Skip to contentSkip to conversation

powerpc/mm: Add translation mode information in /proc/cpuinfo

PATCH In review Aneesh Kumar K.V 4 participants 6 messages Feb 19, 2017 – Feb 27, 2017 PowerPC

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 powerpc/mm: Add translation mode information in /proc/cpuinfo

With this we have on powernv and pseries /proc/cpuinfo reporting

timebase : 512000000
platform : PowerNV
model : 8247-22L
machine : PowerNV 8247-22L
firmware : OPAL
translation : Hash

---
 arch/powerpc/platforms/powernv/setup.c | 4 ++++
 arch/powerpc/platforms/pseries/setup.c | 4 ++++
 2 files changed, 8 insertions(+)
arch/powerpc/platforms/powernv/setup.c +4 -0
index d50c7d99baaf..d38571e289bb 100644
@@ -95,6 +95,10 @@ static void pnv_show_cpuinfo(struct seq_file *m)
 	else
 		seq_printf(m, "firmware\t: BML\n");
 	of_node_put(root);
+	if (radix_enabled())
+		seq_printf(m, "translation\t: Radix\n");
+	else
+		seq_printf(m, "translation\t: Hash\n");
 }
 
 static void pnv_prepare_going_down(void)
arch/powerpc/platforms/pseries/setup.c +4 -0
index 7736352f7279..6576fe306561 100644
@@ -86,6 +86,10 @@ static void pSeries_show_cpuinfo(struct seq_file *m)
 		model = of_get_property(root, "model", NULL);
 	seq_printf(m, "machine\t\t: CHRP %s\n", model);
 	of_node_put(root);
+	if (radix_enabled())
+		seq_printf(m, "translation\t: Radix\n");
+	else
+		seq_printf(m, "translation\t: Hash\n");
 }
 
 /* Initialize firmware assisted non-maskable interrupts if
Signature · 1 line
2.7.4
#2 lore raw
In reply to Aneesh Kumar K.V
Quoted 20 lines · Aneesh Kumar K.V

With this we have on powernv and pseries /proc/cpuinfo reporting

timebase : 512000000
platform : PowerNV
model : 8247-22L
machine : PowerNV 8247-22L
firmware : OPAL
translation : Hash

---
 arch/powerpc/platforms/powernv/setup.c | 4 ++++
 arch/powerpc/platforms/pseries/setup.c | 4 ++++
 2 files changed, 8 insertions(+)
arch/powerpc/platforms/powernv/setup.c +0 -0
index d50c7d99baaf..d38571e289bb 100644

arch/powerpc/platforms/powernv/setup.c static void pnv_show_cpuinfo(struct seq_file *m) · line 102

Quoted 8 lines
arch/powerpc/platforms/powernv/setup.c +4 -0
@@ -95,6 +95,10 @@ static void pnv_show_cpuinfo(struct seq_file *m)
 	else
 		seq_printf(m, "firmware\t: BML\n");
 	of_node_put(root);
+	if (radix_enabled())
+		seq_printf(m, "translation\t: Radix\n");
+	else
+		seq_printf(m, "translation\t: Hash\n");
 }

Can we just call it "MMU" ?

I don't think it's entirely clear what "translation" means here if you don't already know.

cheers

#3 lore raw
Quoted 32 lines · Michael Ellerman
Quoted 20 lines · Aneesh Kumar K.V

With this we have on powernv and pseries /proc/cpuinfo reporting

timebase : 512000000
platform : PowerNV
model : 8247-22L
machine : PowerNV 8247-22L
firmware : OPAL
translation : Hash

---
  arch/powerpc/platforms/powernv/setup.c | 4 ++++
  arch/powerpc/platforms/pseries/setup.c | 4 ++++
  2 files changed, 8 insertions(+)
arch/powerpc/platforms/powernv/setup.c +0 -0
index d50c7d99baaf..d38571e289bb 100644

arch/powerpc/platforms/powernv/setup.c static void pnv_show_cpuinfo(struct seq_file *m) · line 102

Quoted 8 lines
arch/powerpc/platforms/powernv/setup.c +4 -0
@@ -95,6 +95,10 @@ static void pnv_show_cpuinfo(struct seq_file *m)
  	else
  		seq_printf(m, "firmware\t: BML\n");
  	of_node_put(root);
+	if (radix_enabled())
+		seq_printf(m, "translation\t: Radix\n");
+	else
+		seq_printf(m, "translation\t: Hash\n");
  }

Can we just call it "MMU" ?

Quoted 5 lines

I don't think it's entirely clear what "translation" means here if you don't already know.

cheers

I avoided using MMU, because it will confuse hardware guys. Radix is not clearly the full definition of Memory management unit, but rather the translation mode used by memory management unit. But i don't have strong opinion on this.

Do you want me to send an updated patch ? or you can update it when you apply it to your tree ?

-aneesh

#4 lore raw
Quoted 1 line · Aneesh Kumar K.V

I avoided using MMU, because it will confuse hardware guys.

Why would it ? I don't see how anybody would be confused. And they aren't the primary consumers of /proc/cpuinfo anyway. I agree with Michal here.

Quoted 5 lines

Radix is not
clearly the full definition of
Memory management unit, but rather the translation mode used by memory management unit. But
i don't have strong opinion on this.

Who cares ? People understand :-) If you want call it "MMU Mode: if you think it's cleared but I don't like too much having a space on the left in case of stupid parsers...

Quoted 4 lines

Do you want me to send an updated patch ? or you can update it when you apply it to your tree ?

-aneesh

#5 lore raw
In reply to Aneesh Kumar K.V
Quoted 16 lines · Aneesh Kumar K.V
Quoted 15 lines · Michael Ellerman

arch/powerpc/platforms/powernv/setup.c static void pnv_show_cpuinfo(struct seq_file *m) · line 102

Quoted 13 lines · Aneesh Kumar K.V
arch/powerpc/platforms/powernv/setup.c +4 -0
index d50c7d99baaf..d38571e289bb 100644
@@ -95,6 +95,10 @@ static void pnv_show_cpuinfo(struct seq_file *m)
  	else
  		seq_printf(m, "firmware\t: BML\n");
  	of_node_put(root);
+	if (radix_enabled())
+		seq_printf(m, "translation\t: Radix\n");
+	else
+		seq_printf(m, "translation\t: Hash\n");
  }

Can we just call it "MMU" ?

...

Quoted 3 lines

Do you want me to send an updated patch ? or you can update it when you apply it to your tree ?

I'll fix it up.

cheers