PDIR command causes GPF

[Steve Fabian]

The command below, which works in TCC V9, causes consistent GPFs in V10 and
later:

*pdir/s/nej/ou/a:/(a fpn) f:\ > dl.f

F: is a hard drive. File dl.f is created on a different drive (C:\save).
When the GPF occurs, the output file is terminated after the second of the
lines below (the 150648th line of the file):

___A________I F:\PICTURES\200805\28150444.JPG
_H_A________I F:\PICTURES\200805\DESCRIPT.ION
___A_________ F:\PICTURES\200806\02083224.JPG
___A________I F:\PICTURES\200806\02083232.JPG

V09 records 181822 files. Dropping the /Ne flag and redirecting errors still
GPFs, with empty error file.

GPF (with lots of plugins loaded):

TCC 11.00.52
Module=F:\JPSOFT\I11\TakeCmd.dll
Address=10005CD7
Exception=C0000005
EAX=00000000 EBX=00000000 ECX=023D8318 EDX=006B0063
ESI=00000000 EDI=00D70676 EBP=00D746A4 ESP=00D6C64C
CS=0000001B DS=00000023 ES=00000023 SS=00000023
Flags=00010246

Stack:
1 : TakeCmd.dll 0001:00004cd7
2 : TakeCmd.dll 0001:0000ba86
3 : TakeCmd.dll 0001:00009671

GPF (with all plugins unloaded):

TCC 11.00.52
Module=F:\JPSOFT\I11\TakeCmd.dll
Address=10005CD7
Exception=C0000005
EAX=00000000 EBX=00000000 ECX=023D0310 EDX=006B0063
ESI=00000000 EDI=00D70676 EBP=00D746A4 ESP=00D6C64C
CS=0000001B DS=00000023 ES=00000023 SS=00000023
Flags=00010246

Stack:
1 : TakeCmd.dll 0001:00004cd7
2 : TakeCmd.dll 0001:0000ba86
3 : TakeCmd.dll 0001:00009671

 

[Jim Cook]

TCC 12.00.21 Windows Vista [Version 6.0.6002]
TCC Build 21 Windows Vista Build 6002 Service Pack 2

This problem does not manifest for me, but I had to change the F:\ to D:\
for my system testing.


On Fri, Aug 13, 2010 at 12:08 PM, Steve Fábián <[email protected]>wrote:

> The command below, which works in TCC V9, causes consistent GPFs in V10 and
> later:
>
> *pdir/s/nej/ou/a:/(a fpn) f:\ > dl.f
>
> F: is a hard drive. File dl.f is created on a different drive (C:\save).
> When the GPF occurs, the output file is terminated after the second of the
> lines below (the 150648th line of the file):
>
> ___A________I F:\PICTURES\200805\28150444.JPG
> _H_A________I F:\PICTURES\200805\DESCRIPT.ION
> ___A_________ F:\PICTURES\200806\02083224.JPG
> ___A________I F:\PICTURES\200806\02083232.JPG
>
> V09 records 181822 files. Dropping the /Ne flag and redirecting errors
> still
> GPFs, with empty error file.
>
> GPF (with lots of plugins loaded):
>
> TCC 11.00.52
> Module=F:\JPSOFT\I11\TakeCmd.dll
> Address=10005CD7
> Exception=C0000005
> EAX=00000000 EBX=00000000 ECX=023D8318 EDX=006B0063
> ESI=00000000 EDI=00D70676 EBP=00D746A4 ESP=00D6C64C
> CS=0000001B DS=00000023 ES=00000023 SS=00000023
> Flags=00010246
>
> Stack:
> 1 : TakeCmd.dll 0001:00004cd7
> 2 : TakeCmd.dll 0001:0000ba86
> 3 : TakeCmd.dll 0001:00009671
>
> GPF (with all plugins unloaded):
>
> TCC 11.00.52
> Module=F:\JPSOFT\I11\TakeCmd.dll
> Address=10005CD7
> Exception=C0000005
> EAX=00000000 EBX=00000000 ECX=023D0310 EDX=006B0063
> ESI=00000000 EDI=00D70676 EBP=00D746A4 ESP=00D6C64C
> CS=0000001B DS=00000023 ES=00000023 SS=00000023
> Flags=00010246
>
> Stack:
> 1 : TakeCmd.dll 0001:00004cd7
> 2 : TakeCmd.dll 0001:0000ba86
> 3 : TakeCmd.dll 0001:00009671
>
>
>
>
>
>
>
>

--
Jim Cook
2010 Sundays: 4/4, 6/6, 8/8, 10/10, 12/12 and 5/9, 9/5, 7/11, 11/7.
Next year they're Monday.

 

[vefatica]

On Fri, 13 Aug 2010 15:25:32 -0400, Jim Cook <>
wrote:

|This problem does not manifest for me, but I had to change the F:\ to D:\
|for my system testing.

I can't either. I tried my biggest drive ~46000 files, no
descriptions

 

[Steve Fabian]

| On Fri, 13 Aug 2010 15:25:32 -0400, Jim Cook <>
| wrote:
|
|| This problem does not manifest for me, but I had to change the F:\
|| to D:\ for my system testing.
|
| I can't either. I tried my biggest drive ~46000 files, no
| descriptions

In V12 B21 it is inconsistent. I had problems thrice, last two tests
successful. V9 always works, V10 and V11 always fail. Failure location is
consistent across versions. Last entry recorded is last file in a directory,
next entry is from another directory, both subdirectories of the same
subdirectory of the drive's root.
--
Steve

 

[rconn]

> The command below, which works in TCC V9, causes consistent GPFs in V10
> and later:
>
> *pdir/s/nej/ou/a:/(a fpn) f:\ > dl.f
>
> F: is a hard drive. File dl.f is created on a different drive
> (C:\save).
> When the GPF occurs, the output file is terminated after the second of
> the lines below (the 150648th line of the file):

Not reproducible here (with 625,809 files).

That GPF address is in the description handling -- check your descript.ion
file for that directory and see if it's been corrupted.

Rex Conn
JP Software

 

[Steve Fabian]

| ---Quote---
|| The command below, which works in TCC V9, causes consistent GPFs in
|| V10 and later:
||
|| *pdir/s/nej/ou/a:/(a fpn) f:\ > dl.f
||
|| F: is a hard drive. File dl.f is created on a different drive
|| (C:\save).
|| When the GPF occurs, the output file is terminated after the second
|| of the lines below (the 150648th line of the file):
| ---End Quote---
| Not reproducible here (with 625,809 files).
|
| That GPF address is in the description handling -- check your
| descript.ion file for that directory and see if it's been corrupted.

I used LIST to view the DESCRIPT.ION files of the parent directory as
well as both subdirectories (the one actually reported and the one where
reporting fails). None appear to be corrupted.

BTW, when description handling is not requested, as in the failing
command above, why would there be an issue with corrupted description files?
--
Steve