Thread (5 messages) flat view 5 messages, 3 authors, 2008-12-04

Re: [PATCH 2/3] powerpc: Split mmu_context handling

From: Stephen Rothwell <hidden>
Date: 2008-12-04 06:17:40

Hi Ben,

On Thu, 04 Dec 2008 16:41:48 +1100 Benjamin Herrenschmidt [off-list ref] wrote:
quoted hunk ↗ jump to hunk
--- linux-work.orig/arch/powerpc/include/asm/bug.h	2008-12-03 13:50:01.000000000 +1100
+++ linux-work/arch/powerpc/include/asm/bug.h	2008-12-03 13:50:18.000000000 +1100
@@ -26,7 +26,7 @@
 	 .previous
 .endm
 #else
- .macro EMIT_BUG_ENTRY addr,file,line,flags
+.macro EMIT_BUG_ENTRY addr,file,line,flags
Please don't make unrelated trivial fixups - they just make us look hard
to see if something really changed.
-/*
- * switch_mm is the entry point called from the architecture independent
+/* switch_mm is the entry point called from the architecture independent
The same goes for just changing the comment style away from the usual ...
-/*
- * After we have set current->mm to a new value, this activates
+/* After we have set current->mm to a new value, this activates
Ditto
quoted hunk ↗ jump to hunk
--- linux-work.orig/arch/powerpc/kernel/head_32.S	2008-12-03 13:50:01.000000000 +1100
+++ linux-work/arch/powerpc/kernel/head_32.S	2008-12-03 13:50:18.000000000 +1100
@@ -1050,7 +1051,7 @@ start_here:
 	 * We do this here because we know the mmu is disabled, and
 	 * will be enabled for real in just a few instructions.
 	 */
-	lis	r5, abatron_pteptrs@h
+	lis	r5, abatron_pteptrs@h\
                                     ^
Is this right?
quoted hunk ↗ jump to hunk
+++ linux-work/arch/powerpc/mm/mmu_context_hash32.c	2008-12-03 13:50:18.000000000 +1100
+/*
+ * This function defines the mapping from contexts to VSIDs (virtual
+ * segment IDs).  We use a skew on both the context and the high 4 bits
+ * of the 32-bit virtual address (the "effective segment ID") in order
+ * to spread out the entries in the MMU hash table.  Note, if this
+ * function is changed then arch/ppc/mm/hashtable.S will have to be
+ * changed to correspond.
+ */
+#define CTX_TO_VSID(ctx, va)	(((ctx) * (897 * 16) + ((va) >> 28) * 0x111) \
+				 & 0xffffff)
Any reason this is not a static inline function?  Hmmm, actually it
doesn't look like it is used anywhere ... I guess its just for
documentation?
quoted hunk ↗ jump to hunk
+++ linux-work/arch/powerpc/mm/mmu_context_nohash.c	2008-12-03 13:50:18.000000000 +1100
+#ifdef CONFIG_8xx
+#define NO_CONTEXT      	16
+#define LAST_CONTEXT    	15
+#define FIRST_CONTEXT    	0
+
+#elif defined(CONFIG_4xx)
+#define NO_CONTEXT      	256
+#define LAST_CONTEXT    	255
+#define FIRST_CONTEXT    	1
+
+#elif defined(CONFIG_E200) || defined(CONFIG_E500)
+#define NO_CONTEXT      	256
+#define LAST_CONTEXT    	255
+#define FIRST_CONTEXT    	1
Why not combine these last two?
+static unsigned long next_mmu_context;
+	/* free up context `next_mmu_context' */
+	/* if we shouldn't free context 0, don't... */
+	if (next_mmu_context < FIRST_CONTEXT)
If FIRST_CONTEXT is 0, this will generate a compiler warning (as
next_mmu_context can't be < 0).
+config PPC_MMU_NOHASH_32
+	def_bool y
+	depends on PPC_MMU_NOHASH && PPC32
+
+config PPC_MMU_NOHASH_64
+	def_bool y
+	depends on PPC_MMU_NOHASH && PPC64
Neither of these are used anywhere.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

Attachments

Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help