From 06ef66081620e94fe35a518f98624b83a140096e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:10:45 -0700
Subject: [IA64] __ia64_syscall() is no longer used anywhere in the kernel. 
 Remove it.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S      | 12 ------------
 arch/ia64/kernel/ia64_ksyms.c |  3 ---
 2 files changed, 15 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index bd86fea..5b469db 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -470,18 +470,6 @@ ENTRY(load_switch_stack)
 	br.cond.sptk.many b7
 END(load_switch_stack)
 
-GLOBAL_ENTRY(__ia64_syscall)
-	.regstk 6,0,0,0
-	mov r15=in5				// put syscall number in place
-	break __BREAK_SYSCALL
-	movl r2=errno
-	cmp.eq p6,p7=-1,r10
-	;;
-(p6)	st4 [r2]=r8
-(p6)	mov r8=-1
-	br.ret.sptk.many rp
-END(__ia64_syscall)
-
 GLOBAL_ENTRY(execve)
 	mov r15=__NR_execve			// put syscall number in place
 	break __BREAK_SYSCALL
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index 7bbf019..0157281 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -58,9 +58,6 @@ EXPORT_SYMBOL(__strlen_user);
 EXPORT_SYMBOL(__strncpy_from_user);
 EXPORT_SYMBOL(__strnlen_user);
 
-#include <asm/unistd.h>
-EXPORT_SYMBOL(__ia64_syscall);
-
 /* from arch/ia64/lib */
 extern void __divsi3(void);
 extern void __udivsi3(void);
-- 
cgit v1.1


From 9ec1a7ad434fa951ff845dbee3397cf6ad4f76df Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:13:33 -0700
Subject: [IA64] Use dynamic prediction for RSE-clearing branches.

This by itself is good for a 1-2 cycle speed up.  Effect is bigger
when combined with the later patches.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 5b469db..c0f2833 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1012,14 +1012,14 @@ rse_clear_invalid:
 	mov loc5=0
 	mov loc6=0
 	mov loc7=0
-(pRecurse) br.call.sptk.few b0=rse_clear_invalid
+(pRecurse) br.call.dptk.few b0=rse_clear_invalid
 	;;
 	mov loc8=0
 	mov loc9=0
 	cmp.ne pReturn,p0=r0,in1	// if recursion count != 0, we need to do a br.ret
 	mov loc10=0
 	mov loc11=0
-(pReturn) br.ret.sptk.many b0
+(pReturn) br.ret.dptk.many b0
 #endif /* !CONFIG_ITANIUM */
 #	undef pRecurse
 #	undef pReturn
-- 
cgit v1.1


From 3c79c8b1d92a9ae3edf3cbcd2c43c553ee0f1d83 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:15:13 -0700
Subject: [IA64] Schedule fp-clearing insns at least 6 cycles after reading
 ar.bsp.

Decreases syscall overhead by approximately 6 cycles.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index c0f2833..0c84bed 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -705,15 +705,15 @@ ENTRY(ia64_leave_syscall)
 	// start restoring the state saved on the kernel stack (struct pt_regs):
 	ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
 	ld8 r11=[r3],PT(CR_IIP)-PT(R11)
-	mov f6=f0		// clear f6
+	nop.i 0
 	;;
 	invala			// M0|1 invalidate ALAT
 	rsm psr.i | psr.ic	// M2 initiate turning off of interrupt and interruption collection
-	mov f9=f0		// clear f9
+	nop.i 0
 
 	ld8 r29=[r2],16		// load cr.ipsr
 	ld8 r28=[r3],16			// load cr.iip
-	mov f8=f0		// clear f8
+	mov r22=r0		// clear r22
 	;;
 	ld8 r30=[r2],16		// M0|1 load cr.ifs
 	ld8 r25=[r3],16		// M0|1 load ar.unat
@@ -721,15 +721,15 @@ ENTRY(ia64_leave_syscall)
 	;;
 	ld8 r26=[r2],PT(B0)-PT(AR_PFS)	// M0|1 load ar.pfs
 (pKStk)	mov r22=psr		// M2 read PSR now that interrupts are disabled
-	mov f10=f0		// clear f10
+(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
 	;;
 	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
 	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// load ar.rsc
-	mov f11=f0		// clear f11
+	mov f6=f0		// clear f6
 	;;
 	ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT)	// load ar.rnat (may be garbage)
 	ld8 r31=[r3],PT(R1)-PT(PR)		// load predicates
-(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
+	mov f7=f0		// clear f7
 	;;
 	ld8 r20=[r2],PT(R12)-PT(AR_FPSR)	// load ar.fpsr
 	ld8.fill r1=[r3],16	// load r1
@@ -737,24 +737,29 @@ ENTRY(ia64_leave_syscall)
 	;;
 	srlz.d			// M0  ensure interruption collection is off
 	ld8.fill r13=[r3],16
-	mov f7=f0		// clear f7
+	mov f8=f0		// clear f8
 	;;
 	ld8.fill r12=[r2]	// restore r12 (sp)
 	mov.m ar.ssd=r0		// M2 clear ar.ssd
-	mov r22=r0		// clear r22
+	mov b6=r18		// I0  restore b6
 
+	nop.m 0
+	mov f9=f0		// clear f9
+	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
+	;;
 	ld8.fill r15=[r3]	// restore r15
 (pUStk) st1 [r14]=r17
 	addl r3=THIS_CPU(ia64_phys_stacked_size_p8),r0
 	;;
 (pUStk)	ld4 r17=[r3]		// r17 = cpu_data->phys_stacked_size_p8
 	mov.m ar.csd=r0		// M2 clear ar.csd
-	mov b6=r18		// I0  restore b6
+	mov f10=f0		// clear f10
 	;;
 	mov r14=r0		// clear r14
-	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
+	mov f11=f0		// clear f11
 (pKStk) br.cond.dpnt.many skip_rbs_switch
 
+
 	mov.m ar.ccv=r0		// clear ar.ccv
 (pNonSys) br.cond.dpnt.many dont_preserve_current_frame
 	br.cond.sptk.many rbs_switch
-- 
cgit v1.1


From 96e017495e6833adcbad84783e3c5eb685836bbf Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:16:07 -0700
Subject: [IA64] On return from syscall, hint b7 with
 __kernel_syscall_via_epc().

Why is this a good idea?  Clearing b7 to 0 is guaranteed to do us no
good and writing it with __kernel_syscall_via_epc() yields a 6 cycle
improvement _if_ the application performs another EPC-based system-
call without overwriting b7, which is not all that uncommon.  Well
worth the minimal cost of 1 bundle of code.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 0c84bed..6359d7f 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -692,7 +692,7 @@ ENTRY(ia64_leave_syscall)
 	;;
 (p6)	ld4 r31=[r18]				// load current_thread_info()->flags
 	ld8 r19=[r2],PT(B6)-PT(LOADRS)		// load ar.rsc value for "loadrs"
-	mov b7=r0		// clear b7
+	nop.i 0
 	;;
 	ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)	// load ar.bspstore (may be garbage)
 	ld8 r18=[r2],PT(R9)-PT(B6)		// load b6
@@ -754,7 +754,14 @@ ENTRY(ia64_leave_syscall)
 (pUStk)	ld4 r17=[r3]		// r17 = cpu_data->phys_stacked_size_p8
 	mov.m ar.csd=r0		// M2 clear ar.csd
 	mov f10=f0		// clear f10
+
+	nop.m 0
+	movl r14=__kernel_syscall_via_epc	// X
 	;;
+	nop.m 0
+	nop.m 0
+	mov b7=r14		// I0   clear b7 (hint with __kernel_syscall_via_epc)
+
 	mov r14=r0		// clear r14
 	mov f11=f0		// clear f11
 (pKStk) br.cond.dpnt.many skip_rbs_switch
-- 
cgit v1.1


From 060561ff79b01eea58e6d72abfb8e7580ff21f2a Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:17:03 -0700
Subject: [IA64] In syscall-entry, use st8 instead of stf8 to clear pt_regs.r8

Using stf8 seemed like a clever idea at the time, but stf8 forces
the cache-line to be invalidated in the L1D (if it happens to be
there already).  This patch eliminates a guaranteed L1D cache-miss
and, by itself, is good for a 1-2 cycle improvement for heavy-weight
syscalls.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index d9c05d5..dc78c35 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -1,7 +1,7 @@
 /*
  * arch/ia64/kernel/ivt.S
  *
- * Copyright (C) 1998-2001, 2003 Hewlett-Packard Co
+ * Copyright (C) 1998-2001, 2003, 2005 Hewlett-Packard Co
  *	Stephane Eranian <eranian@hpl.hp.com>
  *	David Mosberger <davidm@hpl.hp.com>
  * Copyright (C) 2000, 2002-2003 Intel Co
@@ -918,7 +918,7 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 	tnat.nz p14,p0=in6
 	cmp.lt p10,p9=r11,r8	// frame size can't be more than local+8
 	;;
-	stf8 [r16]=f1		// ensure pt_regs.r8 != 0 (see handle_syscall_error)
+	mov r8=1
 (p9)	tnat.nz p10,p0=r15
 	adds r12=-16,r1		// switch to kernel memory stack (with 16 bytes of scratch)
 
@@ -929,9 +929,9 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 	mov r13=r2				// establish `current'
 	movl r1=__gp				// establish kernel global pointer
 	;;
+	st8 [r16]=r8		// ensure pt_regs.r8 != 0 (see handle_syscall_error)
 (p14)	mov in6=-1
 (p8)	mov in7=-1
-	nop.i 0
 
 	cmp.eq pSys,pNonSys=r0,r0		// set pSys=1, pNonSys=0
 	movl r17=FPSR_DEFAULT
-- 
cgit v1.1


From 87e522a0f7f8a7a5a1a880517989835c4f09c63e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:17:44 -0700
Subject: [IA64] Schedule ia64_leave_syscall() to read ar.bsp earlier

Reschedule code to read ar.bsp as early as possible.  To enable this,
don't bother clearing some of the registers when we're returning to
kernel stacks.  Also, instead of trying to support the pNonSys case
(which makes no sense), do a bugcheck instead (with break 0).  Finally,
remove a clear of r14 which is a left-over from the previous patch.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 43 +++++++++++++++++++------------------------
 1 file changed, 19 insertions(+), 24 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 6359d7f..d97a07c 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -694,22 +694,22 @@ ENTRY(ia64_leave_syscall)
 	ld8 r19=[r2],PT(B6)-PT(LOADRS)		// load ar.rsc value for "loadrs"
 	nop.i 0
 	;;
-	ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)	// load ar.bspstore (may be garbage)
+	mov r16=ar.bsp				// M2  get existing backing store pointer
 	ld8 r18=[r2],PT(R9)-PT(B6)		// load b6
 (p6)	and r15=TIF_WORK_MASK,r31		// any work other than TIF_SYSCALL_TRACE?
 	;;
-	mov r16=ar.bsp				// M2  get existing backing store pointer
+	ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)	// load ar.bspstore (may be garbage)
 (p6)	cmp4.ne.unc p6,p0=r15, r0		// any special work pending?
 (p6)	br.cond.spnt .work_pending_syscall
 	;;
 	// start restoring the state saved on the kernel stack (struct pt_regs):
 	ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
 	ld8 r11=[r3],PT(CR_IIP)-PT(R11)
-	nop.i 0
+(pNonSys) break 0		//      bug check: we shouldn't be here if pNonSys is TRUE!
 	;;
 	invala			// M0|1 invalidate ALAT
 	rsm psr.i | psr.ic	// M2 initiate turning off of interrupt and interruption collection
-	nop.i 0
+	cmp.eq p9,p0=r0,r0	// set p9 to indicate that we should restore cr.ifs
 
 	ld8 r29=[r2],16		// load cr.ipsr
 	ld8 r28=[r3],16			// load cr.iip
@@ -717,11 +717,11 @@ ENTRY(ia64_leave_syscall)
 	;;
 	ld8 r30=[r2],16		// M0|1 load cr.ifs
 	ld8 r25=[r3],16		// M0|1 load ar.unat
-	cmp.eq p9,p0=r0,r0	// set p9 to indicate that we should restore cr.ifs
+(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
 	;;
 	ld8 r26=[r2],PT(B0)-PT(AR_PFS)	// M0|1 load ar.pfs
 (pKStk)	mov r22=psr		// M2 read PSR now that interrupts are disabled
-(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
+	nop 0
 	;;
 	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
 	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// load ar.rsc
@@ -735,40 +735,35 @@ ENTRY(ia64_leave_syscall)
 	ld8.fill r1=[r3],16	// load r1
 (pUStk) mov r17=1
 	;;
-	srlz.d			// M0  ensure interruption collection is off
+(pUStk) st1 [r14]=r17
 	ld8.fill r13=[r3],16
 	mov f8=f0		// clear f8
 	;;
 	ld8.fill r12=[r2]	// restore r12 (sp)
-	mov.m ar.ssd=r0		// M2 clear ar.ssd
+	ld8.fill r15=[r3]	// restore r15
 	mov b6=r18		// I0  restore b6
 
-	nop.m 0
+	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0
 	mov f9=f0		// clear f9
+(pKStk) br.cond.dpnt.many skip_rbs_switch
+
+	srlz.d			// M0  ensure interruption collection is off
 	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
+	cover			// B    add current frame into dirty partition and set cr.ifs
 	;;
-	ld8.fill r15=[r3]	// restore r15
-(pUStk) st1 [r14]=r17
-	addl r3=THIS_CPU(ia64_phys_stacked_size_p8),r0
-	;;
-(pUStk)	ld4 r17=[r3]		// r17 = cpu_data->phys_stacked_size_p8
-	mov.m ar.csd=r0		// M2 clear ar.csd
+(pUStk)	ld4 r17=[r17]		// r17 = cpu_data->phys_stacked_size_p8
+	mov r19=ar.bsp		// M2   get new backing store pointer
 	mov f10=f0		// clear f10
 
 	nop.m 0
 	movl r14=__kernel_syscall_via_epc	// X
 	;;
-	nop.m 0
-	nop.m 0
+	mov.m ar.csd=r0		// M2 clear ar.csd
+	mov.m ar.ccv=r0		// clear ar.ccv
 	mov b7=r14		// I0   clear b7 (hint with __kernel_syscall_via_epc)
 
-	mov r14=r0		// clear r14
+	mov.m ar.ssd=r0		// M2 clear ar.ssd
 	mov f11=f0		// clear f11
-(pKStk) br.cond.dpnt.many skip_rbs_switch
-
-
-	mov.m ar.ccv=r0		// clear ar.ccv
-(pNonSys) br.cond.dpnt.many dont_preserve_current_frame
 	br.cond.sptk.many rbs_switch
 END(ia64_leave_syscall)
 
@@ -946,10 +941,10 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	 */
 (pNonSys) br.cond.dpnt dont_preserve_current_frame
 
-rbs_switch:
 	cover				// add current frame into dirty partition and set cr.ifs
 	;;
 	mov r19=ar.bsp			// get new backing store pointer
+rbs_switch:
 	sub r16=r16,r18			// krbs = old bsp - size of dirty partition
 	cmp.ne p9,p0=r0,r0		// clear p9 to skip restore of cr.ifs
 	;;
-- 
cgit v1.1


From c03f058fbf685f2ff630095d2c1e98d331b81e82 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:18:22 -0700
Subject: [IA64] In ia64_leave_syscall(), fix comments and whitespace only.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 79 ++++++++++++++++++++++++------------------------
 1 file changed, 39 insertions(+), 40 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index d97a07c..a7542c3 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -625,7 +625,7 @@ END(ia64_ret_from_syscall)
  *	      r8-r11: restored (syscall return value(s))
  *		 r12: restored (user-level stack pointer)
  *		 r13: restored (user-level thread pointer)
- *		 r14: cleared
+ *		 r14: set to __kernel_syscall_via_epc
  *		 r15: restored (syscall #)
  *	     r16-r17: cleared
  *		 r18: user-level b6
@@ -646,7 +646,7 @@ END(ia64_ret_from_syscall)
  *		  pr: restored (user-level pr)
  *		  b0: restored (user-level rp)
  *	          b6: restored
- *		  b7: cleared
+ *		  b7: set to __kernel_syscall_via_epc
  *	     ar.unat: restored (user-level ar.unat)
  *	      ar.pfs: restored (user-level ar.pfs)
  *	      ar.rsc: restored (user-level ar.rsc)
@@ -708,63 +708,63 @@ ENTRY(ia64_leave_syscall)
 (pNonSys) break 0		//      bug check: we shouldn't be here if pNonSys is TRUE!
 	;;
 	invala			// M0|1 invalidate ALAT
-	rsm psr.i | psr.ic	// M2 initiate turning off of interrupt and interruption collection
-	cmp.eq p9,p0=r0,r0	// set p9 to indicate that we should restore cr.ifs
+	rsm psr.i | psr.ic	// M2   turn off interrupts and interruption collection
+	cmp.eq p9,p0=r0,r0	// A    set p9 to indicate that we should restore cr.ifs
 
-	ld8 r29=[r2],16		// load cr.ipsr
-	ld8 r28=[r3],16			// load cr.iip
-	mov r22=r0		// clear r22
+	ld8 r29=[r2],16		// M0|1 load cr.ipsr
+	ld8 r28=[r3],16		// M0|1 load cr.iip
+	mov r22=r0		// A    clear r22
 	;;
 	ld8 r30=[r2],16		// M0|1 load cr.ifs
 	ld8 r25=[r3],16		// M0|1 load ar.unat
 (pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
 	;;
 	ld8 r26=[r2],PT(B0)-PT(AR_PFS)	// M0|1 load ar.pfs
-(pKStk)	mov r22=psr		// M2 read PSR now that interrupts are disabled
+(pKStk)	mov r22=psr			// M2   read PSR now that interrupts are disabled
 	nop 0
 	;;
-	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // load b0
-	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// load ar.rsc
-	mov f6=f0		// clear f6
+	ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // M0|1 load b0
+	ld8 r27=[r3],PT(PR)-PT(AR_RSC)	// M0|1 load ar.rsc
+	mov f6=f0			// F    clear f6
 	;;
-	ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT)	// load ar.rnat (may be garbage)
-	ld8 r31=[r3],PT(R1)-PT(PR)		// load predicates
-	mov f7=f0		// clear f7
+	ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT)	// M0|1 load ar.rnat (may be garbage)
+	ld8 r31=[r3],PT(R1)-PT(PR)		// M0|1 load predicates
+	mov f7=f0				// F    clear f7
 	;;
-	ld8 r20=[r2],PT(R12)-PT(AR_FPSR)	// load ar.fpsr
-	ld8.fill r1=[r3],16	// load r1
-(pUStk) mov r17=1
+	ld8 r20=[r2],PT(R12)-PT(AR_FPSR)	// M0|1 load ar.fpsr
+	ld8.fill r1=[r3],16			// M0|1 load r1
+(pUStk) mov r17=1				// A
 	;;
-(pUStk) st1 [r14]=r17
-	ld8.fill r13=[r3],16
-	mov f8=f0		// clear f8
+(pUStk) st1 [r14]=r17				// M2|3
+	ld8.fill r13=[r3],16			// M0|1
+	mov f8=f0				// F    clear f8
 	;;
-	ld8.fill r12=[r2]	// restore r12 (sp)
-	ld8.fill r15=[r3]	// restore r15
-	mov b6=r18		// I0  restore b6
+	ld8.fill r12=[r2]			// M0|1 restore r12 (sp)
+	ld8.fill r15=[r3]			// M0|1 restore r15
+	mov b6=r18				// I0   restore b6
 
-	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0
-	mov f9=f0		// clear f9
-(pKStk) br.cond.dpnt.many skip_rbs_switch
+	addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0 // A
+	mov f9=f0					// F    clear f9
+(pKStk) br.cond.dpnt.many skip_rbs_switch		// B
 
-	srlz.d			// M0  ensure interruption collection is off
-	shr.u r18=r19,16	// I0|1 get byte size of existing "dirty" partition
-	cover			// B    add current frame into dirty partition and set cr.ifs
+	srlz.d				// M0   ensure interruption collection is off (for cover)
+	shr.u r18=r19,16		// I0|1 get byte size of existing "dirty" partition
+	cover				// B    add current frame into dirty partition & set cr.ifs
 	;;
-(pUStk)	ld4 r17=[r17]		// r17 = cpu_data->phys_stacked_size_p8
-	mov r19=ar.bsp		// M2   get new backing store pointer
-	mov f10=f0		// clear f10
+(pUStk) ld4 r17=[r17]			// M0|1 r17 = cpu_data->phys_stacked_size_p8
+	mov r19=ar.bsp			// M2   get new backing store pointer
+	mov f10=f0			// F    clear f10
 
 	nop.m 0
-	movl r14=__kernel_syscall_via_epc	// X
+	movl r14=__kernel_syscall_via_epc // X
 	;;
-	mov.m ar.csd=r0		// M2 clear ar.csd
-	mov.m ar.ccv=r0		// clear ar.ccv
-	mov b7=r14		// I0   clear b7 (hint with __kernel_syscall_via_epc)
+	mov.m ar.csd=r0			// M2   clear ar.csd
+	mov.m ar.ccv=r0			// M2   clear ar.ccv
+	mov b7=r14			// I0   clear b7 (hint with __kernel_syscall_via_epc)
 
-	mov.m ar.ssd=r0		// M2 clear ar.ssd
-	mov f11=f0		// clear f11
-	br.cond.sptk.many rbs_switch
+	mov.m ar.ssd=r0			// M2   clear ar.ssd
+	mov f11=f0			// F    clear f11
+	br.cond.sptk.many rbs_switch	// B
 END(ia64_leave_syscall)
 
 #ifdef CONFIG_IA32_SUPPORT
@@ -940,7 +940,6 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	 * NOTE: alloc, loadrs, and cover can't be predicated.
 	 */
 (pNonSys) br.cond.dpnt dont_preserve_current_frame
-
 	cover				// add current frame into dirty partition and set cr.ifs
 	;;
 	mov r19=ar.bsp			// get new backing store pointer
-- 
cgit v1.1


From f8fa5448fc9b4a7806b1297a0b57808f12fe4d43 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:19:04 -0700
Subject: [IA64] Reschedule break_fault() for better performance.

This patch reorganizes break_fault() to optimistically assume that a
system-call is being performed from user-space (which is almost always
the case).  If it turns out that (a) we're not being called due to a
system call or (b) we're being called from within the kernel, we fixup
the no-longer-valid assumptions in non_syscall() and .break_fixup(),
respectively.

With this approach, there are 3 major phases:

 - Phase 1: Read various control & application registers, in
	    particular the current task pointer from AR.K6.
 - Phase 2: Do all memory loads (load system-call entry,
	    load current_thread_info()->flags, prefetch
	    kernel register-backing store) and switch
	    to kernel register-stack.
 - Phase 3: Call ia64_syscall_setup() and invoke
	    syscall-handler.

Good for 26-30 cycles of improvement on break-based syscall-path.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 173 ++++++++++++++++++++++++++++++-------------------
 1 file changed, 106 insertions(+), 67 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index dc78c35..386087e 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -687,82 +687,118 @@ ENTRY(break_fault)
 	 * to prevent leaking bits from kernel to user level.
 	 */
 	DBG_FAULT(11)
-	mov r16=IA64_KR(CURRENT)		// r16 = current task; 12 cycle read lat.
-	mov r17=cr.iim
-	mov r18=__IA64_BREAK_SYSCALL
-	mov r21=ar.fpsr
-	mov r29=cr.ipsr
-	mov r19=b6
-	mov r25=ar.unat
-	mov r27=ar.rsc
-	mov r26=ar.pfs
-	mov r28=cr.iip
-	mov r31=pr				// prepare to save predicates
-	mov r20=r1
-	;;
+	mov.m r16=IA64_KR(CURRENT)		// M2 r16 <- current task (12 cyc)
+	mov r29=cr.ipsr				// M2 (12 cyc)
+	mov r31=pr				// I0 (2 cyc)
+
+	mov r17=cr.iim				// M2 (2 cyc)
+	mov.m r27=ar.rsc			// M2 (12 cyc)
+	mov r18=__IA64_BREAK_SYSCALL		// A
+
+	mov.m ar.rsc=0				// M2
+	mov.m r21=ar.fpsr			// M2 (12 cyc)
+	mov r19=b6				// I0 (2 cyc)
+	;;
+	mov.m r23=ar.bspstore			// M2 (12 cyc)
+	mov.m r24=ar.rnat			// M2 (5 cyc)
+	mov.i r26=ar.pfs			// I0 (2 cyc)
+
+	invala					// M0|1
+	nop.m 0					// M
+	mov r20=r1				// A			save r1
+
+	nop.m 0
+	movl r30=sys_call_table			// X
+
+	mov r28=cr.iip				// M2 (2 cyc)
+	cmp.eq p0,p7=r18,r17			// I0 is this a system call?
+(p7)	br.cond.spnt non_syscall		// B  no ->
+	//
+	// From this point on, we are definitely on the syscall-path
+	// and we can use (non-banked) scratch registers.
+	//
+///////////////////////////////////////////////////////////////////////
+	mov r1=r16				// A    move task-pointer to "addl"-addressable reg
+	mov r2=r16				// A    setup r2 for ia64_syscall_setup
+	add r9=TI_FLAGS+IA64_TASK_SIZE,r16	// A	r9 = &current_thread_info()->flags
+
 	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
-	cmp.eq p0,p7=r18,r17			// is this a system call? (p7 <- false, if so)
-(p7)	br.cond.spnt non_syscall
+	adds r15=-1024,r15			// A    subtract 1024 from syscall number
+	mov r3=NR_syscalls - 1
 	;;
-	ld1 r17=[r16]				// load current->thread.on_ustack flag
-	st1 [r16]=r0				// clear current->thread.on_ustack flag
-	add r1=-IA64_TASK_THREAD_ON_USTACK_OFFSET,r16	// set r1 for MINSTATE_START_SAVE_MIN_VIRT
+	ld1.bias r17=[r16]			// M0|1 r17 = current->thread.on_ustack flag
+	ld4 r9=[r9]				// M0|1 r9 = current_thread_info()->flags
+	extr.u r8=r29,41,2			// I0   extract ei field from cr.ipsr
+
+	shladd r30=r15,3,r30			// A    r30 = sys_call_table + 8*(syscall-1024)
+	addl r22=IA64_RBS_OFFSET,r1		// A    compute base of RBS
+	cmp.leu p6,p7=r15,r3			// A    syscall number in range?
 	;;
-	invala
 
-	/* adjust return address so we skip over the break instruction: */
+	lfetch.fault.excl.nt1 [r22]		// M0|1 prefetch RBS
+(p6)	ld8 r30=[r30]				// M0|1 load address of syscall entry point
+	tnat.nz.or p7,p0=r15			// I0	is syscall nr a NaT?
 
-	extr.u r8=r29,41,2			// extract ei field from cr.ipsr
-	;;
-	cmp.eq p6,p7=2,r8			// isr.ei==2?
-	mov r2=r1				// setup r2 for ia64_syscall_setup
-	;;
-(p6)	mov r8=0				// clear ei to 0
-(p6)	adds r28=16,r28				// switch cr.iip to next bundle cr.ipsr.ei wrapped
-(p7)	adds r8=1,r8				// increment ei to next slot
-	;;
-	cmp.eq pKStk,pUStk=r0,r17		// are we in kernel mode already?
-	dep r29=r8,r29,41,2			// insert new ei into cr.ipsr
+	mov.m ar.bspstore=r22			// M2   switch to kernel RBS
+	cmp.eq p8,p9=2,r8			// A    isr.ei==2?
 	;;
 
-	// switch from user to kernel RBS:
-	MINSTATE_START_SAVE_MIN_VIRT
-	br.call.sptk.many b7=ia64_syscall_setup
-	;;
-	MINSTATE_END_SAVE_MIN_VIRT		// switch to bank 1
-	ssm psr.ic | PSR_DEFAULT_BITS
-	;;
-	srlz.i					// guarantee that interruption collection is on
-	mov r3=NR_syscalls - 1
-	;;
-(p15)	ssm psr.i				// restore psr.i
-	// p10==true means out registers are more than 8 or r15's Nat is true
-(p10)	br.cond.spnt.many ia64_ret_from_syscall
-	;;
-	movl r16=sys_call_table
+(p8)	mov r8=0				// A    clear ei to 0
+(p7)	movl r30=sys_ni_syscall			// X
 
-	adds r15=-1024,r15			// r15 contains the syscall number---subtract 1024
-	movl r2=ia64_ret_from_syscall
-	;;
-	shladd r20=r15,3,r16			// r20 = sys_call_table + 8*(syscall-1024)
-	cmp.leu p6,p7=r15,r3			// (syscall > 0 && syscall < 1024 + NR_syscalls) ?
-	mov rp=r2				// set the real return addr
+(p8)	adds r28=16,r28				// A    switch cr.iip to next bundle
+(p9)	adds r8=1,r8				// A    increment ei to next slot
+	nop.i 0
 	;;
-(p6)	ld8 r20=[r20]				// load address of syscall entry point
-(p7)	movl r20=sys_ni_syscall
 
-	add r2=TI_FLAGS+IA64_TASK_SIZE,r13
-	;;
-	ld4 r2=[r2]				// r2 = current_thread_info()->flags
-	;;
-	and r2=_TIF_SYSCALL_TRACEAUDIT,r2	// mask trace or audit
+	mov.m r25=ar.unat			// M2 (5 cyc)
+	dep r29=r8,r29,41,2			// I0   insert new ei into cr.ipsr
+	adds r15=1024,r15			// A    restore original syscall number
+	//
+	// If any of the above loads miss in L1D, we'll stall here until
+	// the data arrives.
+	//
+///////////////////////////////////////////////////////////////////////
+	st1 [r16]=r0				// M2|3 clear current->thread.on_ustack flag
+	mov b6=r30				// I0   setup syscall handler branch reg early
+	cmp.eq pKStk,pUStk=r0,r17		// A    were we on kernel stacks already?
+
+	and r9=_TIF_SYSCALL_TRACEAUDIT,r9	// A    mask trace or audit
+	mov r18=ar.bsp				// M2 (12 cyc)
+(pKStk)	br.cond.spnt .break_fixup		// B	we're already in kernel-mode -- fix up RBS
+	;;
+.back_from_break_fixup:
+(pUStk)	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r1 // A    compute base of memory stack
+	cmp.eq p14,p0=r9,r0			// A    are syscalls being traced/audited?
+	br.call.sptk.many b7=ia64_syscall_setup	// B
+1:
+	mov ar.rsc=0x3				// M2   set eager mode, pl 0, LE, loadrs=0
+	nop 0
+	bsw.1					// B (6 cyc) regs are saved, switch to bank 1
 	;;
-	cmp.eq p8,p0=r2,r0
-	mov b6=r20
+
+	ssm psr.ic | PSR_DEFAULT_BITS		// M2	now it's safe to re-enable intr.-collection
+	movl r3=ia64_ret_from_syscall		// X
 	;;
-(p8)	br.call.sptk.many b6=b6			// ignore this return addr
-	br.cond.sptk ia64_trace_syscall
+
+	srlz.i					// M0   ensure interruption collection is on
+	mov rp=r3				// I0   set the real return addr
+(p10)	br.cond.spnt.many ia64_ret_from_syscall	// B    return if bad call-frame or r15 is a NaT
+
+(p15)	ssm psr.i				// M2   restore psr.i
+(p14)	br.call.sptk.many b6=b6			// B    invoke syscall-handker (ignore return addr)
+	br.cond.spnt.many ia64_trace_syscall	// B	do syscall-tracing thingamagic
 	// NOT REACHED
+///////////////////////////////////////////////////////////////////////
+	// On entry, we optimistically assumed that we're coming from user-space.
+	// For the rare cases where a system-call is done from within the kernel,
+	// we fix things up at this point:
+.break_fixup:
+	add r1=-IA64_PT_REGS_SIZE,sp		// A    allocate space for pt_regs structure
+	mov ar.rnat=r24				// M2	restore kernel's AR.RNAT
+	;;
+	mov ar.bspstore=r23			// M2	restore kernel's AR.BSPSTORE
+	br.cond.sptk .back_from_break_fixup
 END(break_fault)
 
 	.org ia64_ivt+0x3000
@@ -837,8 +873,6 @@ END(interrupt)
 	 *	- r31: saved pr
 	 *	-  b0: original contents (to be saved)
 	 * On exit:
-	 *	- executing on bank 1 registers
-	 *	- psr.ic enabled, interrupts restored
 	 *	-  p10: TRUE if syscall is invoked with more than 8 out
 	 *		registers or r15's Nat is true
 	 *	-  r1: kernel's gp
@@ -846,8 +880,11 @@ END(interrupt)
 	 *	-  r8: -EINVAL if p10 is true
 	 *	- r12: points to kernel stack
 	 *	- r13: points to current task
+	 *	- r14: preserved (same as on entry)
+	 *	- p13: preserved
 	 *	- p15: TRUE if interrupts need to be re-enabled
 	 *	- ar.fpsr: set to kernel settings
+	 *	-  b6: preserved (same as on entry)
 	 */
 GLOBAL_ENTRY(ia64_syscall_setup)
 #if PT(B6) != 0
@@ -915,7 +952,7 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 (p13)	mov in5=-1
 	;;
 	st8 [r16]=r21,PT(R8)-PT(AR_FPSR)	// save ar.fpsr
-	tnat.nz p14,p0=in6
+	tnat.nz p13,p0=in6
 	cmp.lt p10,p9=r11,r8	// frame size can't be more than local+8
 	;;
 	mov r8=1
@@ -930,7 +967,7 @@ GLOBAL_ENTRY(ia64_syscall_setup)
 	movl r1=__gp				// establish kernel global pointer
 	;;
 	st8 [r16]=r8		// ensure pt_regs.r8 != 0 (see handle_syscall_error)
-(p14)	mov in6=-1
+(p13)	mov in6=-1
 (p8)	mov in7=-1
 
 	cmp.eq pSys,pNonSys=r0,r0		// set pSys=1, pNonSys=0
@@ -1002,6 +1039,8 @@ END(dispatch_illegal_op_fault)
 	FAULT(17)
 
 ENTRY(non_syscall)
+	mov ar.rsc=r27			// restore ar.rsc before SAVE_MIN_WITH_COVER
+	;;
 	SAVE_MIN_WITH_COVER
 
 	// There is no particular reason for this code to be here, other than that
-- 
cgit v1.1


From 70929a57cfea8c18de13fcea9ae6771018a98949 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:19:37 -0700
Subject: [IA64] Reschedule __kernel_syscall_via_epc().

Avoid some stalls, which is good for about 2 cycles when invoking a
light-weight handler.  When invoking a heavy-weight handler, this
helps by about 7 cycles, with most of the improvement coming from the
improved branch-prediction achieved by splitting the BBB bundle into
two MIB bundles.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/gate.S | 31 ++++++++++++++++++-------------
 1 file changed, 18 insertions(+), 13 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index facf75a..3cd3f2e 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -79,31 +79,34 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	;;
 	rsm psr.be // note: on McKinley "rsm psr.be/srlz.d" is slightly faster than "rum psr.be"
 	LOAD_FSYSCALL_TABLE(r14)
-
+	;;
 	mov r16=IA64_KR(CURRENT)		// 12 cycle read latency
-	tnat.nz p10,p9=r15
+	shladd r18=r17,3,r14
 	mov r19=NR_syscalls-1
 	;;
-	shladd r18=r17,3,r14
-
-	srlz.d
-	cmp.ne p8,p0=r0,r0			// p8 <- FALSE
+	lfetch [r18]				// M0|1
+	mov r29=psr				// read psr (12 cyc load latency)
 	/* Note: if r17 is a NaT, p6 will be set to zero.  */
 	cmp.geu p6,p7=r19,r17			// (syscall > 0 && syscall < 1024+NR_syscalls)?
 	;;
-(p6)	ld8 r18=[r18]
 	mov r21=ar.fpsr
-	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
+	tnat.nz p10,p9=r15
+	mov r26=ar.pfs
 	;;
+	srlz.d
+(p6)	ld8 r18=[r18]
+	nop.i 0
+	;;
+	nop.m 0
 (p6)	mov b7=r18
-(p6)	tbit.z p8,p0=r18,0
+(p6)	tbit.z.unc p8,p0=r18,0
+
+	nop.m 0
+	nop.i 0
 (p8)	br.dptk.many b7
 
-(p6)	rsm psr.i
 	mov r27=ar.rsc
-	mov r26=ar.pfs
-	;;
-	mov r29=psr				// read psr (12 cyc load latency)
+(p6)	rsm psr.i
 /*
  * brl.cond doesn't work as intended because the linker would convert this branch
  * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
@@ -111,6 +114,8 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
  * instead.
  */
 #ifdef CONFIG_ITANIUM
+	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
+	;;
 (p6)	ld8 r14=[r14]				// r14 <- fsys_bubble_down
 	;;
 (p6)	mov b7=r14
-- 
cgit v1.1


From 21bc4f9b34cc1eab3610955207f72c52495ae8ed Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:20:11 -0700
Subject: [IA64] Annotate __kernel_syscall_via_epc() with McKinley dispatch
 info.

Two other very minor changes: use "mov.i" instead of "mov" for reading
ar.pfs (for clarity; doesn't affect the code at all).  Also, predicate
the load of r14 for consistency.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/gate.S | 44 ++++++++++++++++++++++----------------------
 1 file changed, 22 insertions(+), 22 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index 3cd3f2e..272e64c 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -72,41 +72,41 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	 * bundle get executed.  The remaining code must be safe even if
 	 * they do not get executed.
 	 */
-	adds r17=-1024,r15
-	mov r10=0				// default to successful syscall execution
-	epc
+	adds r17=-1024,r15			// A
+	mov r10=0				// A    default to successful syscall execution
+	epc					// B	causes split-issue
 }
 	;;
-	rsm psr.be // note: on McKinley "rsm psr.be/srlz.d" is slightly faster than "rum psr.be"
-	LOAD_FSYSCALL_TABLE(r14)
+	rsm psr.be				// M2 (5 cyc to srlz.d)
+	LOAD_FSYSCALL_TABLE(r14)		// X
 	;;
-	mov r16=IA64_KR(CURRENT)		// 12 cycle read latency
-	shladd r18=r17,3,r14
-	mov r19=NR_syscalls-1
+	mov r16=IA64_KR(CURRENT)		// M2 (12 cyc)
+	shladd r18=r17,3,r14			// A
+	mov r19=NR_syscalls-1			// A
 	;;
 	lfetch [r18]				// M0|1
-	mov r29=psr				// read psr (12 cyc load latency)
-	/* Note: if r17 is a NaT, p6 will be set to zero.  */
-	cmp.geu p6,p7=r19,r17			// (syscall > 0 && syscall < 1024+NR_syscalls)?
+	mov r29=psr				// M2 (12 cyc)
+	// If r17 is a NaT, p6 will be zero
+	cmp.geu p6,p7=r19,r17			// A    (sysnr > 0 && sysnr < 1024+NR_syscalls)?
 	;;
-	mov r21=ar.fpsr
-	tnat.nz p10,p9=r15
-	mov r26=ar.pfs
+	mov r21=ar.fpsr				// M2 (12 cyc)
+	tnat.nz p10,p9=r15			// I0
+	mov.i r26=ar.pfs			// I0 (would stall anyhow due to srlz.d...)
 	;;
-	srlz.d
-(p6)	ld8 r18=[r18]
+	srlz.d					// M0 (forces split-issue) ensure PSR.BE==0
+(p6)	ld8 r18=[r18]				// M0|1
 	nop.i 0
 	;;
 	nop.m 0
-(p6)	mov b7=r18
-(p6)	tbit.z.unc p8,p0=r18,0
+(p6)	mov b7=r18				// I0
+(p6)	tbit.z.unc p8,p0=r18,0			// I0 (dual-issues with "mov b7=r18"!)
 
 	nop.m 0
 	nop.i 0
-(p8)	br.dptk.many b7
+(p8)	br.dptk.many b7				// B
 
-	mov r27=ar.rsc
-(p6)	rsm psr.i
+	mov r27=ar.rsc				// M2 (12 cyc)
+(p6)	rsm psr.i				// M2
 /*
  * brl.cond doesn't work as intended because the linker would convert this branch
  * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
@@ -114,7 +114,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
  * instead.
  */
 #ifdef CONFIG_ITANIUM
-	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
+(p6)	add r14=-8,r14				// r14 <- addr of fsys_bubble_down entry
 	;;
 (p6)	ld8 r14=[r14]				// r14 <- fsys_bubble_down
 	;;
-- 
cgit v1.1


From 1ba7be7d691f6df2557d39c5b1a2e14c32e5dd20 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:20:51 -0700
Subject: [IA64] Reschedule fsys_bubble_down().

Improvements come from eliminating srlz.i, not scheduling AR/CR-reads
too early (while there are others still pending), scheduling the
backing-store switch as well as possible, splitting the BBB bundle
into a MIB/MBB pair.

Why is it safe to eliminate the srlz.i?  Observe
that we used to clear bits ~PSR_PRESERVED_BITS in PSR.L.  Since
PSR_PRESERVED_BITS==PSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we
ended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}.  However,

 PSR.BE : already is turned off in __kernel_syscall_via_epc()
 PSR.AC : don't care (kernel normally turns PSR.AC on)
 PSR.I  : already turned off by the time fsys_bubble_down gets invoked
 PSR.DFL: always 0 (kernel never turns it on)
 PSR.DFH: don't care --- kernel never touches f32-f127 on its own
	  initiative
 PSR.DI : always 0 (kernel never turns it on)
 PSR.SI : always 0 (kernel never turns it on)
 PSR.DB : don't care --- kernel never enables kernel-level breakpoints
 PSR.TB : must be 0 already; if it wasn't zero on entry to
	  __kernel_syscall_via_epc, the branch to fsys_bubble_down
	  will trigger a taken branch; the taken-trap-handler then
	  converts the syscall into a break-based system-call.

In other words: all the bits we're clearying are either 0 already or
are don't cares!  Thus, we don't have to write PSR.L at all and we
don't have to do a srlz.i either.

Good for another ~20 cycle improvement for EPC-based heavy-weight
syscalls.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/fsys.S | 73 ++++++++++++++++++++++---------------------------
 1 file changed, 33 insertions(+), 40 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 0d8650f..57c6556 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -549,9 +549,6 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	 *	- r27: ar.rsc
 	 *	- r29: psr
 	 */
-#	define PSR_PRESERVED_BITS	(IA64_PSR_UP | IA64_PSR_MFL | IA64_PSR_MFH | IA64_PSR_PK \
-					 | IA64_PSR_DT | IA64_PSR_PP | IA64_PSR_SP | IA64_PSR_RT \
-					 | IA64_PSR_IC)
 	/*
 	 * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.  The rest we have
 	 * to synthesize.
@@ -560,62 +557,58 @@ GLOBAL_ENTRY(fsys_bubble_down)
 					 | IA64_PSR_BN | IA64_PSR_I)
 
 	invala
-	movl r8=PSR_ONE_BITS
+	movl r14=ia64_ret_from_syscall
 
-	mov r25=ar.unat			// save ar.unat (5 cyc)
-	movl r9=PSR_PRESERVED_BITS
-
-	mov ar.rsc=0			// set enforced lazy mode, pl 0, little-endian, loadrs=0
+	nop.m 0
 	movl r28=__kernel_syscall_via_break
 	;;
-	mov r23=ar.bspstore		// save ar.bspstore (12 cyc)
-	mov r31=pr			// save pr (2 cyc)
-	mov r20=r1			// save caller's gp in r20
-	;;
+
 	mov r2=r16			// copy current task addr to addl-addressable register
-	and r9=r9,r29
-	mov r19=b6			// save b6 (2 cyc)
+	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
+	mov r31=pr			// save pr (2 cyc)
 	;;
-	mov psr.l=r9			// slam the door (17 cyc to srlz.i)
-	or r29=r8,r29			// construct cr.ipsr value to save
+	st1 [r16]=r0			// clear current->thread.on_ustack flag
 	addl r22=IA64_RBS_OFFSET,r2	// compute base of RBS
+	add r3=TI_FLAGS+IA64_TASK_SIZE,r2
 	;;
-	// GAS reports a spurious RAW hazard on the read of ar.rnat because it thinks
-	// we may be reading ar.itc after writing to psr.l.  Avoid that message with
-	// this directive:
-	dv_serialize_data
-	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
+	ld4 r3=[r3]				// r2 = current_thread_info()->flags
 	lfetch.fault.excl.nt1 [r22]
-	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r2
-
-	// ensure previous insn group is issued before we stall for srlz.i:
+	nop.i 0
 	;;
-	srlz.i				// ensure new psr.l has been established
-	/////////////////////////////////////////////////////////////////////////////
-	////////// from this point on, execution is not interruptible anymore
-	/////////////////////////////////////////////////////////////////////////////
-	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2	// compute base of memory stack
-	cmp.ne pKStk,pUStk=r0,r0	// set pKStk <- 0, pUStk <- 1
+	mov ar.rsc=0			// set enforced lazy mode, pl 0, little-endian, loadrs=0
+	nop.m 0
+	nop.i 0
+	;;
+	mov r23=ar.bspstore		// save ar.bspstore (12 cyc)
+	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
+	nop.i 0
 	;;
-	st1 [r16]=r0			// clear current->thread.on_ustack flag
 	mov ar.bspstore=r22		// switch to kernel RBS
-	mov b6=r18			// copy syscall entry-point to b6 (7 cyc)
-	add r3=TI_FLAGS+IA64_TASK_SIZE,r2
+	movl r8=PSR_ONE_BITS			// X
 	;;
-	ld4 r3=[r3]				// r2 = current_thread_info()->flags
+	mov r25=ar.unat			// save ar.unat (5 cyc)
+	mov r19=b6			// save b6 (2 cyc)
+	mov r20=r1			// save caller's gp in r20
+	;;
+	or r29=r8,r29			// construct cr.ipsr value to save
+	mov b6=r18			// copy syscall entry-point to b6 (7 cyc)
+	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2	// compute base of memory stack
+
 	mov r18=ar.bsp			// save (kernel) ar.bsp (12 cyc)
-	mov ar.rsc=0x3			// set eager mode, pl 0, little-endian, loadrs=0
+	cmp.ne pKStk,pUStk=r0,r0	// set pKStk <- 0, pUStk <- 1
 	br.call.sptk.many b7=ia64_syscall_setup
 	;;
-	ssm psr.i
-	movl r2=ia64_ret_from_syscall
+	mov ar.rsc=0x3			// set eager mode, pl 0, little-endian, loadrs=0
+	mov rp=r14				// set the real return addr
+	nop.i 0
 	;;
-	mov rp=r2				// set the real return addr
+	ssm psr.i
 	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE
-	;;
 (p10)	br.cond.spnt.many ia64_ret_from_syscall	// p10==true means out registers are more than 8
+
+	nop.m 0
 (p8)	br.call.sptk.many b6=b6		// ignore this return addr
-	br.cond.sptk ia64_trace_syscall
+	br.cond.spnt ia64_trace_syscall
 END(fsys_bubble_down)
 
 	.rodata
-- 
cgit v1.1


From fbf7192ba06e4b55022231e5fdb9db8ce3ad6307 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:21:26 -0700
Subject: [IA64] Annotate fsys_bubble_down() with McKinley dispatch info.

This patch changes comments & formatting only.  There is no code
change.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/fsys.S | 118 ++++++++++++++++++++++++++++++------------------
 1 file changed, 74 insertions(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 57c6556..d09a5b8 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -531,84 +531,114 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	.altrp b6
 	.body
 	/*
-	 * We get here for syscalls that don't have a lightweight handler.  For those, we
-	 * need to bubble down into the kernel and that requires setting up a minimal
-	 * pt_regs structure, and initializing the CPU state more or less as if an
-	 * interruption had occurred.  To make syscall-restarts work, we setup pt_regs
-	 * such that cr_iip points to the second instruction in syscall_via_break.
-	 * Decrementing the IP hence will restart the syscall via break and not
-	 * decrementing IP will return us to the caller, as usual.  Note that we preserve
-	 * the value of psr.pp rather than initializing it from dcr.pp.  This makes it
-	 * possible to distinguish fsyscall execution from other privileged execution.
+	 * We get here for syscalls that don't have a lightweight
+	 * handler.  For those, we need to bubble down into the kernel
+	 * and that requires setting up a minimal pt_regs structure,
+	 * and initializing the CPU state more or less as if an
+	 * interruption had occurred.  To make syscall-restarts work,
+	 * we setup pt_regs such that cr_iip points to the second
+	 * instruction in syscall_via_break.  Decrementing the IP
+	 * hence will restart the syscall via break and not
+	 * decrementing IP will return us to the caller, as usual.
+	 * Note that we preserve the value of psr.pp rather than
+	 * initializing it from dcr.pp.  This makes it possible to
+	 * distinguish fsyscall execution from other privileged
+	 * execution.
 	 *
 	 * On entry:
-	 *	- normal fsyscall handler register usage, except that we also have:
+	 *	- normal fsyscall handler register usage, except
+	 *	  that we also have:
 	 *	- r18: address of syscall entry point
 	 *	- r21: ar.fpsr
 	 *	- r26: ar.pfs
 	 *	- r27: ar.rsc
 	 *	- r29: psr
+	 *
+	 * We used to clear some PSR bits here but that requires slow
+	 * serialization.  Fortuntely, that isn't really necessary.
+	 * The rationale is as follows: we used to clear bits
+	 * ~PSR_PRESERVED_BITS in PSR.L.  Since
+	 * PSR_PRESERVED_BITS==PSR.{UP,MFL,MFH,PK,DT,PP,SP,RT,IC}, we
+	 * ended up clearing PSR.{BE,AC,I,DFL,DFH,DI,DB,SI,TB}.
+	 * However,
+	 *
+	 * PSR.BE : already is turned off in __kernel_syscall_via_epc()
+	 * PSR.AC : don't care (kernel normally turns PSR.AC on)
+	 * PSR.I  : already turned off by the time fsys_bubble_down gets
+	 *	    invoked
+	 * PSR.DFL: always 0 (kernel never turns it on)
+	 * PSR.DFH: don't care --- kernel never touches f32-f127 on its own
+	 *	    initiative
+	 * PSR.DI : always 0 (kernel never turns it on)
+	 * PSR.SI : always 0 (kernel never turns it on)
+	 * PSR.DB : don't care --- kernel never enables kernel-level
+	 *	    breakpoints
+	 * PSR.TB : must be 0 already; if it wasn't zero on entry to
+	 *          __kernel_syscall_via_epc, the branch to fsys_bubble_down
+	 *          will trigger a taken branch; the taken-trap-handler then
+	 *          converts the syscall into a break-based system-call.
 	 */
 	/*
-	 * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.  The rest we have
-	 * to synthesize.
+	 * Reading psr.l gives us only bits 0-31, psr.it, and psr.mc.
+	 * The rest we have to synthesize.
 	 */
-#	define PSR_ONE_BITS		((3 << IA64_PSR_CPL0_BIT) | (0x1 << IA64_PSR_RI_BIT) \
+#	define PSR_ONE_BITS		((3 << IA64_PSR_CPL0_BIT)	\
+					 | (0x1 << IA64_PSR_RI_BIT)	\
 					 | IA64_PSR_BN | IA64_PSR_I)
 
-	invala
-	movl r14=ia64_ret_from_syscall
+	invala					// M0|1
+	movl r14=ia64_ret_from_syscall		// X
 
 	nop.m 0
-	movl r28=__kernel_syscall_via_break
+	movl r28=__kernel_syscall_via_break	// X	create cr.iip
 	;;
 
-	mov r2=r16			// copy current task addr to addl-addressable register
-	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
-	mov r31=pr			// save pr (2 cyc)
+	mov r2=r16				// A    get task addr to addl-addressable register
+	adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16 // A
+	mov r31=pr				// I0   save pr (2 cyc)
 	;;
-	st1 [r16]=r0			// clear current->thread.on_ustack flag
-	addl r22=IA64_RBS_OFFSET,r2	// compute base of RBS
-	add r3=TI_FLAGS+IA64_TASK_SIZE,r2
+	st1 [r16]=r0				// M2|3 clear current->thread.on_ustack flag
+	addl r22=IA64_RBS_OFFSET,r2		// A    compute base of RBS
+	add r3=TI_FLAGS+IA64_TASK_SIZE,r2	// A
 	;;
-	ld4 r3=[r3]				// r2 = current_thread_info()->flags
-	lfetch.fault.excl.nt1 [r22]
+	ld4 r3=[r3]				// M0|1 r3 = current_thread_info()->flags
+	lfetch.fault.excl.nt1 [r22]		// M0|1 prefetch register backing-store
 	nop.i 0
 	;;
-	mov ar.rsc=0			// set enforced lazy mode, pl 0, little-endian, loadrs=0
+	mov ar.rsc=0				// M2   set enforced lazy mode, pl 0, LE, loadrs=0
 	nop.m 0
 	nop.i 0
 	;;
-	mov r23=ar.bspstore		// save ar.bspstore (12 cyc)
-	mov.m r24=ar.rnat		// read ar.rnat (5 cyc lat)
+	mov r23=ar.bspstore			// M2 (12 cyc) save ar.bspstore
+	mov.m r24=ar.rnat			// M2 (5 cyc) read ar.rnat (dual-issues!)
 	nop.i 0
 	;;
-	mov ar.bspstore=r22		// switch to kernel RBS
+	mov ar.bspstore=r22			// M2 (6 cyc) switch to kernel RBS
 	movl r8=PSR_ONE_BITS			// X
 	;;
-	mov r25=ar.unat			// save ar.unat (5 cyc)
-	mov r19=b6			// save b6 (2 cyc)
-	mov r20=r1			// save caller's gp in r20
+	mov r25=ar.unat				// M2 (5 cyc) save ar.unat
+	mov r19=b6				// I0   save b6 (2 cyc)
+	mov r20=r1				// A    save caller's gp in r20
 	;;
-	or r29=r8,r29			// construct cr.ipsr value to save
-	mov b6=r18			// copy syscall entry-point to b6 (7 cyc)
-	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2	// compute base of memory stack
+	or r29=r8,r29				// A    construct cr.ipsr value to save
+	mov b6=r18				// I0   copy syscall entry-point to b6 (7 cyc)
+	addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 // A compute base of memory stack
 
-	mov r18=ar.bsp			// save (kernel) ar.bsp (12 cyc)
-	cmp.ne pKStk,pUStk=r0,r0	// set pKStk <- 0, pUStk <- 1
-	br.call.sptk.many b7=ia64_syscall_setup
+	mov r18=ar.bsp				// M2   save (kernel) ar.bsp (12 cyc)
+	cmp.ne pKStk,pUStk=r0,r0		// A    set pKStk <- 0, pUStk <- 1
+	br.call.sptk.many b7=ia64_syscall_setup	// B
 	;;
-	mov ar.rsc=0x3			// set eager mode, pl 0, little-endian, loadrs=0
-	mov rp=r14				// set the real return addr
+	mov ar.rsc=0x3				// M2   set eager mode, pl 0, LE, loadrs=0
+	mov rp=r14				// I0   set the real return addr
 	nop.i 0
 	;;
-	ssm psr.i
-	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE
-(p10)	br.cond.spnt.many ia64_ret_from_syscall	// p10==true means out registers are more than 8
+	ssm psr.i				// M2   we're on kernel stacks now, reenable irqs
+	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE	// I0
+(p10)	br.cond.spnt.many ia64_ret_from_syscall	// B    return if bad call-frame or r15 is a NaT
 
 	nop.m 0
-(p8)	br.call.sptk.many b6=b6		// ignore this return addr
-	br.cond.spnt ia64_trace_syscall
+(p8)	br.call.sptk.many b6=b6			// B    (ignore return address)
+	br.cond.spnt ia64_trace_syscall		// B
 END(fsys_bubble_down)
 
 	.rodata
-- 
cgit v1.1


From e7e965fa1961a8ce32cbbb1bd436c655ad03973e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:22:08 -0700
Subject: [IA64] use srlz.d instead of srlz.i in ia64_leave_kernel()

This patch switches the srlz.i in ia64_leave_kernel() to srlz.d.  As
per architecture manual, the former is needed only to ensure that the
clearing of PSR.IC is seen by the VHPT for subsequent instruction
fetches.  However, since the remainder of the code (up to and
including the RFI instruction) is mapped by a pinned TLB entry, there
is no chance of an iTLB miss and we don't care whether or not the VHPT
sees PSR.IC cleared.  Since srlz.d is substantially cheaper than
srlz.i, this should shave off a few cycles off the interrupt path
(unverified though; I'm not setup to measure this at the moment).

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index a7542c3..4517d4a 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -880,7 +880,7 @@ GLOBAL_ENTRY(ia64_leave_kernel)
 	ldf.fill f7=[r2],PT(F11)-PT(F7)
 	ldf.fill f8=[r3],32
 	;;
-	srlz.i			// ensure interruption collection is off
+	srlz.d	// ensure that inter. collection is off (VHPT is don't care, since text is pinned)
 	mov ar.ccv=r15
 	;;
 	ldf.fill f11=[r2]
-- 
cgit v1.1


From 8e3e50168c8537807b7a6f78588cd72e21363262 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 27 Apr 2005 21:22:40 -0700
Subject: [IA64] need r29=psr *after* rsm psr.i

Yanmin Zhang pointed out a sequence problem when saving the psr.  David
Mosberger provided this patch (which gave up a cycle).

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/gate.S | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/gate.S b/arch/ia64/kernel/gate.S
index 272e64c..86948ce 100644
--- a/arch/ia64/kernel/gate.S
+++ b/arch/ia64/kernel/gate.S
@@ -77,7 +77,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	epc					// B	causes split-issue
 }
 	;;
-	rsm psr.be				// M2 (5 cyc to srlz.d)
+	rsm psr.be | psr.i			// M2 (5 cyc to srlz.d)
 	LOAD_FSYSCALL_TABLE(r14)		// X
 	;;
 	mov r16=IA64_KR(CURRENT)		// M2 (12 cyc)
@@ -98,15 +98,14 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 	nop.i 0
 	;;
 	nop.m 0
-(p6)	mov b7=r18				// I0
 (p6)	tbit.z.unc p8,p0=r18,0			// I0 (dual-issues with "mov b7=r18"!)
-
-	nop.m 0
 	nop.i 0
+	;;
+(p8)	ssm psr.i
+(p6)	mov b7=r18				// I0
 (p8)	br.dptk.many b7				// B
 
 	mov r27=ar.rsc				// M2 (12 cyc)
-(p6)	rsm psr.i				// M2
 /*
  * brl.cond doesn't work as intended because the linker would convert this branch
  * into a branch to a PLT.  Perhaps there will be a way to avoid this with some
@@ -123,7 +122,7 @@ GLOBAL_ENTRY(__kernel_syscall_via_epc)
 #else
 	BRL_COND_FSYS_BUBBLE_DOWN(p6)
 #endif
-
+	ssm psr.i
 	mov r10=-1
 (p10)	mov r8=EINVAL
 (p9)	mov r8=ENOSYS
-- 
cgit v1.1


From ebcc80c1b6629a445f7471cc1ddb48faf8a84e70 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Thu, 5 May 2005 06:40:00 -0700
Subject: [IA64] Merge audit fix for fsyscalls with syscall-optimizations

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/fsys.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index d09a5b8..f566ff4 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -630,10 +630,10 @@ GLOBAL_ENTRY(fsys_bubble_down)
 	;;
 	mov ar.rsc=0x3				// M2   set eager mode, pl 0, LE, loadrs=0
 	mov rp=r14				// I0   set the real return addr
-	nop.i 0
+	and r3=_TIF_SYSCALL_TRACEAUDIT,r3	// A
 	;;
 	ssm psr.i				// M2   we're on kernel stacks now, reenable irqs
-	tbit.z p8,p0=r3,TIF_SYSCALL_TRACE	// I0
+	cmp.eq p8,p0=r3,r0			// A
 (p10)	br.cond.spnt.many ia64_ret_from_syscall	// B    return if bad call-frame or r15 is a NaT
 
 	nop.m 0
-- 
cgit v1.1


From 7f9eaedf894dbaa08c157832e9a6c9c03ffed1ed Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Tue, 10 May 2005 12:49:00 -0700
Subject: [IA64] Fix convert_to_non_syscall() so gdb inferior calls work again

Fix convert_to_non_syscall() so it arranges for the kernel to be left
via ia64_leave_kernel() rather than ia64_leave_syscall().  The latter
no longer tolerates being called with pSys=0 and pNonSys=1.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ptrace.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c
index c253fd5..9e730c7 100644
--- a/arch/ia64/kernel/ptrace.c
+++ b/arch/ia64/kernel/ptrace.c
@@ -705,12 +705,32 @@ convert_to_non_syscall (struct task_struct *child, struct pt_regs  *pt,
 			break;
 	}
 
+	/*
+	 * Note: at the time of this call, the target task is blocked
+	 * in notify_resume_user() and by clearling PRED_LEAVE_SYSCALL
+	 * (aka, "pLvSys") we redirect execution from
+	 * .work_pending_syscall_end to .work_processed_kernel.
+	 */
 	unw_get_pr(&prev_info, &pr);
-	pr &= ~(1UL << PRED_SYSCALL);
+	pr &= ~((1UL << PRED_SYSCALL) | (1UL << PRED_LEAVE_SYSCALL));
 	pr |=  (1UL << PRED_NON_SYSCALL);
 	unw_set_pr(&prev_info, pr);
 
 	pt->cr_ifs = (1UL << 63) | cfm;
+	/*
+	 * Clear the memory that is NOT written on syscall-entry to
+	 * ensure we do not leak kernel-state to user when execution
+	 * resumes.
+	 */
+	pt->r2 = 0;
+	pt->r3 = 0;
+	pt->r14 = 0;
+	memset(&pt->r16, 0, 16*8);	/* clear r16-r31 */
+	memset(&pt->f6, 0, 6*16);	/* clear f6-f11 */
+	pt->b7 = 0;
+	pt->ar_ccv = 0;
+	pt->ar_csd = 0;
+	pt->ar_ssd = 0;
 }
 
 static int
-- 
cgit v1.1


From 66b7f8a30437b8639e798f7db8e9be1da5711efa Mon Sep 17 00:00:00 2001
From: Mark Maule <maule@sgi.com>
Date: Mon, 25 Apr 2005 13:51:00 -0700
Subject: [IA64-SGI] pcdp: add PCDP pci interface support

Resend 2 with changes per Bjorn Helgaas comments.  Changes from original:

+ Change globals to vga_console_iobase/vga_console_membase and make them
  unconditional.
+ Address style-related comments.

Patch to extend the PCDP vga setup code to support PCI io/mem translations
for the legacy vga ioport and ram spaces on architectures (e.g. altix) which
need them.

Summary of the changes:

drivers/firmware/pcdp.c
drivers/firmware/pcdp.h
-----------------------
+ add declaration for the spec-defined PCI interface struct (pcdp_if_pci)
  as well as support macros.

+ extend setup_vga_console() to know about pcdp_if_pci and add a couple of
  globals to hold the io and mem translation offsets if present.

arch/ia64/kernel/setup.c
------------------------
+ tweek early_console_setup() to allow multiple early console setup routines
  to be called.

include/asm-ia64/vga.h
----------------------
+ make VGA_MAP_MEM vga_console_membase aware

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/setup.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index d14692e..2693e15 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -72,6 +72,8 @@ DEFINE_PER_CPU(unsigned long, ia64_phys_stacked_size_p8);
 unsigned long ia64_cycles_per_usec;
 struct ia64_boot_param *ia64_boot_param;
 struct screen_info screen_info;
+unsigned long vga_console_iobase;
+unsigned long vga_console_membase;
 
 unsigned long ia64_max_cacheline_size;
 unsigned long ia64_iobase;	/* virtual address for I/O accesses */
@@ -273,23 +275,25 @@ io_port_init (void)
 static inline int __init
 early_console_setup (char *cmdline)
 {
+	int earlycons = 0;
+
 #ifdef CONFIG_SERIAL_SGI_L1_CONSOLE
 	{
 		extern int sn_serial_console_early_setup(void);
 		if (!sn_serial_console_early_setup())
-			return 0;
+			earlycons++;
 	}
 #endif
 #ifdef CONFIG_EFI_PCDP
 	if (!efi_setup_pcdp_console(cmdline))
-		return 0;
+		earlycons++;
 #endif
 #ifdef CONFIG_SERIAL_8250_CONSOLE
 	if (!early_serial_console_init(cmdline))
-		return 0;
+		earlycons++;
 #endif
 
-	return -1;
+	return (earlycons) ? 0 : -1;
 }
 
 static inline void
-- 
cgit v1.1


From a9f9de7378e3a81f7266f542f2290399a298ef52 Mon Sep 17 00:00:00 2001
From: Mark Maule <maule@sgi.com>
Date: Tue, 26 Apr 2005 08:01:00 -0700
Subject: [IA64-SGI] altix: enable vgacon support

Altix patch to enable use of vgacon driver on that platform.  Depends on the
PCDP generalization patch discussed at:

	http://marc.theaimsgroup.com/?l=linux-ia64&m=111446235101939&w=2

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/sn2_defconfig |  4 ++--
 arch/ia64/sn/kernel/iomv.c      |  6 ++++++
 arch/ia64/sn/kernel/setup.c     | 16 ++++++++++------
 3 files changed, 18 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig
index 487d2e3..c056139 100644
--- a/arch/ia64/configs/sn2_defconfig
+++ b/arch/ia64/configs/sn2_defconfig
@@ -99,7 +99,7 @@ CONFIG_ACPI_DEALLOCATE_IRQ=y
 # Firmware Drivers
 #
 CONFIG_EFI_VARS=y
-# CONFIG_EFI_PCDP is not set
+CONFIG_EFI_PCDP=y
 CONFIG_BINFMT_ELF=y
 # CONFIG_BINFMT_MISC is not set
 
@@ -650,7 +650,7 @@ CONFIG_MMTIMER=y
 #
 # Console display driver support
 #
-# CONFIG_VGA_CONSOLE is not set
+CONFIG_VGA_CONSOLE=y
 CONFIG_DUMMY_CONSOLE=y
 
 #
diff --git a/arch/ia64/sn/kernel/iomv.c b/arch/ia64/sn/kernel/iomv.c
index fec6d8b..7ce3cda 100644
--- a/arch/ia64/sn/kernel/iomv.c
+++ b/arch/ia64/sn/kernel/iomv.c
@@ -9,12 +9,16 @@
 #include <linux/module.h>
 #include <asm/io.h>
 #include <asm/delay.h>
+#include <asm/vga.h>
 #include <asm/sn/nodepda.h>
 #include <asm/sn/simulator.h>
 #include <asm/sn/pda.h>
 #include <asm/sn/sn_cpuid.h>
 #include <asm/sn/shub_mmr.h>
 
+#define IS_LEGACY_VGA_IOPORT(p) \
+	(((p) >= 0x3b0 && (p) <= 0x3bb) || ((p) >= 0x3c0 && (p) <= 0x3df))
+
 /**
  * sn_io_addr - convert an in/out port to an i/o address
  * @port: port to convert
@@ -26,6 +30,8 @@
 void *sn_io_addr(unsigned long port)
 {
 	if (!IS_RUNNING_ON_SIMULATOR()) {
+		if (IS_LEGACY_VGA_IOPORT(port))
+			port += vga_console_iobase;
 		/* On sn2, legacy I/O ports don't point at anything */
 		if (port < (64 * 1024))
 			return NULL;
diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 44bfc7f..3e5ba34 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -36,6 +36,7 @@
 #include <asm/machvec.h>
 #include <asm/system.h>
 #include <asm/processor.h>
+#include <asm/vga.h>
 #include <asm/sn/arch.h>
 #include <asm/sn/addrs.h>
 #include <asm/sn/pda.h>
@@ -273,14 +274,17 @@ void __init sn_setup(char **cmdline_p)
 
 	ia64_sn_plat_set_error_handling_features();
 
+#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
 	/*
-	 * If the generic code has enabled vga console support - lets
-	 * get rid of it again. This is a kludge for the fact that ACPI
-	 * currtently has no way of informing us if legacy VGA is available
-	 * or not.
+	 * If there was a primary vga adapter identified through the
+	 * EFI PCDP table, make it the preferred console.  Otherwise
+	 * zero out conswitchp.
 	 */
-#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
-	if (conswitchp == &vga_con) {
+
+	if (vga_console_membase) {
+		/* usable vga ... make tty0 the preferred default console */
+		add_preferred_console("tty", 0, NULL);
+	} else {
 		printk(KERN_DEBUG "SGI: Disabling VGA console\n");
 #ifdef CONFIG_DUMMY_CONSOLE
 		conswitchp = &dummy_con;
-- 
cgit v1.1


From e15da40176f16050c8d92b36387a49370317e276 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Tue, 3 May 2005 12:07:00 -0700
Subject: [IA64] Recognize HP sx2000 chipset

No functional change, just identify the device nicely:

  -IOC: Unknown (103c:12ec) 0.1 HPA 0xf8020002000 IOVA space 1024Mb at 0x40000000
  +IOC: sx2000 0.1 HPA 0xf8020002000 IOVA space 1024Mb at 0x40000000

We used to create fake PCI devices for these chips, but we no longer do that.
So I don't think there's any reason to touch pci_ids.h now.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/hp/common/sba_iommu.c | 4 ++++
 1 file changed, 4 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index b8db6e3..1195759 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -156,10 +156,13 @@
 */
 #define DELAYED_RESOURCE_CNT	64
 
+#define PCI_DEVICE_ID_HP_SX2000_IOC	0x12ec
+
 #define ZX1_IOC_ID	((PCI_DEVICE_ID_HP_ZX1_IOC << 16) | PCI_VENDOR_ID_HP)
 #define ZX2_IOC_ID	((PCI_DEVICE_ID_HP_ZX2_IOC << 16) | PCI_VENDOR_ID_HP)
 #define REO_IOC_ID	((PCI_DEVICE_ID_HP_REO_IOC << 16) | PCI_VENDOR_ID_HP)
 #define SX1000_IOC_ID	((PCI_DEVICE_ID_HP_SX1000_IOC << 16) | PCI_VENDOR_ID_HP)
+#define SX2000_IOC_ID	((PCI_DEVICE_ID_HP_SX2000_IOC << 16) | PCI_VENDOR_ID_HP)
 
 #define ZX1_IOC_OFFSET	0x1000	/* ACPI reports SBA, we want IOC */
 
@@ -1726,6 +1729,7 @@ static struct ioc_iommu ioc_iommu_info[] __initdata = {
 	{ ZX1_IOC_ID, "zx1", ioc_zx1_init },
 	{ ZX2_IOC_ID, "zx2", NULL },
 	{ SX1000_IOC_ID, "sx1000", NULL },
+	{ SX2000_IOC_ID, "sx2000", NULL },
 };
 
 static struct ioc * __init
-- 
cgit v1.1


From 458f935527372499b714bf4f8e646a68bb0f52e3 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Wed, 4 May 2005 13:25:00 -0700
Subject: [IA64] Speed up lfetch.fault [NULL]

This patch greatly speeds up the handling of lfetch.fault instructions
which result in NaT consumption. Due to the NaT-page mapped at address
0, this is guaranteed to happen when lfetch.fault'ing a NULL pointer.
With this patch in place, we can even define prefetch()/prefetchw() as
lfetch.fault without significant performance degradation.  More
importantly, it allows compilers to be more aggressive with using
lfetch.fault on pointers that might be NULL.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ivt.S | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S
index b28d221..3bb3a13 100644
--- a/arch/ia64/kernel/ivt.S
+++ b/arch/ia64/kernel/ivt.S
@@ -1243,6 +1243,25 @@ END(disabled_fp_reg)
 // 0x5600 Entry 26 (size 16 bundles) Nat Consumption (11,23,37,50)
 ENTRY(nat_consumption)
 	DBG_FAULT(26)
+
+	mov r16=cr.ipsr
+	mov r17=cr.isr
+	mov r31=pr				// save PR
+	;;
+	and r18=0xf,r17				// r18 = cr.ipsr.code{3:0}
+	tbit.z p6,p0=r17,IA64_ISR_NA_BIT
+	;;
+	cmp.ne.or p6,p0=IA64_ISR_CODE_LFETCH,r18
+	dep r16=-1,r16,IA64_PSR_ED_BIT,1
+(p6)	br.cond.spnt 1f		// branch if (cr.ispr.na == 0 || cr.ipsr.code{3:0} != LFETCH)
+	;;
+	mov cr.ipsr=r16		// set cr.ipsr.na
+	mov pr=r31,-1
+	;;
+	rfi
+
+1:	mov pr=r31,-1
+	;;
 	FAULT(26)
 END(nat_consumption)
 
-- 
cgit v1.1


From 8e4641b3ee8073a46995cf97e0875df25b88d776 Mon Sep 17 00:00:00 2001
From: Prarit Bhargava <prarit@sgi.com>
Date: Mon, 9 May 2005 05:07:00 -0700
Subject: [IA64] sparse cleanup of shub_mmr.h

This patch is a sparse compile cleanup of shub_mmr.h using both the defconfig
and the sn2_defconfig config files.

The issue with this file was the missing usage of __IA64_UL_CONST wrapper.
This wrapper is defined in include/asm-ia64/types.h and wraps a long
constant definition with UL or with nothing depending on its usage in the
kernel.  The missing wrapper caused many sparse compile errors like

        warning: constant 0x0x0000000010000380 so big it is long

Signed-off-by: Prarit Bhargava <prarit@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/sn2/ptc_deadlock.S | 1 +
 1 file changed, 1 insertion(+)

(limited to 'arch')

diff --git a/arch/ia64/sn/kernel/sn2/ptc_deadlock.S b/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
index 7947312..96cb71d 100644
--- a/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
+++ b/arch/ia64/sn/kernel/sn2/ptc_deadlock.S
@@ -6,6 +6,7 @@
  * Copyright (C) 2000-2004 Silicon Graphics, Inc. All rights reserved.
  */
 
+#include <asm/types.h>
 #include <asm/sn/shub_mmr.h>
 
 #define DEADLOCKBIT	SH_PIO_WRITE_STATUS_WRITE_DEADLOCK_SHFT
-- 
cgit v1.1


From 71a5d027c90dd7e14ddaaf22358ff5138ba89ec9 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Tue, 10 May 2005 08:01:00 -0700
Subject: [IA64-SGI] - new macros for SGI SN simulator

This patch changes some macros that are used when running kernel on the
SGI simulator.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/io_init.c |  2 +-
 arch/ia64/sn/kernel/setup.c   | 27 +++++++++++++++++++--------
 2 files changed, 20 insertions(+), 9 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index 9e07f54..783eb43 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -384,7 +384,7 @@ static int __init sn_pci_init(void)
 	extern void register_sn_procfs(void);
 #endif
 
-	if (!ia64_platform_is("sn2") || IS_RUNNING_ON_SIMULATOR())
+	if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM())
 		return 0;
 
 	/*
diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 3e5ba34..22e10d2 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -96,6 +96,7 @@ u8 sn_coherency_id;
 EXPORT_SYMBOL(sn_coherency_id);
 u8 sn_region_size;
 EXPORT_SYMBOL(sn_region_size);
+int sn_prom_type;	/* 0=hardware, 1=medusa/realprom, 2=medusa/fakeprom */
 
 short physical_node_map[MAX_PHYSNODE_ID];
 
@@ -354,7 +355,7 @@ void __init sn_setup(char **cmdline_p)
 
 	ia64_mark_idle = &snidle;
 
-	/* 
+	/*
 	 * For the bootcpu, we do this here. All other cpus will make the
 	 * call as part of cpu_init in slave cpu initialization.
 	 */
@@ -401,7 +402,7 @@ static void __init sn_init_pdas(char **cmdline_p)
 		nodepdaindr[cnode] =
 		    alloc_bootmem_node(NODE_DATA(cnode), sizeof(nodepda_t));
 		memset(nodepdaindr[cnode], 0, sizeof(nodepda_t));
-		memset(nodepdaindr[cnode]->phys_cpuid, -1, 
+		memset(nodepdaindr[cnode]->phys_cpuid, -1,
 		    sizeof(nodepdaindr[cnode]->phys_cpuid));
 	}
 
@@ -431,7 +432,7 @@ static void __init sn_init_pdas(char **cmdline_p)
 	}
 
 	/*
-	 * Initialize the per node hubdev.  This includes IO Nodes and 
+	 * Initialize the per node hubdev.  This includes IO Nodes and
 	 * headless/memless nodes.
 	 */
 	for (cnode = 0; cnode < numionodes; cnode++) {
@@ -459,6 +460,14 @@ void __init sn_cpu_init(void)
 	int i;
 	static int wars_have_been_checked;
 
+	if (smp_processor_id() == 0 && IS_MEDUSA()) {
+		if (ia64_sn_is_fake_prom())
+			sn_prom_type = 2;
+		else
+			sn_prom_type = 1;
+		printk("Running on medusa with %s PROM\n", (sn_prom_type == 1) ? "real" : "fake");
+	}
+
 	memset(pda, 0, sizeof(pda));
 	if (ia64_sn_get_sn_info(0, &sn_hub_info->shub2, &sn_hub_info->nasid_bitmask, &sn_hub_info->nasid_shift,
 				&sn_system_size, &sn_sharing_domain_size, &sn_partition_id,
@@ -524,7 +533,7 @@ void __init sn_cpu_init(void)
 	 */
 	{
 		u64 pio1[] = {SH1_PIO_WRITE_STATUS_0, 0, SH1_PIO_WRITE_STATUS_1, 0};
-		u64 pio2[] = {SH2_PIO_WRITE_STATUS_0, SH2_PIO_WRITE_STATUS_1, 
+		u64 pio2[] = {SH2_PIO_WRITE_STATUS_0, SH2_PIO_WRITE_STATUS_1,
 			SH2_PIO_WRITE_STATUS_2, SH2_PIO_WRITE_STATUS_3};
 		u64 *pio;
 		pio = is_shub1() ? pio1 : pio2;
@@ -556,6 +565,10 @@ static void __init scan_for_ionodes(void)
 	int nasid = 0;
 	lboard_t *brd;
 
+	/* fakeprom does not support klgraph */
+	if (IS_RUNNING_ON_FAKE_PROM())
+		return;
+
 	/* Setup ionodes with memory */
 	for (nasid = 0; nasid < MAX_PHYSNODE_ID; nasid += 2) {
 		char *klgraph_header;
@@ -567,8 +580,6 @@ static void __init scan_for_ionodes(void)
 		cnodeid = -1;
 		klgraph_header = __va(ia64_sn_get_klconfig_addr(nasid));
 		if (!klgraph_header) {
-			if (IS_RUNNING_ON_SIMULATOR())
-				continue;
 			BUG();	/* All nodes must have klconfig tables! */
 		}
 		cnodeid = nasid_to_cnodeid(nasid);
@@ -634,8 +645,8 @@ int
 nasid_slice_to_cpuid(int nasid, int slice)
 {
 	long cpu;
-	
-	for (cpu=0; cpu < NR_CPUS; cpu++) 
+
+	for (cpu=0; cpu < NR_CPUS; cpu++)
 		if (cpuid_to_nasid(cpu) == nasid &&
 					cpuid_to_slice(cpu) == slice)
 			return cpu;
-- 
cgit v1.1


From 92a582ed2757456ca9599f8b4ea2064f2154eb02 Mon Sep 17 00:00:00 2001
From: Prarit Bhargava <prarit@sgi.com>
Date: Tue, 10 May 2005 12:40:00 -0700
Subject: [IA64] sparse cleanup of TIOCA files

This patch is a sparse compile cleanup of tioca_provider.c, sn_hwperf.h, and
tioca_provider.h.  Each of these files had sparse warnings when
compiled.

Signed-off-by: Prarit Bhargava <prarit@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/pci/tioca_provider.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c
index 8dae9eb..05aa8c2f 100644
--- a/arch/ia64/sn/pci/tioca_provider.c
+++ b/arch/ia64/sn/pci/tioca_provider.c
@@ -336,7 +336,7 @@ tioca_dma_d48(struct pci_dev *pdev, uint64_t paddr)
 	if (!ct_addr)
 		return 0;
 
-	bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffff);
+	bus_addr = (dma_addr_t) (ct_addr & 0xffffffffffffUL);
 	node_upper = ct_addr >> 48;
 
 	if (node_upper > 64) {
@@ -464,7 +464,7 @@ map_return:
  * For mappings created using the direct modes (64 or 48) there are no
  * resources to release.
  */
-void
+static void
 tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir)
 {
 	int i, entry;
@@ -514,7 +514,7 @@ tioca_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir)
  * The mapping mode used is based on the devices dma_mask.  As a last resort
  * use the GART mapped mode.
  */
-uint64_t
+static uint64_t
 tioca_dma_map(struct pci_dev *pdev, uint64_t paddr, size_t byte_count)
 {
 	uint64_t mapaddr;
@@ -580,7 +580,7 @@ tioca_error_intr_handler(int irq, void *arg, struct pt_regs *pt)
  * On successful setup, returns the kernel version of tioca_common back to
  * the caller.
  */
-void *
+static void *
 tioca_bus_fixup(struct pcibus_bussoft *prom_bussoft)
 {
 	struct tioca_common *tioca_common;
-- 
cgit v1.1


From c1ffb6a8aaed45ba2dc0a8f09241b0a96f9955ba Mon Sep 17 00:00:00 2001
From: Bruce Losure <blosure@sgi.com>
Date: Tue, 24 May 2005 08:30:00 -0700
Subject: [IA64-SGI] Altix patch to tiocx, add subsys_initcall

This patch fixes an ordering issue between the init code for the
tiocx bus driver and tiocx-related device drivers.   Also adds
a new brick to the list of known FPGA bricks.

Signed-off-by: Bruce Losure <blosure@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/tiocx.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index a087b27..8716f4d 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -204,8 +204,8 @@ cx_device_register(nasid_t nasid, int part_num, int mfg_num,
 	cx_dev->dev.parent = NULL;
 	cx_dev->dev.bus = &tiocx_bus_type;
 	cx_dev->dev.release = tiocx_bus_release;
-	snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d.0x%x",
-		 cx_dev->cx_id.nasid, cx_dev->cx_id.part_num);
+	snprintf(cx_dev->dev.bus_id, BUS_ID_SIZE, "%d",
+		 cx_dev->cx_id.nasid);
 	device_register(&cx_dev->dev);
 	get_device(&cx_dev->dev);
 
@@ -236,7 +236,6 @@ int cx_device_unregister(struct cx_dev *cx_dev)
  */
 static int cx_device_reload(struct cx_dev *cx_dev)
 {
-	device_remove_file(&cx_dev->dev, &dev_attr_cxdev_control);
 	cx_device_unregister(cx_dev);
 	return cx_device_register(cx_dev->cx_id.nasid, cx_dev->cx_id.part_num,
 				  cx_dev->cx_id.mfg_num, cx_dev->hubdev);
@@ -383,6 +382,7 @@ static int is_fpga_brick(int nasid)
 	switch (tiocx_btchar_get(nasid)) {
 	case L1_BRICKTYPE_SA:
 	case L1_BRICKTYPE_ATHENA:
+	case L1_BRICKTYPE_DAYTONA:
 		return 1;
 	}
 	return 0;
@@ -409,7 +409,7 @@ static int tiocx_reload(struct cx_dev *cx_dev)
 		uint64_t cx_id;
 
 		cx_id =
-		    *(volatile int32_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) +
+		    *(volatile uint64_t *)(TIO_SWIN_BASE(nasid, TIOCX_CORELET) +
 					  WIDGET_ID);
 		part_num = XWIDGET_PART_NUM(cx_id);
 		mfg_num = XWIDGET_MFG_NUM(cx_id);
@@ -458,6 +458,10 @@ static ssize_t store_cxdev_control(struct device *dev, struct device_attribute *
 
 	switch (n) {
 	case 1:
+		tio_corelet_reset(cx_dev->cx_id.nasid, TIOCX_CORELET);
+		tiocx_reload(cx_dev);
+		break;
+	case 2:
 		tiocx_reload(cx_dev);
 		break;
 	case 3:
@@ -537,7 +541,7 @@ static void __exit tiocx_exit(void)
 	bus_unregister(&tiocx_bus_type);
 }
 
-module_init(tiocx_init);
+subsys_initcall(tiocx_init);
 module_exit(tiocx_exit);
 
 /************************************************************************
-- 
cgit v1.1


From 819c67e69c4e0062787e27dd989f5f9d00d4834e Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <davidm@hpl.hp.com>
Date: Thu, 9 Jun 2005 22:40:00 -0700
Subject: [IA64] Replace stale KDB-code with useful MAGIC_SYSRQ code in
 simserial.c

Patch makes it possible to use the "F4" function key to do
magic sysrq in the HP Ski simulator.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/hp/sim/simserial.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 786e707..7a8ae0f 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -30,6 +30,7 @@
 #include <linux/module.h>
 #include <linux/serial.h>
 #include <linux/serialP.h>
+#include <linux/sysrq.h>
 
 #include <asm/irq.h>
 #include <asm/hw_irq.h>
@@ -149,12 +150,17 @@ static  void receive_chars(struct tty_struct *tty, struct pt_regs *regs)
 				seen_esc = 2;
 				continue;
 			} else if ( seen_esc == 2 ) {
-				if ( ch == 'P' ) show_state();		/* F1 key */
-#ifdef CONFIG_KDB
-				if ( ch == 'S' )
-					kdb(KDB_REASON_KEYBOARD, 0, (kdb_eframe_t) regs);
+				if ( ch == 'P' ) /* F1 */
+					show_state();
+#ifdef CONFIG_MAGIC_SYSRQ
+				if ( ch == 'S' ) { /* F4 */
+					do
+						ch = ia64_ssc(0, 0, 0, 0,
+							      SSC_GETCHAR);
+					while (!ch);
+					handle_sysrq(ch, regs, NULL);
+				}
 #endif
-
 				seen_esc = 0;
 				continue;
 			}
-- 
cgit v1.1


From a68db763af9b676590c3fe9ec3f17bf18015eb2f Mon Sep 17 00:00:00 2001
From: Peter Chubb <peterc@gelato.unsw.edu.au>
Date: Thu, 23 Jun 2005 21:14:00 -0700
Subject: [IA64] Fix another IA64 preemption problem

There's another problem shown up by Ingo's recent patch to make
smp_processor_id() complain if it's called with preemption enabled.
local_finish_flush_tlb_mm() calls activate_context() in a situation
where it could be rescheduled to another processor.  This patch
disables preemption around the call.

Signed-off-by: Peter Chubb <peterc@gelato.unsw.edu.au>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smp.c | 3 +++
 1 file changed, 3 insertions(+)

(limited to 'arch')

diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c
index b49d4dd..0166a98 100644
--- a/arch/ia64/kernel/smp.c
+++ b/arch/ia64/kernel/smp.c
@@ -231,13 +231,16 @@ smp_flush_tlb_all (void)
 void
 smp_flush_tlb_mm (struct mm_struct *mm)
 {
+	preempt_disable();
 	/* this happens for the common case of a single-threaded fork():  */
 	if (likely(mm == current->active_mm && atomic_read(&mm->mm_users) == 1))
 	{
 		local_finish_flush_tlb_mm(mm);
+		preempt_enable();
 		return;
 	}
 
+	preempt_enable();
 	/*
 	 * We could optimize this further by using mm->cpu_vm_mask to track which CPUs
 	 * have been running in the address space.  It's not clear that this is worth the
-- 
cgit v1.1


From 06697eea4ce18870fbce82dffbb375e216145f47 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Wed, 29 Jun 2005 15:45:41 -0700
Subject: [IA64] Update tiger_defconfig

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/tiger_defconfig | 39 +++++++++++++++++++++++++++++++--------
 1 file changed, 31 insertions(+), 8 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig
index 47f4534..73454ee 100644
--- a/arch/ia64/configs/tiger_defconfig
+++ b/arch/ia64/configs/tiger_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.12-20050621
-# Tue Jun 21 14:03:24 2005
+# Linux kernel version: 2.6.13-rc1-20050629
+# Wed Jun 29 15:28:12 2005
 #
 
 #
@@ -80,18 +80,29 @@ CONFIG_MCKINLEY=y
 # CONFIG_IA64_PAGE_SIZE_8KB is not set
 CONFIG_IA64_PAGE_SIZE_16KB=y
 # CONFIG_IA64_PAGE_SIZE_64KB is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
 CONFIG_IA64_L1_CACHE_SHIFT=7
 # CONFIG_NUMA is not set
 CONFIG_VIRTUAL_MEM_MAP=y
 CONFIG_HOLES_IN_ZONE=y
 CONFIG_IA64_CYCLONE=y
 CONFIG_IOSAPIC=y
+# CONFIG_IA64_SGI_SN_XP is not set
 CONFIG_FORCE_MAX_ZONEORDER=18
 CONFIG_SMP=y
 CONFIG_NR_CPUS=4
 CONFIG_HOTPLUG_CPU=y
 # CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_HAVE_DEC_LOCK=y
 CONFIG_IA32_SUPPORT=y
 CONFIG_COMPAT=y
@@ -257,6 +268,7 @@ CONFIG_BLK_DEV_CMD64X=y
 # CONFIG_BLK_DEV_HPT366 is not set
 # CONFIG_BLK_DEV_SC1200 is not set
 CONFIG_BLK_DEV_PIIX=y
+# CONFIG_BLK_DEV_IT821X is not set
 # CONFIG_BLK_DEV_NS87415 is not set
 # CONFIG_BLK_DEV_PDC202XX_OLD is not set
 # CONFIG_BLK_DEV_PDC202XX_NEW is not set
@@ -395,6 +407,7 @@ CONFIG_UNIX=y
 CONFIG_INET=y
 CONFIG_IP_MULTICAST=y
 # CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
 # CONFIG_IP_PNP is not set
 # CONFIG_NET_IPIP is not set
 # CONFIG_NET_IPGRE is not set
@@ -407,6 +420,8 @@ CONFIG_SYN_COOKIES=y
 # CONFIG_INET_TUNNEL is not set
 CONFIG_IP_TCPDIAG=y
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
 # CONFIG_IPV6 is not set
 # CONFIG_NETFILTER is not set
 
@@ -598,9 +613,7 @@ CONFIG_GAMEPORT=m
 # CONFIG_GAMEPORT_NS558 is not set
 # CONFIG_GAMEPORT_L4 is not set
 # CONFIG_GAMEPORT_EMU10K1 is not set
-# CONFIG_GAMEPORT_VORTEX is not set
 # CONFIG_GAMEPORT_FM801 is not set
-# CONFIG_GAMEPORT_CS461X is not set
 
 #
 # Character devices
@@ -629,7 +642,6 @@ CONFIG_SERIAL_8250_NR_UARTS=6
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
 # CONFIG_SERIAL_8250_DETECT_IRQ is not set
-# CONFIG_SERIAL_8250_MULTIPORT is not set
 # CONFIG_SERIAL_8250_RSA is not set
 
 #
@@ -743,6 +755,7 @@ CONFIG_USB_DEVICEFS=y
 CONFIG_USB_EHCI_HCD=m
 # CONFIG_USB_EHCI_SPLIT_ISO is not set
 # CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
 CONFIG_USB_OHCI_HCD=m
 # CONFIG_USB_OHCI_BIG_ENDIAN is not set
 CONFIG_USB_OHCI_LITTLE_ENDIAN=y
@@ -779,9 +792,11 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_USB_HIDDEV is not set
 # CONFIG_USB_AIPTEK is not set
 # CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
 # CONFIG_USB_KBTAB is not set
 # CONFIG_USB_POWERMATE is not set
 # CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
 # CONFIG_USB_EGALAX is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
@@ -838,7 +853,7 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_USB_TEST is not set
 
 #
-# USB ATM/DSL drivers
+# USB DSL modem support
 #
 
 #
@@ -857,12 +872,17 @@ CONFIG_USB_HIDINPUT=y
 # CONFIG_INFINIBAND is not set
 
 #
+# SN Devices
+#
+
+#
 # File systems
 #
 CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 CONFIG_EXT2_FS_POSIX_ACL=y
 CONFIG_EXT2_FS_SECURITY=y
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 CONFIG_EXT3_FS_POSIX_ACL=y
@@ -922,7 +942,6 @@ CONFIG_NTFS_FS=m
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_DEVFS_FS is not set
 # CONFIG_DEVPTS_FS_XATTR is not set
 CONFIG_TMPFS=y
 CONFIG_TMPFS_XATTR=y
@@ -953,15 +972,18 @@ CONFIG_RAMFS=y
 #
 CONFIG_NFS_FS=m
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
 CONFIG_NFS_V4=y
 CONFIG_NFS_DIRECTIO=y
 CONFIG_NFSD=m
 CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
 CONFIG_NFSD_V4=y
 CONFIG_NFSD_TCP=y
 CONFIG_LOCKD=m
 CONFIG_LOCKD_V4=y
 CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=m
 CONFIG_SUNRPC_GSS=m
 CONFIG_RPCSEC_GSS_KRB5=m
@@ -1069,6 +1091,7 @@ CONFIG_LOG_BUF_SHIFT=20
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
 # CONFIG_DEBUG_FS is not set
+# CONFIG_KPROBES is not set
 CONFIG_IA64_GRANULE_16MB=y
 # CONFIG_IA64_GRANULE_64MB is not set
 # CONFIG_IA64_PRINT_HAZARDS is not set
@@ -1090,7 +1113,7 @@ CONFIG_CRYPTO=y
 # CONFIG_CRYPTO_HMAC is not set
 # CONFIG_CRYPTO_NULL is not set
 # CONFIG_CRYPTO_MD4 is not set
-CONFIG_CRYPTO_MD5=m
+CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_SHA1 is not set
 # CONFIG_CRYPTO_SHA256 is not set
 # CONFIG_CRYPTO_SHA512 is not set
-- 
cgit v1.1


From 0647d8cfd40f7974335884d73deef83feeffc2a0 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Wed, 29 Jun 2005 16:01:40 -0700
Subject: [IA64] Update zx1_defconfig

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/configs/zx1_defconfig | 166 +++++++++++++++++++++++++++++-----------
 1 file changed, 122 insertions(+), 44 deletions(-)

(limited to 'arch')

diff --git a/arch/ia64/configs/zx1_defconfig b/arch/ia64/configs/zx1_defconfig
index 21d6f9b..b7755e4 100644
--- a/arch/ia64/configs/zx1_defconfig
+++ b/arch/ia64/configs/zx1_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.10
-# Wed Dec 29 09:05:48 2004
+# Linux kernel version: 2.6.13-rc1-20050629
+# Wed Jun 29 15:31:11 2005
 #
 
 #
@@ -12,6 +12,7 @@ CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
@@ -24,23 +25,26 @@ CONFIG_BSD_PROCESS_ACCT=y
 # CONFIG_BSD_PROCESS_ACCT_V3 is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_LOG_BUF_SHIFT=17
 CONFIG_HOTPLUG=y
 CONFIG_KOBJECT_UEVENT=y
 # CONFIG_IKCONFIG is not set
+# CONFIG_CPUSETS is not set
 # CONFIG_EMBEDDED is not set
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
 CONFIG_SHMEM=y
 CONFIG_CC_ALIGN_FUNCTIONS=0
 CONFIG_CC_ALIGN_LABELS=0
 CONFIG_CC_ALIGN_LOOPS=0
 CONFIG_CC_ALIGN_JUMPS=0
 # CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
 
 #
 # Loadable module support
@@ -59,12 +63,15 @@ CONFIG_IA64=y
 CONFIG_64BIT=y
 CONFIG_MMU=y
 CONFIG_RWSEM_XCHGADD_ALGORITHM=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_TIME_INTERPOLATION=y
 CONFIG_EFI=y
 CONFIG_GENERIC_IOMAP=y
+CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 # CONFIG_IA64_GENERIC is not set
 # CONFIG_IA64_DIG is not set
 CONFIG_IA64_HP_ZX1=y
+# CONFIG_IA64_HP_ZX1_SWIOTLB is not set
 # CONFIG_IA64_SGI_SN2 is not set
 # CONFIG_IA64_HP_SIM is not set
 # CONFIG_ITANIUM is not set
@@ -73,22 +80,36 @@ CONFIG_MCKINLEY=y
 # CONFIG_IA64_PAGE_SIZE_8KB is not set
 CONFIG_IA64_PAGE_SIZE_16KB=y
 # CONFIG_IA64_PAGE_SIZE_64KB is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
 CONFIG_IA64_L1_CACHE_SHIFT=7
 # CONFIG_NUMA is not set
 CONFIG_VIRTUAL_MEM_MAP=y
+CONFIG_HOLES_IN_ZONE=y
 # CONFIG_IA64_CYCLONE is not set
 CONFIG_IOSAPIC=y
+# CONFIG_IA64_SGI_SN_XP is not set
 CONFIG_FORCE_MAX_ZONEORDER=18
 CONFIG_SMP=y
 CONFIG_NR_CPUS=16
 # CONFIG_HOTPLUG_CPU is not set
+# CONFIG_SCHED_SMT is not set
 # CONFIG_PREEMPT is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
 CONFIG_HAVE_DEC_LOCK=y
 CONFIG_IA32_SUPPORT=y
 CONFIG_COMPAT=y
 CONFIG_IA64_MCA_RECOVERY=y
 CONFIG_PERFMON=y
 CONFIG_IA64_PALINFO=y
+CONFIG_ACPI_DEALLOCATE_IRQ=y
 
 #
 # Firmware Drivers
@@ -120,6 +141,7 @@ CONFIG_ACPI_BUS=y
 CONFIG_ACPI_POWER=y
 CONFIG_ACPI_PCI=y
 CONFIG_ACPI_SYSTEM=y
+# CONFIG_ACPI_CONTAINER is not set
 
 #
 # Bus options (PCI, PCMCIA)
@@ -129,6 +151,7 @@ CONFIG_PCI_DOMAINS=y
 # CONFIG_PCI_MSI is not set
 CONFIG_PCI_LEGACY_PROC=y
 CONFIG_PCI_NAMES=y
+# CONFIG_PCI_DEBUG is not set
 
 #
 # PCI Hotplug Support
@@ -138,7 +161,6 @@ CONFIG_HOTPLUG_PCI=y
 CONFIG_HOTPLUG_PCI_ACPI=y
 # CONFIG_HOTPLUG_PCI_ACPI_IBM is not set
 # CONFIG_HOTPLUG_PCI_CPCI is not set
-# CONFIG_HOTPLUG_PCI_PCIE is not set
 # CONFIG_HOTPLUG_PCI_SHPC is not set
 
 #
@@ -147,10 +169,6 @@ CONFIG_HOTPLUG_PCI_ACPI=y
 # CONFIG_PCCARD is not set
 
 #
-# PC-card bridges
-#
-
-#
 # Device Drivers
 #
 
@@ -184,6 +202,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
 # CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
 CONFIG_BLK_DEV_LOOP=y
 # CONFIG_BLK_DEV_CRYPTOLOOP is not set
 # CONFIG_BLK_DEV_NBD is not set
@@ -203,6 +222,7 @@ CONFIG_IOSCHED_NOOP=y
 CONFIG_IOSCHED_AS=y
 CONFIG_IOSCHED_DEADLINE=y
 CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
 
 #
 # ATA/ATAPI/MFM/RLL support
@@ -246,6 +266,7 @@ CONFIG_BLK_DEV_CMD64X=y
 # CONFIG_BLK_DEV_HPT366 is not set
 # CONFIG_BLK_DEV_SC1200 is not set
 # CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
 # CONFIG_BLK_DEV_NS87415 is not set
 # CONFIG_BLK_DEV_PDC202XX_OLD is not set
 # CONFIG_BLK_DEV_PDC202XX_NEW is not set
@@ -275,6 +296,7 @@ CONFIG_CHR_DEV_OSST=y
 CONFIG_BLK_DEV_SR=y
 CONFIG_BLK_DEV_SR_VENDOR=y
 CONFIG_CHR_DEV_SG=y
+# CONFIG_CHR_DEV_SCH is not set
 
 #
 # Some SCSI devices (e.g. CD jukebox) support multiple LUNs
@@ -288,6 +310,7 @@ CONFIG_SCSI_LOGGING=y
 #
 CONFIG_SCSI_SPI_ATTRS=y
 # CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
 
 #
 # SCSI low-level drivers
@@ -303,13 +326,10 @@ CONFIG_SCSI_SPI_ATTRS=y
 # CONFIG_MEGARAID_NEWGEN is not set
 # CONFIG_MEGARAID_LEGACY is not set
 # CONFIG_SCSI_SATA is not set
-# CONFIG_SCSI_BUSLOGIC is not set
 # CONFIG_SCSI_CPQFCTS is not set
 # CONFIG_SCSI_DMX3191D is not set
-# CONFIG_SCSI_EATA is not set
 # CONFIG_SCSI_EATA_PIO is not set
 # CONFIG_SCSI_FUTURE_DOMAIN is not set
-# CONFIG_SCSI_GDTH is not set
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
@@ -319,8 +339,6 @@ CONFIG_SCSI_SYM53C8XX_DEFAULT_TAGS=16
 CONFIG_SCSI_SYM53C8XX_MAX_TAGS=64
 # CONFIG_SCSI_SYM53C8XX_IOMAPPED is not set
 # CONFIG_SCSI_IPR is not set
-# CONFIG_SCSI_PCI2000 is not set
-# CONFIG_SCSI_PCI2220I is not set
 # CONFIG_SCSI_QLOGIC_ISP is not set
 # CONFIG_SCSI_QLOGIC_FC is not set
 CONFIG_SCSI_QLOGIC_1280=y
@@ -331,7 +349,7 @@ CONFIG_SCSI_QLA2XXX=y
 # CONFIG_SCSI_QLA2300 is not set
 # CONFIG_SCSI_QLA2322 is not set
 # CONFIG_SCSI_QLA6312 is not set
-# CONFIG_SCSI_QLA6322 is not set
+# CONFIG_SCSI_LPFC is not set
 # CONFIG_SCSI_DC395x is not set
 # CONFIG_SCSI_DC390T is not set
 # CONFIG_SCSI_DEBUG is not set
@@ -344,9 +362,9 @@ CONFIG_SCSI_QLA2XXX=y
 #
 # Fusion MPT device support
 #
-CONFIG_FUSION=y
-CONFIG_FUSION_MAX_SGE=40
-# CONFIG_FUSION_CTL is not set
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
 
 #
 # IEEE 1394 (FireWire) support
@@ -368,12 +386,12 @@ CONFIG_NET=y
 #
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
-# CONFIG_NETLINK_DEV is not set
 CONFIG_UNIX=y
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
 CONFIG_IP_MULTICAST=y
 # CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
 # CONFIG_IP_PNP is not set
 # CONFIG_NET_IPIP is not set
 # CONFIG_NET_IPGRE is not set
@@ -386,6 +404,8 @@ CONFIG_IP_MULTICAST=y
 # CONFIG_INET_TUNNEL is not set
 # CONFIG_IP_TCPDIAG is not set
 # CONFIG_IP_TCPDIAG_IPV6 is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
 
 #
 # IP: Virtual Server Configuration
@@ -405,8 +425,6 @@ CONFIG_NETFILTER=y
 CONFIG_IP_NF_ARPTABLES=y
 # CONFIG_IP_NF_ARPFILTER is not set
 # CONFIG_IP_NF_ARP_MANGLE is not set
-# CONFIG_IP_NF_COMPAT_IPCHAINS is not set
-# CONFIG_IP_NF_COMPAT_IPFWADM is not set
 
 #
 # SCTP Configuration (EXPERIMENTAL)
@@ -483,7 +501,6 @@ CONFIG_NET_PCI=y
 # CONFIG_DGRS is not set
 # CONFIG_EEPRO100 is not set
 CONFIG_E100=y
-# CONFIG_E100_NAPI is not set
 # CONFIG_FEALNX is not set
 # CONFIG_NATSEMI is not set
 # CONFIG_NE2K_PCI is not set
@@ -505,9 +522,11 @@ CONFIG_E1000=y
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
 # CONFIG_R8169 is not set
+# CONFIG_SKGE is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
+# CONFIG_BNX2 is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -565,18 +584,6 @@ CONFIG_INPUT_EVDEV=y
 # CONFIG_INPUT_EVBUG is not set
 
 #
-# Input I/O drivers
-#
-# CONFIG_GAMEPORT is not set
-CONFIG_SOUND_GAMEPORT=y
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_PCIPS2 is not set
-# CONFIG_SERIO_RAW is not set
-
-#
 # Input Device Drivers
 #
 # CONFIG_INPUT_KEYBOARD is not set
@@ -586,6 +593,16 @@ CONFIG_SERIO=y
 # CONFIG_INPUT_MISC is not set
 
 #
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_PCIPS2 is not set
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
+#
 # Character devices
 #
 CONFIG_VT=y
@@ -603,7 +620,6 @@ CONFIG_SERIAL_8250_NR_UARTS=8
 CONFIG_SERIAL_8250_EXTENDED=y
 CONFIG_SERIAL_8250_SHARE_IRQ=y
 # CONFIG_SERIAL_8250_DETECT_IRQ is not set
-# CONFIG_SERIAL_8250_MULTIPORT is not set
 # CONFIG_SERIAL_8250_RSA is not set
 
 #
@@ -611,6 +627,7 @@ CONFIG_SERIAL_8250_SHARE_IRQ=y
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -644,6 +661,12 @@ CONFIG_DRM_RADEON=y
 # CONFIG_DRM_SIS is not set
 # CONFIG_RAW_DRIVER is not set
 # CONFIG_HPET is not set
+# CONFIG_HANGCHECK_TIMER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
 
 #
 # I2C support
@@ -668,6 +691,7 @@ CONFIG_I2C_ALGOPCF=y
 # CONFIG_I2C_AMD8111 is not set
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
+# CONFIG_I2C_PIIX4 is not set
 # CONFIG_I2C_ISA is not set
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
@@ -691,10 +715,14 @@ CONFIG_I2C_ALGOPCF=y
 # CONFIG_SENSORS_ADM1025 is not set
 # CONFIG_SENSORS_ADM1026 is not set
 # CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
 # CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_FSCHER is not set
+# CONFIG_SENSORS_FSCPOS is not set
 # CONFIG_SENSORS_GL518SM is not set
+# CONFIG_SENSORS_GL520SM is not set
 # CONFIG_SENSORS_IT87 is not set
 # CONFIG_SENSORS_LM63 is not set
 # CONFIG_SENSORS_LM75 is not set
@@ -705,21 +733,29 @@ CONFIG_I2C_ALGOPCF=y
 # CONFIG_SENSORS_LM85 is not set
 # CONFIG_SENSORS_LM87 is not set
 # CONFIG_SENSORS_LM90 is not set
+# CONFIG_SENSORS_LM92 is not set
 # CONFIG_SENSORS_MAX1619 is not set
 # CONFIG_SENSORS_PC87360 is not set
+# CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_SIS5595 is not set
 # CONFIG_SENSORS_SMSC47M1 is not set
 # CONFIG_SENSORS_VIA686A is not set
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83L785TS is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 
 #
 # Other I2C Chip support
 #
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
 # CONFIG_SENSORS_EEPROM is not set
 # CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_RTC8564 is not set
+# CONFIG_SENSORS_MAX6875 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -746,6 +782,7 @@ CONFIG_VIDEO_DEV=y
 #
 # Video Adapters
 #
+# CONFIG_TUNER_MULTI_I2C is not set
 # CONFIG_VIDEO_BT848 is not set
 # CONFIG_VIDEO_CPIA is not set
 # CONFIG_VIDEO_SAA5246A is not set
@@ -778,6 +815,11 @@ CONFIG_VIDEO_DEV=y
 # Graphics support
 #
 CONFIG_FB=y
+CONFIG_FB_CFB_FILLRECT=y
+CONFIG_FB_CFB_COPYAREA=y
+CONFIG_FB_CFB_IMAGEBLIT=y
+CONFIG_FB_SOFT_CURSOR=y
+# CONFIG_FB_MACMODES is not set
 CONFIG_FB_MODE_HELPERS=y
 # CONFIG_FB_TILEBLITTING is not set
 # CONFIG_FB_CIRRUS is not set
@@ -785,6 +827,7 @@ CONFIG_FB_MODE_HELPERS=y
 # CONFIG_FB_CYBER2000 is not set
 # CONFIG_FB_ASILIANT is not set
 # CONFIG_FB_IMSTT is not set
+# CONFIG_FB_NVIDIA is not set
 # CONFIG_FB_RIVA is not set
 # CONFIG_FB_MATROX is not set
 # CONFIG_FB_RADEON_OLD is not set
@@ -801,6 +844,7 @@ CONFIG_FB_RADEON_DEBUG=y
 # CONFIG_FB_VOODOO1 is not set
 # CONFIG_FB_TRIDENT is not set
 # CONFIG_FB_PM3 is not set
+# CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
 
 #
@@ -820,6 +864,7 @@ CONFIG_LOGO=y
 # CONFIG_LOGO_LINUX_MONO is not set
 # CONFIG_LOGO_LINUX_VGA16 is not set
 CONFIG_LOGO_LINUX_CLUT224=y
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
 
 #
 # Sound
@@ -869,6 +914,8 @@ CONFIG_SND_AC97_CODEC=y
 # CONFIG_SND_CS46XX is not set
 # CONFIG_SND_CS4281 is not set
 # CONFIG_SND_EMU10K1 is not set
+# CONFIG_SND_EMU10K1X is not set
+# CONFIG_SND_CA0106 is not set
 # CONFIG_SND_KORG1212 is not set
 # CONFIG_SND_MIXART is not set
 # CONFIG_SND_NM256 is not set
@@ -876,6 +923,7 @@ CONFIG_SND_AC97_CODEC=y
 # CONFIG_SND_RME96 is not set
 # CONFIG_SND_RME9652 is not set
 # CONFIG_SND_HDSP is not set
+# CONFIG_SND_HDSPM is not set
 # CONFIG_SND_TRIDENT is not set
 # CONFIG_SND_YMFPCI is not set
 # CONFIG_SND_ALS4000 is not set
@@ -893,13 +941,14 @@ CONFIG_SND_FM801_TEA575X=y
 # CONFIG_SND_INTEL8X0M is not set
 # CONFIG_SND_SONICVIBES is not set
 # CONFIG_SND_VIA82XX is not set
+# CONFIG_SND_VIA82XX_MODEM is not set
 # CONFIG_SND_VX222 is not set
+# CONFIG_SND_HDA_INTEL is not set
 
 #
 # USB devices
 #
 # CONFIG_SND_USB_AUDIO is not set
-# CONFIG_SND_USB_USX2Y is not set
 
 #
 # Open Sound System
@@ -909,6 +958,8 @@ CONFIG_SND_FM801_TEA575X=y
 #
 # USB support
 #
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
 CONFIG_USB=y
 # CONFIG_USB_DEBUG is not set
 
@@ -920,8 +971,6 @@ CONFIG_USB_BANDWIDTH=y
 # CONFIG_USB_DYNAMIC_MINORS is not set
 # CONFIG_USB_SUSPEND is not set
 # CONFIG_USB_OTG is not set
-CONFIG_USB_ARCH_HAS_HCD=y
-CONFIG_USB_ARCH_HAS_OHCI=y
 
 #
 # USB Host Controller Drivers
@@ -929,7 +978,10 @@ CONFIG_USB_ARCH_HAS_OHCI=y
 CONFIG_USB_EHCI_HCD=y
 # CONFIG_USB_EHCI_SPLIT_ISO is not set
 # CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
 CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
 CONFIG_USB_UHCI_HCD=y
 # CONFIG_USB_SL811_HCD is not set
 
@@ -947,12 +999,11 @@ CONFIG_USB_UHCI_HCD=y
 #
 CONFIG_USB_STORAGE=y
 # CONFIG_USB_STORAGE_DEBUG is not set
-# CONFIG_USB_STORAGE_RW_DETECT is not set
 # CONFIG_USB_STORAGE_DATAFAB is not set
 # CONFIG_USB_STORAGE_FREECOM is not set
 # CONFIG_USB_STORAGE_ISD200 is not set
 # CONFIG_USB_STORAGE_DPCM is not set
-# CONFIG_USB_STORAGE_HP8200e is not set
+# CONFIG_USB_STORAGE_USBAT is not set
 # CONFIG_USB_STORAGE_SDDR09 is not set
 # CONFIG_USB_STORAGE_SDDR55 is not set
 # CONFIG_USB_STORAGE_JUMPSHOT is not set
@@ -966,9 +1017,11 @@ CONFIG_USB_HIDINPUT=y
 CONFIG_USB_HIDDEV=y
 # CONFIG_USB_AIPTEK is not set
 # CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
 # CONFIG_USB_KBTAB is not set
 # CONFIG_USB_POWERMATE is not set
 # CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
 # CONFIG_USB_EGALAX is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
@@ -978,7 +1031,6 @@ CONFIG_USB_HIDDEV=y
 #
 # CONFIG_USB_MDC800 is not set
 # CONFIG_USB_MICROTEK is not set
-# CONFIG_USB_HPUSBSCSI is not set
 
 #
 # USB Multimedia devices
@@ -992,6 +1044,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_SE401 is not set
 # CONFIG_USB_SN9C102 is not set
 # CONFIG_USB_STV680 is not set
+# CONFIG_USB_PWC is not set
 
 #
 # USB Network Adapters
@@ -1001,6 +1054,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_PEGASUS is not set
 # CONFIG_USB_RTL8150 is not set
 # CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=y
 
 #
 # USB port drivers
@@ -1016,7 +1070,6 @@ CONFIG_USB_HIDDEV=y
 #
 # CONFIG_USB_EMI62 is not set
 # CONFIG_USB_EMI26 is not set
-# CONFIG_USB_TIGL is not set
 # CONFIG_USB_AUERSWALD is not set
 # CONFIG_USB_RIO500 is not set
 # CONFIG_USB_LEGOTOWER is not set
@@ -1025,9 +1078,11 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_CYTHERM is not set
 # CONFIG_USB_PHIDGETKIT is not set
 # CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_SISUSBVGA is not set
 
 #
-# USB ATM/DSL drivers
+# USB DSL modem support
 #
 
 #
@@ -1041,12 +1096,22 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_MMC is not set
 
 #
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
 # File systems
 #
 CONFIG_EXT2_FS=y
 CONFIG_EXT2_FS_XATTR=y
 # CONFIG_EXT2_FS_POSIX_ACL is not set
 # CONFIG_EXT2_FS_SECURITY is not set
+# CONFIG_EXT2_FS_XIP is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT3_FS_XATTR=y
 # CONFIG_EXT3_FS_POSIX_ACL is not set
@@ -1056,6 +1121,10 @@ CONFIG_JBD=y
 CONFIG_FS_MBCACHE=y
 # CONFIG_REISERFS_FS is not set
 # CONFIG_JFS_FS is not set
+
+#
+# XFS support
+#
 # CONFIG_XFS_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_ROMFS_FS is not set
@@ -1089,7 +1158,6 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_DEVFS_FS is not set
 # CONFIG_DEVPTS_FS_XATTR is not set
 CONFIG_TMPFS=y
 CONFIG_TMPFS_XATTR=y
@@ -1120,15 +1188,18 @@ CONFIG_RAMFS=y
 #
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
 CONFIG_NFS_V4=y
 # CONFIG_NFS_DIRECTIO is not set
 CONFIG_NFSD=y
 CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
 # CONFIG_NFSD_V4 is not set
 # CONFIG_NFSD_TCP is not set
 CONFIG_LOCKD=y
 CONFIG_LOCKD_V4=y
 CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
 CONFIG_SUNRPC=y
 CONFIG_SUNRPC_GSS=y
 CONFIG_RPCSEC_GSS_KRB5=y
@@ -1209,6 +1280,8 @@ CONFIG_NLS_UTF8=y
 # CONFIG_CRC_CCITT is not set
 CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
 
 #
 # Profiling support
@@ -1218,14 +1291,18 @@ CONFIG_CRC32=y
 #
 # Kernel hacking
 #
+# CONFIG_PRINTK_TIME is not set
 CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
+CONFIG_LOG_BUF_SHIFT=17
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
+CONFIG_KPROBES=y
 CONFIG_IA64_GRANULE_16MB=y
 # CONFIG_IA64_GRANULE_64MB is not set
 CONFIG_IA64_PRINT_HAZARDS=y
@@ -1252,6 +1329,7 @@ CONFIG_CRYPTO_MD5=y
 # CONFIG_CRYPTO_SHA256 is not set
 # CONFIG_CRYPTO_SHA512 is not set
 # CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
 CONFIG_CRYPTO_DES=y
 # CONFIG_CRYPTO_BLOWFISH is not set
 # CONFIG_CRYPTO_TWOFISH is not set
-- 
cgit v1.1