* Re: [ECOS] system ASSERT problem
@ 2007-03-21 17:23 bob.koninckx
2007-03-26 15:39 ` Tomasz Kutyla
0 siblings, 1 reply; 11+ messages in thread
From: bob.koninckx @ 2007-03-21 17:23 UTC (permalink / raw)
To: Tomasz Kutyla, bob.koninckx; +Cc: Kurt Siedenburg, ecos-discuss
>-----Original Message-----
>From: Tomasz Kutyla [mailto:tkutyla@marp.pl]
>Sent: Wednesday, March 21, 2007 12:01 PM
>To: bob.koninckx@o-3s.com
>Cc: 'Kurt Siedenburg', ecos-discuss@ecos.sourceware.org
>Subject: Re: [ECOS] system ASSERT problem
>
>Below is the code for mcard_open:
>
>int mcard_open(mcard_t *card)
>{
> CYG_CHECK_DATA_PTR(card, "Valid card pointer required");
>
> if( !card ) return -EINVAL;
>
> if( card->opened ) return 0;
>
> card->bmask = NULL;
> card->mcard_hdl = 0;
>
> int e;
>
> e = cyg_io_lookup ("/dev/i2c/m14c64", &card->mcard_hdl);
>
> if( !card->mcard_hdl ) return e;
This could be it. If e == 0 and mcard->mcard_hdl == 0 (which probably means there's something wrong with your driver), you'll get the assert afterwards.
>
> e = mcard_read_header(card);
> if( e ) return e;
>
> MCARD_PRINT_HEADER(&card->header);
>
> e = mcard_read_bmask(card);
> if( e ) return e;
>
> MCARD_PRINT_BMASK(card);
>
> card->opened = true;
>
> return 0;
>}
>
>Tomasz
>
>
>
>bob.koninckx@o-3s.com napisaÂł(a):
>>> -----Original Message-----
>>> From: Kurt Siedenburg [mailto:kurt.siedenburg@vicom.com]
>>> Sent: Wednesday, March 21, 2007 11:07 AM
>>> To: 'Tomasz KutyÂła', ecos-discuss@ecos.sourceware.org
>>> Subject: RE: [ECOS] system ASSERT problem
>>>
>>> My best guess is that due to advanced compiler optimization the compiler determines that the line
>>> e = mcard_open(&mcard);
>>> either always sets mcard.mcard_hdl to NULL.
>>> Or the whole call to mcard_open gets optimized away.
>>>
>>
>> We did not see the code of mcard_open. Does it actually change the value of mcard.mcard_hdl? If it doesn't, the assert just does what it is supposed to do.
>>
>>
>>> Kurt
>>>
>>
>> Bob
>>
>>
>>> -----Original Message-----
>>> From: ecos-discuss-owner@ecos.sourceware.org [mailto:ecos-discuss-owner@ecos.sourceware.org] On Behalf Of Tomasz Kutyla
>>> Sent: Wednesday, March 21, 2007 3:32 AM
>>> To: ecos-discuss@ecos.sourceware.org
>>> Subject: Re: [ECOS] system ASSERT problem
>>>
>>>
>>>
>>>>> I got strange problem with my application. When I compile system with
>>>>> global compiler flags
>>>>>
>>>>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline
>>>>> -Wundef -Woverloaded-virtual -ggdb -ffunction-sections
>>>>> -fdata-sections -fno-rtti -fno-exceptions -fvtable-gc -finit-priority
>>>>> -save-temps
>>>>>
>>>>> I got info on terminal:
>>>>>
>>>>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card
>>>>> handler required
>>>>> ASSERT FAIL: mcard.c [ 193] int
>>>>> mcard_write_bmask()
>>>>> Valid card handler required
>>>>>
>>>>> But when I remove '-O2' flag the problems doesn't occur. What can be
>>>>> the problem???
>>>>>
>>>>>
>>>> There is no mcard.c in the repository, so i guess it is something you
>>>> have added. Without the code it is hard to guess what is happening.
>>>>
>>> Below is the code that generates the ASSERT:
>>>
>>> int mcard_write_bmask(mcard_t *card)
>>> {
>>> CYG_CHECK_DATA_PTR(card, "valid card pointer required");
>>> CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
>>> CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
>>>
>>> if( !card->mcard_hdl ) return -EINVAL;
>>> if( !card->bmask ) return -EINVAL;
>>>
>>> int e;
>>> e = mcard_verify_card(card);
>>> if( e ) return e;
>>>
>>> e = cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_LOCK,0,0);
>>> if( e ) return -EIO;
>>>
>>> cyg_uint32 address = card->header.bmask_offset;
>>> cyg_uint32 len = sizeof(cyg_uint32);
>>>
>>> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_WRITE_POS,&address,&len);
>>> len = card->header.bmask_size;
>>>
>>> e = cyg_io_write(card->mcard_hdl,card->bmask,&len);
>>>
>>> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_UNLOCK,0,0);
>>>
>>> if( e ) return -EIO;
>>>
>>> return ENOERR;
>>> }
>>>
>>> **************************
>>>
>>> and the mcard_write_bmask function is called from function below:
>>>
>>> typedef struct mcard_t
>>> {
>>> cyg_io_handle_t mcard_hdl;
>>> mcard_header_t header;
>>> mcard_bmask_t *bmask;
>>> cyg_uint8 opened;
>>> } mcard_t;
>>>
>>> int I::formatCard()
>>> {
>>> int e;
>>>
>>> mcard_t mcard;
>>> mcard.mcard_hdl = 0;
>>> e = mcard_open(&mcard);
>>> if( e ) return -1;
>>>
>>> MCARD_PRINT_HEADER(&mcard.header);
>>>
>>> CLEAR_BUF(mcard.bmask,mcard.header.bmask_size);
>>> e = mcard_write_bmask(&mcard);
>>> if( e ) return e;
>>>
>>> MCARD_PRINT_BMASK(&mcard);
>>>
>>> mcard_pdata_t pdata;
>>> CLEAR_BUF(&pdata,sizeof(mcard_pdata_t))
>>> e = mcard_write_pdata(&mcard, &pdata);
>>> if( e ) return e;
>>>
>>> MCARD_PRINT_PDATA(&pdata);
>>>
>>> return 0;
>>> }
>>>
>>> --
>>> Tomasz Kutyla, Design Engineer
>>> MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191 fax +48 12 415-86-80
>>>
>>> e-mail: tkutyla@marp.pl; http://www.marp.pl
>>>
>>>
>>> --
>>> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>>> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>>>
>>>
>>> --
>>> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>>> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>>>
>>>
>>>
>>
>>
>>
>>
>
>--
>Tomasz Kutyla, Design Engineer
>MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
>tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
>fax +48 12 415-86-80
>
>e-mail: tkutyla@marp.pl; http://www.marp.pl
>
>
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 17:23 [ECOS] system ASSERT problem bob.koninckx
@ 2007-03-26 15:39 ` Tomasz Kutyla
0 siblings, 0 replies; 11+ messages in thread
From: Tomasz Kutyla @ 2007-03-26 15:39 UTC (permalink / raw)
To: bob.koninckx; +Cc: Kurt Siedenburg, ecos-discuss
>> Below is the code for mcard_open:
>>
>> int mcard_open(mcard_t *card)
>> {
>> CYG_CHECK_DATA_PTR(card, "Valid card pointer required");
>>
>> if( !card ) return -EINVAL;
>>
>> if( card->opened ) return 0;
>>
>> card->bmask = NULL;
>> card->mcard_hdl = 0;
>>
>> int e;
>>
>> e = cyg_io_lookup ("/dev/i2c/m14c64", &card->mcard_hdl);
>>
>> if( !card->mcard_hdl ) return e;
>>
>
> This could be it. If e == 0 and mcard->mcard_hdl == 0 (which probably means there's something wrong with your driver), you'll get the assert afterwards.
>
You were right about the condition. I changed it to: check e and then I
check card->mcard_hdl. Besides there was another problem:
int I::formatCard()
{
int e;
mcard_t mcard;
mcard.mcard_hdl = 0;
e = mcard_open(&mcard);
if( e ) return -1;
...
}
When I create variable mcard it includes one element that is checked in mcard_open function. While it is created it isn't set to false. mcard_open checks the element if it is false. I found that during creation element is set to random value, so mcard_open return before completion.
Thanks for help!!;)
Tomasz
PS. Andrew thanks also for your tips;)
>
>> e = mcard_read_header(card);
>> if( e ) return e;
>>
>> MCARD_PRINT_HEADER(&card->header);
>>
>> e = mcard_read_bmask(card);
>> if( e ) return e;
>>
>> MCARD_PRINT_BMASK(card);
>>
>> card->opened = true;
>>
>> return 0;
>> }
>>
>> Tomasz
>>
>>
>>
>> bob.koninckx@o-3s.com napisaÂł(a):
>>
>>>> -----Original Message-----
>>>> From: Kurt Siedenburg [mailto:kurt.siedenburg@vicom.com]
>>>> Sent: Wednesday, March 21, 2007 11:07 AM
>>>> To: 'Tomasz KutyÂła', ecos-discuss@ecos.sourceware.org
>>>> Subject: RE: [ECOS] system ASSERT problem
>>>>
>>>> My best guess is that due to advanced compiler optimization the compiler determines that the line
>>>> e = mcard_open(&mcard);
>>>> either always sets mcard.mcard_hdl to NULL.
>>>> Or the whole call to mcard_open gets optimized away.
>>>>
>>>>
>>> We did not see the code of mcard_open. Does it actually change the value of mcard.mcard_hdl? If it doesn't, the assert just does what it is supposed to do.
>>>
>>>
>>>
>>>> Kurt
>>>>
>>>>
>>> Bob
>>>
>>>
>>>
>>>> -----Original Message-----
>>>> From: ecos-discuss-owner@ecos.sourceware.org [mailto:ecos-discuss-owner@ecos.sourceware.org] On Behalf Of Tomasz Kutyla
>>>> Sent: Wednesday, March 21, 2007 3:32 AM
>>>> To: ecos-discuss@ecos.sourceware.org
>>>> Subject: Re: [ECOS] system ASSERT problem
>>>>
>>>>
>>>>
>>>>
>>>>>> I got strange problem with my application. When I compile system with
>>>>>> global compiler flags
>>>>>>
>>>>>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline
>>>>>> -Wundef -Woverloaded-virtual -ggdb -ffunction-sections
>>>>>> -fdata-sections -fno-rtti -fno-exceptions -fvtable-gc -finit-priority
>>>>>> -save-temps
>>>>>>
>>>>>> I got info on terminal:
>>>>>>
>>>>>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card
>>>>>> handler required
>>>>>> ASSERT FAIL: mcard.c [ 193] int
>>>>>> mcard_write_bmask()
>>>>>> Valid card handler required
>>>>>>
>>>>>> But when I remove '-O2' flag the problems doesn't occur. What can be
>>>>>> the problem???
>>>>>>
>>>>>>
>>>>>>
>>>>> There is no mcard.c in the repository, so i guess it is something you
>>>>> have added. Without the code it is hard to guess what is happening.
>>>>>
>>>>>
>>>> Below is the code that generates the ASSERT:
>>>>
>>>> int mcard_write_bmask(mcard_t *card)
>>>> {
>>>> CYG_CHECK_DATA_PTR(card, "valid card pointer required");
>>>> CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
>>>> CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
>>>>
>>>> if( !card->mcard_hdl ) return -EINVAL;
>>>> if( !card->bmask ) return -EINVAL;
>>>>
>>>> int e;
>>>> e = mcard_verify_card(card);
>>>> if( e ) return e;
>>>>
>>>> e = cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_LOCK,0,0);
>>>> if( e ) return -EIO;
>>>>
>>>> cyg_uint32 address = card->header.bmask_offset;
>>>> cyg_uint32 len = sizeof(cyg_uint32);
>>>>
>>>> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_WRITE_POS,&address,&len);
>>>> len = card->header.bmask_size;
>>>>
>>>> e = cyg_io_write(card->mcard_hdl,card->bmask,&len);
>>>>
>>>> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_UNLOCK,0,0);
>>>>
>>>> if( e ) return -EIO;
>>>>
>>>> return ENOERR;
>>>> }
>>>>
>>>> **************************
>>>>
>>>> and the mcard_write_bmask function is called from function below:
>>>>
>>>> typedef struct mcard_t
>>>> {
>>>> cyg_io_handle_t mcard_hdl;
>>>> mcard_header_t header;
>>>> mcard_bmask_t *bmask;
>>>> cyg_uint8 opened;
>>>> } mcard_t;
>>>>
>>>> int I::formatCard()
>>>> {
>>>> int e;
>>>>
>>>> mcard_t mcard;
>>>> mcard.mcard_hdl = 0;
>>>> e = mcard_open(&mcard);
>>>> if( e ) return -1;
>>>>
>>>> MCARD_PRINT_HEADER(&mcard.header);
>>>>
>>>> CLEAR_BUF(mcard.bmask,mcard.header.bmask_size);
>>>> e = mcard_write_bmask(&mcard);
>>>> if( e ) return e;
>>>>
>>>> MCARD_PRINT_BMASK(&mcard);
>>>>
>>>> mcard_pdata_t pdata;
>>>> CLEAR_BUF(&pdata,sizeof(mcard_pdata_t))
>>>> e = mcard_write_pdata(&mcard, &pdata);
>>>> if( e ) return e;
>>>>
>>>> MCARD_PRINT_PDATA(&pdata);
>>>>
>>>> return 0;
>>>> }
>>>>
>>>> --
>>>> Tomasz Kutyla, Design Engineer
>>>> MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191 fax +48 12 415-86-80
>>>>
>>>> e-mail: tkutyla@marp.pl; http://www.marp.pl
>>>>
>>>>
>>>> --
>>>> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>>>> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>>>>
>>>>
>>>> --
>>>> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>>>> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>> --
>> Tomasz Kutyla, Design Engineer
>> MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
>> tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
>> fax +48 12 415-86-80
>>
>> e-mail: tkutyla@marp.pl; http://www.marp.pl
>>
>>
>>
>
>
>
>
--
Tomasz Kutyla, Design Engineer
MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
fax +48 12 415-86-80
e-mail: tkutyla@marp.pl; http://www.marp.pl
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 16:00 ` Tomasz Kutyla
@ 2007-03-21 16:03 ` Andrew Lunn
0 siblings, 0 replies; 11+ messages in thread
From: Andrew Lunn @ 2007-03-21 16:03 UTC (permalink / raw)
To: Tomasz Kutyla; +Cc: Andrew Lunn, ecos-discuss
> I use Cofig Tool for windows. And only thing I did was to delete -O2
> flag from global compiler flags. I didn't change any ASSERTS flags.
> Besides I can see that all actions on card were done without any error.
> >If it is reproducible the way forward is to see the assembly language
> >the compiler produces for both the working and none working case.
> >
> Where can I find assembly language after compilation?
You need to pass -S to gcc. It will then stop after producing assembly
code, not object code.
Andrew
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 15:47 bob.koninckx
@ 2007-03-21 16:02 ` Tomasz Kutyla
0 siblings, 0 replies; 11+ messages in thread
From: Tomasz Kutyla @ 2007-03-21 16:02 UTC (permalink / raw)
To: bob.koninckx; +Cc: Kurt Siedenburg, ecos-discuss
Below is the code for mcard_open:
int mcard_open(mcard_t *card)
{
CYG_CHECK_DATA_PTR(card, "Valid card pointer required");
if( !card ) return -EINVAL;
if( card->opened ) return 0;
card->bmask = NULL;
card->mcard_hdl = 0;
int e;
e = cyg_io_lookup ("/dev/i2c/m14c64", &card->mcard_hdl);
if( !card->mcard_hdl ) return e;
e = mcard_read_header(card);
if( e ) return e;
MCARD_PRINT_HEADER(&card->header);
e = mcard_read_bmask(card);
if( e ) return e;
MCARD_PRINT_BMASK(card);
card->opened = true;
return 0;
}
Tomasz
bob.koninckx@o-3s.com napisaÂł(a):
>> -----Original Message-----
>> From: Kurt Siedenburg [mailto:kurt.siedenburg@vicom.com]
>> Sent: Wednesday, March 21, 2007 11:07 AM
>> To: 'Tomasz KutyÂła', ecos-discuss@ecos.sourceware.org
>> Subject: RE: [ECOS] system ASSERT problem
>>
>> My best guess is that due to advanced compiler optimization the compiler determines that the line
>> e = mcard_open(&mcard);
>> either always sets mcard.mcard_hdl to NULL.
>> Or the whole call to mcard_open gets optimized away.
>>
>
> We did not see the code of mcard_open. Does it actually change the value of mcard.mcard_hdl? If it doesn't, the assert just does what it is supposed to do.
>
>
>> Kurt
>>
>
> Bob
>
>
>> -----Original Message-----
>> From: ecos-discuss-owner@ecos.sourceware.org [mailto:ecos-discuss-owner@ecos.sourceware.org] On Behalf Of Tomasz Kutyla
>> Sent: Wednesday, March 21, 2007 3:32 AM
>> To: ecos-discuss@ecos.sourceware.org
>> Subject: Re: [ECOS] system ASSERT problem
>>
>>
>>
>>>> I got strange problem with my application. When I compile system with
>>>> global compiler flags
>>>>
>>>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline
>>>> -Wundef -Woverloaded-virtual -ggdb -ffunction-sections
>>>> -fdata-sections -fno-rtti -fno-exceptions -fvtable-gc -finit-priority
>>>> -save-temps
>>>>
>>>> I got info on terminal:
>>>>
>>>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card
>>>> handler required
>>>> ASSERT FAIL: mcard.c [ 193] int
>>>> mcard_write_bmask()
>>>> Valid card handler required
>>>>
>>>> But when I remove '-O2' flag the problems doesn't occur. What can be
>>>> the problem???
>>>>
>>>>
>>> There is no mcard.c in the repository, so i guess it is something you
>>> have added. Without the code it is hard to guess what is happening.
>>>
>> Below is the code that generates the ASSERT:
>>
>> int mcard_write_bmask(mcard_t *card)
>> {
>> CYG_CHECK_DATA_PTR(card, "valid card pointer required");
>> CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
>> CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
>>
>> if( !card->mcard_hdl ) return -EINVAL;
>> if( !card->bmask ) return -EINVAL;
>>
>> int e;
>> e = mcard_verify_card(card);
>> if( e ) return e;
>>
>> e = cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_LOCK,0,0);
>> if( e ) return -EIO;
>>
>> cyg_uint32 address = card->header.bmask_offset;
>> cyg_uint32 len = sizeof(cyg_uint32);
>>
>> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_WRITE_POS,&address,&len);
>> len = card->header.bmask_size;
>>
>> e = cyg_io_write(card->mcard_hdl,card->bmask,&len);
>>
>> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_UNLOCK,0,0);
>>
>> if( e ) return -EIO;
>>
>> return ENOERR;
>> }
>>
>> **************************
>>
>> and the mcard_write_bmask function is called from function below:
>>
>> typedef struct mcard_t
>> {
>> cyg_io_handle_t mcard_hdl;
>> mcard_header_t header;
>> mcard_bmask_t *bmask;
>> cyg_uint8 opened;
>> } mcard_t;
>>
>> int I::formatCard()
>> {
>> int e;
>>
>> mcard_t mcard;
>> mcard.mcard_hdl = 0;
>> e = mcard_open(&mcard);
>> if( e ) return -1;
>>
>> MCARD_PRINT_HEADER(&mcard.header);
>>
>> CLEAR_BUF(mcard.bmask,mcard.header.bmask_size);
>> e = mcard_write_bmask(&mcard);
>> if( e ) return e;
>>
>> MCARD_PRINT_BMASK(&mcard);
>>
>> mcard_pdata_t pdata;
>> CLEAR_BUF(&pdata,sizeof(mcard_pdata_t))
>> e = mcard_write_pdata(&mcard, &pdata);
>> if( e ) return e;
>>
>> MCARD_PRINT_PDATA(&pdata);
>>
>> return 0;
>> }
>>
>> --
>> Tomasz Kutyla, Design Engineer
>> MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191 fax +48 12 415-86-80
>>
>> e-mail: tkutyla@marp.pl; http://www.marp.pl
>>
>>
>> --
>> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>>
>>
>> --
>> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>>
>>
>>
>
>
>
>
--
Tomasz Kutyla, Design Engineer
MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
fax +48 12 415-86-80
e-mail: tkutyla@marp.pl; http://www.marp.pl
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 13:36 ` Andrew Lunn
@ 2007-03-21 16:00 ` Tomasz Kutyla
2007-03-21 16:03 ` Andrew Lunn
0 siblings, 1 reply; 11+ messages in thread
From: Tomasz Kutyla @ 2007-03-21 16:00 UTC (permalink / raw)
To: Andrew Lunn; +Cc: ecos-discuss
> On Wed, Mar 21, 2007 at 11:32:02AM +0100, Tomasz Kuty??a wrote:
>
>>>> I got strange problem with my application. When I compile system with
>>>> global compiler flags
>>>>
>>>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
>>>> -Woverloaded-virtual -ggdb -ffunction-sections -fdata-sections -fno-rtti
>>>> -fno-exceptions -fvtable-gc -finit-priority -save-temps
>>>>
>>>> I got info on terminal:
>>>>
>>>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card handler
>>>> required
>>>> ASSERT FAIL: mcard.c [ 193] int
>>>> mcard_write_bmask()
>>>> Valid card handler required
>>>>
>>>> But when I remove '-O2' flag the problems doesn't occur. What can be the
>>>> problem???
>>>>
>>>>
>>> There is no mcard.c in the repository, so i guess it is something you
>>> have added. Without the code it is hard to guess what is happening.
>>>
>> Below is the code that generates the ASSERT:
>>
>> int mcard_write_bmask(mcard_t *card)
>> {
>> CYG_CHECK_DATA_PTR(card, "valid card pointer required");
>> CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
>> CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
>>
>> if( !card->mcard_hdl ) return -EINVAL;
>> if( !card->bmask ) return -EINVAL;
>>
>
> Well the asserts are wrong anyway. The code correctly handle the cases
> which the asserts are testing for, so the asserts are not needed.
>
> As to why this happens now, are you really sure the asserts were
> enabled in your old configuration without -O2? Is it really
> reproducible?
>
I use Cofig Tool for windows. And only thing I did was to delete -O2
flag from global compiler flags. I didn't change any ASSERTS flags.
Besides I can see that all actions on card were done without any error.
> If it is reproducible the way forward is to see the assembly language
> the compiler produces for both the working and none working case.
>
Where can I find assembly language after compilation?
Tomasz
--
Tomasz Kutyla, Design Engineer
MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
fax +48 12 415-86-80
e-mail: tkutyla@marp.pl; http://www.marp.pl
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
@ 2007-03-21 15:47 bob.koninckx
2007-03-21 16:02 ` Tomasz Kutyla
0 siblings, 1 reply; 11+ messages in thread
From: bob.koninckx @ 2007-03-21 15:47 UTC (permalink / raw)
To: Kurt Siedenburg, Tomasz Kutyła, ecos-discuss
>-----Original Message-----
>From: Kurt Siedenburg [mailto:kurt.siedenburg@vicom.com]
>Sent: Wednesday, March 21, 2007 11:07 AM
>To: 'Tomasz KutyÂła', ecos-discuss@ecos.sourceware.org
>Subject: RE: [ECOS] system ASSERT problem
>
>My best guess is that due to advanced compiler optimization the compiler determines that the line
> e = mcard_open(&mcard);
>either always sets mcard.mcard_hdl to NULL.
>Or the whole call to mcard_open gets optimized away.
We did not see the code of mcard_open. Does it actually change the value of mcard.mcard_hdl? If it doesn't, the assert just does what it is supposed to do.
> Kurt
Bob
>
>-----Original Message-----
>From: ecos-discuss-owner@ecos.sourceware.org [mailto:ecos-discuss-owner@ecos.sourceware.org] On Behalf Of Tomasz Kutyla
>Sent: Wednesday, March 21, 2007 3:32 AM
>To: ecos-discuss@ecos.sourceware.org
>Subject: Re: [ECOS] system ASSERT problem
>
>
>>> I got strange problem with my application. When I compile system with
>>> global compiler flags
>>>
>>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline
>>> -Wundef -Woverloaded-virtual -ggdb -ffunction-sections
>>> -fdata-sections -fno-rtti -fno-exceptions -fvtable-gc -finit-priority
>>> -save-temps
>>>
>>> I got info on terminal:
>>>
>>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card
>>> handler required
>>> ASSERT FAIL: mcard.c [ 193] int
>>> mcard_write_bmask()
>>> Valid card handler required
>>>
>>> But when I remove '-O2' flag the problems doesn't occur. What can be
>>> the problem???
>>>
>>
>> There is no mcard.c in the repository, so i guess it is something you
>> have added. Without the code it is hard to guess what is happening.
>Below is the code that generates the ASSERT:
>
>int mcard_write_bmask(mcard_t *card)
>{
> CYG_CHECK_DATA_PTR(card, "valid card pointer required");
> CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
> CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
>
> if( !card->mcard_hdl ) return -EINVAL;
> if( !card->bmask ) return -EINVAL;
>
> int e;
> e = mcard_verify_card(card);
> if( e ) return e;
>
> e = cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_LOCK,0,0);
> if( e ) return -EIO;
>
> cyg_uint32 address = card->header.bmask_offset;
> cyg_uint32 len = sizeof(cyg_uint32);
>
>cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_WRITE_POS,&address,&len);
> len = card->header.bmask_size;
>
> e = cyg_io_write(card->mcard_hdl,card->bmask,&len);
>
> cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_UNLOCK,0,0);
>
> if( e ) return -EIO;
>
> return ENOERR;
>}
>
>**************************
>
>and the mcard_write_bmask function is called from function below:
>
>typedef struct mcard_t
>{
> cyg_io_handle_t mcard_hdl;
> mcard_header_t header;
> mcard_bmask_t *bmask;
> cyg_uint8 opened;
>} mcard_t;
>
>int I::formatCard()
>{
> int e;
>
> mcard_t mcard;
> mcard.mcard_hdl = 0;
> e = mcard_open(&mcard);
> if( e ) return -1;
>
> MCARD_PRINT_HEADER(&mcard.header);
>
> CLEAR_BUF(mcard.bmask,mcard.header.bmask_size);
> e = mcard_write_bmask(&mcard);
> if( e ) return e;
>
> MCARD_PRINT_BMASK(&mcard);
>
> mcard_pdata_t pdata;
> CLEAR_BUF(&pdata,sizeof(mcard_pdata_t))
> e = mcard_write_pdata(&mcard, &pdata);
> if( e ) return e;
>
> MCARD_PRINT_PDATA(&pdata);
>
> return 0;
>}
>
>--
>Tomasz Kutyla, Design Engineer
>MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191 fax +48 12 415-86-80
>
>e-mail: tkutyla@marp.pl; http://www.marp.pl
>
>
>--
>Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>
>
>--
>Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
>and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>
>
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* RE: [ECOS] system ASSERT problem
2007-03-21 10:32 ` Tomasz Kutyła
2007-03-21 13:36 ` Andrew Lunn
@ 2007-03-21 15:09 ` Kurt Siedenburg
1 sibling, 0 replies; 11+ messages in thread
From: Kurt Siedenburg @ 2007-03-21 15:09 UTC (permalink / raw)
To: Tomasz Kutyła, ecos-discuss
My best guess is that due to advanced compiler optimization the compiler determines that the line
e = mcard_open(&mcard);
either always sets mcard.mcard_hdl to NULL.
Or the whole call to mcard_open gets optimized away.
Kurt
-----Original Message-----
From: ecos-discuss-owner@ecos.sourceware.org [mailto:ecos-discuss-owner@ecos.sourceware.org] On Behalf Of Tomasz Kutyla
Sent: Wednesday, March 21, 2007 3:32 AM
To: ecos-discuss@ecos.sourceware.org
Subject: Re: [ECOS] system ASSERT problem
>> I got strange problem with my application. When I compile system with
>> global compiler flags
>>
>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline
>> -Wundef -Woverloaded-virtual -ggdb -ffunction-sections
>> -fdata-sections -fno-rtti -fno-exceptions -fvtable-gc -finit-priority
>> -save-temps
>>
>> I got info on terminal:
>>
>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card
>> handler required
>> ASSERT FAIL: mcard.c [ 193] int
>> mcard_write_bmask()
>> Valid card handler required
>>
>> But when I remove '-O2' flag the problems doesn't occur. What can be
>> the problem???
>>
>
> There is no mcard.c in the repository, so i guess it is something you
> have added. Without the code it is hard to guess what is happening.
Below is the code that generates the ASSERT:
int mcard_write_bmask(mcard_t *card)
{
CYG_CHECK_DATA_PTR(card, "valid card pointer required");
CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
if( !card->mcard_hdl ) return -EINVAL;
if( !card->bmask ) return -EINVAL;
int e;
e = mcard_verify_card(card);
if( e ) return e;
e = cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_LOCK,0,0);
if( e ) return -EIO;
cyg_uint32 address = card->header.bmask_offset;
cyg_uint32 len = sizeof(cyg_uint32);
cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_WRITE_POS,&address,&len);
len = card->header.bmask_size;
e = cyg_io_write(card->mcard_hdl,card->bmask,&len);
cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_UNLOCK,0,0);
if( e ) return -EIO;
return ENOERR;
}
**************************
and the mcard_write_bmask function is called from function below:
typedef struct mcard_t
{
cyg_io_handle_t mcard_hdl;
mcard_header_t header;
mcard_bmask_t *bmask;
cyg_uint8 opened;
} mcard_t;
int I::formatCard()
{
int e;
mcard_t mcard;
mcard.mcard_hdl = 0;
e = mcard_open(&mcard);
if( e ) return -1;
MCARD_PRINT_HEADER(&mcard.header);
CLEAR_BUF(mcard.bmask,mcard.header.bmask_size);
e = mcard_write_bmask(&mcard);
if( e ) return e;
MCARD_PRINT_BMASK(&mcard);
mcard_pdata_t pdata;
CLEAR_BUF(&pdata,sizeof(mcard_pdata_t))
e = mcard_write_pdata(&mcard, &pdata);
if( e ) return e;
MCARD_PRINT_PDATA(&pdata);
return 0;
}
--
Tomasz Kutyla, Design Engineer
MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191 fax +48 12 415-86-80
e-mail: tkutyla@marp.pl; http://www.marp.pl
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 10:32 ` Tomasz Kutyła
@ 2007-03-21 13:36 ` Andrew Lunn
2007-03-21 16:00 ` Tomasz Kutyla
2007-03-21 15:09 ` Kurt Siedenburg
1 sibling, 1 reply; 11+ messages in thread
From: Andrew Lunn @ 2007-03-21 13:36 UTC (permalink / raw)
To: Tomasz Kuty??a; +Cc: ecos-discuss
On Wed, Mar 21, 2007 at 11:32:02AM +0100, Tomasz Kuty??a wrote:
>
> >>I got strange problem with my application. When I compile system with
> >>global compiler flags
> >>
> >>-ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
> >>-Woverloaded-virtual -ggdb -ffunction-sections -fdata-sections -fno-rtti
> >>-fno-exceptions -fvtable-gc -finit-priority -save-temps
> >>
> >>I got info on terminal:
> >>
> >>ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card handler
> >>required
> >>ASSERT FAIL: mcard.c [ 193] int
> >>mcard_write_bmask()
> >>Valid card handler required
> >>
> >>But when I remove '-O2' flag the problems doesn't occur. What can be the
> >>problem???
> >>
> >
> >There is no mcard.c in the repository, so i guess it is something you
> >have added. Without the code it is hard to guess what is happening.
> Below is the code that generates the ASSERT:
>
> int mcard_write_bmask(mcard_t *card)
> {
> CYG_CHECK_DATA_PTR(card, "valid card pointer required");
> CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
> CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
>
> if( !card->mcard_hdl ) return -EINVAL;
> if( !card->bmask ) return -EINVAL;
Well the asserts are wrong anyway. The code correctly handle the cases
which the asserts are testing for, so the asserts are not needed.
As to why this happens now, are you really sure the asserts were
enabled in your old configuration without -O2? Is it really
reproducible?
If it is reproducible the way forward is to see the assembly language
the compiler produces for both the working and none working case.
Andrew
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 10:14 ` Andrew Lunn
@ 2007-03-21 10:32 ` Tomasz Kutyła
2007-03-21 13:36 ` Andrew Lunn
2007-03-21 15:09 ` Kurt Siedenburg
0 siblings, 2 replies; 11+ messages in thread
From: Tomasz Kutyła @ 2007-03-21 10:32 UTC (permalink / raw)
To: ecos-discuss
>> I got strange problem with my application. When I compile system with
>> global compiler flags
>>
>> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
>> -Woverloaded-virtual -ggdb -ffunction-sections -fdata-sections -fno-rtti
>> -fno-exceptions -fvtable-gc -finit-priority -save-temps
>>
>> I got info on terminal:
>>
>> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card handler
>> required
>> ASSERT FAIL: mcard.c [ 193] int
>> mcard_write_bmask()
>> Valid card handler required
>>
>> But when I remove '-O2' flag the problems doesn't occur. What can be the
>> problem???
>>
>
> There is no mcard.c in the repository, so i guess it is something you
> have added. Without the code it is hard to guess what is happening.
Below is the code that generates the ASSERT:
int mcard_write_bmask(mcard_t *card)
{
CYG_CHECK_DATA_PTR(card, "valid card pointer required");
CYG_CHECK_DATA_PTR(card->bmask, "valid mask pointer required");
CYG_ASSERT(card->mcard_hdl,"Valid card handler required");
if( !card->mcard_hdl ) return -EINVAL;
if( !card->bmask ) return -EINVAL;
int e;
e = mcard_verify_card(card);
if( e ) return e;
e = cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_LOCK,0,0);
if( e ) return -EIO;
cyg_uint32 address = card->header.bmask_offset;
cyg_uint32 len = sizeof(cyg_uint32);
cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_WRITE_POS,&address,&len);
len = card->header.bmask_size;
e = cyg_io_write(card->mcard_hdl,card->bmask,&len);
cyg_io_set_config(card->mcard_hdl,CYG_IO_SET_CONFIG_CARD_UNLOCK,0,0);
if( e ) return -EIO;
return ENOERR;
}
**************************
and the mcard_write_bmask function is called from function below:
typedef struct mcard_t
{
cyg_io_handle_t mcard_hdl;
mcard_header_t header;
mcard_bmask_t *bmask;
cyg_uint8 opened;
} mcard_t;
int I::formatCard()
{
int e;
mcard_t mcard;
mcard.mcard_hdl = 0;
e = mcard_open(&mcard);
if( e ) return -1;
MCARD_PRINT_HEADER(&mcard.header);
CLEAR_BUF(mcard.bmask,mcard.header.bmask_size);
e = mcard_write_bmask(&mcard);
if( e ) return e;
MCARD_PRINT_BMASK(&mcard);
mcard_pdata_t pdata;
CLEAR_BUF(&pdata,sizeof(mcard_pdata_t))
e = mcard_write_pdata(&mcard, &pdata);
if( e ) return e;
MCARD_PRINT_PDATA(&pdata);
return 0;
}
--
Tomasz Kutyla, Design Engineer
MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
fax +48 12 415-86-80
e-mail: tkutyla@marp.pl; http://www.marp.pl
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [ECOS] system ASSERT problem
2007-03-21 10:09 Tomasz Kutyła
@ 2007-03-21 10:14 ` Andrew Lunn
2007-03-21 10:32 ` Tomasz Kutyła
0 siblings, 1 reply; 11+ messages in thread
From: Andrew Lunn @ 2007-03-21 10:14 UTC (permalink / raw)
To: Tomasz Kuty?a; +Cc: ecos-discuss
On Wed, Mar 21, 2007 at 11:09:09AM +0100, Tomasz Kuty?a wrote:
> Hi,
>
> I got strange problem with my application. When I compile system with
> global compiler flags
>
> -ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
> -Woverloaded-virtual -ggdb -ffunction-sections -fdata-sections -fno-rtti
> -fno-exceptions -fvtable-gc -finit-priority -save-temps
>
> I got info on terminal:
>
> ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card handler
> required
> ASSERT FAIL: mcard.c [ 193] int
> mcard_write_bmask()
> Valid card handler required
>
> But when I remove '-O2' flag the problems doesn't occur. What can be the
> problem???
There is no mcard.c in the repository, so i guess it is something you
have added. Without the code it is hard to guess what is happening.
Andrew
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
* [ECOS] system ASSERT problem
@ 2007-03-21 10:09 Tomasz Kutyła
2007-03-21 10:14 ` Andrew Lunn
0 siblings, 1 reply; 11+ messages in thread
From: Tomasz Kutyła @ 2007-03-21 10:09 UTC (permalink / raw)
To: ecos-discuss
Hi,
I got strange problem with my application. When I compile system with
global compiler flags
-ml -m4 -O2 -Wall -Wpointer-arith -Wstrict-prototypes -Winline -Wundef
-Woverloaded-virtual -ggdb -ffunction-sections -fdata-sections -fno-rtti
-fno-exceptions -fvtable-gc -finit-priority -save-temps
I got info on terminal:
ASSERT FAIL: <3>mcard.c[193]int mcard_write_bmask() Valid card handler
required
ASSERT FAIL: mcard.c [ 193] int
mcard_write_bmask()
Valid card handler required
But when I remove '-O2' flag the problems doesn't occur. What can be the
problem???
Thanks in advance for your help;)
Regards
Tommy
--
Tomasz Kutyla, Design Engineer
MARP Electronic Sp. z o.o., ul. Pachonskiego 9, 31-223 Krakow
tel. +48 12 415-87-29 (ext. 35), mob.: +48 693 112 191
fax +48 12 415-86-80
e-mail: tkutyla@marp.pl; http://www.marp.pl
--
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2007-03-26 6:59 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-03-21 17:23 [ECOS] system ASSERT problem bob.koninckx
2007-03-26 15:39 ` Tomasz Kutyla
-- strict thread matches above, loose matches on Subject: below --
2007-03-21 15:47 bob.koninckx
2007-03-21 16:02 ` Tomasz Kutyla
2007-03-21 10:09 Tomasz Kutyła
2007-03-21 10:14 ` Andrew Lunn
2007-03-21 10:32 ` Tomasz Kutyła
2007-03-21 13:36 ` Andrew Lunn
2007-03-21 16:00 ` Tomasz Kutyla
2007-03-21 16:03 ` Andrew Lunn
2007-03-21 15:09 ` Kurt Siedenburg
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).