Skip Menu |
 

Date: Tue, 20 Dec 2005 17:20:25 -0800
From: Quanah Gibson-Mount <quanah@stanford.edu>
To: krb5-bugs@mit.edu
Subject: Re: Alpha support is missing in KRB5-1.4.3
Download (untitled) / with headers
text/plain 2.8KiB


--On Tuesday, December 20, 2005 4:36 PM -0800 Quanah Gibson-Mount
<quanah@stanford.edu> wrote:

Show quoted text
>
>
> --On Tuesday, December 20, 2005 4:30 PM -0800 Quanah Gibson-Mount
> <quanah@stanford.edu> wrote:
>
>> 2) There is no way to disable re-entrancy from configure. Although
>> thread safety is great, I'd just like this to compile on the alpha. If I
>> could turn off re-entrancy from configure, then this should be
>> accomplished. Defining --disable-thread-support still uses the re-entrant
>> thread functions, so this does not help me get around this problem.
>
> Actually, this appears to have done what I want, but I'd still rather
> just be able to keep threads, even if they weren't threadsafe.

There are, however, other problems remaining:

/usr/pubsw/bin/gcc -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\"
-DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\"
-DKRB5_KRB4_COMPAT=1 -DHAVE_BT_RSEQ=1 -DKRB5_PRIVATE=1 -DKRB5_DEPRECATED=1
-DKRB5_DNS_LOOKUP_KDC=1 -DKRB5_DNS_LOOKUP=1 -DHAVE_LIBRESOLV=1
-DHAVE_DN_SKIPNAME=1 -DHAVE_RES_SEARCH=1 -DDELAY_INITIALIZER=1
-DCONSTRUCTOR_ATTR_WORKS=1 -DDESTRUCTOR_ATTR_WORKS=1
-DUSE_LINKER_INIT_OPTION=1 -DUSE_LINKER_FINI_OPTION=1 -DSTDC_HEADERS=1
-DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1
-DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_UNISTD_H=1
-DHAVE_STDLIB_H=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_LIMITS_H=1 -DHAVE_MEMORY_H=1
-DSIZEOF_SHORT=2 -DSIZEOF_INT=4 -DSIZEOF_LONG=8 -I../../../include
-I../../../../../src/lib/gssapi/krb5/../../../include
-I../../../include/krb5
-I../../../../../src/lib/gssapi/krb5/../../../include/krb5 -I.
-I../../../../../src/lib/gssapi/krb5
-I../../../../../src/lib/gssapi/krb5/.. -I../generic
-I../../../../../src/lib/gssapi/krb5/../generic -O2 -DSTANFORD -c
../../../../../src/lib/gssapi/krb5/accept_sec_context.c
In file included from ../../../../../src/lib/gssapi/krb5/gssapiP_krb5.h:69,
from
../../../../../src/lib/gssapi/krb5/accept_sec_context.c:74:
./gssapi_krb5.h:98:22: inttypes.h: No such file or directory
In file included from ../../../../../src/lib/gssapi/krb5/gssapiP_krb5.h:69,
from
../../../../../src/lib/gssapi/krb5/accept_sec_context.c:74:
./gssapi_krb5.h:99: error: parse error before "gss_uint64"
./gssapi_krb5.h:99: warning: data definition has no type or storage class
./gssapi_krb5.h:134: error: parse error before "gss_uint64"
./gssapi_krb5.h:134: warning: no semicolon at end of struct or union
./gssapi_krb5.h:135: warning: data definition has no type or storage class
./gssapi_krb5.h:146: error: parse error before '}' token
./gssapi_krb5.h:146: warning: data definition has no type or storage class


--Quanah

--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html
Date: Tue, 20 Dec 2005 16:30:36 -0800
From: Quanah Gibson-Mount <quanah@stanford.edu>
To: krb5-bugs@mit.edu
Subject: Alpha support is missing in KRB5-1.4.3
Download (untitled) / with headers
text/plain 1.6KiB
In attempting to compile krb5-1.4.3 on our Alpha, I found that there really
is no support for this operating system.
I note the following:

1) Alpha's threadsafe functions have different arguments than what MIT Krb5
assigns to it in fake-addrinfo.h. Some of this is easy to fix:

I changed line 177 to:

177 #if (defined (_AIX) || defined (_OSF)) /* XXX should have a
feature test! */

and added #define _OSF 1 to autoconf.h after configure finished.

However, this still leaves problems with:

getservbyname_r
and
getservbyport_r


On the Alpha, these are defined as:

extern int getservbyname_r __((const char *, const char *, struct servent
*, \
struct servent_data *));
extern int getservbyport_r __((int, const char *, struct servent *, \
struct servent_data *));


There are no function definitions that match these number of arguments in
the fake-addrinfo.h file. It also doesn't even closely match either of the
two definitions for getservbyname_r (7 args or 5 args). I think it most
matches the 7 args case, however, since it does return an int, and it takes
two structs of data.


2) There is no way to disable re-entrancy from configure. Although thread
safety is great, I'd just like this to compile on the alpha. If I could
turn off re-entrancy from configure, then this should be accomplished.
Defining --disable-thread-support still uses the re-entrant thread
functions, so this does not help me get around this problem.


--Quanah

--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html
Date: Tue, 20 Dec 2005 16:36:25 -0800
From: Quanah Gibson-Mount <quanah@stanford.edu>
To: krb5-bugs@mit.edu
Subject: Re: Alpha support is missing in KRB5-1.4.3


--On Tuesday, December 20, 2005 4:30 PM -0800 Quanah Gibson-Mount
<quanah@stanford.edu> wrote:

Show quoted text
> 2) There is no way to disable re-entrancy from configure. Although
> thread safety is great, I'd just like this to compile on the alpha. If I
> could turn off re-entrancy from configure, then this should be
> accomplished. Defining --disable-thread-support still uses the re-entrant
> thread functions, so this does not help me get around this problem.

Actually, this appears to have done what I want, but I'd still rather just
be able to keep threads, even if they weren't threadsafe.

--Quanah


--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html
From: Russ Allbery <rra@stanford.edu>
To: rt@krbdev.mit.edu
Cc: quanah@stanford.edu
Subject: Re: [krbdev.mit.edu #3303] Alpha support is missing in KRB5-1.4.3
Date: Tue, 20 Dec 2005 17:45:15 -0800
RT-Send-Cc:
Quanah Gibson-Mount via RT <rt-comment@krbdev.mit.edu> writes:

Show quoted text
> In attempting to compile krb5-1.4.3 on our Alpha, I found that there
> really is no support for this operating system.

Er, the helpful missing data point here is that this is Tru64 4.0G (in
other words, not Tru64 5.x, which is probably where this has been tested).

Alpha isn't very specific. :) I'm sure Linux on Alpha works fine.

--
Russ Allbery (rra@stanford.edu) <http://www.eyrie.org/~eagle/>
From: Ken Raeburn <raeburn@MIT.EDU>
Subject: Re: [krbdev.mit.edu #3303] Alpha support is missing in KRB5-1.4.3
Date: Tue, 20 Dec 2005 22:39:10 -0500
To: rt@krbdev.mit.edu
RT-Send-Cc:
On Dec 20, 2005, at 20:45, Russ Allbery via RT wrote:
Show quoted text
> Quanah Gibson-Mount via RT <rt-comment@krbdev.mit.edu> writes:
>> In attempting to compile krb5-1.4.3 on our Alpha, I found that there
>> really is no support for this operating system.
>
> Er, the helpful missing data point here is that this is Tru64 4.0G (in
> other words, not Tru64 5.x, which is probably where this has been
> tested).

Tru64 5.1A, I think, is our only Tru64 test system.

Show quoted text
> Alpha isn't very specific. :) I'm sure Linux on Alpha works fine.

NetBSD/Alpha is not in terrible shape either.

Ken
Date: Tue, 20 Dec 2005 17:32:45 -0800
From: Quanah Gibson-Mount <quanah@stanford.edu>
To: krb5-bugs@mit.edu
Subject: Re: Alpha support is missing in KRB5-1.4.3
Download (untitled) / with headers
text/plain 2.2KiB


--On Tuesday, December 20, 2005 5:20 PM -0800 Quanah Gibson-Mount
<quanah@stanford.edu> wrote:

/usr/pubsw/bin/gcc -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\"
Show quoted text
> -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\"
> -DKRB5_KRB4_COMPAT=1 -DHAVE_BT_RSEQ=1 -DKRB5_PRIVATE=1
> -DKRB5_DEPRECATED=1 -DKRB5_DNS_LOOKUP_KDC=1 -DKRB5_DNS_LOOKUP=1
> -DHAVE_LIBRESOLV=1 -DHAVE_DN_SKIPNAME=1 -DHAVE_RES_SEARCH=1
> -DDELAY_INITIALIZER=1 -DCONSTRUCTOR_ATTR_WORKS=1
> -DDESTRUCTOR_ATTR_WORKS=1 -DUSE_LINKER_INIT_OPTION=1
> -DUSE_LINKER_FINI_OPTION=1 -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1
> -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1
> -DHAVE_STRINGS_H=1 -DHAVE_UNISTD_H=1 -DHAVE_STDLIB_H=1
> -DHAVE_SYS_TYPES_H=1 -DHAVE_LIMITS_H=1 -DHAVE_MEMORY_H=1 -DSIZEOF_SHORT=2
> -DSIZEOF_INT=4 -DSIZEOF_LONG=8 -I../../../include
> -I../../../../../src/lib/gssapi/krb5/../../../include
> -I../../../include/krb5
> -I../../../../../src/lib/gssapi/krb5/../../../include/krb5 -I.
> -I../../../../../src/lib/gssapi/krb5
> -I../../../../../src/lib/gssapi/krb5/.. -I../generic
> -I../../../../../src/lib/gssapi/krb5/../generic -O2 -DSTANFORD -c
> ../../../../../src/lib/gssapi/krb5/accept_sec_context.c
> In file included from
> ../../../../../src/lib/gssapi/krb5/gssapiP_krb5.h:69,
> from
> ../../../../../src/lib/gssapi/krb5/accept_sec_context.c:74:
> ./gssapi_krb5.h:98:22: inttypes.h: No such file or directory
> In file included from
> ../../../../../src/lib/gssapi/krb5/gssapiP_krb5.h:69,
> from
> ../../../../../src/lib/gssapi/krb5/accept_sec_context.c:74:
> ./gssapi_krb5.h:99: error: parse error before "gss_uint64"
> ./gssapi_krb5.h:99: warning: data definition has no type or storage class
> ./gssapi_krb5.h:134: error: parse error before "gss_uint64"
> ./gssapi_krb5.h:134: warning: no semicolon at end of struct or union
> ./gssapi_krb5.h:135: warning: data definition has no type or storage class
> ./gssapi_krb5.h:146: error: parse error before '}' token
> ./gssapi_krb5.h:146: warning: data definition has no type or storage class

I note that I find that u_int64_t on Alpha is an "unsigned long long".

--Quanah

--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html