[En-Nut-Discussion] Preview Fix 060626

Michael Jones Michael.e.Jones at web.de
Mon Jun 26 22:21:31 CEST 2006


Hello,

I've had a good look into these new files and that what I see looks good! I
like the way the NutTimerProcessElapsed(...) is shrunk and all...

But, it does not work for me. The Ethernet "path" (don't know where yet)
dies so fast that I can't even upload new firmware before it happens. :-(

But there is a distinctive difference that I can observe using
NutDumpTimerList(...) which is called every 5 seconds from the main thread:

"my" working version (Hendrik's and my patches)

Heap: 10947<VT>
Addr Ticks  Left Callback<LF><CR>
4DB9     0    46 00000(3DAC)<LF><CR>
499F     0    14 NutEventTimeout(54EC)<LF><CR>
553E     0    96 NutEventTimeout(3DAC)<LF><CR>
451C     0    33 NutEventTimeout(4874)<LF><CR>
552A     0   184 NutThreadWake(4630)<LF><CR>
5516     0  4299 NutThreadWake(5260)<LF><CR>
4DA5     0927537 NutEventTimeout(1134)<LF><CR>
Heap: 10973<VT>
Addr Ticks  Left Callback<LF><CR>
499F     0    72 NutEventTimeout(54EC)<LF><CR>
553E     0    92 NutEventTimeout(3DAC)<LF><CR>
552A     0   219 NutThreadWake(4630)<LF><CR>
5516     0  2303 NutThreadWake(5260)<LF><CR>
451C     0 25457 NutEventTimeout(4874)<LF><CR>
4DA5     0902080 NutEventTimeout(1134)<LF><CR>
Heap: 10949<VT>
Addr Ticks  Left Callback<LF><CR>
4DB9     0    40 00000(3DAC)<LF><CR>
499F     0    43 NutEventTimeout(54EC)<LF><CR>
553E     0    67 NutEventTimeout(3DAC)<LF><CR>
552A     0   242 NutThreadWake(4630)<LF><CR>
5516     0   307 NutThreadWake(5260)<LF><CR>
451C     0 25457 NutEventTimeout(4874)<LF><CR>
4DA5     0902080 NutEventTimeout(1134)<LF><CR>
Heap: 11017<VT>
Addr Ticks  Left Callback<LF><CR>
553E     0    57 00000(3DAC)<LF><CR>
499F     0    30 NutEventTimeout(54EC)<LF><CR>
4DB9     0    66 NutEventTimeout(3DAC)<LF><CR>
552A     0   249 NutThreadWake(4630)<LF><CR>
5516     0  3301 NutThreadWake(5260)<LF><CR>
451C     0 20467 NutEventTimeout(4874)<LF><CR>
4DA5     0902080 NutEventTimeout(1134)<LF><CR>

The preview Fix version:

Heap: 11017<VT>
Addr Ticks  Left Callback<LF><CR>
499E     0    81 NutEventTimeout(3DA9)<LF><CR>
4CA3     0    91 NutThreadWake(462F)<LF><CR>
4C8D     0  3690 NutThreadWake(5261)<LF><CR>
4519     0 15318 NutEventTimeout(4873)<LF><CR>
4C77     0152713 NutEventTimeout(1134)<LF><CR>
Heap: 11017<VT>
Addr Ticks  Left Callback<LF><CR>
499E     0    10 NutEventTimeout(3DA9)<LF><CR>
4CA3     0   171 NutThreadWake(462F)<LF><CR>
4C8D     0  1694 NutThreadWake(5261)<LF><CR>
4519     0 15318 NutEventTimeout(4873)<LF><CR>
4C77     0152713 NutEventTimeout(1134)<LF><CR>
Heap: 11017<VT>
Addr Ticks  Left Callback<LF><CR>
499E     0   150 NutEventTimeout(3DA9)<LF><CR>
4CA3     0    41 NutThreadWake(462F)<LF><CR>
4C8D     0  4688 NutThreadWake(5261)<LF><CR>
4519     0 10328 NutEventTimeout(4873)<LF><CR>
4C77     0152713 NutEventTimeout(1134)<LF><CR>
Heap: 10828<VT>
Addr Ticks  Left Callback<LF><CR>
549F     0    81 NutEventTimeout(1168)<LF><CR>
4CA3     0   419 NutThreadWake(462F)<LF><CR>
4C8D     0  2393 NutThreadWake(5261)<LF><CR>
4519     0 10328 NutEventTimeout(4873)<LF><CR>
4C77     0152713 NutEventTimeout(1134)<LF><CR>

You may note that there is one NutEventTimeout less in the Timer list...

Any Ideas?

Cu,
Michael


-----Original Message-----
From: en-nut-discussion-bounces at egnite.de
[mailto:en-nut-discussion-bounces at egnite.de] On Behalf Of Harald Kipp
Sent: Monday, June 26, 2006 8:44 PM
To: Ethernut User Chat (English)
Subject: [En-Nut-Discussion] Preview Fix 060626

My version of a fix
ethernut-fix060626.zip
is available at
http://www.ethernut.de/arc/

It contains a number of files to be replaced by 4.1.5 or
later or CVS HEAD.

It should solve the disappearing thread issue, the timers-
eating-all-memory issue and it should further reduce critical
sections.

Please try this with one or two applications and let me know
any problems.

Harald

_______________________________________________
En-Nut-Discussion mailing list
En-Nut-Discussion at egnite.de
http://www.egnite.de/mailman/listinfo.cgi/en-nut-discussion





More information about the En-Nut-Discussion mailing list