Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
ck-sources : All about choice !
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3 ... 6, 7, 8, 9, 10, 11  Next  
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware
View previous topic :: View next topic  
Author Message
thunderrd
n00b
n00b


Joined: 20 Aug 2010
Posts: 59

PostPosted: Wed Jul 24, 2013 2:43 pm    Post subject: Reply with quote

Seems to be a problem here:

Code:
root@Q6600: ~# emerge -avuDN @world

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild     U  ] x11-proto/videoproto-2.3.2 [2.3.1-r1] ABI_X86="32 (64) (-x32)" 117 kB
[ebuild     U  ] x11-libs/libpciaccess-0.13.2 [0.13.1-r1] USE="zlib -minimal -static-libs" ABI_X86="32 (64) (-x32)" 351 kB
[ebuild  r  U  ] dev-libs/icu-51.2:0/51.2 [51.1:0/51.1] USE="-debug -doc -examples -static-libs" 21,331 kB
[ebuild     U  ] x11-libs/libXfont-1.4.6 [1.4.5-r1] USE="bzip2 ipv6 truetype -doc -static-libs" 478 kB
[ebuild     U  ] dev-libs/nss-3.15.1 [3.15] USE="-utils" 6,140 kB
[ebuild   R    ] x11-misc/xmobar-0.18  USE="alsa dbus -mail -mpd -mpris -threaded -timezone -wifi -xft (-inotify%)" 0 kB
[ebuild  NS    ] sys-kernel/ck-sources-3.9.11:3.9.11 [3.8.3:3.8.3, 3.9.2:3.9.2] USE="-bfsonly -build -deblob -symlink" 302 kB
[ebuild  rR    ] sys-apps/gptfdisk-0.8.6  USE="icu ncurses -static" 0 kB

Total: 8 packages (5 upgrades, 1 in new slot, 2 reinstalls), Size of downloads: 28,715 kB

Would you like to merge these packages? [Yes/No]

>>> Verifying ebuild manifests

!!! Digest verification failed:
!!! /mnt/ext3_STORAGE/usr/portage/sys-kernel/ck-sources/metadata.xml
!!! Reason: Filesize does not match recorded size
!!! Got: 812
!!! Expected: 0


I re-created the manifest with 'ebuild ck-sources-3.9.11.ebuild manifest', and things seem to have straightened out, but I don't know why it happened to begin with.

Just a heads-up, if anyone else has the same trouble.
Back to top
View user's profile Send private message
aCOSwt
Bodhisattva
Bodhisattva


Joined: 19 Oct 2007
Posts: 2537
Location: Hilbert space

PostPosted: Wed Jul 24, 2013 4:43 pm    Post subject: Reply with quote

thunderrd wrote:
I re-created the manifest with 'ebuild ck-sources-3.9.11.ebuild manifest', and things seem to have straightened out, but I don't know why it happened to begin with.

You did right.
Most generally, that sort of error is corrected after re-syncing the portage tree.
_________________
Back to top
View user's profile Send private message
TomWij
Retired Dev
Retired Dev


Joined: 04 Jul 2012
Posts: 1553

PostPosted: Wed Jul 24, 2013 7:46 pm    Post subject: Reply with quote

No, apparently something did go wrong with that commit; I am not able to pinpoint why, but it recorded the metadata.xml having size 0 for some reason.

Quote:
revision 1.345
date: 2013-07-24 21:41:39 +0200; author: tomwij; state: Exp; lines: +8 -8; commitid: 533851f02df34567;
Updated manifest; in the previous commit, the size for metadata.xml somehow got recorded as 0.

(Portage version: 2.1.12.13/cvs/Linux x86_64, signed Manifest commit with key 6D34E57D)


So, syncing around an hour from now; that should be resolved, but you can indeed do something like `cd /usr/portage/sys-kernel/ck-sources ; repoman manifest` to fix it up.
Back to top
View user's profile Send private message
ulenrich
Veteran
Veteran


Joined: 10 Oct 2010
Posts: 1480

PostPosted: Thu Jul 25, 2013 7:22 pm    Post subject: Reply with quote

ulenrich wrote:
fact the very first stable and tested kernel is a point.six :( :( :( :( :( :(

This time I seemingly loose my bet:
I just run a recently compiled linux-3.10.3rc1 with BFS enabled - without problems.
But I have disabled nearly all of the newer kernel features ...

@aCOSwt, what patch did you talk about eventually needed?

[edit] Uuups, after 2 hours crashed and totally freezed while heavy duty compiling in a chroot. And I must mention I tainted the kernel using the proprietary nvidia with the inofficial patch ...
Back to top
View user's profile Send private message
Chiitoo
Administrator
Administrator


Joined: 28 Feb 2010
Posts: 2550
Location: Here and Away Again

PostPosted: Sun Jul 28, 2013 1:24 am    Post subject: ><)))°€ Reply with quote

Teegrins!


Awesome work, Con Kolivas, aCOSwt, et al. I've been using these sources for quite a while, and they are quite impressive indeed! Reading about the past of the sources was very interesting, and I am glad that Kolivas returned to it.
Hopefully it wont be as stressful as it seemed to once be (settling to a patch-set is no doubt helping that, or so I'd imagine).


Anyblue, I've come upon some weirdness with the 3.9.11, and wondering now if it's just me (can't find it mentioned anywhere):

Code:
net/bridge/br_multicast.c: In function ‘br_multicast_del_pg’:
net/bridge/br_multicast.c:272:38: error: ‘struct net_bridge_mdb_entry’ has no member named ‘timer_armed’
make[2]: *** [net/bridge/br_multicast.o] Error 1
make[1]: *** [net/bridge] Error 2
make: *** [net] Error 2

Now, I am a complete noob with these things still, but I tried my best to make sense of things as I delved deep into the source code while I also sought the internet for any related information (losing a good portion of the day to being amused by mailing list posts from Torvalds), and came to a conclusion in the end. Would I be completely off the map if I were to claim that this patch:
    http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=c7e8e8a8f7a70b343ca1e0f90a31e35ab2d16de1
Code:
diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 19942e3..0daae3e 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -447,7 +447,7 @@ static int __br_mdb_del(struct net_bridge *br, struct br_mdb_entry *entry)
       call_rcu_bh(&p->rcu, br_multicast_free_pg);
       err = 0;
 
-      if (!mp->ports && !mp->mglist &&
+      if (!mp->ports && !mp->mglist && mp->timer_armed &&
           netif_running(br->dev))
          mod_timer(&mp->timer, jiffies);
       break;
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 81befac..69af490 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -270,7 +270,7 @@ static void br_multicast_del_pg(struct net_bridge *br,
       del_timer(&p->timer);
       call_rcu_bh(&p->rcu, br_multicast_free_pg);
 
-      if (!mp->ports && !mp->mglist &&
+      if (!mp->ports && !mp->mglist && mp->timer_armed &&
           netif_running(br->dev))
          mod_timer(&mp->timer, jiffies);

has been back-ported, but the cause has not:
    http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=9f00b2e7cf241fa389733d41b6
Code:
diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c
index 2475147..40bda80 100644
--- a/net/bridge/br_multicast.c
+++ b/net/bridge/br_multicast.c
@@ -617,8 +617,6 @@ rehash:
 
    mp->br = br;
    mp->addr = *group;
-   setup_timer(&mp->timer, br_multicast_group_expired,
-          (unsigned long)mp);
 
    hlist_add_head_rcu(&mp->hlist[mdb->ver], &mdb->mhash[hash]);
    mdb->size++;
@@ -656,7 +654,6 @@ static int br_multicast_add_group(struct net_bridge *br,
    struct net_bridge_mdb_entry *mp;
    struct net_bridge_port_group *p;
    struct net_bridge_port_group __rcu **pp;
-   unsigned long now = jiffies;
    int err;
 
    spin_lock(&br->multicast_lock);
@@ -671,7 +668,6 @@ static int br_multicast_add_group(struct net_bridge *br,
 
    if (!port) {
       mp->mglist = true;
-      mod_timer(&mp->timer, now + br->multicast_membership_interval);
       goto out;
    }
 
@@ -679,7 +675,7 @@ static int br_multicast_add_group(struct net_bridge *br,
         (p = mlock_dereference(*pp, br)) != NULL;
         pp = &p->next) {
       if (p->port == port)
-         goto found;
+         goto out;
       if ((unsigned long)p->port < (unsigned long)port)
          break;
    }
@@ -690,8 +686,6 @@ static int br_multicast_add_group(struct net_bridge *br,
    rcu_assign_pointer(*pp, p);
    br_mdb_notify(br->dev, port, group, RTM_NEWMDB);
 
-found:
-   mod_timer(&p->timer, now + br->multicast_membership_interval);
 out:
    err = 0;
 
@@ -1131,6 +1125,10 @@ static int br_ip4_multicast_query(struct net_bridge *br,
    if (!mp)
       goto out;
 
+   setup_timer(&mp->timer, br_multicast_group_expired, (unsigned long)mp);
+   mod_timer(&mp->timer, now + br->multicast_membership_interval);
+   mp->timer_armed = true;
+
    max_delay *= br->multicast_last_member_count;
 
    if (mp->mglist &&
@@ -1205,6 +1203,10 @@ static int br_ip6_multicast_query(struct net_bridge *br,
    if (!mp)
       goto out;
 
+   setup_timer(&mp->timer, br_multicast_group_expired, (unsigned long)mp);
+   mod_timer(&mp->timer, now + br->multicast_membership_interval);
+   mp->timer_armed = true;
+
    max_delay *= br->multicast_last_member_count;
    if (mp->mglist &&
        (timer_pending(&mp->timer) ?
@@ -1263,7 +1265,7 @@ static void br_multicast_leave_group(struct net_bridge *br,
          call_rcu_bh(&p->rcu, br_multicast_free_pg);
          br_mdb_notify(br->dev, port, group, RTM_DELMDB);
 
-         if (!mp->ports && !mp->mglist &&
+         if (!mp->ports && !mp->mglist && mp->timer_armed &&
              netif_running(br->dev))
             mod_timer(&mp->timer, jiffies);
       }
@@ -1275,30 +1277,12 @@ static void br_multicast_leave_group(struct net_bridge *br,
            br->multicast_last_member_interval;
 
    if (!port) {
-      if (mp->mglist &&
+      if (mp->mglist && mp->timer_armed &&
           (timer_pending(&mp->timer) ?
            time_after(mp->timer.expires, time) :
            try_to_del_timer_sync(&mp->timer) >= 0)) {
          mod_timer(&mp->timer, time);
       }
-
-      goto out;
-   }
-
-   for (p = mlock_dereference(mp->ports, br);
-        p != NULL;
-        p = mlock_dereference(p->next, br)) {
-      if (p->port != port)
-         continue;
-
-      if (!hlist_unhashed(&p->mglist) &&
-          (timer_pending(&p->timer) ?
-           time_after(p->timer.expires, time) :
-           try_to_del_timer_sync(&p->timer) >= 0)) {
-         mod_timer(&p->timer, time);
-      }
-
-      break;
    }
 
 out:
@@ -1674,6 +1658,7 @@ void br_multicast_stop(struct net_bridge *br)
       hlist_for_each_entry_safe(mp, n, &mdb->mhash[i],
                  hlist[ver]) {
          del_timer(&mp->timer);
+         mp->timer_armed = false;
          call_rcu_bh(&mp->rcu, br_multicast_free_group);
       }
    }
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index e260710..1b0ac95 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -112,6 +112,7 @@ struct net_bridge_mdb_entry
    struct timer_list      timer;
    struct br_ip         addr;
    bool            mglist;
+   bool            timer_armed;
 };
 
 struct net_bridge_mdb_htable

I figured the error message makes sense since the “timer_armed” doesn't seem to exist at all. The sources indeed build after reverting the commit: c7e8e8a8f7a70b343ca1e0f90a31e35ab2d16de1

Perhaps my kernel configuration is just out of whack, but this really seems wrong to me. ^^;


Again, thank you, and I would appreciate if someone wiser could confirm and/or point me my error(s)!
_________________
Kindest of regardses.
Back to top
View user's profile Send private message
aCOSwt
Bodhisattva
Bodhisattva


Joined: 19 Oct 2007
Posts: 2537
Location: Hilbert space

PostPosted: Sun Jul 28, 2013 11:01 am    Post subject: Re: ><)))°€ Reply with quote

Chiitoo wrote:
Anyblue, I've come upon some weirdness with the 3.9.11, and wondering now if it's just me (can't find it mentioned anywhere):

It's actually... not just you!
Chiitoo wrote:
Would I be completely off the map if I were to claim that this patch:
    http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=c7e8e8a8f7a70b343ca1e0f90a31e35ab2d16de1

has been back-ported

And is responsible for the problem you experience. You are correct.

I'll make my best to push a ck-sources-3.9.11-r1 (in wich this patch will have been removed) before tonight.

In the mean time, if you are in a hurry, you can :

1/ #cp ck-sources-3.9.11.ebuild ck-sources-3.9.11-r1.ebuild
2/ Edit the ck-sources-3.9.11-r1.ebuild and change the line reading K_GENPATCHES_VER="16" against K_GENPATCHES_VER="17"
3/ Save and exit
4/ #repoman manifest

Then emerge this package.
_________________
Back to top
View user's profile Send private message
SayusiAndo
n00b
n00b


Joined: 11 Jun 2013
Posts: 30
Location: Budapest, Hungary

PostPosted: Mon Jul 29, 2013 8:54 am    Post subject: ck-sources-3.9.11 failed to compile with genkernel Reply with quote

Hi All,

When I want to compile the ck-sources-3.9.11 using genkernel - it is comfortable for me - then I got the errors below:

Code:

configuration written to .config

*** End of the configuration.
*** Execute 'make' to start the build or try 'make help'.

*         >> Compiling 3.9.11-ck bzImage...
*         >> Not installing firmware as it's included in the kernel already (CONFIG_FIRMWARE_IN_KERNEL=y)...
*         >> Compiling 3.9.11-ck modules...
* ERROR: Failed to compile the "modules" target...
*
* -- Grepping log... --
*
*  SHIPPED scripts/kconfig/zconf.lex.c
*  SHIPPED scripts/kconfig/zconf.hash.c
*  HOSTCC  scripts/kconfig/zconf.tab.o
*  HOSTLD  scripts/kconfig/conf
*scripts/kconfig/conf --oldconfig Kconfig
*.config:407:warning: symbol value 'm' invalid for ACPI_CONTAINER
*.config:2863:warning: symbol value 'm' invalid for RTC_LIB
*.config:2864:warning: symbol value 'm' invalid for RTC_CLASS


Should I attach the full genkernel log? It is huge...
_________________
- -
-- Csanyi Andras/Sayusi Ando --
-- "Trust in God and keep the gunpowder dry!!" - Cromwell
Back to top
View user's profile Send private message
aCOSwt
Bodhisattva
Bodhisattva


Joined: 19 Oct 2007
Posts: 2537
Location: Hilbert space

PostPosted: Mon Jul 29, 2013 9:32 am    Post subject: Re: ck-sources-3.9.11 failed to compile with genkernel Reply with quote

SayusiAndo wrote:
When I want to compile the ck-sources-3.9.11 using genkernel - it is comfortable for me - then I got the errors below

:? I am very sorry : I know absolutely nothing of genkernel. :oops:
I'd suggest you open a new thread with your question, so that should attract the attention of genkernel's gurus (JRG, NeddySeagoon...)
_________________
Back to top
View user's profile Send private message
SayusiAndo
n00b
n00b


Joined: 11 Jun 2013
Posts: 30
Location: Budapest, Hungary

PostPosted: Mon Jul 29, 2013 9:35 am    Post subject: Re: ck-sources-3.9.11 failed to compile with genkernel Reply with quote

aCOSwt wrote:
SayusiAndo wrote:
When I want to compile the ck-sources-3.9.11 using genkernel - it is comfortable for me - then I got the errors below

:? I am very sorry : I know absolutely nothing of genkernel. :oops:
I'd suggest you open a new thread with your question, so that should attract the attention of genkernel's gurus (JRG, NeddySeagoon...)


There is no problem at all. Btw, I should try it manually, but I'm a lazy human... :) I'll do accordingly.
_________________
- -
-- Csanyi Andras/Sayusi Ando --
-- "Trust in God and keep the gunpowder dry!!" - Cromwell
Back to top
View user's profile Send private message
kernelOfTruth
Watchman
Watchman


Joined: 20 Dec 2005
Posts: 6111
Location: Vienna, Austria; Germany; hello world :)

PostPosted: Tue Jul 30, 2013 9:58 pm    Post subject: Reply with quote

this thread is about choice + ck stuff

so I figured the following might be interesting:

http://marc.info/?l=linux-kernel&m=137521302202537&w=2

latency cut-down by 80% :!:


it's still RFC - so let's see how it develops ...

BFQ occassionally tends to hardlock for me - seems like they haven't found the cause yet :(

so I'll meanwhile use CFQ instead ...
_________________
https://github.com/kernelOfTruth/ZFS-for-SystemRescueCD/tree/ZFS-for-SysRescCD-4.9.0
https://github.com/kernelOfTruth/pulseaudio-equalizer-ladspa

Hardcore Gentoo Linux user since 2004 :D
Back to top
View user's profile Send private message
SayusiAndo
n00b
n00b


Joined: 11 Jun 2013
Posts: 30
Location: Budapest, Hungary

PostPosted: Wed Jul 31, 2013 8:26 am    Post subject: Reply with quote

Hi Guys,

I got this error message when I want to compile ck-kernel-3.9.11 without genkernel.

Code:

sayusi-desktop linux # make
make[1]: Nothing to be done for `all'.
make[1]: Nothing to be done for `relocs'.
  CHK     include/generated/uapi/linux/version.h
  CHK     include/generated/utsrelease.h
  CALL    scripts/checksyscalls.sh
  CC      scripts/mod/devicetable-offsets.s
  GEN     scripts/mod/devicetable-offsets.h
  HOSTCC  scripts/mod/file2alias.o
  HOSTLD  scripts/mod/modpost
  CHK     include/generated/compile.h
make[3]: `arch/x86/realmode/rm/realmode.bin' is up to date.
  CHK     kernel/config_data.h
  CC [M]  net/bridge/br_multicast.o
net/bridge/br_multicast.c: In function ‘br_multicast_del_pg’:
net/bridge/br_multicast.c:272:38: error: ‘struct net_bridge_mdb_entry’ has no member named ‘timer_armed’
make[2]: *** [net/bridge/br_multicast.o] Error 1
make[1]: *** [net/bridge] Error 2
make: *** [net] Error 2
sayusi-desktop linux # eselect kernel list
Available kernel symlink targets:
  [1]   linux-3.9.7-ck
  [2]   linux-3.9.11-ck *
  [3]   linux-3.10.3-gentoo-r1
  [4]   linux-3.10.4-gentoo
sayusi-desktop linux #


Any suggestion what should I do? What information is needed for you guys to figure out what the problem is?
_________________
- -
-- Csanyi Andras/Sayusi Ando --
-- "Trust in God and keep the gunpowder dry!!" - Cromwell
Back to top
View user's profile Send private message
aCOSwt
Bodhisattva
Bodhisattva


Joined: 19 Oct 2007
Posts: 2537
Location: Hilbert space

PostPosted: Wed Jul 31, 2013 8:44 am    Post subject: Reply with quote

SayusiAndo wrote:
Any suggestion what should I do? What information is needed for you guys to figure out what the problem is?

Please read my answer to Chiitoo, 6 posts above.

I must apologize as I still did not push the 3.9.11-r1 as promised. I have been bothered by a MySQL upgrade then 3.10.3 and 4 came quickly... with their loads of problems...
Act as I suggested to Chiitoo or... be patient.
_________________
Back to top
View user's profile Send private message
SayusiAndo
n00b
n00b


Joined: 11 Jun 2013
Posts: 30
Location: Budapest, Hungary

PostPosted: Wed Jul 31, 2013 8:49 am    Post subject: Reply with quote

aCOSwt wrote:
SayusiAndo wrote:
Any suggestion what should I do? What information is needed for you guys to figure out what the problem is?

Please read my answer to Chiitoo, 6 posts above.

I must apologize as I still did not push the 3.9.11-r1 as promised. I have been bothered by a MySQL upgrade then 3.10.3 and 4 came quickly... with their loads of problems...
Act as I suggested to Chiitoo or... be patient.


Ahh, thank you very much!
_________________
- -
-- Csanyi Andras/Sayusi Ando --
-- "Trust in God and keep the gunpowder dry!!" - Cromwell
Back to top
View user's profile Send private message
aCOSwt
Bodhisattva
Bodhisattva


Joined: 19 Oct 2007
Posts: 2537
Location: Hilbert space

PostPosted: Thu Aug 01, 2013 11:15 am    Post subject: Reply with quote

3.9.11-r1 and 3.10.4 are on their way
_________________
Back to top
View user's profile Send private message
SayusiAndo
n00b
n00b


Joined: 11 Jun 2013
Posts: 30
Location: Budapest, Hungary

PostPosted: Thu Aug 01, 2013 11:26 am    Post subject: Reply with quote

aCOSwt wrote:
3.9.11-r1 and 3.10.4 are on their way


Rispect maximus!
_________________
- -
-- Csanyi Andras/Sayusi Ando --
-- "Trust in God and keep the gunpowder dry!!" - Cromwell
Back to top
View user's profile Send private message
Ant P.
Watchman
Watchman


Joined: 18 Apr 2009
Posts: 6920

PostPosted: Wed Sep 11, 2013 1:18 am    Post subject: Reply with quote

The 3.11-ck1 patch's out as of yesterday.
Back to top
View user's profile Send private message
thunderrd
n00b
n00b


Joined: 20 Aug 2010
Posts: 59

PostPosted: Wed Sep 11, 2013 8:10 am    Post subject: Reply with quote

Yup. 3.11/BFS 0.442

From IRC:
Quote:
Topic for 22#ck is: http://ck-hack.blogspot.com ; Latest -ck: 3.11-ck1 | http://kernel.kolivas.org | lrzip 0.616 http://lrzip.kolivas.org | latest BFS: 0.442



and:
http://ck-hack.blogspot.com/2013/09/bfs-0441-311-ck1.html
Back to top
View user's profile Send private message
ulenrich
Veteran
Veteran


Joined: 10 Oct 2010
Posts: 1480

PostPosted: Wed Oct 09, 2013 12:16 am    Post subject: Reply with quote

I made a backport of bfs-442
for use with LTS kernel Linux-3.10:

3.10-sched-bfs-442-new.patch
Found in attachement of bug
https://bugs.gentoo.org/show_bug.cgi?id=487362
Back to top
View user's profile Send private message
PaulBredbury
Watchman
Watchman


Joined: 14 Jul 2005
Posts: 7310

PostPosted: Wed Oct 09, 2013 8:09 am    Post subject: Reply with quote

ulenrich wrote:
3.10-sched-bfs-442-new.patch

Thanks, it seems to be running OK :)

I was interested in the new check for throttling:

Quote:
The other significant change is to check for throttled CPUs when choosing an idle CPU to move a process to, which should impact the behaviour and possibly throughput when using a scaling CPU governor, such as ondemand.


For (hopefully) convenience, here's a diff from BFS 440 to your 442, created by applying both to vanilla kernel 3.10.15.

Edit: And here's Alfred Chen's patch, in proper format.


Last edited by PaulBredbury on Wed Oct 09, 2013 11:29 am; edited 1 time in total
Back to top
View user's profile Send private message
ulenrich
Veteran
Veteran


Joined: 10 Oct 2010
Posts: 1480

PostPosted: Wed Oct 09, 2013 8:47 am    Post subject: Reply with quote

@Paul, three hours later than me at ck-hack.blogspot.com "Alfred Chen" showd a patch where he relocated mutex_lock and mutex_unlock. What do you think of it:

Alfred Chen puts the
mutex_lock (respective mutex_unlock)

outside of
for_each_online_cpu(cpu)

If mutex_lock is meant for one cpu
this sound reasonable? Is it?
Back to top
View user's profile Send private message
PaulBredbury
Watchman
Watchman


Joined: 14 Jul 2005
Posts: 7310

PostPosted: Wed Oct 09, 2013 12:08 pm    Post subject: Reply with quote

Alfred Chen's patch runs OK too - that's about all I'm qualified to say.

I kinda noticed a hang at startup, a couple of times with kernel 3.10 & BFS, but it coincided with me playing with kernel & ACPI settings, so I can't reproduce it for a proper test.
Back to top
View user's profile Send private message
PaulBredbury
Watchman
Watchman


Joined: 14 Jul 2005
Posts: 7310

PostPosted: Fri Oct 18, 2013 8:36 am    Post subject: Reply with quote

And here's another patch for improved ondemand (main patch).

So, to recap, the patches to apply to kernel 3.10.x are:

    BFS 442
    Alfred Chen's mutex patch
    This ondemand recalc patch

It's all working stably for me.

Edit on 20140109: kernel 3.10.26 had a trivial change, causing these patches to need refreshing. Here's those 3 patches, merged into 1 patch to apply to kernel 3.10.26 :)


Last edited by PaulBredbury on Thu Jan 09, 2014 11:56 pm; edited 1 time in total
Back to top
View user's profile Send private message
ulenrich
Veteran
Veteran


Joined: 10 Oct 2010
Posts: 1480

PostPosted: Fri Oct 18, 2013 11:42 am    Post subject: Reply with quote

@Paul +1
Yes runs well for me. But
There is a BFS patch of
+#define DEF_FREQUENCY_UP_THRESHOLD (63)

For me the ondemand patch runs better with the original value
-#define DEF_FREQUENCY_UP_THRESHOLD (80)

The DEF_FREQUENCY_DOWN_DIFFERENTIAL value was obosleted anyway...

Should we ask Con Kolivas to outsource the patching of
drivers/cpufreq/cpufreq_ondemand.c
into the ck1 patchset?
Back to top
View user's profile Send private message
PaulBredbury
Watchman
Watchman


Joined: 14 Jul 2005
Posts: 7310

PostPosted: Fri Oct 18, 2013 12:13 pm    Post subject: Reply with quote

It's just a default. I go in the opposite direction, to keep gaming smooth:

Code:
echo 50 > /sys/devices/system/cpu/cpufreq/ondemand/up_threshold
echo 50 > /sys/devices/system/cpu/cpufreq/ondemand/sampling_down_factor
Back to top
View user's profile Send private message
thunderrd
n00b
n00b


Joined: 20 Aug 2010
Posts: 59

PostPosted: Sun Nov 10, 2013 5:26 pm    Post subject: Reply with quote

@aCOSwt:

Is there any hope that there will be further updates, or are we now stalled at 3.9? Things have been pretty silent here.

I saw the problems in Bug #479382, and am thinking maybe you have had enough proxy maintaining :)

I, for one, appreciate all that you have done, Eric. Thank you.
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Kernel & Hardware All times are GMT
Goto page Previous  1, 2, 3 ... 6, 7, 8, 9, 10, 11  Next
Page 7 of 11

 
Jump to:  
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