Re: [PATCH -V3 11/11] arch/powerpc: Add 64TB support

["Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>]

Paul Mackerras <paulus@samba.org> writes:

> On Mon, Jul 23, 2012 at 03:52:05PM +0530, Aneesh Kumar K.V wrote:
>> Paul Mackerras <paulus@samba.org> writes:
>> 
>> > On Mon, Jul 09, 2012 at 06:43:41PM +0530, Aneesh Kumar K.V wrote:
>> >
>> >> -#define USER_ESID_BITS		16
>> >> -#define USER_ESID_BITS_1T	4
>> >> +#define USER_ESID_BITS		18
>> >> +#define USER_ESID_BITS_1T	6
>> >
>> > You also need to change the proto-VSID generation for kernel addresses
>> > when you do this.  If you don't you'll end up with some user processes
>> > using the same VSIDs as we use for the kernel addresses, meaning that
>> > those processes won't run very well...
>> >
>> 
>> Can you explain this more. right now we generate vsid as below
>> 
>> vsid_scramble(ea >> SID_SHIFT, 256M) for kernel
>> 
>> vsid_scramble((context << USER_ESID_BITS) | (ea >> SID_SHIFT), 256M);
>> for user
>> 
>> what changes are you suggesting ?
>
> Think about it.  With the current values of USER_ESID_BITS and
> CONTEXT_BITS, and the addresses we use for kernel mappings, there are
> no values of context, user_ea and kernel_ea for which
>
> kernel_ea >> SID_SHIFT == (context << USER_ESID_BITS) | (user_ea >> SID_SHIFT)
>
> If you increase USER_ESID_BITS, then there will be some context values
> for which that equation becomes true.  For example, if you increase
> USER_ESID_BITS to 18, then context 0x30000 will generate the same
> proto-VSIDs as the kernel linear mapping.  Since we can hand out
> contexts up to 0x7ffff (with CONTEXT_BITS = 19), there is a collision.
>
> In other words, the proto-VSID space (the space of values that are
> input to vsid_scramble) is currently divided into two mutually
> exclusive regions: from 0 to 2^35 - 1 for user processes, and from
> 2^35 to 2^36 - 1 for kernel addresses.  You are wanting to expand the
> amount of proto-VSID space that user processes can use, but you need
> either to move the kernel portion of the space, or to make sure that
> the context allocator doesn't hand out context values that would
> collide with the kernel portion of the space (or both).

How about this ?. There is a FIXME in there which is not related.

commit 4881230dc34cbe24141f06004371082da6b121a5
Author: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Date:   Tue Jul 24 11:40:29 2012 +0530

    proto vsid fixes

diff --git a/arch/powerpc/include/asm/mmu-hash64.h b/arch/powerpc/include/asm/mmu-hash64.h
index daa3e4b..736df15 100644
--- a/arch/powerpc/include/asm/mmu-hash64.h
+++ b/arch/powerpc/include/asm/mmu-hash64.h
@@ -516,9 +516,19 @@ typedef struct {
 /* This is only valid for addresses >= PAGE_OFFSET */
 static inline unsigned long get_kernel_vsid(unsigned long ea, int ssize)
 {
-	if (ssize == MMU_SEGSIZE_256M)
-		return vsid_scramble(ea >> SID_SHIFT, 256M);
-	return vsid_scramble(ea >> SID_SHIFT_1T, 1T);
+	unsigned long proto_vsid;
+	/*
+	 * We need to make sure proto_vsid for the kernel is
+	 * >= 2^(CONTEXT_BITS + USER_ESID_BITS[_1T])
+	 */
+	if (ssize == MMU_SEGSIZE_256M) {
+		proto_vsid = ea >> SID_SHIFT;
+		proto_vsid |= (1UL << (CONTEXT_BITS + USER_ESID_BITS));
+		return vsid_scramble(proto_vsid, 256M);
+	}
+	proto_vsid = ea >> SID_SHIFT_1T;
+	proto_vsid |= (1UL << (CONTEXT_BITS + USER_ESID_BITS_1T));
+	return vsid_scramble(proto_vsid, 1T);
 }
 
 /* Returns the segment size indicator for a user address */
@@ -534,11 +544,17 @@ static inline int user_segment_size(unsigned long addr)
 static inline unsigned long get_vsid(unsigned long context, unsigned long ea,
 				     int ssize)
 {
-	if (ssize == MMU_SEGSIZE_256M)
-		return vsid_scramble((context << USER_ESID_BITS)
-				     | (ea >> SID_SHIFT), 256M);
-	return vsid_scramble((context << USER_ESID_BITS_1T)
-			     | (ea >> SID_SHIFT_1T), 1T);
+	unsigned long proto_vsid;
+	if (ssize == MMU_SEGSIZE_256M) {
+		proto_vsid = ((context << USER_ESID_BITS) |(ea >> SID_SHIFT));
+		/* truncate this to 37 bits */
+		proto_vsid &= (1UL << (CONTEXT_BITS + USER_ESID_BITS)) - 1;
+		return vsid_scramble(proto_vsid, 256M);
+	}
+	proto_vsid = ((context << USER_ESID_BITS_1T) | (ea >> SID_SHIFT_1T));
+	/* truncate this to 25 bits */
+	proto_vsid &= (1UL << (CONTEXT_BITS + USER_ESID_BITS_1T)) - 1;
+	return vsid_scramble( proto_vsid, 1T);
 }
 
 #endif /* __ASSEMBLY__ */
diff --git a/arch/powerpc/kernel/exceptions-64s.S b/arch/powerpc/kernel/exceptions-64s.S
index 1c06d29..40ed208 100644
--- a/arch/powerpc/kernel/exceptions-64s.S
+++ b/arch/powerpc/kernel/exceptions-64s.S
@@ -958,7 +958,9 @@ _GLOBAL(do_stab_bolted)
 	rldimi	r10,r11,7,52	/* r10 = first ste of the group */
 
 	/* Calculate VSID */
-	/* This is a kernel address, so protovsid = ESID */
+	/* This is a kernel address, so protovsid = ESID | 1 << 37 */
+	li	r9,0x1
+	rldimi  r11,r9,(CONTEXT_BITS + USER_ESID_BITS),0
 	ASM_VSID_SCRAMBLE(r11, r9, 256M)
 	rldic	r9,r11,12,16	/* r9 = vsid << 12 */
 
diff --git a/arch/powerpc/mm/slb_low.S b/arch/powerpc/mm/slb_low.S
index db2cb3f..792bfa8 100644
--- a/arch/powerpc/mm/slb_low.S
+++ b/arch/powerpc/mm/slb_low.S
@@ -57,8 +57,13 @@ _GLOBAL(slb_allocate_realmode)
 _GLOBAL(slb_miss_kernel_load_linear)
 	li	r11,0
 BEGIN_FTR_SECTION
+	li	r9,0x1
+	rldimi  r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0
 	b	slb_finish_load
 END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
+	srdi	r10,r10,40-28		/* get 1T ESID */
+	li	r9,0x1
+	rldimi  r10,r9,(CONTEXT_BITS + USER_ESID_BITS_1T),0
 	b	slb_finish_load_1T
 
 1:
@@ -86,8 +91,13 @@ _GLOBAL(slb_miss_kernel_load_vmemmap)
 	li	r11,0
 6:
 BEGIN_FTR_SECTION
+	li	r9,0x1
+	rldimi  r10,r9,(CONTEXT_BITS + USER_ESID_BITS),0
 	b	slb_finish_load
 END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
+	srdi	r10,r10,40-28		/* get 1T ESID */
+	li	r9,0x1
+	rldimi  r10,r9,(CONTEXT_BITS + USER_ESID_BITS_1T),0
 	b	slb_finish_load_1T
 
 0:	/* user address: proto-VSID = context << 15 | ESID. First check
@@ -156,10 +166,13 @@ END_MMU_FTR_SECTION_IFCLR(MMU_FTR_1T_SEGMENT)
 BEGIN_FTR_SECTION
 	cmpldi	r10,0x1000
 END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT)
-	rldimi	r10,r9,USER_ESID_BITS,0
+	rldimi	r10,r9,USER_ESID_BITS,0 /* FIXME!! ? We should do it after srdi ? */
 BEGIN_FTR_SECTION
+	srdi	r10,r10,40-28		/* get 1T ESID */
+	clrldi	r10,r10,(64 - (CONTEXT_BITS + USER_ESID_BITS_1T))
 	bge	slb_finish_load_1T
 END_MMU_FTR_SECTION_IFSET(MMU_FTR_1T_SEGMENT)
+	clrldi	r10,r10,(64 - (CONTEXT_BITS + USER_ESID_BITS))
 	b	slb_finish_load
 
 8:	/* invalid EA */
@@ -292,7 +305,6 @@ _GLOBAL(slb_compare_rr_to_size)
  * r3 = EA, r10 = proto-VSID, r11 = flags, clobbers r9
  */
 slb_finish_load_1T:
-	srdi	r10,r10,40-28		/* get 1T ESID */
 	ASM_VSID_SCRAMBLE(r10,r9,1T)
 	/*
 	 * bits above VSID_BITS_1T need to be ignored from r10