[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:01449] Re: a error occured at the same place as v2.1rc1


Hi,

I think you're turning on optimization while compiling HTS.
Please turn off it and report values of these variables again.

Regards,

Heiga ZEN (Byung Ha CHUN)


Yufeng Hao wrote (2008/06/18 15:34):

(gdb) bt
#0  0x000000000042453b in ZeroDMatrix (m=0x10585d8) at HMath.c:350
#1 0x000000000040a51e in CreateAccStruct (x=0x784a80, xfinfo=Variable "xfinfo" is not available.
) at HAdapt.c:1775
#2 0x00000000004131a0 in GenAdaptXForm (hset=0x7fbfffefe0, xfinfo=0x588100) at HAdapt.c:4598 #3 0x0000000000414a68 in UpdateSemiTiedModels (hset=0x7fbfffefe0, xfinfo=0x588100) at HAdapt.c:5558 #4 0x00000000004057c5 in UpdateModels (hset=0x7fbfffefe0, xfinfo=0x588100, pbuf2=0x0, uFlags=UPSEMIT)
    at HERest.c:1935
#5  0x0000000000406b1d in main (argc=Variable "argc" is not available.
) at HERest.c:805
(gdb) frame 1 #1 0x000000000040a51e in CreateAccStruct (x=0x784a80, xfinfo=Variable "xfinfo" is not available.
) at HAdapt.c:1775
1775                    ZeroDMatrix(accs->G[cnti]);
(gdb) p b
$4 = 1
(gdb) p cnti
$5 = 2
(gdb) p bsize
$6 = 1
(gdb) p vsize
$7 = 0
(gdb) frame 2
#2 0x00000000004131a0 in GenAdaptXForm (hset=0x7fbfffefe0, xfinfo=0x588100) at HAdapt.c:4598 4598 accs = CreateAccStruct(&gstack,xfinfo,GetBaseClassVSize(bclass,b),blockSize,bandWidth);
(gdb) p b
$8 = 0
configuration list as below: HTK Configuration Parameters[18]
  Module/Tool     Parameter                  Value
# BANDWIDTH IntVec 3 1 1 1 IntVec 1 1 IntVec 1 1 IntVec 1 1 # BLOCKSIZE IntVec 3 25 25 25 IntVec 1 1 IntVec 1 1 IntVec 1 1
#                 ADAPTKIND                   BASE
#                 USEBIAS                    FALSE
#                 TRANSKIND                  SEMIT
# BASECLASS /home/hyf/work/HTS-demo_CMU-ARCTIC-SLT/models/qst001/ver1/cmp/stc.base
#                 SAVEFULLC                   TRUE
#                 SEMITIEDMACRO                cmp
#                 MAXSEMITIEDITER                20
#                 MINDUR                         5
#                 MAXSTDDEVCOEF                 10
#                 APPLYDURVARFLOOR              TRUE
#                 DURVARFLOORPERCENTILE          1.000000
#                 VFLOORSCALESTR  Vector 4 0.01 0.01 0.01 0.01
#                 MINLEAFOCC              5.000000
#                 NATURALWRITEORDER              TRUE
#                 NATURALREADORDER              TRUE
#                 APPLYVFLOOR                 TRUE

-yufeng
On Wed, Jun 18, 2008 at 1:45 PM, Heiga ZEN (Byung Ha CHUN) <zen@xxxxxxxxxxxxxxx <mailto:zen@xxxxxxxxxxxxxxx>> wrote:

    Hi,

    Yufeng Hao wrote (2008/06/18 14:25):


        Iteration 20
         Floored 665 component elements
         LogLikelihood = 2.360921
        Program received signal SIGSEGV, Segmentation fault.
        0x000000000042453b in ZeroDMatrix (m=0x10585d8) at HMath.c:350
        350           for (j=1;j<=nc;j++) m[i][j]=0.0;
        (gdb) bt
        #0  0x000000000042453b in ZeroDMatrix (m=0x10585d8) at HMath.c:350
        #1  0x000000000040a51e in CreateAccStruct (x=0x784a80,
        xfinfo=Variable "xfinfo" is not available.
        ) at HAdapt.c:1775
        #2  0x00000000004131a0 in GenAdaptXForm (hset=0x7fbfffefe0,
        xfinfo=0x588100) at HAdapt.c:4598
        #3  0x0000000000414a68 in UpdateSemiTiedModels
        (hset=0x7fbfffefe0, xfinfo=0x588100) at HAdapt.c:5558
        #4  0x00000000004057c5 in UpdateModels (hset=0x7fbfffefe0,
        xfinfo=0x588100, pbuf2=0x0, uFlags=UPSEMIT)
           at HERest.c:1935
        #5  0x0000000000406b1d in main (argc=Variable "argc" is not
        available.
        ) at HERest.c:805
        (gdb)
        (gdb) p i
        $1 = 2
        (gdb) p j
        $2 = 2
        (gdb) p nc
        $3 = 1
        (gdb) p m
        $4 = 0x10585d8
        (gdb) p m[2][1]
        Cannot access memory at address 0x9


    Please also report values of the following variables when it stopped:

    CreateAccStruct(): b, cnti, bsize, vsize


Please also show the values of configuration variables (as HERest -D shows).



    Regards,

    Heiga ZEN (Byung Ha CHUN)

-- ------------------------------------------------
    Heiga ZEN     (in Japanese pronunciation)
    Byung Ha CHUN (in Korean pronunciation)

    Department of Computer Science and Engineering
    Nagoya Institute of Technology
    Gokiso-cho, Showa-ku, Nagoya 466-8555 Japan

    http://www.sp.nitech.ac.jp/~zen
    ------------------------------------------------




--
------------------------------------------------
Heiga ZEN     (in Japanese pronunciation)
Byung Ha CHUN (in Korean pronunciation)

Department of Computer Science and Engineering
Nagoya Institute of Technology
Gokiso-cho, Showa-ku, Nagoya 466-8555 Japan

http://www.sp.nitech.ac.jp/~zen
------------------------------------------------

References
[hts-users:01444] a error occured at the same place as v2.1rc1, Yufeng Hao
[hts-users:01445] Re: a error occured at the same place as v2.1rc1, Heiga ZEN (Byung Ha CHUN)
[hts-users:01446] Re: a error occured at the same place as v2.1rc1, Yufeng Hao
[hts-users:01447] Re: a error occured at the same place as v2.1rc1, Heiga ZEN (Byung Ha CHUN)
[hts-users:01448] Re: a error occured at the same place as v2.1rc1, Yufeng Hao