adding to the cflags list: in custom.py
customCCFLAGS = ['-D_GNU_SOURCE']
                  ^^^^^^^^^^^^^^^
completes build in gentoo. haven't tested yet. too sleepy.

kelly


On Mon, 27 Sep 2010, jpff@cs.bath.ac.uk wrote:

>
> The problem is in linux/if.h and net/if.h, both part of gentoo
>
> The Debian nad SuSE ones lok different; thi sis from memory as i am
> drowning in email at present.
>
> I think the problem is that linux/if.h include net/if.h depending on a
> macro, to stop it being loaded twice, but this has the effect of not
> loading the second part of linux/if.h as the macro is set in net/if.h
>
> Over to you.  I thought this was raised with gentoo with resulting
> silence.  When I tried to install Sentoo it computed olidly for over 30
> hrs and then seg faulted.  I did get a follow up to the bug report within
> 12 months, but I had lost interst by then
>
> ==John
>
>
>> On Mon, Sep 27, 2010 at 1:32 AM,   wrote:
>>> To elaborate, this is a Gentoo problem.  We spent a long tinme going
>>> through it and the bottom lin eis that there are two headters in Gentoo
>>> tghat cannot be loaded in eitherorder to define everything needed.  The
>>> xolution is to use a non-broken Linux distrinution!
>>
>> Well, I might not look like it, but I'm a *massive* masochist. :3
>>
>> That ebuild works-ish. I split out the sed invocations into patches,
>> and I'll be opening a Gentoo bug to get the entire thing into Portage.
>>
>> Since one of the headers is in Linux, what's the other header? Glibc?
>> Is there a Gentoo bug open? Is the incompatible header upstream
>> somewhere, or is it from a Gentoo-specific patch?
>>
>> ~ C.
>>
>> --
>> When the facts change, I change my mind. What do you do, sir? ~ Keynes
>>
>> Corbin Simpson
>> 
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
>>
>>
>>
>
>
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

-D_BSD_SOURCE also succeded compiling. i'm only using the wii opcodes and 
loscil, i'm in a more music theoretical phase of development right now so 
not relying heavly on novel signal generation.

to keep the wii opcodes from segfaulting  in the wiiuse src: 
wiiuse/src/io_nix.c:173 insert the line
        memset(&addr, 0 , sizeof addr);

haven't i tried the ebuild yet. if you want me to look at some things, i 
have both ccmra fc 11,13, and gentoo systems to try things on. wrap it up 
and i'll have a look at it.

is there a 'scons test' target?

kelly

  On Fri, 1 Oct 2010, Corbin 
Simpson wrote:

> Okay, I've done a bit more work on this.

For 5.12.1, there are interesting issues. I have successful builds
from source, which either throw a wall of errors and fail to load
(Fedora), or segfault (Gentoo) during orchestra parsing. I think the
problem is my use of continuation characters, but I don't know why it
wouldn't work.

Fedora's csound is apparently 5.10 plus a handful of patches. I
haven't had a chance to look at it deeply. If I can compile that on
Gentoo, and if it works, I may package that instead for now.

Also, I worked on the 5.12.1 ebuild a bit, splitting out the sed
invocations to patches, and adding EABI checks. I'll submit this all
to Gentoo when I've got something that works.

~ C.

On Sun, Sep 19, 2010 at 11:34 AM, kelly  hirai
 wrote:
> adding to the cflags list: in custom.py
> customCCFLAGS = ['-D_GNU_SOURCE']
>                 ^^^^^^^^^^^^^^^
> completes build in gentoo. haven't tested yet. too sleepy.
>
> kelly
>
>
> On Mon, 27 Sep 2010, jpff@cs.bath.ac.uk wrote:
>
>>
>> The problem is in linux/if.h and net/if.h, both part of gentoo
>>
>> The Debian nad SuSE ones lok different; thi sis from memory as i am
>> drowning in email at present.
>>
>> I think the problem is that linux/if.h include net/if.h depending on a
>> macro, to stop it being loaded twice, but this has the effect of not
>> loading the second part of linux/if.h as the macro is set in net/if.h
>>
>> Over to you.  I thought this was raised with gentoo with resulting
>> silence.  When I tried to install Sentoo it computed olidly for over 30
>> hrs and then seg faulted.  I did get a follow up to the bug report within
>> 12 months, but I had lost interst by then
>>
>> ==John
>>
>>
>>> On Mon, Sep 27, 2010 at 1:32 AM,   wrote:
>>>>
>>>> To elaborate, this is a Gentoo problem.  We spent a long tinme going
>>>> through it and the bottom lin eis that there are two headters in Gentoo
>>>> tghat cannot be loaded in eitherorder to define everything needed.  The
>>>> xolution is to use a non-broken Linux distrinution!
>>>
>>> Well, I might not look like it, but I'm a *massive* masochist. :3
>>>
>>> That ebuild works-ish. I split out the sed invocations into patches,
>>> and I'll be opening a Gentoo bug to get the entire thing into Portage.
>>>
>>> Since one of the headers is in Linux, what's the other header? Glibc?
>>> Is there a Gentoo bug open? Is the incompatible header upstream
>>> somewhere, or is it from a Gentoo-specific patch?
>>>
>>> ~ C.
>>>
>>> --
>>> When the facts change, I change my mind. What do you do, sir? ~ Keynes
>>>
>>> Corbin Simpson
>>> 
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>

i'm not sure if it is nin the latest sources. my recollection is that 
there hasn'e been much movement on libwiiuse. i have a modification date 
of apr 1 2008 v12  on the archive i'm using, which i compared to the 
sourceforge time stamp apr 3 2008 and assumed it hasn't been bumped.

laforest wrote me sept 17 2009 suggesting the change. i zapped back 'works 
for me'. still working.

k.

On Sun, 3 Oct 2010, jpff@cs.bath.ac.uk wrote:

>> -D_BSD_SOURCE also succeded compiling. i'm only using the wii opcodes and
>> loscil, i'm in a more music theoretical phase of development right now so
>> not relying heavly on novel signal generation.
>>
>> to keep the wii opcodes from segfaulting  in the wiiuse src:
>> wiiuse/src/io_nix.c:173 insert the line
>>         memset(&addr, 0 , sizeof addr);
>>
>
> That line i sin my sources
>
> static int wiiuse_connect_single(struct wiimote_t* wm, char* address) {
> 	struct sockaddr_l2 addr;
>
>        memset(&addr, 0, sizeof (addr)); /* Sanity */
> 	if (!wm || WIIMOTE_IS_CONNECTED(wm))
> 		return 0;
>
> Is thi soething I just did and forgot, or is it in the lastest version?
>
> ==John ff
>
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

In file included from OOps/remote.c:31:
/usr/include/linux/if.h:180: error: field ‘ifru_map’ has incomplete type
OOps/remote.c: In function ‘getIpAddress’:
OOps/remote.c:70: error: storage size of ‘ifr’ isn’t known

This is with the 5.12.1 source tarball, on Gentoo. (Gentoo has no
ebuild for csound.)

Suggestions? Looking at the code, it appears to be related to the
kernel headers somehow, but I'm not familiar with that part of the
kernel.

~ C.

-- 
When the facts change, I change my mind. What do you do, sir? ~ Keynes

Corbin Simpson



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

To elaborate, this is a Gentoo problem.  We spent a long tinme going
through it and the bottom lin eis that there are two headters in Gentoo
tghat cannot be loaded in eitherorder to define everything needed.  The
xolution is to use a non-broken Linux distrinution!

>
> on 2010-09-26 at 14:47 Corbin Simpson wrote:
>
>>In file included from OOps/remote.c:31:
>>/usr/include/linux/if.h:180: error: field ‘ifru_map’ has incomplete
>> type
>>OOps/remote.c: In function ‘getIpAddress’:
>>OOps/remote.c:70: error: storage size of ‘ifr’ isn’t known
>
> yes, this problem has been addressed many times but apparently no solution
> has been found.
>
>
>>This is with the 5.12.1 source tarball, on Gentoo. (Gentoo has no
>>ebuild for csound.)
>
> there are some unofficial ebuilds out there. try the attached one. it's
> very basic and crude, but it does the job. at least it does for me, let me
> know if it works for you too. i haven't been keeping up with the new scons
> options to edit the use flags accordingly, but the most important options
> are covered.
>
> good luck,
>
> lj
>
>
> Send bugs reports to the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>




Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Mon, Sep 27, 2010 at 1:32 AM,   wrote:
> To elaborate, this is a Gentoo problem.  We spent a long tinme going
> through it and the bottom lin eis that there are two headters in Gentoo
> tghat cannot be loaded in eitherorder to define everything needed.  The
> xolution is to use a non-broken Linux distrinution!

Well, I might not look like it, but I'm a *massive* masochist. :3

That ebuild works-ish. I split out the sed invocations into patches,
and I'll be opening a Gentoo bug to get the entire thing into Portage.

Since one of the headers is in Linux, what's the other header? Glibc?
Is there a Gentoo bug open? Is the incompatible header upstream
somewhere, or is it from a Gentoo-specific patch?

~ C.

-- 
When the facts change, I change my mind. What do you do, sir? ~ Keynes

Corbin Simpson



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

The problem is in linux/if.h and net/if.h, both part of gentoo

The Debian nad SuSE ones lok different; thi sis from memory as i am
drowning in email at present.

I think the problem is that linux/if.h include net/if.h depending on a
macro, to stop it being loaded twice, but this has the effect of not
loading the second part of linux/if.h as the macro is set in net/if.h

Over to you.  I thought this was raised with gentoo with resulting
silence.  When I tried to install Sentoo it computed olidly for over 30
hrs and then seg faulted.  I did get a follow up to the bug report within
12 months, but I had lost interst by then

==John


> On Mon, Sep 27, 2010 at 1:32 AM,   wrote:
>> To elaborate, this is a Gentoo problem.  We spent a long tinme going
>> through it and the bottom lin eis that there are two headters in Gentoo
>> tghat cannot be loaded in eitherorder to define everything needed.  The
>> xolution is to use a non-broken Linux distrinution!
>
> Well, I might not look like it, but I'm a *massive* masochist. :3
>
> That ebuild works-ish. I split out the sed invocations into patches,
> and I'll be opening a Gentoo bug to get the entire thing into Portage.
>
> Since one of the headers is in Linux, what's the other header? Glibc?
> Is there a Gentoo bug open? Is the incompatible header upstream
> somewhere, or is it from a Gentoo-specific patch?
>
> ~ C.
>
> --
> When the facts change, I change my mind. What do you do, sir? ~ Keynes
>
> Corbin Simpson
> 
>
>
> Send bugs reports to the Sourceforge bug tracker
>             https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>
>
>




Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Okay, I've done a bit more work on this.

For 5.12.1, there are interesting issues. I have successful builds
from source, which either throw a wall of errors and fail to load
(Fedora), or segfault (Gentoo) during orchestra parsing. I think the
problem is my use of continuation characters, but I don't know why it
wouldn't work.

Fedora's csound is apparently 5.10 plus a handful of patches. I
haven't had a chance to look at it deeply. If I can compile that on
Gentoo, and if it works, I may package that instead for now.

Also, I worked on the 5.12.1 ebuild a bit, splitting out the sed
invocations to patches, and adding EABI checks. I'll submit this all
to Gentoo when I've got something that works.

~ C.

On Sun, Sep 19, 2010 at 11:34 AM, kelly  hirai
 wrote:
> adding to the cflags list: in custom.py
> customCCFLAGS = ['-D_GNU_SOURCE']
>                 ^^^^^^^^^^^^^^^
> completes build in gentoo. haven't tested yet. too sleepy.
>
> kelly
>
>
> On Mon, 27 Sep 2010, jpff@cs.bath.ac.uk wrote:
>
>>
>> The problem is in linux/if.h and net/if.h, both part of gentoo
>>
>> The Debian nad SuSE ones lok different; thi sis from memory as i am
>> drowning in email at present.
>>
>> I think the problem is that linux/if.h include net/if.h depending on a
>> macro, to stop it being loaded twice, but this has the effect of not
>> loading the second part of linux/if.h as the macro is set in net/if.h
>>
>> Over to you.  I thought this was raised with gentoo with resulting
>> silence.  When I tried to install Sentoo it computed olidly for over 30
>> hrs and then seg faulted.  I did get a follow up to the bug report within
>> 12 months, but I had lost interst by then
>>
>> ==John
>>
>>
>>> On Mon, Sep 27, 2010 at 1:32 AM,   wrote:
>>>>
>>>> To elaborate, this is a Gentoo problem.  We spent a long tinme going
>>>> through it and the bottom lin eis that there are two headters in Gentoo
>>>> tghat cannot be loaded in eitherorder to define everything needed.  The
>>>> xolution is to use a non-broken Linux distrinution!
>>>
>>> Well, I might not look like it, but I'm a *massive* masochist. :3
>>>
>>> That ebuild works-ish. I split out the sed invocations into patches,
>>> and I'll be opening a Gentoo bug to get the entire thing into Portage.
>>>
>>> Since one of the headers is in Linux, what's the other header? Glibc?
>>> Is there a Gentoo bug open? Is the incompatible header upstream
>>> somewhere, or is it from a Gentoo-specific patch?
>>>
>>> ~ C.
>>>
>>> --
>>> When the facts change, I change my mind. What do you do, sir? ~ Keynes
>>>
>>> Corbin Simpson
>>> 
>>>
>>>
>>> Send bugs reports to the Sourceforge bug tracker
>>>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
>>> Discussions of bugs and features can be posted here
>>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>>> csound"
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>> Send bugs reports to the Sourceforge bug tracker
>>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
>> Discussions of bugs and features can be posted here
>> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
>> csound"
>>
> Send bugs reports to the Sourceforge bug tracker
>           https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>



-- 
When the facts change, I change my mind. What do you do, sir? ~ Keynes

Corbin Simpson



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> -D_BSD_SOURCE also succeded compiling. i'm only using the wii opcodes and
> loscil, i'm in a more music theoretical phase of development right now so
> not relying heavly on novel signal generation.
>
> to keep the wii opcodes from segfaulting  in the wiiuse src:
> wiiuse/src/io_nix.c:173 insert the line
>         memset(&addr, 0 , sizeof addr);
>

That line i sin my sources

static int wiiuse_connect_single(struct wiimote_t* wm, char* address) {
	struct sockaddr_l2 addr;

        memset(&addr, 0, sizeof (addr)); /* Sanity */
	if (!wm || WIIMOTE_IS_CONNECTED(wm))
		return 0;

Is thi soething I just did and forgot, or is it in the lastest version?

==John ff



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"