patch for "diff in charge" errors
Subsystems (2)
Hi All,
in htb 3.3 I turned on more debugs for bad timestamps. As result we found another bug in all 2.4 (and probably 2.2) kernels. Because the bug is from today I attached patch. It is no harm to apply it in every case.
It will get rid of these:
May 14 09:31:00 kernel: HTB: bad diff in charge, diff=1310000 now=6873088 then=4281851904 j=525127 May 14 09:31:00 kernel: HTB: bad diff in charge, diff=1310000 now=6873088 then=4281851904 j=525127
(hopefuly). I'll probably make it part of future HTB patches. devik
include/net/pkt_sched.h +1 -1
@@ -221,7 +221,7 @@#define PSCHED_EXPORTLIST_2 -#if ~0UL ==0xFFFFFFFF +#if BITS_PER_LONG <=32 #define PSCHED_WATCHER unsigned long
Hi Devik!
Quoted 2 lines · Martin Devera
in htb 3.3 I turned on more debugs for bad timestamps. As result we found another bug in all 2.4 (and probably 2.2) kernels.
Great to hear that. Just curious, what effect could this bug have (e.g. htb not measuring right, performance, etc.)?
Could this bug have an impact on systems running HTB2 & SFQ?
Thank you!
Greetings, Nils
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/
Its effect is periodic rate error each 40 minutes. In 2.4.19preX it cause total instability of CBQ, HTB and TBF.
devik
Quoted 6 lines · Nils Lichtenfeld
Hi Devik!
Quoted 2 lines · Martin Devera
in htb 3.3 I turned on more debugs for bad timestamps. As result we found another bug in all 2.4 (and probably 2.2) kernels.
Quoted 10 lines
Great to hear that. Just curious, what effect could this bug have (e.g. htb not measuring right, performance, etc.)?
Could this bug have an impact on systems running HTB2 & SFQ?
Thank you!
Greetings, Nils
_______________________________________________
LARTC mailing list / [email protected]
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://lartc.org/