Skip to contentSkip to conversation

thread_info implementation

DISCUSSION Arkadiy Chapkis - Arc 6 participants 9 messages Feb 11, 2002 – Feb 12, 2002

lore
Share & tools
lore
#0 lore raw

[1.] One line summary of the problem: Can't compile kernel 2.5.4

[2.] Full description of the problem/report: After configuring and starting
compilation, there are messages:

In file included from /usr/local/src/test/linux-2.5.4/include/linux/spinlock.h:7,
from
/usr/local/src/test/linux-2.5.4/include/linux/module.h:11,

                 from loop.c:55:

/usr/local/src/test/linux-2.5.4/include/linux/thread_info.h:10:29:
asm/thread_info.h: No such file or directory

[4.] Kernel version (from /proc/version): Current is 2.4.16, downloaded is
2.5.4

# cat /proc/version
Linux version 2.4.16 ([email protected]) (gcc version 2.96 20000731 (Red Hat
Linux 7.1 2.96-99))

[7.] Environment AlphaStation 200 4/233

[7.2.] Processor information (from /proc/cpuinfo):

# cat /proc/cpuinfo
cpu                     : Alpha
cpu model               : EV4
cpu variation           : 0
cpu revision            : 0
cpu serial number       : Linux_is_Great!
system type             : Avanti
system variation        : 0
system revision         : 0
system serial number    : MILO-2.0.35-c5.
cycle frequency [Hz]    : 233313780
timer frequency [Hz]    : 1024.00
page size [bytes]       : 8192
phys. address bits      : 34
max. addr. space #      : 63
BogoMIPS                : 458.36
kernel unaligned acc    : 90143620 (pc=fffffc00004db7b0,va=fffffc000056830a)
user unaligned acc      : 90 (pc=1200bb3f0,va=12028b87a)
platform string         : N/A
cpus detected           : 0

[X.] Other notes, patches, fixes, workarounds: Apperently the thread_info implementation for sparc platform broke alpha platform as there is no asm/thread_info.h file.

Thanks you,

Arc C. [email protected] [email protected]

#1 lore raw
Quoted 7 lines · Arkadiy Chapkis - Arc

In file included from /usr/local/src/test/linux-2.5.4/include/linux/spinlock.h:7,
from
/usr/local/src/test/linux-2.5.4/include/linux/module.h:11,

                 from loop.c:55:

/usr/local/src/test/linux-2.5.4/include/linux/thread_info.h:10:29:
asm/thread_info.h: No such file or directory

This is known. I don't believe any arches except i386 and SPARC64 are using the new thread_info / task_struct implementation introduced during 2.5.4-pre.

I think I saw some changesets from Jeff Garzik wrt alpha thread_info support, so perhaps in 2.5.5-pre1.

Robert Love

#2 lore raw
Quoted 10 lines · Robert Love
Quoted 7 lines · Arkadiy Chapkis - Arc

In file included from /usr/local/src/test/linux-2.5.4/include/linux/spinlock.h:7,
from
/usr/local/src/test/linux-2.5.4/include/linux/module.h:11,

                 from loop.c:55:

/usr/local/src/test/linux-2.5.4/include/linux/thread_info.h:10:29:
asm/thread_info.h: No such file or directory

Quoted 3 lines

This is known. I don't believe any arches except i386 and SPARC64 are using the new thread_info / task_struct implementation introduced during 2.5.4-pre.

You are optimist.
I could not manage to make my sparc64 boot with 2.5.3+ kernels. Now, Actually my problem is reiserFS on sparc64 (I already posted about this). Let's hope I could run 2.5 on sparc64 soon ;)

Luigi

Quoted 12 lines

I think I saw some changesets from Jeff Garzik wrt alpha thread_info support, so perhaps in 2.5.5-pre1.

Robert Love

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#3 lore raw
Quoted 1 line · Luigi Genoni

You are optimist.

The glass is always half full ;)

Quoted 3 lines

I could not manage to make my sparc64 boot with 2.5.3+ kernels. Now, Actually my problem is reiserFS on sparc64 (I already posted about this). Let's hope I could run 2.5 on sparc64 soon ;)

I know Dave Miller merged a lot of SPARC64 code into 2.5.4 (including preemptible kernel support for SPARC64!) ... at least it compiles. I suspect no other arches do right now.

Robert Love

#4 lore raw
In reply to Luigi Genoni
Quoted 7 lines · Robert Love
Quoted 1 line · Luigi Genoni

You are optimist.

The glass is always half full ;)

Quoted 7 lines
Quoted 3 lines

I could not manage to make my sparc64 boot with 2.5.3+ kernels. Now, Actually my problem is reiserFS on sparc64 (I already posted about this). Let's hope I could run 2.5 on sparc64 soon ;)

I know Dave Miller merged a lot of SPARC64 code into 2.5.4 (including preemptible kernel support for SPARC64!) ... at least it compiles. I suspect no other arches do right now.

That computing hotshot otherwise known as Richard Henderon got alpha axp going with thread_info, though only in UP so far. So, i386, sparc64, and alpha have been hacking into working shape.

:Jeff

Signature · 3 lines
Jeff Garzik      | "I went through my candy like hot oatmeal
Building 1024    |  through an internally-buttered weasel."
MandrakeSoft     |             - goats.com
#5 lore raw
In reply to Luigi Genoni
Quoted 2 lines · Jeff Garzik

... Richard Henderon got alpha axp going with thread_info, though only in UP so far.

The SMP patch had one stupid missing line. So that's working now too.

Though I seem to be having some problems with NFS. Mount goes into D state for quite some time and the portmapper complains about timeouts connecting to localhost. Anyone else see anything like that? I suppose I'll build an x86 kernel from the same source and see what I can find...

r~

#6 lore raw
In reply to Luigi Genoni
Quoted 4 lines · Richard Henderson

Though I seem to be having some problems with NFS. Mount goes into D state for quite some time and the portmapper complains about timeouts connecting to localhost. Anyone else see anything like that? I suppose I'll build an x86 kernel from the same source and see what I can find...

Yes. I saw this start happening in my tree when I merged 2.5.4pre2 When I tried a 2.5.4pre2 vanilla, the problem was gone, so I put it down to a problem in my tree. When I rebooted back into it, the problem was gone. Aren't heisenbugs fun?
Exactly the same symptoms, portmap borken, NIS/NFS subsequently fail. This was on x86 btw.

Signature · 2 lines
| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs
#7 lore raw
In reply to Luigi Genoni
Quoted 8 lines · Dave Jones
Quoted 4 lines · Richard Henderson

Though I seem to be having some problems with NFS. Mount goes into D state for quite some time and the portmapper complains about timeouts connecting to localhost. Anyone else see anything like that? I suppose I'll build an x86 kernel from the same source and see what I can find...

Quoted 5 lines

Yes. I saw this start happening in my tree when I merged 2.5.4pre2 When I tried a 2.5.4pre2 vanilla, the problem was gone, so I put it down to a problem in my tree. When I rebooted back into it, the problem was gone. Aren't heisenbugs fun?
Exactly the same symptoms, portmap borken, NIS/NFS subsequently fail.

/etc/nsswitch.conf set up correctly? /etc/host.conf?

I notice that newer RH and MDK initscripts require a bunch of stuff like netlink devices and ipv6 support, make sure you have those enabled too...

Jeff

Signature · 3 lines
Jeff Garzik      | "I went through my candy like hot oatmeal
Building 1024    |  through an internally-buttered weasel."
MandrakeSoft     |             - goats.com
#8 lore raw
Quoted 2 lines · Jeff Garzik
Quoted 1 line

Exactly the same symptoms, portmap borken, NIS/NFS subsequently fail.

/etc/nsswitch.conf set up correctly? /etc/host.conf?

No problems there. This box worked fine on any other kernel, and that problem only showed up the once..

Quoted 3 lines

I notice that newer RH and MDK initscripts require a bunch of stuff like netlink devices and ipv6 support, make sure you have those enabled too...

Yup.

Signature · 2 lines
| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs