[PATCH]: load_mmu for SMP systems

[PATCH]: load_mmu for SMP systems

[Kip Walker]

[-- Attachment #1: Type: text/plain, Size: 481 bytes --]

In SMP systems, each CPU needs to set up "current_cpu_data.tlbsize". 
Some CPUs do this initialization in cpu_probe, which is called both by
init_arch and start_secondary.  However, some CPUs do this in their TLB
setup code, which is called via load_mmu.  The SMP boot code doesn't
currently call load_mmu() for the secondary CPUs.  Here's a simple fix
for the 2.4 tree.

TLB flush routines that have loops running up to tlbsize will lose if
it's not set properly on all CPUs!

Kip

[-- Attachment #2: loadmmu.diff --]
[-- Type: text/plain, Size: 936 bytes --]

Index: arch/mips/kernel/smp.c
===================================================================
RCS file: /home/cvs/linux/arch/mips/kernel/smp.c,v
retrieving revision 1.10.2.24
diff -u -r1.10.2.24 smp.c
--- arch/mips/kernel/smp.c	23 Apr 2003 20:10:49 -0000	1.10.2.24
+++ arch/mips/kernel/smp.c	25 Apr 2003 18:19:55 -0000
@@ -99,6 +99,7 @@
 	unsigned int cpu = smp_processor_id();
 
 	cpu_probe();
+	load_mmu();
 	prom_init_secondary();
 	per_cpu_trap_init();
 
Index: arch/mips64/kernel/smp.c
===================================================================
RCS file: /home/cvs/linux/arch/mips64/kernel/smp.c,v
retrieving revision 1.26.2.22
diff -u -r1.26.2.22 smp.c
--- arch/mips64/kernel/smp.c	23 Apr 2003 20:10:49 -0000	1.26.2.22
+++ arch/mips64/kernel/smp.c	25 Apr 2003 18:19:55 -0000
@@ -103,6 +103,7 @@
 	unsigned int cpu = smp_processor_id();
 
 	cpu_probe();
+	load_mmu();
 	prom_init_secondary();
 	per_cpu_trap_init();
 

Re: [PATCH]: load_mmu for SMP systems

[Kip Walker]

[-- Attachment #1: Type: text/plain, Size: 568 bytes --]


OK, I'll add some declarations make it tidier.

Kip

Kip Walker wrote:
> 
> In SMP systems, each CPU needs to set up "current_cpu_data.tlbsize".
> Some CPUs do this initialization in cpu_probe, which is called both by
> init_arch and start_secondary.  However, some CPUs do this in their TLB
> setup code, which is called via load_mmu.  The SMP boot code doesn't
> currently call load_mmu() for the secondary CPUs.  Here's a simple fix
> for the 2.4 tree.
> 
> TLB flush routines that have loops running up to tlbsize will lose if
> it's not set properly on all CPUs!

[-- Attachment #2: loadmmu.diff --]
[-- Type: text/plain, Size: 1507 bytes --]

Index: arch/mips/kernel/smp.c
===================================================================
RCS file: /home/cvs/linux/arch/mips/kernel/smp.c,v
retrieving revision 1.10.2.24
diff -u -r1.10.2.24 smp.c
--- arch/mips/kernel/smp.c	23 Apr 2003 20:10:49 -0000	1.10.2.24
+++ arch/mips/kernel/smp.c	25 Apr 2003 18:27:05 -0000
@@ -39,6 +39,9 @@
 #include <asm/mmu_context.h>
 #include <asm/smp.h>
 
+extern void load_mmu(void);
+extern void per_cpu_trap_init(void);
+
 /* The 'big kernel lock' */
 spinlock_t kernel_flag __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
 int smp_threads_ready;	/* Not used */
@@ -99,6 +102,7 @@
 	unsigned int cpu = smp_processor_id();
 
 	cpu_probe();
+	load_mmu();
 	prom_init_secondary();
 	per_cpu_trap_init();
 
Index: arch/mips64/kernel/smp.c
===================================================================
RCS file: /home/cvs/linux/arch/mips64/kernel/smp.c,v
retrieving revision 1.26.2.22
diff -u -r1.26.2.22 smp.c
--- arch/mips64/kernel/smp.c	23 Apr 2003 20:10:49 -0000	1.26.2.22
+++ arch/mips64/kernel/smp.c	25 Apr 2003 18:27:05 -0000
@@ -39,6 +39,9 @@
 #include <asm/mmu_context.h>
 #include <asm/smp.h>
 
+extern void load_mmu(void);
+extern void per_cpu_trap_init(void);      
+
 /* The 'big kernel lock' */
 spinlock_t kernel_flag __cacheline_aligned_in_smp = SPIN_LOCK_UNLOCKED;
 int smp_threads_ready;	/* Not used */
@@ -103,6 +106,7 @@
 	unsigned int cpu = smp_processor_id();
 
 	cpu_probe();
+	load_mmu();
 	prom_init_secondary();
 	per_cpu_trap_init();
 

Re: [PATCH]: load_mmu for SMP systems

[Ralf Baechle]

On Fri, Apr 25, 2003 at 11:24:20AM -0700, Kip Walker wrote:

> In SMP systems, each CPU needs to set up "current_cpu_data.tlbsize". 
> Some CPUs do this initialization in cpu_probe, which is called both by
> init_arch and start_secondary.  However, some CPUs do this in their TLB
> setup code, which is called via load_mmu.  The SMP boot code doesn't
> currently call load_mmu() for the secondary CPUs.  Here's a simple fix
> for the 2.4 tree.

I instead changed cpu-probe to set tlbsize properly.  Nothing wrong with
your patch, it just fits better into my Grand Plan (TM) :-)

> TLB flush routines that have loops running up to tlbsize will lose if
> it's not set properly on all CPUs!

Yeah, they're going to be sort of slow.  There must be a reason for all
those GHz processors ;-)

  Ralf

Re: [PATCH]: load_mmu for SMP systems

[Kip Walker]

Ralf Baechle wrote:
> 
> > TLB flush routines that have loops running up to tlbsize will lose if
> > it's not set properly on all CPUs!
> 
> Yeah, they're going to be sort of slow.  There must be a reason for all
> those GHz processors ;-)

Um, it was worse than that if (for example) a complete TLB flush has a
"for (i=0; i<0; i++)" loop around it.  My board was experiencing
occasional userland segfaults thanks to bogus TLB flushing.

Kip