| View previous topic :: View next topic |
| Author |
Message |
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
 Posted: Wed Jul 16, 2003 7:14 pm Post subject: |
 |
|
| Compile broke on compiling my scsi controller, an adaptec using aic7xxx. Reverting the wpadded.patch fixed it (it only seemed to die because of too many warnings). Just a note for anyone else. |
|
| Back to top |
|
 |
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
| Posted: Wed Jul 16, 2003 7:15 pm Post subject: |
|
|
| Wedge_ wrote: | Is anyone else getting lots of warning messages while compiling mm1? Stuff like this: | Code: | include/linux/sched.h:215: warning: padding struct to align `context'
include/linux/sched.h:363: warning: padding struct to align `pid'
include/linux/sched.h:405: warning: padding struct to align `user'
include/linux/sched.h:411: warning: padding struct to align `link_count'
include/linux/sched.h:470: warning: padding struct size to alignment boundary
|
From the changelog, I guess it's because of the wpadding patch. It still seems to work fine once compiled, but should I be getting all the warnings? |
The patch wasn't tested on gcc 3.x, judging by Andrew's comments on the matter. I just reverted the patch because I had to (see above) |
|
| Back to top |
|
|
Bobbie
Tux's lil' helper
Joined: 21 Jul 2002
Posts: 77
|
| Posted: Wed Jul 16, 2003 7:30 pm Post subject: |
|
|
I compiled 2.6.0-test1-ac1 without much trouble and here's my observations so far:
- I can't enable dma on hd drives with hdparm:
hdparm -d1 /dev/hda
/dev/hda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma = 0 (off)
- when the CPU activity is high (emerging for example) , mozilla scrolling is very jerky. I'm using the nvidia binary drivers. If no activity scrolling is very fast
- does lirc modules support 2.6.0 ?
- it is not related to this kernel since i have the problem with 2.4 : at boot time I have the message:
VFS: Mounted root (ext3 filesystem) readonly.
Freeing unused kernel memory: 152k freed
Warning: unable to open an initial console.
issing device in /dev . It happended when I move my gentoo install from a parttition to another (I had to recreate /dev)
- the new kernel is able to use IDE CDRW drive whitout scsi emulation...anybody tested that ? is there anything special to do (update programs, etc) ? |
|
| Back to top |
|
|
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
| Posted: Wed Jul 16, 2003 7:45 pm Post subject: |
|
|
Right. Scheduler bludgeoning.
The first thing I did was start Quark, turn opaque dragging on and start a make -j20 on the kernel. All my apps (Mozilla, xchat) were responsive. Dragging doesn't stall xfree, and was for the majority of the time smooth. Next I opened Evolution and gconf-editor, and was surprised: They loaded a lot faster than I think they used to with other patchsets.
Configuration editor was pretty snappy, Evolution's splash screen took a bit, but was good under the circumstances. xterm takes 1~2 seconds to give me a bash prompt. Quark hasn't skipped yet, even during the snappy Evolution component load while make -j20
Now for non-insane workload. Opening links in new Mozilla tabs no longer chops the mouse for a split-second, like it used to. Other than that, uh... everything's spiffy. O1-5int+softrr on vanilla -test1 took care of all the issues I can remember, excepting artifacts in other windows for a short time when opaque dragging (which i still blame xfree for, for varying reasons). So now it's pretty much just the faster the better. :) |
|
| Back to top |
|
|
Wedge_
Advocate
Joined: 08 Aug 2002
Posts: 3614
Location: Scotland
|
| Posted: Wed Jul 16, 2003 7:48 pm Post subject: |
|
|
| Quote: | | The patch wasn't tested on gcc 3.x, judging by Andrew's comments on the matter. I just reverted the patch because I had to (see above) |
Well, it seems to work for me, despite all the warnings, but I'll get rid of it anyway. Thanks for that.
_________________
Per Ardua Ad Astra
The Earth is the cradle of the mind, but we cannot live forever in a cradle - Konstantin E. Tsiolkovsky
Gentoo Radeon FAQ |
|
| Back to top |
|
|
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
| Posted: Wed Jul 16, 2003 8:04 pm Post subject: |
|
|
| Bobbie wrote: |
- the new kernel is able to use IDE CDRW drive whitout scsi emulation...anybody tested that ? is there anything special to do (update programs, etc) ? |
There are plenty of topics on the matter, do a search for ATAPI burning or something similiar (key word is ATAPI). I have it working fine here. |
|
| Back to top |
|
|
doalwa
Tux's lil' helper
Joined: 03 Mar 2003
Posts: 142
Location: Old Europe
|
| Posted: Wed Jul 16, 2003 8:15 pm Post subject: |
|
|
Well, so far, so pleased..just installed the new kernel, compiled it..nvidia kernel works, my printer works, cd bruning works...BUT...my mouse doesn't 
I'm using a regular optical MS INtelliMouse Explorer USB which is connected to my PS/2 port using an adapter. THe mouse worked flawlessly with 2.4.21. I'm sure there is something missing in my .config.
Under "Input device support" i checked
<Input Devices>
<Mouse Interface>
<Provide legacy psaux Interface>
<Serial i/o support>
<i8042 pc keyboard controller>
<Keyboards>
<AT Keyboard Support>
<Mice>
<PS/2 Mice Support>
So far, so self-explanatory  But when I boot my freshly compiled kernel and issue an XFree86 -configure, he detects the mouse on /dev/mouse as usual. /dev/mouse is linked to /dev/psaux, okay. But no matter what i tried, it just won't work. Any of you guys have a hint for me? |
|
| Back to top |
|
|
nillekind
n00b
Joined: 05 Oct 2002
Posts: 64
|
| Posted: Wed Jul 16, 2003 8:37 pm Post subject: |
|
|
| bssteph wrote: |
Okey dokey. Two patches, as promised. They both patch cleanly against a vanilla linux-2.6.0-test1 kernel.
|
Would you please make a split out version for the softrr patch? I want to use mm1+softrr+O6. But the original softrr doesn't apply to 2.6.0.
Thanks,
Conner |
|
| Back to top |
|
|
syscrash
Guru
Joined: 14 Apr 2003
Posts: 541
|
| Posted: Wed Jul 16, 2003 8:58 pm Post subject: |
|
|
Where's the USB mouse support? I can't use it without my mouse  |
|
| Back to top |
|
|
dsd
Developer
Joined: 30 Mar 2003
Posts: 2162
Location: nr London
|
| Posted: Wed Jul 16, 2003 8:59 pm Post subject: |
|
|
well.. this is the first kernel that ive tried in the "development range" (only ever used 2.4 before).
im impressed so far, i have overcome some issues with sound too. but, i'm getting no output from rear speakers. the card is a SB Live, and im using the ALSA emu10k1 driver from the kernel.
I'm guessing I need some sort of alsa configuration tool such as alsamixergui, but "emerge alsamixergui" also wants to install:
media-sound/alsa-driver-0.9.2
media-libs/alsa-lib-0.9.2
media-sound/alsa-utils-0.9.2
media-sound/alsamixergui-0.9.0.1.2-r1
which of these are necessary to be installed, and which arent relevant (i.e. designed for 2.4 kernels as mentioned earlier in the thread).
thanks,
dsd.
_________________
http://dev.gentoo.org/~dsd |
|
| Back to top |
|
|
bssteph
l33t
Joined: 26 Feb 2003
Posts: 652
Location: Wisconsin
|
| Posted: Wed Jul 16, 2003 9:01 pm Post subject: |
|
|
| nillekind wrote: | | bssteph wrote: |
Okey dokey. Two patches, as promised. They both patch cleanly against a vanilla linux-2.6.0-test1 kernel.
|
Would you please make a split out version for the softrr patch? I want to use mm1+softrr+O6. But the original softrr doesn't apply to 2.6.0.
Thanks,
Conner |
Sorry but I'm busy right now and can't work on it. I could tonight but that's a while from now.
If you want to do it yourself, apply O6int (which should work fine), and then softrr. softrr will fail in 3 hunks: #1, #4, and #6 (iirc). All three fail just because what the diff looks for has changed. Edit kernel/sched.c by:
- adding the three #define lines at the end of the #define block
- search for void scheduler_tick( and add the timeslice line after the cpu line as in the diff
- search for enqueue_task(p, rq->active) (be mindful of the order as seen in the diff) and add the two lines right after it (what you add is before an } else ... line
patches are plain-text. use it as your guide and edit accordingly |
|
| Back to top |
|
|
wilburpan
l33t
Joined: 21 Jan 2003
Posts: 977
|
| Posted: Wed Jul 16, 2003 11:00 pm Post subject: |
|
|
| lannie wrote: | Just compiled the new kernel... the desktop beautifully on the dell inspiron 4000 .. expect for a few problems
1) My Synaptics touchpad is still stuffed  .. got to use a USB mouse to roam around |
I have an Inspiron 5000e, and managed to get my touchpad to finally work with 2.6.0-test1, after failing to do so with 2.5.70-75. I put the info here. |
|
| Back to top |
|
|
alexmaz
n00b
Joined: 07 Jul 2002
Posts: 35
Location: Milano - Italia
|
| Posted: Wed Jul 16, 2003 11:54 pm Post subject: |
|
|
2.6.0-test1-ac2 have a possible fix for synaptics touchpad
_________________
Italy rejects war as an instrument of aggression against the freedoms of others peoples and as a means for settling international controversies |
|
| Back to top |
|
|
jrz
Apprentice
Joined: 19 Mar 2003
Posts: 272
Location: Sacramento, CA USA
|
| Posted: Thu Jul 17, 2003 12:36 am Post subject: |
|
|
i fixed my mouse prob by building mouse support as a module and loading that module before hid
mousedev
hid
s'all good now, the /dev/input/mouse0 link is back
glhf!!1
_________________
Press F1 for Help. |
|
| Back to top |
|
|
jimlynch11
Guru
Joined: 21 Feb 2003
Posts: 590
Location: massachusetts
|
| Posted: Thu Jul 17, 2003 12:43 am Post subject: |
|
|
whenever trying to do anything alsa, i get weird error messages about no such device.
| Quote: |
root@lyncjp0-pc linux-2.6.0-test1-mm1 # alsamixer
alsamixer: function snd_ctl_open failed for default: No such device
root@lyncjp0-pc linux-2.6.0-test1-mm1 #
|
| Quote: | root@lyncjp0-pc linux-2.6.0-test1-mm1 # amixer
amixer: Mixer attach default error: No such device
root@lyncjp0-pc linux-2.6.0-test1-mm1 #
|
i know i have the right driver compiled in, so what do i do? |
|
| Back to top |
|
|
jrz
Apprentice
Joined: 19 Mar 2003
Posts: 272
Location: Sacramento, CA USA
|
| Posted: Thu Jul 17, 2003 1:08 am Post subject: |
|
|
when in doubt, try modules? (someone comment on this, it's worked for me in every case)
_________________
Press F1 for Help. |
|
| Back to top |
|
|
jimlynch11
Guru
Joined: 21 Feb 2003
Posts: 590
Location: massachusetts
|
| Posted: Thu Jul 17, 2003 1:11 am Post subject: |
|
|
| jrz wrote: | | when in doubt, try modules? (someone comment on this, it's worked for me in every case) |
i tried recompiling it as a module. the module is running, but still no sound |
|
| Back to top |
|
|
Safrax
Guru
Joined: 23 Apr 2002
Posts: 422
|
| Posted: Thu Jul 17, 2003 1:23 am Post subject: |
|
|
| jimlynch11 wrote: | | jrz wrote: | | when in doubt, try modules? (someone comment on this, it's worked for me in every case) |
i tried recompiling it as a module. the module is running, but still no sound |
What type of sound card is it? and can you paste the output of lsmod in a reply? |
|
| Back to top |
|
|
jimlynch11
Guru
Joined: 21 Feb 2003
Posts: 590
Location: massachusetts
|
| Posted: Thu Jul 17, 2003 1:53 am Post subject: |
|
|
| Safrax wrote: | | jimlynch11 wrote: | | jrz wrote: | | when in doubt, try modules? (someone comment on this, it's worked for me in every case) |
i tried recompiling it as a module. the module is running, but still no sound |
What type of sound card is it? and can you paste the output of lsmod in a reply? |
its an intel8x0 driver...
| Quote: | lyncjp0@lyncjp0-pc lyncjp0 $ lsmod
Module Size Used by
snd_pcm 93024 0
snd_timer 24352 0
snd_page_alloc 9028 0
snd_seq_device 7812 0
snd_intel8x0 29444 0
snd_ac97_codec 50660 1 snd_intel8x0
snd_mpu401_uart 7392 1 snd_intel8x0
snd_rawmidi 21920 1 snd_mpu401_uart
lyncjp0@lyncjp0-pc lyncjp0 $ |
|
|
| Back to top |
|
|
Aonoa
Guru
Joined: 23 May 2002
Posts: 589
|
| Posted: Thu Jul 17, 2003 4:54 am Post subject: |
|
|
Seems there are a lot of people with mouse issues, USB mostly ?
Including me, I'll try to load mousedev before hid as you said; jrz
Hope it's a fix  |
|
| Back to top |
|
|
butters
Guru
Joined: 13 May 2002
Posts: 427
Location: Poughkeepsie, NY
|
| Posted: Thu Jul 17, 2003 5:08 am Post subject: |
|
|
Is the 2.6.0-test1 kernel in the portage tree? It seemed to be yesterday, but not anymore. Is there anything I should know about compiling kernels not from portage?
_________________
If tugboats were bigger, they'd be the ones getting tugged. |
|
| Back to top |
|
|
nillekind
n00b
Joined: 05 Oct 2002
Posts: 64
|
| Posted: Thu Jul 17, 2003 8:42 am Post subject: |
|
|
| bssteph wrote: |
Sorry but I'm busy right now and can't work on it. I could tonight but that's a while from now.
If you want to do it yourself, apply O6int (which should work fine), and then softrr. |
Thanks for your advice. I applied MM1 O6int and O6.1int and applied the softrr patch manually. So here's the diff:
| Code: |
--- linux-2.6.0-test1.virgin/kernel/sched.c 2003-07-17 10:30:23.000000000 +0200
+++ linux-2.6.0-test1.mod/kernel/sched.c 2003-07-17 10:29:46.000000000 +0200
@@ -79,6 +79,9 @@
#define SLEEP_BUFFER (HZ/100)
#define NODE_THRESHOLD 125
#define MAX_BONUS (MAX_USER_PRIO * PRIO_BONUS_RATIO / 100)
+#define SCHED_TS_KSOFTRR 5
+#define MIN_SRT_TIMESLICE (50 * HZ / 1000)
+#define MAX_SRT_TIMESLICE (150 * HZ / 1000)
/*
* If a task is 'interactive' then we reinsert it in the active
@@ -132,9 +135,12 @@
#define BASE_TIMESLICE(p) (MIN_TIMESLICE + \
((MAX_TIMESLICE - MIN_TIMESLICE) * (MAX_PRIO-1-(p)->static_prio)/(MAX_USER_PRIO - 1)))
+#define BASE_SRT_TIMESLICE(p) (MIN_SRT_TIMESLICE + \
+ ((MAX_SRT_TIMESLICE - MIN_SRT_TIMESLICE) * (MAX_RT_PRIO-1-(p)->static_prio)/(MAX_RT_PRIO - 1)))
+
static inline unsigned int task_timeslice(task_t *p)
{
- return BASE_TIMESLICE(p);
+ return p->policy != SCHED_SOFTRR ? BASE_TIMESLICE(p): BASE_SRT_TIMESLICE(p);
}
/*
@@ -161,7 +167,7 @@ struct prio_array {
struct runqueue {
spinlock_t lock;
unsigned long nr_running, nr_switches, expired_timestamp,
- nr_uninterruptible;
+ nr_uninterruptible, ts_timestamp;
task_t *curr, *idle;
struct mm_struct *prev_mm;
prio_array_t *active, *expired, arrays[2];
@@ -1245,6 +1251,7 @@ DEFINE_PER_CPU(struct kernel_stat, kstat
void scheduler_tick(int user_ticks, int sys_ticks)
{
int cpu = smp_processor_id();
+ unsigned int time_slice;
struct cpu_usage_stat *cpustat = &kstat_this_cpu.cpustat;
runqueue_t *rq = this_rq();
task_t *p = current;
@@ -1283,17 +1290,32 @@ void scheduler_tick(int user_ticks, int
p->sleep_avg--;
if (unlikely(rt_task(p))) {
/*
- * RR tasks need a special form of timeslice management.
+ * RR and SOFTRR tasks need a special form of timeslice management.
* FIFO tasks have no timeslices.
*/
- if ((p->policy == SCHED_RR) && !--p->time_slice) {
- p->time_slice = task_timeslice(p);
+ if ((p->policy == SCHED_RR || p->policy == SCHED_SOFTRR) &&
+ !--p->time_slice) {
+ p->time_slice = time_slice = task_timeslice(p);
p->first_time_slice = 0;
set_tsk_need_resched(p);
- /* put it at the end of the queue: */
+ /*
+ * We rotate SCHED_RR like POSIX states. On the
+ * contrary, SCHED_SOFTRR are real-time tasks without
+ * attitude and we do not want them to starve other
+ * tasks while we want them to be able to preempt
+ * SCHED_NORMAL tasks. The rule is that SCHED_SOFTRR
+ * will be expired if they require roughly more then
+ * 1/SCHED_TS_KSOFTRR percent of CPU time.
+ */
dequeue_task(p, rq->active);
- enqueue_task(p, rq->active);
+ if (p->policy == SCHED_RR ||
+ (jiffies - rq->ts_timestamp) > SCHED_TS_KSOFTRR * time_slice)
+ enqueue_task(p, rq->active);
+ else
+ enqueue_task(p, rq->expired);
+
+ rq->ts_timestamp = jiffies;
}
goto out_unlock;
}
@@ -1310,6 +1332,9 @@ void scheduler_tick(int user_ticks, int
enqueue_task(p, rq->expired);
} else
enqueue_task(p, rq->active);
+
+ rq->ts_timestamp = jiffies;
+
} else if (p->prio < effective_prio(p)){
/*
* Tasks that have lowered their priority are put to the end
@@ -1823,12 +1848,22 @@ static int setscheduler(pid_t pid, int p
else {
retval = -EINVAL;
if (policy != SCHED_FIFO && policy != SCHED_RR &&
- policy != SCHED_NORMAL)
+ policy != SCHED_NORMAL && policy != SCHED_SOFTRR)
goto out_unlock;
}
/*
- * Valid priorities for SCHED_FIFO and SCHED_RR are
+ * If the caller requested a SCHED_RR policy without having the
+ * necessary rights, we downgrade the policy to SCHED_SOFTRR. This
+ * is currrently here to enable to test the new SOFTRR realtime
+ * policy with existing programs that try to ask for SCHED_RR. Not
+ * sure if this should remain as permanent feature.
+ */
+ if (policy == SCHED_RR && !capable(CAP_SYS_NICE))
+ policy = SCHED_SOFTRR;
+
+ /*
+ * Valid priorities for SCHED_FIFO, SCHED_RR and SCHED_SOFTRR are
* 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL is 0.
*/
retval = -EINVAL;
@@ -2152,6 +2187,7 @@ asmlinkage long sys_sched_get_priority_m
switch (policy) {
case SCHED_FIFO:
case SCHED_RR:
+ case SCHED_SOFTRR:
ret = MAX_USER_RT_PRIO-1;
break;
case SCHED_NORMAL:
@@ -2175,6 +2211,7 @@ asmlinkage long sys_sched_get_priority_m
switch (policy) {
case SCHED_FIFO:
case SCHED_RR:
+ case SCHED_SOFTRR:
ret = 1;
break;
case SCHED_NORMAL:
|
Have fun,
Conner |
|
| Back to top |
|
|
asimon
l33t
Joined: 27 Jun 2002
Posts: 979
Location: Germany, Old Europe
|
| Posted: Thu Jul 17, 2003 11:04 am Post subject: |
|
|
| Bobbie wrote: |
- the new kernel is able to use IDE CDRW drive whitout scsi emulation...anybody tested that ? is there anything special to do (update programs, etc) ? |
I burned one CD with k3b, cdrtools-2.01_alpha16, and mm-sources-2.6.0_beta1-r1. I did compiled the ide stuff including ide-cd into the kernel but no ide-scsi emulation. It worked as good as with scsi emulation previously.
But I used a k3b cvs snapshot which already supports ATAPI burning (it detected the ATAPI cdwriter automatically), I dunno if the last stable version does already support it.
Have fun,
Andreas |
|
| Back to top |
|
|
Safrax
Guru
Joined: 23 Apr 2002
Posts: 422
|
| Posted: Thu Jul 17, 2003 11:38 am Post subject: |
|
|
| dsd wrote: | well.. this is the first kernel that ive tried in the "development range" (only ever used 2.4 before).
im impressed so far, i have overcome some issues with sound too. but, i'm getting no output from rear speakers. the card is a SB Live, and im using the ALSA emu10k1 driver from the kernel.
I'm guessing I need some sort of alsa configuration tool such as alsamixergui, but "emerge alsamixergui" also wants to install:
media-sound/alsa-driver-0.9.2
media-libs/alsa-lib-0.9.2
media-sound/alsa-utils-0.9.2
media-sound/alsamixergui-0.9.0.1.2-r1
which of these are necessary to be installed, and which arent relevant (i.e. designed for 2.4 kernels as mentioned earlier in the thread).
thanks,
dsd. |
alsa-driver is the only thing in that list that doesn't need to be installed. The kernel ALSA implementation is much newer than what portage is wanting to install. |
|
| Back to top |
|
|
ahr
n00b
Joined: 14 Jul 2002
Posts: 31
Location: Reykjavik, Iceland
|
| Posted: Thu Jul 17, 2003 12:53 pm Post subject: |
|
|
Same here.. Except I got SB 64 AWE Gold, ISA card. Works with 2.4.x and ALSA.
As for the other stuff; My system works just dandy. -- very responsive
| jimlynch11 wrote: | | jrz wrote: | | when in doubt, try modules? (someone comment on this, it's worked for me in every case) |
i tried recompiling it as a module. the module is running, but still no sound |
|
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
