headers
Marc Michalewicz | 1 Jun 2005 09:42
Picon

undefined reference to `__impure_ptr' during linking test

Hello again !

puhh, I am getting closer to my first kernel - i think and hope.
So, now I succees in building a library, but when trying to link tests
I am getting now strange messages about undefined
refernence:

"
/usr/share/gnuh8300_v0501_elf-1/lib/gcc/h8300-elf/3.4-GNUH8_v0501/../../../../h8300-elf/lib/h8300s/int32/libsupc++.a(vterminate.o)(.text+0xca):
In function `.L2':
: undefined reference to `__impure_ptr'
"

It is only this "impure_ptr", and only in "vterminate.o" nothing else
which is not found.

Did anybody experience this before and has a solution ?

Thanks in advance,
Marc
Permalink | Reply | 0 comments |
headers
Neundorf, Alexander | 1 Jun 2005 10:33

AW: undefined reference to `__impure_ptr' during linking test

Hi,

> Von: ecos-devel-owner <at> sources.redhat.com
> 
> Hello again !
> 
> puhh, I am getting closer to my first kernel - i think and hope.
> So, now I succees in building a library, but when trying to link tests
> I am getting now strange messages about undefined refernence:
> 
> "
> /usr/share/gnuh8300_v0501_elf-1/lib/gcc/h8300-elf/3.4-GNUH8_v0
> 501/../../../../h8300-elf/lib/h8300s/int32/libsupc++.a(vtermin
> ate.o)(.text+0xca): In function `.L2':
> : undefined reference to `__impure_ptr'
> "
> 
> It is only this "impure_ptr", and only in "vterminate.o" nothing else
> which is not found.
> 
> Did anybody experience this before and has a solution ?

I'm using gcc 3.4.1 and I think I had the same problem. Since then I compile and link the following reent.c,
taken from newlib:

---------8<-----------8<---------------8<----------------

#if (__GNUC__ == 3) && (__GNUC_MINOR__ == 4)

#include <reent.h>
(Continue reading)

Permalink | Reply | 501 comments |
headers
Marc Michalewicz | 1 Jun 2005 12:05
Picon

Re: AW: undefined reference to `__impure_ptr' during linking test

Thanks a lot,

this helped; i can build all tests now !

Next question is addressed to ecos-discuss - I just saw that this is the
right list for things like that.

Cheers,
Marc

Am Mittwoch, den 01.06.2005, 10:33 +0200 schrieb Neundorf, Alexander:
> Hi,
> 
> > Von: ecos-devel-owner <at> sources.redhat.com
> > 
> > Hello again !
> > 
> > puhh, I am getting closer to my first kernel - i think and hope.
> > So, now I succees in building a library, but when trying to link tests
> > I am getting now strange messages about undefined refernence:
> > 
> > "
> > /usr/share/gnuh8300_v0501_elf-1/lib/gcc/h8300-elf/3.4-GNUH8_v0
> > 501/../../../../h8300-elf/lib/h8300s/int32/libsupc++.a(vtermin
> > ate.o)(.text+0xca): In function `.L2':
> > : undefined reference to `__impure_ptr'
> > "
> > 
> > It is only this "impure_ptr", and only in "vterminate.o" nothing else
> > which is not found.
(Continue reading)

Permalink | Reply | 501 comments |
headers
Jonathan Larmour | 3 Jun 2005 14:13

ecos-* mailing list headers

I mentioned in passing to a sourceware administrator that the ecos mailing 
lists other than ecos-discuss still had "sources.redhat.com" in the 
headers rather than "ecos.sourceware.org".

I didn't intend him to fix it but he went ahead and did so. Which is fine 
in itself, but apologies for the lack of notice if anyone has any mail 
filters that are now going awry! It's a one off and now permanent change 
though.

Jifl
--

-- 
--["No sense being pessimistic, it wouldn't work anyway"]-- Opinions==mine
Permalink | Reply | 0 comments |
headers
Jani Monoses | 10 Jun 2005 15:44
Picon

status of flash_v2 branch?

Hello

Andrew, Bart, could one of you describe shortly what the current state of the flash v2 
work is?
I am considerig using it along with Savin's dataflash driver, as well as writing a lpc2xxx 
flash driver for the v2 API.
I checked out the branch but it seems it is not in sync with CVS head, some redboot and 
flash/io changes in HEAD which could be relevant to it are not merged.

Also this update
http://ecos.sourceware.org/ml/ecos-patches/2004-10/msg00000.html
is somewhat different from what is in CVS flash_v2.

thanks
Jani
Permalink | Reply | 1 comment |
headers
Andrew Lunn | 10 Jun 2005 16:36
Picon

Re: status of flash_v2 branch?

On Fri, Jun 10, 2005 at 04:44:12PM +0300, Jani Monoses wrote:
> Hello
> 
> Andrew, Bart, could one of you describe shortly what the current state of 
> the flash v2 work is?

It basically works. Bart once said Jifl had some fixes but i cannot
remember if they ever got merged in. We are working on the assumption
that eCos 3.0 will have it merged in. Until then its bit rotting in
the branch.

> I am considerig using it along with Savin's dataflash driver, as well as 
> writing a lpc2xxx flash driver for the v2 API.
> I checked out the branch but it seems it is not in sync with CVS head, some 
> redboot and flash/io changes in HEAD which could be relevant to it are not 
> merged.

There has been no attempt to keep it in sync with HEAD. You might want
to try to merge the branch into the trunk for everything except
io/flash. I suggest you just use io/flash directly from the branch. Or
you could try merging the HEAD into the branch and see what happens.

What ever way you go, it would be nice to hear the results...

> Also this update
> http://ecos.sourceware.org/ml/ecos-patches/2004-10/msg00000.html
> is somewhat different from what is in CVS flash_v2.

The dataflash all gets a bit messy because it needs the SPI code which
Bart did not contribute until after the flash_v2 branch. So when i did
(Continue reading)

Permalink | Reply | 0 comments |
headers
oliver munz @ s p e a g | 27 Jun 2005 14:46
Picon

why use the SA1110-USB-driver a static /dev/usbs... definition ???

Hello

I'm writing a device-driver on basis of the SA1110-USB-driver.
First I made a package without functionality. Then I tryed to get a handel
of my driver. It dosent worked. There was also no entry in the
..._install\lib\\libextras.a.

After houres of searching, I found, that the reason is the static definition
of the ..._devtab_entry. After removing the "static" key-word, my
test-program found his handels and does some pseudo inits...

The question for me is now, what is the reason for a static devtab?

The original source is:

"...\ecos\packages\devs\usb\sa11x0\current\src\usbs_sa11x0_data.cxx"

static CHAR_DEVIO_TABLE(usbs_sa11x0_ep0_devtab_functions,
    &cyg_devio_cwrite,
    &cyg_devio_cread,
    &cyg_devio_select,
    &usbs_devtab_get_config,
    &usbs_devtab_set_config);

static CHAR_DEVTAB_ENTRY(usbs_sa11x0_ep0_devtab_entry,
    CYGDAT_DEVS_USB_SA11X0_DEVTAB_BASENAME "0c",
    0,
    &usbs_sa11x0_ep0_devtab_functions,
    &usbs_sa11x0_devtab_ep0_init,
    0,
(Continue reading)

Permalink | Reply | 3 comments |
headers
Andrew Lunn | 27 Jun 2005 15:02
Picon

Re: why use the SA1110-USB-driver a static /dev/usbs... definition ???

On Mon, Jun 27, 2005 at 02:46:23PM +0200, oliver munz  <at>  s p e a g wrote:
> Hello
> 
> I'm writing a device-driver on basis of the SA1110-USB-driver.
> First I made a package without functionality. Then I tryed to get a handel
> of my driver. It dosent worked. There was also no entry in the
> ..._install\lib\\libextras.a.
> 
> After houres of searching, I found, that the reason is the static definition
> of the ..._devtab_entry. After removing the "static" key-word, my
> test-program found his handels and does some pseudo inits...
> 
> The question for me is now, what is the reason for a static devtab?
> 
> The original source is:
> 
> "...\ecos\packages\devs\usb\sa11x0\current\src\usbs_sa11x0_data.cxx"
> 
> static CHAR_DEVIO_TABLE(usbs_sa11x0_ep0_devtab_functions,
>    &cyg_devio_cwrite,
>    &cyg_devio_cread,
>    &cyg_devio_select,
>    &usbs_devtab_get_config,
>    &usbs_devtab_set_config);
> 
> static CHAR_DEVTAB_ENTRY(usbs_sa11x0_ep0_devtab_entry,
>    CYGDAT_DEVS_USB_SA11X0_DEVTAB_BASENAME "0c",
>    0,
>    &usbs_sa11x0_ep0_devtab_functions,
>    &usbs_sa11x0_devtab_ep0_init,
(Continue reading)

Permalink | Reply | 2 comments |
headers
Manfred Gruber | 27 Jun 2005 15:06
Picon

repositority question

Hi !

I used v2_0 version of ecos.
Because I want AM29XXXXX_V2 driver i checked out the flash_v2 tree. And now 
work on on this tree.

How are new things handled? My question is when there became new features to 
trunk of ecos are the also then in the flash_v2 tree?

Or do I have to update to the trunk tree and then patch this on my own to 
flash_v2 tree ?

thanks regards manfred
Permalink | Reply | 3 comments |
headers
Andrew Lunn | 27 Jun 2005 15:10
Picon

Re: repositority question

On Mon, Jun 27, 2005 at 03:06:34PM +0200, Manfred Gruber wrote:
> Hi !
> 
> I used v2_0 version of ecos.
> Because I want AM29XXXXX_V2 driver i checked out the flash_v2 tree. And now 
> work on on this tree.
> 
> How are new things handled? My question is when there became new features to 
> trunk of ecos are the also then in the flash_v2 tree?
> 
> Or do I have to update to the trunk tree and then patch this on my own to 
> flash_v2 tree ?

We don't keep the branch up to date with the trunk. When eCos 3.0
comes along we will merge the branch into the trunk, but until then
you need to do this yourself.

        Andrew
Permalink | Reply | 2 comments |

Gmane