Click here to go to the forum index Click here for the home page
 
Author Message

<  MyStuff  ~  EPG2MEI -- new event API?

Page 34 of 39
Goto page Previous  1, 2, 3 ... 33, 34, 35 ... 37, 38, 39  Next
pedrovski
Posted: Thu Jan 28, 2010 5:02 pm Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
R2-D2 wrote:
Not sure if/how well the Toppy copes with USB reading back the file while it's open elsewhere.
To clarify, the corruption happens when the file is written to the toppy, e.g. it happens regardless of whether I subsequently copy the file back from the toppy or not. (EPG2MEI's logs confirm this).

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
R2-D2
Posted: Thu Jan 28, 2010 9:04 pm Reply with quote
Frequent contributor Joined: 18 Dec 2006 Posts: 12148
pedrovski wrote:
To clarify, the corruption happens when the file is written to the toppy
This looks to me to be the classic Toppy issue where it reports file sizes that are off by 512 bytes (see [Fl], but don't think about using that patch!), causing your USB connection to over-read by 512 bytes. But it doesn't occur on a 512-byte boundary, so I'm still investigating...

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches
View user's profile Send private message Visit poster's website
pedrovski
Posted: Thu Jan 28, 2010 10:30 pm Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
I'd still love to hear if anyone else is seeing this, if it's reproducible for anyone else. I can't shake the strong suspicion that it's something I'm doing in my USB file transfer code.

Is anyone else seeing "MEI error" entries in their EPG2MEI.log?

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
R2-D2
Posted: Thu Jan 28, 2010 11:05 pm Reply with quote
Frequent contributor Joined: 18 Dec 2006 Posts: 12148
pedrovski wrote:
I can't shake the strong suspicion that it's something I'm doing in my USB file transfer code.
Your code? Yes, I suppose it's possible you've missed one of the known bugs in the USB hardware/software if you've got your own driver, but then you'd probably get a similar thing on other files, or the same file under a different name...

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches
View user's profile Send private message Visit poster's website
pedrovski
Posted: Fri Jan 29, 2010 12:19 am Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
R2-D2 wrote:
Your code?

This is why I'm asking for validation from someone else. My driver is using an import of Steve Woodford's code*. I'm painfully aware that makes my situation very different.

R2-D2 wrote:
Yes, I suppose it's possible you've missed one of the known bugs in the USB hardware/software if you've got your own driver, but then you'd probably get a similar thing on other files, or the same file under a different name...

It only seems to occur in this one very specific case. I've not seen anything like it and can't reproduce it under any other circumstances.

* http://forum.toppy.org.uk/forum/viewtopic.php?p=50071

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
pedrovski
Posted: Fri Jan 29, 2010 2:35 am Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
R2-D2 wrote:
This looks to me to be the classic Toppy issue where it reports file sizes that are off by 512 bytes

-rw-r--r-- 1 pedrovski pedrovski 65476 Jan 27 23:12 mei.diff

http://rapidshare.com/files/342670972/mei.diff.html

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
R2-D2
Posted: Fri Jan 29, 2010 11:05 am Reply with quote
Frequent contributor Joined: 18 Dec 2006 Posts: 12148
The diff isn't particularly helpful -- I think it's the same as last time with sections repeated on the end. EPG2MEI does hook the USB file receiver, and specifically on the Close(), but I've doubled checked the code (and the disassembly) and I can't see why it wouldn't work transparently. As hinted previously, does it do the same thing if you change the name of the file (so EPG2MEI doesn't do any other processing on it)?

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches
View user's profile Send private message Visit poster's website
pedrovski
Posted: Fri Jan 29, 2010 12:34 pm Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
R2-D2 wrote:
The diff isn't particularly helpful -- I think it's the same as last time with sections repeated on the end.

I thought the size might be significant, it's more than off-by-512.
R2-D2 wrote:
EPG2MEI does hook the USB file receiver, and specifically on the Close(), but I've doubled checked the code (and the disassembly) and I can't see why it wouldn't work transparently. As hinted previously, does it do the same thing if you change the name of the file (so EPG2MEI doesn't do any other processing on it)?

No, I have to write to the file EPG2MEI has open or no problem occurs. I can e.g.do this:
cp /tmp/RT.mei ProgramFiles/tmp
rm ProgramFiles/RT.mei
rename ProgramFiles/tmp ProgramFiles/RT.mei
This does not trigger the problem. EPG2MEI does not report that RT.mei has changed (but does subsequently read it without reporting any errors) and ProgramFiles/RT.mei is not corrupt.

I'm rebuilding the kernel with debug on in the toppy driver. This will provide a full trace of the USB activity.

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
R2-D2
Posted: Fri Jan 29, 2010 12:39 pm Reply with quote
Frequent contributor Joined: 18 Dec 2006 Posts: 12148
pedrovski wrote:
No, I have to write to the file EPG2MEI has open or no problem occurs.
I think it's your USB code then, because every USB input file closure goes through pretty much the same hook code in EPG2MEI -- I don't see why that specific one would work differently unless it's connected to bad sharing. It certainly doesn't happen on my TF5810 (or TF5800).

_________________
Troubleshooting -- User Manual -- Dark Side of the Matrix: Firmwares and Patches
View user's profile Send private message Visit poster's website
pedrovski
Posted: Fri Jan 29, 2010 2:34 pm Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
R2-D2 wrote:
I think it's your USB code then, because every USB input file closure goes through pretty much the same hook code in EPG2MEI -- I don't see why that specific one would work differently unless it's connected to bad sharing.

Oh I see. I thought it might be because you had the equivalent of open() on the file. If you're hooking the firmware to call your code on USB file close() events instead, I can see that would affect all transfers, which is not the case here.

R2-D2 wrote:
It certainly doesn't happen on my TF5810 (or TF5800).


Thanks, that's the vital bit of information I need. I'll see what the USB trace throws up.

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
pedrovski
Posted: Mon Feb 08, 2010 1:27 am Reply with quote
Regular contributor Joined: 16 Jul 2008 Posts: 66
R2-D2 wrote:
I think it's your USB code then
Me too.

Long story.

http://www.openbsd.org/loongson.html

_________________
TF5810, TS On, F/W: 15.09AbAfBfBqC0CeCkCpCwDeEcEfEgEmEpErEsEvFFmGmHHeKtMMhNfOtPcPePsRRaRhRmRpRsSScSdSfSlSrSxTdTeTfTpTqTtUVrXwXl
TAPs: Extend v1.7; MPDisplayLITE V1.2; EIT Sub v0.6; EPG2MEI v0.96; SecCache (UK) v0.4; MyStuff 6.1; MyInfo B4.1a; Font Manager 1.0d;
Sig generated by MyInfo on 24/1/10
View user's profile Send private message
ChunkyBloke
Posted: Thu Mar 11, 2010 10:50 pm Reply with quote
Regular contributor Joined: 14 Jul 2006 Posts: 84
Shoot me down if this is in the wrong place but I think this is an EPG2MEI issue.

Every now and again (maybe once every 3 months or so) I get periods of missing EPG data in Mystuff. This is happening now and appears to be all MUX2 channels where the data is missing. In the native Toppy EPG the data is still there, all present and correct - it just seems as if EPG2MEI is ignoring MUX2 for some reason. Deleting .mei files to force them to be be recreated just reproduces the problem i.e. after 15 mins or so of no button pressing, the EPG is populated with the exception of MUX2 channels.

When this has happened before, it just seems to sort itself out after a few days with no further interference on my part.

Not that I think it makes any difference, but I'm on the Waltham transmitter.

Any ideas anyone?

_________________
Ex-Toppy (TF5800x2) now Enigma2 based Vu+ Duo2 4xDVB-C/T/T2 tuners, 1 TB HDD running OpenVix
View user's profile Send private message
juwlz
Posted: Thu Mar 11, 2010 11:16 pm Reply with quote
MyStuff Team Joined: 12 Aug 2005 Posts: 10802 Location: Wokingham, Berkshire (Hannington transmitter)
Do you leave your Toppy on 24/7? I've seen something similar (although I from memory it was on all channels), but only if the Toppy hadn't been in standby for several days. And of course, when I tried to reproduce it deliberately, I couldn't.

Julie

_________________
5800, TS On, F/W: MS6 Recommended F/W 12/9/2009 -Sy+BmC0CfCtFsIMPePsR3UUuUxZ
5810, TS On, F/W: F/W: MS6 Recommended F/W 12/9/2009 +BmCfCtFsR3Z
MyStuff info; Help!; Reference
Harmony 885 remote. Sig date 8 April 2012
View user's profile Send private message Visit poster's website
ChunkyBloke
Posted: Thu Mar 11, 2010 11:33 pm Reply with quote
Regular contributor Joined: 14 Jul 2006 Posts: 84
Hi Juwlz, no it never gets left on 24/7 - it usually goes to bed when I do ! Very Happy

Sod's law means that after a couple of days of MUX2 data being missing and me posting here, it's gone and fixed itself again!

Fortunately this doesn't happen too often, so it's not too much of an issue, but it is always MUX2 which seems a bit weird - could it somehow be associated with an issue I read somewhere else about the EPG.MEI file getting truncated somehow?

_________________
Ex-Toppy (TF5800x2) now Enigma2 based Vu+ Duo2 4xDVB-C/T/T2 tuners, 1 TB HDD running OpenVix
View user's profile Send private message
juwlz
Posted: Fri Mar 12, 2010 8:55 am Reply with quote
MyStuff Team Joined: 12 Aug 2005 Posts: 10802 Location: Wokingham, Berkshire (Hannington transmitter)
ChunkyBloke wrote:
Fortunately this doesn't happen too often, so it's not too much of an issue, but it is always MUX2 which seems a bit weird - could it somehow be associated with an issue I read somewhere else about the EPG.MEI file getting truncated somehow?
I wouldn't have thought so - usually that results in a crash, not just missing EPG data, because the data gets tructated randomly in the middle of a record.

Next time it happens, have a look at the native EPG and see if the data is missing there, or whether it's just missing in the MyStuff EPG. If the former, it's the broadcaster's fault; if the latter, R2-D2 will probably want to look at your EPG2MEI log Wink.

Julie

_________________
5800, TS On, F/W: MS6 Recommended F/W 12/9/2009 -Sy+BmC0CfCtFsIMPePsR3UUuUxZ
5810, TS On, F/W: F/W: MS6 Recommended F/W 12/9/2009 +BmCfCtFsR3Z
MyStuff info; Help!; Reference
Harmony 885 remote. Sig date 8 April 2012
View user's profile Send private message Visit poster's website

Display posts from previous:  

All times are GMT + 1 Hour
Page 34 of 39
Goto page Previous  1, 2, 3 ... 33, 34, 35 ... 37, 38, 39  Next

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