https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
Eugene Grosbein changed:
What|Removed |Added
Status|Open|Closed
See Also|
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #26 from h...@restart.be ---
I try with 12.0-ALPHA2 r338128M with a10_timer - see
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=229644#c27
The throughput is back to normal :-)
I think this bug is solved.
--
You are
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #25 from h...@restart.be ---
I try with 12.0-CURRENT r336112 and patch from
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=229644
The throughput regression is still there
--
You are receiving this mail because:
You are on
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
Eugene Grosbein changed:
What|Removed |Added
Keywords|patch |
--- Comment #24 from Eugene
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
Mark Linimon changed:
What|Removed |Added
Keywords||patch, regression
--
You are
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #23 from Eugene Grosbein ---
Timing problems seems to be first suspect for me. We need more eyes of ARM
people here.
--
You are receiving this mail because:
You are on the CC list for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #22 from h...@restart.be ---
Created attachment 194167
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194167=edit
diff -u vmstat r334436 r320599 after fetch of 10329777 bytes
--
You are receiving this mail because:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #21 from h...@restart.be ---
Created attachment 194166
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194166=edit
vmstat -ai of r320559 after fetch of 10329777 bytes
--
You are receiving this mail because:
You are
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #19 from h...@restart.be ---
(In reply to Eugene Grosbein from comment #16)
Test are run on the same piece of hardware: boot rpool/ROOT/current (r334436)
and rpool/ROOT/default (r320599).
There is a time management problem
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #18 from h...@restart.be ---
Created attachment 194156
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194156=edit
diff -u verbose boot r334436 r320599
--
You are receiving this mail because:
You are on the CC list
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
Eugene Grosbein changed:
What|Removed |Added
Assignee|n...@freebsd.org |freebsd-...@freebsd.org
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #16 from Eugene Grosbein ---
Unified diff for dmesg.boot files is still useful and lack of it won't make it
easier to understand the problem for an arm64 developer.
Interrupts are assigned in very different ways for those
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #15 from h...@restart.be ---
(In reply to Eugene Grosbein from comment #10)
On r334436 I cancel the fetch after more than 10MB downloaded
On r320599 I fetch the complete 854MB
The files seems too different for a diff to be
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #14 from h...@restart.be ---
Created attachment 194153
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194153=edit
vmstat -ai of r320599
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #12 from h...@restart.be ---
Created attachment 194151
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194151=edit
Verbose boot of r320599
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #13 from h...@restart.be ---
Created attachment 194152
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194152=edit
vmstat -ai of r334436
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #11 from h...@restart.be ---
Created attachment 194150
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=194150=edit
Verbose boot of r334436
--
You are receiving this mail because:
You are the assignee for the bug.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #10 from Eugene Grosbein ---
(In reply to hlh from comment #9)
I suspect there is some kind of hardware driver problem, e.g. delayed or missed
interrupts etc. Please collect more information:
1) perform verbose boot, if
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
--- Comment #9 from h...@restart.be ---
(In reply to Eugene Grosbein from comment #8)
Strange results during fetch:
With r334436 the cpu is all the time idle: 99%
With r320599 the cpu is user: 3% system: 8% interrupt: 11%
--
You are
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
Eugene Grosbein changed:
What|Removed |Added
CC||eu...@freebsd.org
--- Comment
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=228802
Eugene Grosbein changed:
What|Removed |Added
Component|Individual Port(s) |kern
Version|Latest
21 matches
Mail list logo
