Mailing List Archive

> On 1 Feb 2021, at 9:44 pm, John Hoyt <john.hoyt@gmail.com> wrote:
>
> process launch

Despite trying a dozen times, including the same recording that I used to make the trace I posted, both under lldb and native I have not suceeded in another core dump.
I'll keep trying.
James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

On Tue, 2 Feb 2021 at 00:08, James Linder <jam@tigger.ws> wrote:
>
> Looks like a good start point, say if you want more:
>
> * thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x16c13e008)
> * frame #0: 0x00000001044689a8 myth`AudioOutputGraph::GetImage(long long) const + 2008
> myth`AudioOutputGraph::GetImage:
> -> 0x1044689a8 <+2008>: movswl (%rax), %ecx
> 0x1044689ab <+2011>: movl %ecx, -0x168(%rbp)
> 0x1044689b1 <+2017>: cmpl $0x1, -0x154(%rbp)
> 0x1044689b8 <+2024>: jle 0x1044689d4 ; <+2052>
>
If I were a betting man, I'd guess that:-

https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L287

is the problem. e.g. the timestamps are bogus (due to the problematic
stream) and the error is not picked up and/or checked, we then convert
that timestamp to an offset into a buffer - and don't bounds check the
offset in:-

https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L123

However - I would expect the issue to be 100% repeatable if that were
the case...

James - when it doesn't crash - do you sometimes see garbage in the
audio graph or does it look 'normal'?

regards
Mark
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

On 02/02/2021 13:52, Mark Kendall wrote:
> On Tue, 2 Feb 2021 at 00:08, James Linder <jam@tigger.ws> wrote:
>>
>> Looks like a good start point, say if you want more:
>>
>> * thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x16c13e008)
>> * frame #0: 0x00000001044689a8 myth`AudioOutputGraph::GetImage(long long) const + 2008
>> myth`AudioOutputGraph::GetImage:
>> -> 0x1044689a8 <+2008>: movswl (%rax), %ecx
>> 0x1044689ab <+2011>: movl %ecx, -0x168(%rbp)
>> 0x1044689b1 <+2017>: cmpl $0x1, -0x154(%rbp)
>> 0x1044689b8 <+2024>: jle 0x1044689d4 ; <+2052>
>>
> If I were a betting man, I'd guess that:-
>
> https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L287
>
> is the problem. e.g. the timestamps are bogus (due to the problematic
> stream) and the error is not picked up and/or checked, we then convert
> that timestamp to an offset into a buffer - and don't bounds check the
> offset in:-
>
> https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L123
>
> However - I would expect the issue to be 100% repeatable if that were
> the case...
>
> James - when it doesn't crash - do you sometimes see garbage in the
> audio graph or does it look 'normal'?
>
> regards
> Mark

I had an encounter with the AOG code at Trac #12901, but gave up. IIRC
the audiotimecode there is not rigidly connected to the value at which
the display routine is entered...

John P
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> On 2 Feb 2021, at 10:57 pm, John Pilkington <johnpilk222@gmail.com> wrote:
>
> On 02/02/2021 13:52, Mark Kendall wrote:
>> On Tue, 2 Feb 2021 at 00:08, James Linder <jam@tigger.ws> wrote:
>>>
>>> Looks like a good start point, say if you want more:
>>>
>>> * thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x16c13e008)
>>> * frame #0: 0x00000001044689a8 myth`AudioOutputGraph::GetImage(long long) const + 2008
>>> myth`AudioOutputGraph::GetImage:
>>> -> 0x1044689a8 <+2008>: movswl (%rax), %ecx
>>> 0x1044689ab <+2011>: movl %ecx, -0x168(%rbp)
>>> 0x1044689b1 <+2017>: cmpl $0x1, -0x154(%rbp)
>>> 0x1044689b8 <+2024>: jle 0x1044689d4 ; <+2052>
>>>
>> If I were a betting man, I'd guess that:-
>> https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L287

Actually that makes sense

if *most* timestamps are valid and you choose those by chance then all is well
but if you skip to a bad one then tilt.I never get a crash on E or <space> only on skip
Also editing good streams (either fixed or good luck recording) then no problems

>> is the problem. e.g. the timestamps are bogus (due to the problematic
>> stream) and the error is not picked up and/or checked, we then convert
>> that timestamp to an offset into a buffer - and don't bounds check the
>> offset in:-
>> https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L123
>> However - I would expect the issue to be 100% repeatable if that were
>> the case...

As I said if you choose the same point then it is repeatable (no doubt) but for testing Johns builg I marked and cut at random :-)

>> James - when it doesn't crash - do you sometimes see garbage in the
>> audio graph or does it look 'normal'?

I'll pay attention and report back

Thanks Mark, just one niggling point why the mac frontend but never my linux front ends. For past ages I have done my HD cutlist on linux and have never had a crash (SuSE and ubuntu, only built from source) crash.
On my laptop (mac) I run an ubuntu VM just for editing! (ansible just works on ubuntu, but needs handholding on SuSE)

(A comment: To some extent I find ubuntu leading the charge in "You WILL do what WE think is best" eg mythtv user for mythtv, sillypasswords not mythtv for sql, you want gsudo? forget it! It's a bad idea and you dont need it.
at least SuSE have xdg_menu_su)


james
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> On 3 Feb 2021, at 6:43 am, James Linder <jam@tigger.ws> wrote:
>
>
>
>> On 2 Feb 2021, at 10:57 pm, John Pilkington <johnpilk222@gmail.com> wrote:
>>
>> On 02/02/2021 13:52, Mark Kendall wrote:
>>> On Tue, 2 Feb 2021 at 00:08, James Linder <jam@tigger.ws> wrote:
>>>>
>>>> Looks like a good start point, say if you want more:
>>>>
>>>> * thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=1, address=0x16c13e008)
>>>> * frame #0: 0x00000001044689a8 myth`AudioOutputGraph::GetImage(long long) const + 2008
>>>> myth`AudioOutputGraph::GetImage:
>>>> -> 0x1044689a8 <+2008>: movswl (%rax), %ecx
>>>> 0x1044689ab <+2011>: movl %ecx, -0x168(%rbp)
>>>> 0x1044689b1 <+2017>: cmpl $0x1, -0x154(%rbp)
>>>> 0x1044689b8 <+2024>: jle 0x1044689d4 ; <+2052>
>>>>
>>> If I were a betting man, I'd guess that:-
>>> https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L287
>
> Actually that makes sense
>
> if *most* timestamps are valid and you choose those by chance then all is well
> but if you skip to a bad one then tilt.I never get a crash on E or <space> only on skip
> Also editing good streams (either fixed or good luck recording) then no problems
>
>>> is the problem. e.g. the timestamps are bogus (due to the problematic
>>> stream) and the error is not picked up and/or checked, we then convert
>>> that timestamp to an offset into a buffer - and don't bounds check the
>>> offset in:-
>>> https://github.com/MythTV/mythtv/blob/fixes/31/mythtv/libs/libmyth/audio/audiooutputgraph.cpp#L123
>>> However - I would expect the issue to be 100% repeatable if that were
>>> the case...
>
> As I said if you choose the same point then it is repeatable (no doubt) but for testing Johns builg I marked and cut at random :-)
>
>>> James - when it doesn't crash - do you sometimes see garbage in the
>>> audio graph or does it look 'normal'?
>
> I'll pay attention and report back
>
> Thanks Mark, just one niggling point why the mac frontend but never my linux front ends. For past ages I have done my HD cutlist on linux and have never had a crash (SuSE and ubuntu, only built from source) crash.
> On my laptop (mac) I run an ubuntu VM just for editing! (ansible just works on ubuntu, but needs handholding on SuSE)
>
> (A comment: To some extent I find ubuntu leading the charge in "You WILL do what WE think is best" eg mythtv user for mythtv, sillypasswords not mythtv for sql, you want gsudo? forget it! It's a bad idea and you dont need it.
> at least SuSE have xdg_menu_su)

John is the git repo clean for all your fiddling (presumably debug, not release hence the debug version)

v31-016630a35c

Is mostly working for me, I’d like to build it for High Sierra. Are the fixes John P suggested comitted?
Does your build-script address the python packages oops?

Thanks
James


_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

>
> John is the git repo clean for all your fiddling (presumably debug, not
> release hence the debug version)
>

Fiddling (according to Oxford - annoyingly trivial or petty) would not have
been the word I would personally have selected to describe my activity.
...but I guess everyone has a different perspective on life.

I'm not sure I understand what you are asking. I don't have a public git
repo at the moment, but I'll try to answer what I believe is your question.

The mythtv fixes/31 github repo currently has the John P suggested fix to
MythUIButtonListItem (which fixed your first cutlist issue) committed.

The mythtv packaging github repo's fixes/31 and master branches both have
all of my recent updates to the compile script committed (including the fix
for my python path error)

The mythtv ansible github repo has all of the latest updates to ansible
(including the new user selectable database_vsersion flag committed).

The compile script clones the mythtv github repo at HEAD based on the
default for the script or user specified branch. The current script does
not have any switches for debug/profile builds so if you are going to
compile on your own you'll need to add it to the script's config flags.

v31-016630a35c
>
Is mostly working for me, I’d like to build it for High Sierra. Are the
> fixes John P suggested comitted?
>

The posted binaries for v31-016630a35c and v31-1aff6fd2db (the latest
release) both have the MythUIButtonListItem patch and the python path
fix. v31-1aff6fd2db also has some additional fixes to the metadata
grabbers while "might" fix some of your frustrations - or not...

Does your build-script address the python packages oops?
>

Yes (covered above)

If you are going to build for High Sierra - you will need to build on a
High Sierra computer or VM. Macports likes to only install ports for the
macOS version you are running.

Also, if you are running Catalina, you'll do better to build off of a
Catalina machine. There were many graphics performance enhancements in
Catalina that will let video playback work better and use less CPU.

BTW - I believe the rest of us are waiting on you providing an update on
where / how your cutlist editor experience occurs (i.e. responses to Mark's
questions in a previous email).

> On 5 Feb 2021, at 8:11 pm, John Hoyt <john.hoyt@gmail.com> wrote:
>
> John is the git repo clean for all your fiddling (presumably debug, not release hence the debug version)
>
> Fiddling (according to Oxford - annoyingly trivial or petty) would not have been the word I would personally have selected to describe my activity. ...but I guess everyone has a different perspective on life.

I did not mean that, or imply that <smile> I used fiddling in context of very detailed work requiring great skill and patience ie a fiddly job is one where considerable care is required as well as skill to achive a successful outcome. I'm sorry that my venacular conveyed the Oxford meaning.

> I'm not sure I understand what you are asking. I don't have a public git repo at the moment, but I'll try to answer what I believe is your question.
>
> The mythtv fixes/31 github repo currently has the John P suggested fix to MythUIButtonListItem (which fixed your first cutlist issue) committed.
>
> The mythtv packaging github repo's fixes/31 and master branches both have all of my recent updates to the compile script committed (including the fix for my python path error)
>
> The mythtv ansible github repo has all of the latest updates to ansible (including the new user selectable database_vsersion flag committed).
>
> The compile script clones the mythtv github repo at HEAD based on the default for the script or user specified branch. The current script does not have any switches for debug/profile builds so if you are going to compile on your own you'll need to add it to the script's config flags.
>
> v31-016630a35c
> Is mostly working for me, I’d like to build it for High Sierra. Are the fixes John P suggested comitted?
>
> The posted binaries for v31-016630a35c and v31-1aff6fd2db (the latest release) both have the MythUIButtonListItem patch and the python path fix. v31-1aff6fd2db also has some additional fixes to the metadata grabbers while "might" fix some of your frustrations - or not...
>
> Does your build-script address the python packages oops?
>
> Yes (covered above)
>
> If you are going to build for High Sierra - you will need to build on a High Sierra computer or VM. Macports likes to only install ports for the macOS version you are running.
>
> Also, if you are running Catalina, you'll do better to build off of a Catalina machine. There were many graphics performance enhancements in Catalina that will let video playback work better and use less CPU.

I have a 2011 iMac 27. It can't go past Migh Sierra.
So I want a frontend on my catelina macbook and one on the iMac.
But, as you said "covered above" so thanks.

> BTW - I believe the rest of us are waiting on you providing an update on where / how your cutlist editor experience occurs (i.e. responses to Mark's questions in a previous email).

I posted the lldg trace of a crash, and posted that I had been unable to get another. Perhaps fine weather makes even the worst recordings look good. Rain (40 mm tomorrow) so I'll try then.

ciao n thanks
James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

>
> I did not mean that, or imply that <smile> I used fiddling in context of
> very detailed work requiring great skill and patience ie a fiddly job is
> one where considerable care is required as well as skill to achive a
> successful outcome. I'm sorry that my venacular conveyed the Oxford meaning.
>

The english language is a rather poor way of communicating. Adding in
local vernacular can often makes things even more troublesome. I'll try to
take that into account in the future.


> I have a 2011 iMac 27. It can't go past Migh Sierra.
> So I want a frontend on my catelina macbook and one on the iMac.
> But, as you said "covered above" so thanks.
>

The are High Sierra builds on the sourceforge site with the latest updates,
but feel free to build them on your own. Having someone else use the build
script will only make it better (and will uncover bugs that I haven't
considered / been able to test). If you do build on your own, please let
me know of any fixes you might incur (maybe off list) that way I can get
them worked into the sourceforge/github versions.

I posted the lldg trace of a crash, and posted that I had been unable to
> get another. Perhaps fine weather makes even the worst recordings look
> good. Rain (40 mm tomorrow) so I'll try then.
>

I saw that one, but the repeatability issue was where the comment came
from. Was it crashing in the same place, same file / memory access error.
In the bad file was there any other things being tosed out when it
magically worked. Sorry for the rain, but glad you'll be getting
additional data points to test.

I've tried to get the cutlist editor to crash locally to no avail. I even
fired up a fresh Catalona and High Sierra VM - and I just can't get it to
segfault.

> On 5 Feb 2021, at 9:18 pm, John Hoyt <john.hoyt@gmail.com> wrote:
>
> I saw that one, but the repeatability issue was where the comment came from. Was it crashing in the same place, same file / memory access error. In the bad file was there any other things being tosed out when it magically worked. Sorry for the rain, but glad you'll be getting additional data points to test.

What I did was very slack -sorry

Select a movie
play some seconds
skip forward a few times
E
<space>
<up><up> to 1 min
<rt><rt> - crash

mythffprobe listed lots of errors

OK bizare:

I started with a 1080 h264 movie
mythffprobe showed lots of errors
using your JP fixes I cut the movie, and spliced it together with my h264cut script.

I fiddled (there is that word again, I played around) with handbrake, shotcut to try fixing the errors.

My cut and trancoded video has no errors.
My original movie has no errors. What did I do? In particular I've been trying in vain to get a core dump using the "original" movie that no longer has errors !!!

OK valid conclusion "Damaged recording causes core dump"
I shall make some recordings in the rain and try to cause a core dump on the morrow. I'll report what happens.

James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

So I finally managed to fix some storage problems on my server and
added 10074_20210123140800.ts to my videos, built a seek table for it
it and started to investigate.

Start editing and there is obvious garbage data in the audio graph
after some seeks and a little extra logging shows that
AudioOutputGraph is indeed going 'out of bounds'. Still trying to work
out the exact root cause but the symptoms are an easy fix - so should
be sorted in a few days (and yes - I'll backport to fixes/31).

Regards
Mark
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

On Fri, Feb 5, 2021 at 12:23 PM Mark Kendall <mark.kendall@gmail.com> wrote:

> So I finally managed to fix some storage problems on my server and
> added 10074_20210123140800.ts to my videos, built a seek table for it
> it and started to investigate.
>
> Start editing and there is obvious garbage data in the audio graph
> after some seeks and a little extra logging shows that
> AudioOutputGraph is indeed going 'out of bounds'. Still trying to work
> out the exact root cause but the symptoms are an easy fix - so should
> be sorted in a few days (and yes - I'll backport to fixes/31).
>

Thanks for looking into this Mark!

> On 6 Feb 2021, at 1:37 am, John Hoyt <john.hoyt@gmail.com> wrote:
>
> On Fri, Feb 5, 2021 at 12:23 PM Mark Kendall <mark.kendall@gmail.com> wrote:
> So I finally managed to fix some storage problems on my server and
> added 10074_20210123140800.ts to my videos, built a seek table for it
> it and started to investigate.
>
> Start editing and there is obvious garbage data in the audio graph
> after some seeks and a little extra logging shows that
> AudioOutputGraph is indeed going 'out of bounds'. Still trying to work
> out the exact root cause but the symptoms are an easy fix - so should
> be sorted in a few days (and yes - I'll backport to fixes/31).
>
> Thanks for looking into this Mark!

Indeed thanks.
I will still record-in-the-rain but the issue is confirmed already.
James

_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> On 6 Feb 2021, at 10:19 am, James Linder <jam@tigger.ws> wrote:
>
>
>
>> On 6 Feb 2021, at 1:37 am, John Hoyt <john.hoyt@gmail.com> wrote:
>>
>> On Fri, Feb 5, 2021 at 12:23 PM Mark Kendall <mark.kendall@gmail.com> wrote:
>> So I finally managed to fix some storage problems on my server and
>> added 10074_20210123140800.ts to my videos, built a seek table for it
>> it and started to investigate.
>>
>> Start editing and there is obvious garbage data in the audio graph
>> after some seeks and a little extra logging shows that
>> AudioOutputGraph is indeed going 'out of bounds'. Still trying to work
>> out the exact root cause but the symptoms are an easy fix - so should
>> be sorted in a few days (and yes - I'll backport to fixes/31).
>>
>> Thanks for looking into this Mark!
>

I just grabbed
1aff6fd2db
but I guess I ought to wait for Mark. So when he commit I download or build which ever is easier
James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> On 6 Feb 2021, at 11:03 am, James Linder <jam@tigger.ws> wrote:
>
>
>
>> On 6 Feb 2021, at 10:19 am, James Linder <jam@tigger.ws> wrote:
>>
>>
>>
>>> On 6 Feb 2021, at 1:37 am, John Hoyt <john.hoyt@gmail.com> wrote:
>>>
>>> On Fri, Feb 5, 2021 at 12:23 PM Mark Kendall <mark.kendall@gmail.com> wrote:
>>> So I finally managed to fix some storage problems on my server and
>>> added 10074_20210123140800.ts to my videos, built a seek table for it
>>> it and started to investigate.
>>>
>>> Start editing and there is obvious garbage data in the audio graph
>>> after some seeks and a little extra logging shows that
>>> AudioOutputGraph is indeed going 'out of bounds'. Still trying to work
>>> out the exact root cause but the symptoms are an easy fix - so should
>>> be sorted in a few days (and yes - I'll backport to fixes/31).
>>>
>>> Thanks for looking into this Mark!

An observation that may be of use here:

Out of the debugger the HD movie I recorded in the rain core dumps often
In the debugger (lldb) I hardly ever core dump
In linux the Edit->cut>move core dump never happens

In my experience a pointer that has gone awry:
Valid pointer -> my address space
Invalid pointer (debugger)-> somewhere in my address space
Invalid pointer (no debugger)-> outside my address space so core dump

linux: invalid-> somewhere in my address space
mac: invalid-> outside my address space

(all the above just by chance)
The visulation garbage supports this argument

James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> On 6 Feb 2021, at 11:03 am, James Linder <jam@tigger.ws> wrote:
>
> I just grabbed
> 1aff6fd2db
> but I guess I ought to wait for Mark. So when he commit I download or build which ever is easier

I kept pulling the dragon's tail until it bit :-)

here is quite a chunk of trace: http://tigger.ws/downloads/dump
Again it shows audio visulization.

James

_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

On 06/02/2021 03:03, James Linder wrote:
>
>
>> On 6 Feb 2021, at 10:19 am, James Linder <jam@tigger.ws> wrote:
>>
>>
>>
>>> On 6 Feb 2021, at 1:37 am, John Hoyt <john.hoyt@gmail.com> wrote:
>>>
>>> On Fri, Feb 5, 2021 at 12:23 PM Mark Kendall <mark.kendall@gmail.com> wrote:
>>> So I finally managed to fix some storage problems on my server and
>>> added 10074_20210123140800.ts to my videos, built a seek table for it
>>> it and started to investigate.
>>>
>>> Start editing and there is obvious garbage data in the audio graph
>>> after some seeks and a little extra logging shows that
>>> AudioOutputGraph is indeed going 'out of bounds'. Still trying to work
>>> out the exact root cause but the symptoms are an easy fix - so should
>>> be sorted in a few days (and yes - I'll backport to fixes/31).
>>>
>>> Thanks for looking into this Mark!
>>
>
> I just grabbed
> 1aff6fd2db
> but I guess I ought to wait for Mark. So when he commit I download or build which ever is easier
> James

After Mark's commit "AudioOutputGraph: Fix buffer overflow" yesterday at
0c76409, for the first time, returning to an edit point regenerates the
same audio graph. I look forward to the further cleanup and have
scheduled a recording of the time signal :-)

And FWIW https://lists.archive.carbon60.com/mythtv/dev/614955#614955

John P
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

>
> > I just grabbed
> > 1aff6fd2db
> > but I guess I ought to wait for Mark. So when he commit I download or
> build which ever is easier
>

James, I'll post fresh builds for High Sierra, Catalina, and Big Sur once
the updates get backported to v31.


> After Mark's commit "AudioOutputGraph: Fix buffer overflow" yesterday at
> 0c76409, for the first time, returning to an edit point regenerates the
> same audio graph. I look forward to the further cleanup and have
> scheduled a recording of the time signal :-)
>

Glad to hear Mark's fix seems to be working.

On 06/02/2021 14:58, John Hoyt wrote:
> > I just grabbed
> > 1aff6fd2db
> > but I guess I ought to wait for Mark. So when he commit I
> download or build which ever is easier
>
>
> James, I'll post fresh builds for High Sierra, Catalina, and Big Sur
> once the updates get backported to v31.
>
> After Mark's commit "AudioOutputGraph: Fix buffer overflow"
> yesterday at
> 0c76409, for the first time, returning to an edit point regenerates the
> same audio graph.  I look forward to the further cleanup and have
> scheduled a recording of the time signal  :-)
>
>
> Glad to hear Mark's fix seems to be working.
>

Unfortunately it's still not repeatable in that radio recording, but
progress all the same.
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> On 6 Feb 2021, at 11:08 pm, John Pilkington <johnpilk222@gmail.com> wrote:
>
> On 06/02/2021 14:58, John Hoyt wrote:
>> > I just grabbed
>> > 1aff6fd2db
>> > but I guess I ought to wait for Mark. So when he commit I
>> download or build which ever is easier
>> James, I'll post fresh builds for High Sierra, Catalina, and Big Sur once the updates get backported to v31.
>> After Mark's commit "AudioOutputGraph: Fix buffer overflow"
>> yesterday at
>> 0c76409, for the first time, returning to an edit point regenerates the
>> same audio graph. I look forward to the further cleanup and have
>> scheduled a recording of the time signal :-)
>> Glad to hear Mark's fix seems to be working.
>
> Unfortunately it's still not repeatable in that radio recording, but progress all the same.

John I don't think it's your ansible setup, I'll spelunk and also I'll build a version I've built in the past

. . .
clang -F/opt/local/libexec/qt5/lib -o obj/audiooutput.o audio/audiooutput.cpp
ccache clang++ -c -pipe -stdlib=libc++ -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -DPIC -I /Users/jam/mythtv-31/mythtv/mythtv/external -I /opt/local/include -std=c++11 -DNDEBUG -fomit-frame-pointer -fPIC -DQT_DISABLE_DEPRECATED_BEFORE=0x050700 -msse -pthread -g -Wall -Wextra -Wpointer-arith -D__STDC_CONSTANT_MACROS -D__STDC_LIMIT_MACROS -Wno-constant-logical-operand -Wno-unused-value -Qunused-arguments -Wimplicit-fallthrough -std=gnu++11 -arch x86_64 -isysroot /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk -mmacosx-version-min=10.13 -fPIC -DMMX -DUSING_APPLEREMOTE -D_GNU_SOURCE -DMYTH_API -DQT_NO_DEBUG -DQT_WIDGETS_LIB -DQT_GUI_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_SQL_LIB -DQT_SCRIPT_LIB -DQT_DBUS_LIB -DQT_CORE_LIB -I. -I/opt/local/include/libxml2 -I../../external/libmythsoundtouch -I../libmythfreesurround -I../libmythbase -I../.. -I.. -I. -I../libmythupnp -I../libmythui -I../.. -I../../external/FFmpeg -I../libmythservicecontrac
ts -I/opt/local/libexec/qt5/lib/QtWidgets.framework/Headers -I/opt/local/libexec/qt5/lib/QtGui.framework/Headers -I/opt/local/libexec/qt5/lib/QtNetwork.framework/Headers -I/opt/local/libexec/qt5/lib/QtXml.framework/Headers -I/opt/local/libexec/qt5/lib/QtSql.framework/Headers -I/opt/local/libexec/qt5/lib/QtScript.framework/Headers -I/opt/local/libexec/qt5/lib/QtDBus.framework/Headers -I/opt/local/libexec/qt5/lib/QtCore.framework/Headers -Imoc -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenGL.framework/Headers -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/AGL.framework/Headers -I/opt/local/libexec/qt5/mkspecs/macx-clang -F/opt/local/libexec/qt5/lib -o obj/audiooutputbase.o audio/audiooutputbase.cpp
audio/audiooutputbase.cpp:630:18: error: use of undeclared identifier 'ff_codec_id_string'
.arg(ff_codec_id_string(m_codec))
^
1 error generated.
make[2]: *** [obj/audiooutputbase.o] Error 1

but since the error is right where we have been f...um...trying to solve a problem it is suspicious

James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

> audio/audiooutputbase.cpp:630:18: error: use of undeclared identifier
> 'ff_codec_id_string'
> .arg(ff_codec_id_string(m_codec))
> ^


What I believe is happening is the compiler is trying to build off of
ffmpeg's libraries and not the ones compiled and used internally by
mythtv. I never did sort this pathing issue out myself, but the easy
solution is to uninstall ffmpeg.

> On 7 Feb 2021, at 1:31 pm, James Linder <jam@tigger.ws> wrote:
>
>
>
>> On 6 Feb 2021, at 11:08 pm, John Pilkington <johnpilk222@gmail.com> wrote:
>>
>> On 06/02/2021 14:58, John Hoyt wrote:
>>>> I just grabbed
>>>> 1aff6fd2db
>>>> but I guess I ought to wait for Mark. So when he commit I
>>> download or build which ever is easier
>>> James, I'll post fresh builds for High Sierra, Catalina, and Big Sur once the updates get backported to v31.
>>> After Mark's commit "AudioOutputGraph: Fix buffer overflow"
>>> yesterday at
>>> 0c76409, for the first time, returning to an edit point regenerates the
>>> same audio graph. I look forward to the further cleanup and have
>>> scheduled a recording of the time signal :-)
>>> Glad to hear Mark's fix seems to be working.
>>
>> Unfortunately it's still not repeatable in that radio recording, but progress all the same.
>
> John I don't think it's your ansible setup, I'll spelunk and also I'll build a version I've built in the past
>
> . . .
> clang -F/opt/local/libexec/qt5/lib -o obj/audiooutput.o audio/audiooutput.cpp
> ccache clang++ -c -pipe -stdlib=libc++ -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -DPIC -I /Users/jam/mythtv-31/mythtv/mythtv/external -I /opt/local/include -std=c++11 -DNDEBUG -fomit-frame-pointer -fPIC -DQT_DISABLE_DEPRECATED_BEFORE=0x050700 -msse -pthread -g -Wall -Wextra -Wpointer-arith -D__STDC_CONSTANT_MACROS -D__STDC_LIMIT_MACROS -Wno-constant-logical-operand -Wno-unused-value -Qunused-arguments -Wimplicit-fallthrough -std=gnu++11 -arch x86_64 -isysroot /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk -mmacosx-version-min=10.13 -fPIC -DMMX -DUSING_APPLEREMOTE -D_GNU_SOURCE -DMYTH_API -DQT_NO_DEBUG -DQT_WIDGETS_LIB -DQT_GUI_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_SQL_LIB -DQT_SCRIPT_LIB -DQT_DBUS_LIB -DQT_CORE_LIB -I. -I/opt/local/include/libxml2 -I../../external/libmythsoundtouch -I../libmythfreesurround -I../libmythbase -I../.. -I.. -I. -I../libmythupnp -I../libmythui -I../.. -I../../external/FFmpeg -I../libmythservicecontr
ac
> ts -I/opt/local/libexec/qt5/lib/QtWidgets.framework/Headers -I/opt/local/libexec/qt5/lib/QtGui.framework/Headers -I/opt/local/libexec/qt5/lib/QtNetwork.framework/Headers -I/opt/local/libexec/qt5/lib/QtXml.framework/Headers -I/opt/local/libexec/qt5/lib/QtSql.framework/Headers -I/opt/local/libexec/qt5/lib/QtScript.framework/Headers -I/opt/local/libexec/qt5/lib/QtDBus.framework/Headers -I/opt/local/libexec/qt5/lib/QtCore.framework/Headers -Imoc -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenGL.framework/Headers -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/AGL.framework/Headers -I/opt/local/libexec/qt5/mkspecs/macx-clang -F/opt/local/libexec/qt5/lib -o obj/audiooutputbase.o audio/audiooutputbase.cpp
> audio/audiooutputbase.cpp:630:18: error: use of undeclared identifier 'ff_codec_id_string'
> .arg(ff_codec_id_string(m_codec))
> ^
> 1 error generated.
> make[2]: *** [obj/audiooutputbase.o] Error 1

To err is human
To really muck things up you need a computer

I'm using parallels as a VM and it does not like more than 2 cores. Since 6 cores is LOTS quicker I built native, not in a VM.

I tried to build 102dbe673325fccbc5bf073f94e4 in the past and it worked, but now that fails too so clearly my setup and most likely my mackports is causing a problem.

I ran my fe normally to watch the news
/Applications/Mythfrontend.app/Contents/MacOS/mythfrontend
dyld: Library not loaded: /opt/local/lib/libjpeg.9.dylib
Referenced from: /Applications/Mythfrontend.app/Contents/Frameworks/QtWebKit.framework/Versions/5/QtWebKit
Reason: image not found
/Applications/Mythfrontend.app/Contents/MacOS/mythfrontend: line 15: 401 Abort trap: 6 ./mythfrontend.real $@

I'll like my wounds, and re-install macos, but the bundle reffering to a macports dylib looks wrong.

James
_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

>
> I ran my fe normally to watch the news
> /Applications/Mythfrontend.app/Contents/MacOS/mythfrontend
> dyld: Library not loaded: /opt/local/lib/libjpeg.9.dylib
> Referenced from:
> /Applications/Mythfrontend.app/Contents/Frameworks/QtWebKit.framework/Versions/5/QtWebKit
> Reason: image not found
> /Applications/Mythfrontend.app/Contents/MacOS/mythfrontend: line 15: 401
> Abort trap: 6 ./mythfrontend.real $@
>
> I'll like my wounds, and re-install macos, but the bundle reffering to a
> macports dylib looks wrong.
>

Was this the version you downloaded, previously built yourself, or tried to
build providing this error?

If it's the downloaded version, that's a head scratcher.

If it's a previous built version - there could be a host of problems
depending on how old it was and if the version had all of the libraries
bundled correctly.

Reinstalling macOS is a bit extreme. I usually start with removing
macports and reinstalling that in case macports messed up the underlying
system.

https://guide.macports.org/chunked/installing.macports.uninstalling.html

BTW - when you reinstall macports, make sure to switch it to git for the
ports tree. You'll get the updated ports faster.

https://guide.macports.org/#installing.macports.git

On Sun, Feb 7, 2021 at 7:23 AM John Hoyt <john.hoyt@gmail.com> wrote:

> audio/audiooutputbase.cpp:630:18: error: use of undeclared identifier
>> 'ff_codec_id_string'
>> .arg(ff_codec_id_string(m_codec))
>> ^
>
>
> What I believe is happening is the compiler is trying to build off of
> ffmpeg's libraries and not the ones compiled and used internally by
> mythtv. I never did sort this pathing issue out myself, but the easy
> solution is to uninstall ffmpeg.
>

Another possibility is that you're compiling with a case-sensitive
filesystem. I found the case-sensitive filesystem to be troublesome and
moved my build systems to case-insensitive systems (which is what Apple
recommends for developers and regular users).

> On 7 Feb 2021, at 8:33 pm, James Linder <jam@tigger.ws> wrote:
>
>
>
>> On 7 Feb 2021, at 1:31 pm, James Linder <jam@tigger.ws> wrote:
>>
>>
>>
>>> On 6 Feb 2021, at 11:08 pm, John Pilkington <johnpilk222@gmail.com> wrote:
>>>
>>> On 06/02/2021 14:58, John Hoyt wrote:
>>>>> I just grabbed
>>>>> 1aff6fd2db
>>>>> but I guess I ought to wait for Mark. So when he commit I
>>>> download or build which ever is easier
>>>> James, I'll post fresh builds for High Sierra, Catalina, and Big Sur once the updates get backported to v31.
>>>> After Mark's commit "AudioOutputGraph: Fix buffer overflow"
>>>> yesterday at
>>>> 0c76409, for the first time, returning to an edit point regenerates the
>>>> same audio graph. I look forward to the further cleanup and have
>>>> scheduled a recording of the time signal :-)
>>>> Glad to hear Mark's fix seems to be working.
>>>
>>> Unfortunately it's still not repeatable in that radio recording, but progress all the same.
>>
>> John I don't think it's your ansible setup, I'll spelunk and also I'll build a version I've built in the past
>>
>> . . .
>> clang -F/opt/local/libexec/qt5/lib -o obj/audiooutput.o audio/audiooutput.cpp
>> ccache clang++ -c -pipe -stdlib=libc++ -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -DPIC -I /Users/jam/mythtv-31/mythtv/mythtv/external -I /opt/local/include -std=c++11 -DNDEBUG -fomit-frame-pointer -fPIC -DQT_DISABLE_DEPRECATED_BEFORE=0x050700 -msse -pthread -g -Wall -Wextra -Wpointer-arith -D__STDC_CONSTANT_MACROS -D__STDC_LIMIT_MACROS -Wno-constant-logical-operand -Wno-unused-value -Qunused-arguments -Wimplicit-fallthrough -std=gnu++11 -arch x86_64 -isysroot /Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk -mmacosx-version-min=10.13 -fPIC -DMMX -DUSING_APPLEREMOTE -D_GNU_SOURCE -DMYTH_API -DQT_NO_DEBUG -DQT_WIDGETS_LIB -DQT_GUI_LIB -DQT_NETWORK_LIB -DQT_XML_LIB -DQT_SQL_LIB -DQT_SCRIPT_LIB -DQT_DBUS_LIB -DQT_CORE_LIB -I. -I/opt/local/include/libxml2 -I../../external/libmythsoundtouch -I../libmythfreesurround -I../libmythbase -I../.. -I.. -I. -I../libmythupnp -I../libmythui -I../.. -I../../external/FFmpeg -I../libmythservicecont
r
> ac
>> ts -I/opt/local/libexec/qt5/lib/QtWidgets.framework/Headers -I/opt/local/libexec/qt5/lib/QtGui.framework/Headers -I/opt/local/libexec/qt5/lib/QtNetwork.framework/Headers -I/opt/local/libexec/qt5/lib/QtXml.framework/Headers -I/opt/local/libexec/qt5/lib/QtSql.framework/Headers -I/opt/local/libexec/qt5/lib/QtScript.framework/Headers -I/opt/local/libexec/qt5/lib/QtDBus.framework/Headers -I/opt/local/libexec/qt5/lib/QtCore.framework/Headers -Imoc -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/OpenGL.framework/Headers -I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX10.15.sdk/System/Library/Frameworks/AGL.framework/Headers -I/opt/local/libexec/qt5/mkspecs/macx-clang -F/opt/local/libexec/qt5/lib -o obj/audiooutputbase.o audio/audiooutputbase.cpp
>> audio/audiooutputbase.cpp:630:18: error: use of undeclared identifier 'ff_codec_id_string'
>> .arg(ff_codec_id_string(m_codec))
>> ^
>> 1 error generated.
>> make[2]: *** [obj/audiooutputbase.o] Error 1
>
> To err is human
> To really muck things up you need a computer
>
> I'm using parallels as a VM and it does not like more than 2 cores. Since 6 cores is LOTS quicker I built native, not in a VM.
>
> I tried to build 102dbe673325fccbc5bf073f94e4 in the past and it worked, but now that fails too so clearly my setup and most likely my mackports is causing a problem.
>
> I ran my fe normally to watch the news
> /Applications/Mythfrontend.app/Contents/MacOS/mythfrontend
> dyld: Library not loaded: /opt/local/lib/libjpeg.9.dylib
> Referenced from: /Applications/Mythfrontend.app/Contents/Frameworks/QtWebKit.framework/Versions/5/QtWebKit
> Reason: image not found
> /Applications/Mythfrontend.app/Contents/MacOS/mythfrontend: line 15: 401 Abort trap: 6 ./mythfrontend.real $@
>
> I'll like my wounds, and re-install macos, but the bundle reffering to a macports dylib looks wrong.
>

When Jack got out of the bottomless pit ...
[plasma] /Users/jam/mythtv-31/mythtv [550]% /Applications/Mythfrontend.app/Contents/MacOS/mythfrontend --version
dyld: Library not loaded: /opt/local/lib/libjpeg.9.dylib
Referenced from: /Applications/Mythfrontend.app/Contents/Frameworks/QtWebKit.framework/Versions/5/QtWebKit
Reason: image not found
/Applications/Mythfrontend.app/Contents/MacOS/mythfrontend: line 15: 503 Abort trap: 6 ./mythfrontend.real $@

but I'm sure it was 1aff6fd2db

Sorry about the spelling . . . typing in bed lick not like, macports not mackports :-)
I'll try remove-reinstall macports then bite the bullet. (Gee that sounds like the typical windows fix - reinstall (and yes it is not the same))
ciao
James

_______________________________________________
mythtv-users mailing list
mythtv-users@mythtv.org
http://lists.mythtv.org/mailman/listinfo/mythtv-users
http://wiki.mythtv.org/Mailing_List_etiquette
MythTV Forums: https://forum.mythtv.org

Well, I just recreated the compile error locally. Installing ffmpeg is
definitely the culprit. I'll try hacking at this issue to see if I can
figure out what in the include pathing is causing the error.

ccache clang++ -c -pipe -stdlib=libc++ -D_FILE_OFFSET_BITS=64
-D_LARGEFILE_SOURCE -DPIC -I /Users/jhoyt/mythtv-31/mythtv/mythtv/external
-I /opt/local/include -std=c++11 -DNDEBUG -fomit-frame-pointer -fPIC
-DQT_DISABLE_DEPRECATED_BEFORE=0x050700 -msse -pthread -g -Wall -Wextra
-Wpointer-arith -D__STDC_CONSTANT_MACROS -D__STDC_LIMIT_MACROS
-Wno-constant-logical-operand -Wno-unused-value -Qunused-arguments
-Wimplicit-fallthrough -std=gnu++11 -arch x86_64 -isysroot
/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX11.1.sdk
-mmacosx-version-min=10.13 -fPIC -DMMX -DUSING_APPLEREMOTE -D_GNU_SOURCE
-DMYTH_API -DQT_NO_DEBUG -DQT_WIDGETS_LIB -DQT_GUI_LIB -DQT_NETWORK_LIB
-DQT_XML_LIB -DQT_SQL_LIB -DQT_SCRIPT_LIB -DQT_DBUS_LIB -DQT_CORE_LIB -I.
-I/opt/local/include/libxml2 -I../../external/libmythsoundtouch
-I../libmythfreesurround -I../libmythbase -I../.. -I.. -I. -I../libmythupnp
-I../libmythui -I../.. -I../../external/FFmpeg -I../libmythservicecontracts
-I/opt/local/libexec/qt5/lib/QtWidgets.framework/Headers
-I/opt/local/libexec/qt5/lib/QtGui.framework/Headers
-I/opt/local/libexec/qt5/lib/QtNetwork.framework/Headers
-I/opt/local/libexec/qt5/lib/QtXml.framework/Headers
-I/opt/local/libexec/qt5/lib/QtSql.framework/Headers
-I/opt/local/libexec/qt5/lib/QtScript.framework/Headers
-I/opt/local/libexec/qt5/lib/QtDBus.framework/Headers
-I/opt/local/libexec/qt5/lib/QtCore.framework/Headers -Imoc
-I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX11.1.sdk/System/Library/Frameworks/OpenGL.framework/Headers
-I/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX11.1.sdk/System/Library/Frameworks/AGL.framework/Headers
-I/opt/local/libexec/qt5/mkspecs/macx-clang -F/opt/local/libexec/qt5/lib -o
obj/audiooutputbase.o audio/audiooutputbase.cpp
audio/audiooutputbase.cpp:630:18: error: use of undeclared identifier
'ff_codec_id_string'
.arg(ff_codec_id_string(m_codec))