper-1.38 failed.
Call stack:
ebuild.sh, line 1614: Called dyn_compile
ebuild.sh, line 971: Called qa_call 'src_compile'
environment, line 4074: Called src_compile
ndiswrapper-1.38.ebuild, line 59: Called linux-mod_src_compile
linux-mod.eclass, line 511: Called die
!!! Unable to make KSRC=/usr/src/linux KVERS=2.6.21-rc2-skunk2.3-genetic_rsdl KBUILD=/usr/src/linux DISABLE_USB=1 -j1 all.
!!! If you need support, post the topmost build error, and the call stack if relevant.
!!! A complete build log is located at '/var/log/portage/net-wireless:ndiswrapper-1.38:20070312-034602.log'.
|
On a side note, did the wireless-dev tree get merged with -mm? I think I read that somewhere, and if so I will try out the bcm43xx driver again. |
|
Back to top |
|
|
|