* [uml-devel] [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken
@ 2005-08-29 20:07 ` Jeff Dike
0 siblings, 0 replies; 6+ messages in thread
From: Jeff Dike @ 2005-08-29 20:07 UTC (permalink / raw)
To: akpm; +Cc: linux-kernel, user-mode-linux-devel, Al Viro
Noticed by Al Viro - SMP on x86_64 is fundamentally broken due to
UML's reuse of the host arch's percpu stuff. This is OK on x86, but
the x86_64 pda stuff just won't work for UML.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Index: linux-2.6.13-rc6/arch/um/Kconfig
===================================================================
--- linux-2.6.13-rc6.orig/arch/um/Kconfig 2005-08-15 12:03:07.000000000 -0400
+++ linux-2.6.13-rc6/arch/um/Kconfig 2005-08-15 14:04:23.000000000 -0400
@@ -196,7 +196,7 @@
config SMP
bool "Symmetric multi-processing support (EXPERIMENTAL)"
default n
- depends on MODE_TT && EXPERIMENTAL
+ depends on (MODE_TT && EXPERIMENTAL && !SMP_BROKEN) || (BROKEN && SMP_BROKEN)
help
This option enables UML SMP support.
It is NOT related to having a real SMP box. Not directly, at least.
Index: linux-2.6.13-rc6/arch/um/Kconfig_x86_64
===================================================================
--- linux-2.6.13-rc6.orig/arch/um/Kconfig_x86_64 2005-08-08 12:11:18.000000000 -0400
+++ linux-2.6.13-rc6/arch/um/Kconfig_x86_64 2005-08-15 14:04:27.000000000 -0400
@@ -33,3 +33,7 @@
config ARCH_REUSE_HOST_VSYSCALL_AREA
bool
default n
+
+config SMP_BROKEN
+ bool
+ default y
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken
@ 2005-08-29 20:07 ` Jeff Dike
0 siblings, 0 replies; 6+ messages in thread
From: Jeff Dike @ 2005-08-29 20:07 UTC (permalink / raw)
To: akpm; +Cc: linux-kernel, user-mode-linux-devel, Al Viro
Noticed by Al Viro - SMP on x86_64 is fundamentally broken due to
UML's reuse of the host arch's percpu stuff. This is OK on x86, but
the x86_64 pda stuff just won't work for UML.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Index: linux-2.6.13-rc6/arch/um/Kconfig
===================================================================
--- linux-2.6.13-rc6.orig/arch/um/Kconfig 2005-08-15 12:03:07.000000000 -0400
+++ linux-2.6.13-rc6/arch/um/Kconfig 2005-08-15 14:04:23.000000000 -0400
@@ -196,7 +196,7 @@
config SMP
bool "Symmetric multi-processing support (EXPERIMENTAL)"
default n
- depends on MODE_TT && EXPERIMENTAL
+ depends on (MODE_TT && EXPERIMENTAL && !SMP_BROKEN) || (BROKEN && SMP_BROKEN)
help
This option enables UML SMP support.
It is NOT related to having a real SMP box. Not directly, at least.
Index: linux-2.6.13-rc6/arch/um/Kconfig_x86_64
===================================================================
--- linux-2.6.13-rc6.orig/arch/um/Kconfig_x86_64 2005-08-08 12:11:18.000000000 -0400
+++ linux-2.6.13-rc6/arch/um/Kconfig_x86_64 2005-08-15 14:04:27.000000000 -0400
@@ -33,3 +33,7 @@
config ARCH_REUSE_HOST_VSYSCALL_AREA
bool
default n
+
+config SMP_BROKEN
+ bool
+ default y
^ permalink raw reply [flat|nested] 6+ messages in thread
* [uml-devel] Re: [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken
2005-08-29 20:07 ` Jeff Dike
@ 2005-08-30 20:42 ` Andi Kleen
-1 siblings, 0 replies; 6+ messages in thread
From: Andi Kleen @ 2005-08-30 20:42 UTC (permalink / raw)
To: Jeff Dike; +Cc: linux-kernel, user-mode-linux-devel, Al Viro
Jeff Dike <jdike@addtoit.com> writes:
> Noticed by Al Viro - SMP on x86_64 is fundamentally broken due to
> UML's reuse of the host arch's percpu stuff. This is OK on x86, but
> the x86_64 pda stuff just won't work for UML.
The generic one should work too, it's just less efficient.
So you can probably easily replace them.
-Andi
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken
@ 2005-08-30 20:42 ` Andi Kleen
0 siblings, 0 replies; 6+ messages in thread
From: Andi Kleen @ 2005-08-30 20:42 UTC (permalink / raw)
To: Jeff Dike; +Cc: linux-kernel, user-mode-linux-devel, Al Viro
Jeff Dike <jdike@addtoit.com> writes:
> Noticed by Al Viro - SMP on x86_64 is fundamentally broken due to
> UML's reuse of the host arch's percpu stuff. This is OK on x86, but
> the x86_64 pda stuff just won't work for UML.
The generic one should work too, it's just less efficient.
So you can probably easily replace them.
-Andi
^ permalink raw reply [flat|nested] 6+ messages in thread
* [uml-devel] Re: [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken
2005-08-30 20:42 ` Andi Kleen
@ 2005-08-30 21:28 ` Jeff Dike
-1 siblings, 0 replies; 6+ messages in thread
From: Jeff Dike @ 2005-08-30 21:28 UTC (permalink / raw)
To: Andi Kleen; +Cc: linux-kernel, user-mode-linux-devel, Al Viro
On Tue, Aug 30, 2005 at 10:42:15PM +0200, Andi Kleen wrote:
> The generic one should work too, it's just less efficient.
> So you can probably easily replace them.
Yup, just something that hasn't been done yet.
Jeff
-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
User-mode-linux-devel mailing list
User-mode-linux-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken
@ 2005-08-30 21:28 ` Jeff Dike
0 siblings, 0 replies; 6+ messages in thread
From: Jeff Dike @ 2005-08-30 21:28 UTC (permalink / raw)
To: Andi Kleen; +Cc: linux-kernel, user-mode-linux-devel, Al Viro
On Tue, Aug 30, 2005 at 10:42:15PM +0200, Andi Kleen wrote:
> The generic one should work too, it's just less efficient.
> So you can probably easily replace them.
Yup, just something that hasn't been done yet.
Jeff
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2005-08-30 22:22 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-08-29 20:07 [uml-devel] [PATCH 4/9] UML - Mark SMP on UML/x86_64 as broken Jeff Dike
2005-08-29 20:07 ` Jeff Dike
2005-08-30 20:42 ` [uml-devel] " Andi Kleen
2005-08-30 20:42 ` Andi Kleen
2005-08-30 21:28 ` [uml-devel] " Jeff Dike
2005-08-30 21:28 ` Jeff Dike
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.