Skip to contentSkip to conversation

2.4.x and SMP fails to compile (`current' undefined)

DISCUSSION David Ford 10 participants 21 messages Jan 31, 2001 – Feb 2, 2001

lore
Share & tools
lore

David Ford

author
#0 lore raw

A person just brought up a problem in #kernelnewbies, building an SMP kernel doesn't work very well, current is undefined. I don't have more time to debug it but I'll strip the config and put it up at
http://stuph.org/smp-config

-d

--
There is a natural aristocracy among men. The grounds of this are virtue and talents. Thomas Jefferson The good thing about standards is that there are so many to choose from. Andrew S. Tanenbaum

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

#1 lore raw
Quoted 4 lines · * David Ford ([email protected]) wrote:

A person just brought up a problem in #kernelnewbies, building an SMP kernel doesn't work very well, current is undefined. I don't have more time to debug it but I'll strip the config and put it up at
http://stuph.org/smp-config

They're trying to compile SMP for Athlon/K7 (CONFIG_MK7=y).

Stephen

author
#2 lore raw
In reply to David Ford

Mhm. Is it worth the effort to make a dependancy on the CPU type for SMP?

</idle questions>

-d

Quoted 7 lines · Stephen Frost
Quoted 4 lines · * David Ford ([email protected]) wrote:

A person just brought up a problem in #kernelnewbies, building an SMP kernel doesn't work very well, current is undefined. I don't have more time to debug it but I'll strip the config and put it up at
http://stuph.org/smp-config

They're trying to compile SMP for Athlon/K7 (CONFIG_MK7=y).

--
There is a natural aristocracy among men. The grounds of this are virtue and talents. Thomas Jefferson The good thing about standards is that there are so many to choose from. Andrew S. Tanenbaum

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

#5 lore raw
In reply to David Ford

[David Ford]

Quoted 2 lines

Mhm. Is it worth the effort to make a dependancy on the CPU type for SMP?

Yes. 'make config' should not allow unsupported configurations, at least where convenient.

So - are any of the other chip types also incompatible with SMP support (the Winchips, maybe)?

Peter

arch/i386/config.in renamed from arch/i386/config.in~ +4 -1
@@ -154,7 +154,10 @@
 
 bool 'Math emulation' CONFIG_MATH_EMULATION
 bool 'MTRR (Memory Type Range Register) support' CONFIG_MTRR
-bool 'Symmetric multi-processing support' CONFIG_SMP
+# AMD SMP - not yet
+if [ "$CONFIG_MK6" != "y" -a "$CONFIG_MK7" != "y" ]; then
+   bool 'Symmetric multi-processing support' CONFIG_SMP
+fi
 if [ "$CONFIG_SMP" != "y" ]; then
    bool 'APIC and IO-APIC support on uniprocessors' CONFIG_X86_UP_IOAPIC
    if [ "$CONFIG_X86_UP_IOAPIC" = "y" ]; then

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

#6 lore raw
In reply to David Ford
Quoted 17 lines · David Ford

Mhm. Is it worth the effort to make a dependancy on the CPU type for SMP?

</idle questions>

-d

Quoted 7 lines · Stephen Frost
Quoted 4 lines · * David Ford ([email protected]) wrote:

A person just brought up a problem in #kernelnewbies, building an SMP kernel doesn't work very well, current is undefined. I don't have more time to debug it but I'll strip the config and put it up at
http://stuph.org/smp-config

They're trying to compile SMP for Athlon/K7 (CONFIG_MK7=y).

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h. in_interrupt() cannot be called from there.

I have posted a patch here many times since last May. Most recent was Saturday.

Tom
--
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/

#14 lore raw
In reply to David Ford
Quoted 8 lines · Peter Samuelson

[Tom Leete]

Quoted 2 lines

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h.

So you're saying SMP *is* supported on Athlon? Do motherboards exist?

Peter

No, I'm saying that SMP locking etc. is compatible with Athlon. The failure to build is not a workaround but a coding error. SMP builds for UP machines are supposed to work.

Tom

Signature · 5 lines
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/
#18 lore raw
In reply to David Ford

[Tom Leete]

Quoted 3 lines

No, I'm saying that SMP locking etc. is compatible with Athlon. The failure to build is not a workaround but a coding error. SMP builds for UP machines are supposed to work.

Yes, but if SMP for Athlons is not supported, what is the point in allowing CONFIG_SMP + CONFIG_MK7 (or CONFIG_SMP + CONFIG_MK6)? Such a kernel will not run on *any* SMP system, since AMD kernels do not work on Intel. If an AMD user really wants to carry around SMP baggage for no reason, let him use CONFIG_M586TSC.

This is like offering MCA-bus drivers for 'make config' on SPARC. Sure, they might compile, but nobody will have any use for them. Thus, we ensure that they never appear on the menus in the first place.

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

#7 lore raw
In reply to Tom Leete
Quoted 2 lines

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h. in_interrupt() cannot be called from there.

The string.h code was fine, someone came along and put in a ridiculous loop in the include dependancies and broke it. Nobody has had the time to untangle it cleanly since

Quoted 2 lines

I have posted a patch here many times since last May. Most recent was Saturday.

uninlining the code is too high a cost.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/

#19 lore raw
In reply to David Ford
Quoted 5 lines

Yes, but if SMP for Athlons is not supported, what is the point in allowing CONFIG_SMP + CONFIG_MK7 (or CONFIG_SMP + CONFIG_MK6)? Such a kernel will not run on *any* SMP system, since AMD kernels do not work on Intel. If an AMD user really wants to carry around SMP baggage for no reason, let him use CONFIG_M586TSC.

Authors of drivers or other kernel add-on code often have to test their stuff on a variety of configurations. _Especially_ on an SMP kernel even if all they have is UP.

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

#3 lore raw
In reply to David Ford

probably not now that SMP athlon boards are supposed to be starting to be available.

David Lang

Quoted 28 lines · David Ford

Date: Tue, 30 Jan 2001 20:37:02 -0800
From: David Ford <[email protected]>
To: Stephen Frost <[email protected]>
Cc: LKML <[email protected]>
Subject: Re: 2.4.x and SMP fails to compile (`current' undefined)

Mhm. Is it worth the effort to make a dependancy on the CPU type for SMP?

</idle questions>

-d

Quoted 7 lines · Stephen Frost
Quoted 4 lines · * David Ford ([email protected]) wrote:

A person just brought up a problem in #kernelnewbies, building an SMP kernel doesn't work very well, current is undefined. I don't have more time to debug it but I'll strip the config and put it up at
http://stuph.org/smp-config

They're trying to compile SMP for Athlon/K7 (CONFIG_MK7=y).

--
There is a natural aristocracy among men. The grounds of this are virtue and talents. Thomas Jefferson The good thing about standards is that there are so many to choose from. Andrew S. Tanenbaum

Quoted 5 lines

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

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

#15 lore raw
In reply to Tom Leete

about the third story down is one mentioning SMP athlon boards actually starting to show up

http://www.aceshardware.com/Spades/list_news.php?category=AMD

David Lang

Quoted 21 lines · Tom Leete

Date: Wed, 31 Jan 2001 05:48:31 -0500
From: Tom Leete <[email protected]>
To: Peter Samuelson <[email protected]>
Cc: David Ford <[email protected]>, Stephen Frost <[email protected]>,
LKML <[email protected]>
Subject: Re: 2.4.x and SMP fails to compile (`current' undefined)

Quoted 8 lines · Peter Samuelson

[Tom Leete]

Quoted 2 lines

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h.

So you're saying SMP *is* supported on Athlon? Do motherboards exist?

Peter

No, I'm saying that SMP locking etc. is compatible with Athlon. The failure to build is not a workaround but a coding error. SMP builds for UP machines are supposed to work.

Quoted 8 lines

Tom
--
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/

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

#16 lore raw
In reply to Tom Leete

[David Lang]

Quoted 2 lines

about the third story down is one mentioning SMP athlon boards actually starting to show up

Cool! I want one. So ... is there any way to know if and when Linux kernel support will be available? I assume AMD doesn't follow the MPS 1.x standards.

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

#17 lore raw
Quoted 3 lines

Cool! I want one. So ... is there any way to know if and when Linux kernel support will be available? I assume AMD doesn't follow the MPS 1.x standards.

As far as I am aware they will be following the software interface.

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

#8 lore raw
In reply to Alan Cox
Quoted 7 lines · Alan Cox
Quoted 3 lines

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h. in_interrupt() cannot be called from there.

The string.h code was fine, someone came along and put in a ridiculous loop in the include dependancies and broke it. Nobody has had the time to untangle it cleanly since

Yes, bitrot. I don't see a rearrangement of system headers happening in 2.4. I'm pretty sure if I committed such a patch it would have no measurable lifetime.

Quoted 5 lines
Quoted 3 lines

I have posted a patch here many times since last May. Most recent was Saturday.

uninlining the code is too high a cost.

I question that. Athlon does branch prediction on call targets, function calls are cheap. 3dnow saves 25%-50% of cycles on a copy. How many function calls can be paid for with 1000 cycles or so?

My patch still inlines the standard string const_memcpy for the case of small known length.

If I configure SMP for a UP box, performance is clearly not my first concern. If I have a real SMP Athlon system, performance should not improve by only using one processor.

How about we get it to build before we optimize it?

Regards,
Tom

Signature · 5 lines
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/
#9 lore raw

Make it and I will care and post it on kernel.org for you. I need that patch soon.

Quoted 26 lines · Tom Leete
Quoted 7 lines · Alan Cox
Quoted 3 lines

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h. in_interrupt() cannot be called from there.

The string.h code was fine, someone came along and put in a ridiculous loop in the include dependancies and broke it. Nobody has had the time to untangle it cleanly since

Yes, bitrot. I don't see a rearrangement of system headers happening in 2.4. I'm pretty sure if I committed such a patch it would have no measurable lifetime.

Quoted 5 lines
Quoted 3 lines

I have posted a patch here many times since last May. Most recent was Saturday.

uninlining the code is too high a cost.

I question that. Athlon does branch prediction on call targets, function calls are cheap. 3dnow saves 25%-50% of cycles on a copy. How many function calls can be paid for with 1000 cycles or so?

My patch still inlines the standard string const_memcpy for the case of small known length.

Quoted 16 lines

If I configure SMP for a UP box, performance is clearly not my first concern. If I have a real SMP Athlon system, performance should not improve by only using one processor.

How about we get it to build before we optimize it?

Regards,
Tom

Signature · 5 lines
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/

Andre Hedrick
Linux ATA Development

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

#11 lore raw
In reply to Tom Leete
Quoted 8 lines
Quoted 1 line

uninlining the code is too high a cost.

I question that. Athlon does branch prediction on call targets, function calls are cheap. 3dnow saves 25%-50% of cycles on a copy. How many function calls can be paid for with 1000 cycles or so?

My patch still inlines the standard string const_memcpy for the case of small known length.

We have a very large number of memcpy's of unknown short length (often in interrupts) that are close to branches. A lot of

	if(foo==NULL)
		return
	memcpy(..

stuff for example.

Im more than happy for someone to do the benches and prove me wrong

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

#10 lore raw
In reply to Andre Hedrick
Quoted 14 lines · Andre Hedrick

Make it and I will care and post it on kernel.org for you. I need that patch soon.

Quoted 8 lines · Tom Leete
Quoted 3 lines · Alan Cox

The string.h code was fine, someone came along and put in a ridiculous loop in the include dependancies and broke it. Nobody has had the time to untangle it cleanly since

Yes, bitrot. I don't see a rearrangement of system headers happening in 2.4. I'm pretty sure if I committed such a patch it would have no measurable lifetime.

Hi Andre,

I meant that nobody should be reshuffling 2.4 headers now, didn't intend to sound like I take that personally.

I'll take a look. I may be able to do something with include guards or other #defines + multiple passes. We already have the multiple passes.

I think my arguments for the present patch are good. I'm making a mod of Arjan's athlon.c to see if I'm right. If you have a suggestion for another benchmark, I'd like to hear about it. Whatever the results, I'll post them here.

Glad if whatever comes out is useful to you.

Cheers,
Tom

Signature · 5 lines
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/
#12 lore raw
In reply to Alan Cox
Quoted 11 lines · Alan Cox

We have a very large number of memcpy's of unknown short length (often in interrupts) that are close to branches. A lot of

        if(foo==NULL)
                return
        memcpy(..

stuff for example.

Im more than happy for someone to do the benches and prove me wrong

Agreed, that is a bad case, and there is overhead for it in my patch. I'm putting together some metrics, will post results here.

Regards,
Tom

Signature · 5 lines
The Daemons lurk and are dumb. -- Emerson
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/
#4 lore raw
In reply to David Lang
Quoted 2 lines · David Lang

probably not now that SMP athlon boards are supposed to be starting to be available.

They already are... there is a dual Athlon box at the MDK booth here at LinuxWorld.

Jeff

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

#20 lore raw
In reply to David Ford

Hi!

Quoted 4 lines
Quoted 2 lines

It's not an incompatibility with the k7 chip, just bad code in include/asm-i386/string.h.

So you're saying SMP *is* supported on Athlon? Do motherboards exist?

Check today's slashdot ;-).
Pavel

Signature · 6 lines
I'm [email protected]. "In my country we have almost anarchy and I don't care."
Panos Katsaloulis describing me w.r.t. patents at [email protected]
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/