Sophie: kernel-2.6.18-128.1.10.el5 src

kernel-2.6.18-128.1.10.el5.src.rpm

From: AMEET M. PARANJAPE <aparanja@redhat.com>
Date: Wed, 29 Oct 2008 14:53:09 -0400
Subject: [ppc64] clock_gettime is not incrementing nanoseconds
Message-id: 20081029185243.31108.40899.sendpatchset@squad5-lp1.lab.bos.redhat.com
O-Subject: [PATCH RHEL5.3 BZ469073] clock_gettime is not incrementing nanoseconds in RHEL5.2 on power 64
Bugzilla: 469073
RH-Acked-by: David Howells <dhowells@redhat.com>
RH-Acked-by: Brian Maly <bmaly@redhat.com>

RHBZ#:
======
https://bugzilla.redhat.com/show_bug.cgi?id=469073

Description:
===========
This changes the code in the Virtual Dynamically-linked Shared Object (VDSO) so
that clock_gettime has nanosecond resolution for CLOCK_REALTIME and
CLOCK_MONOTONIC.  It adds a field to the VDSO data page containing the xtime
value corresponding to tb_orig_stamp so that we can avoid the possibility of
seeing time go backwards slightly (because stamp_xsec only has 2^-20 second
resolution, there is a possibility that a nanosecond-resolution value computed
using a new value of stamp_xsec might be slightly less than a value computed
previously using an old value).

This fixes both 32-bit and 64-bit.

RHEL Version Found:
================
RHEL 5.2

kABI Status:
============
No symbols were harmed.

Brew:
=====
Built on all platforms.
http://brewweb.devel.redhat.com/brew/taskinfo?taskID=1545373

Upstream Status:
================
This patch has been submitted by Paul Mackerras to the linux community.  I will
update the RKML post as soon as the GIT ID is available.

Test Status:
============
Compile attached test program in the Bugzilla.  The Count in output indicates
the number of times clock_gettime() was called without the result incrementing.
The output timer values show that the value is only incrementing microseconds,
not nanoseconds.

Running this testcase again with the patch applied does not show this problem.

===============================================================
Ameet Paranjape 978-392-3903 ext 23903
IBM on-site partner

Proposed Patch:
===============

diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index e85b792..7f3f54d 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -264,6 +264,7 @@ int main(void)
 	DEFINE(CFG_SYSCALL_MAP32, offsetof(struct vdso_data, syscall_map_32));
 	DEFINE(WTOM_CLOCK_SEC, offsetof(struct vdso_data, wtom_clock_sec));
 	DEFINE(WTOM_CLOCK_NSEC, offsetof(struct vdso_data, wtom_clock_nsec));
+	DEFINE(STAMP_XTIME, offsetof(struct vdso_data, stamp_xtime));
 #ifdef CONFIG_PPC64
 	DEFINE(CFG_SYSCALL_MAP64, offsetof(struct vdso_data, syscall_map_64));
 	DEFINE(TVAL64_TV_SEC, offsetof(struct timeval, tv_sec));
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 8108a7d..310f23f 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -477,6 +477,7 @@ static inline void update_gtod(u64 new_tb_stamp, u64 new_stamp_xsec,
 	vdso_data->tb_to_xs = new_tb_to_xs;
 	vdso_data->wtom_clock_sec = wall_to_monotonic.tv_sec;
 	vdso_data->wtom_clock_nsec = wall_to_monotonic.tv_nsec;
+	vdso_data->stamp_xtime = xtime;
 	smp_wmb();
 	++(vdso_data->tb_update_count);
 }
diff --git a/arch/powerpc/kernel/vdso32/gettimeofday.S b/arch/powerpc/kernel/vdso32/gettimeofday.S
index 05909f7..cf75705 100644
--- a/arch/powerpc/kernel/vdso32/gettimeofday.S
+++ b/arch/powerpc/kernel/vdso32/gettimeofday.S
@@ -16,6 +16,13 @@
 #include <asm/asm-offsets.h>
 #include <asm/unistd.h>
 
+/* Offset for the low 32-bit part of a field of long type */
+#ifdef CONFIG_PPC64
+#define LOPART	4
+#else
+#define LOPART	0
+#endif
+
 	.text
 /*
  * Exact prototype of gettimeofday
@@ -88,101 +95,53 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 
 	mflr	r12			/* r12 saves lr */
   .cfi_register lr,r12
-	mr	r10,r3			/* r10 saves id */
 	mr	r11,r4			/* r11 saves tp */
 	bl	__get_datapage@local	/* get data page */
 	mr	r9,r3			/* datapage ptr in r9 */
-	beq	cr1,50f			/* if monotonic -> jump there */
-
-	/*
-	 * CLOCK_REALTIME
-	 */
-
-	bl	__do_get_xsec@local	/* get xsec from tb & kernel */
-	bne-	98f			/* out of line -> do syscall */
 
-	/* seconds are xsec >> 20 */
-	rlwinm	r5,r4,12,20,31
-	rlwimi	r5,r3,12,0,19
-	stw	r5,TSPC32_TV_SEC(r11)
-
-	/* get remaining xsec and convert to nsec. we scale
-	 * up remaining xsec by 12 bits and get the top 32 bits
-	 * of the multiplication, then we multiply by 1000
-	 */
-	rlwinm	r5,r4,12,0,19
-	lis	r6,1000000@h
-	ori	r6,r6,1000000@l
-	mulhwu	r5,r5,r6
-	mulli	r5,r5,1000
-	stw	r5,TSPC32_TV_NSEC(r11)
-	mtlr	r12
-	crclr	cr0*4+so
-	li	r3,0
-	blr
+50:	bl	__do_get_tspec@local	/* get sec/nsec from tb & kernel */
+	bne	cr1,80f			/* not monotonic -> all done */
 
 	/*
 	 * CLOCK_MONOTONIC
 	 */
 
-50:	bl	__do_get_xsec@local	/* get xsec from tb & kernel */
-	bne-	98f			/* out of line -> do syscall */
-
-	/* seconds are xsec >> 20 */
-	rlwinm	r6,r4,12,20,31
-	rlwimi	r6,r3,12,0,19
-
-	/* get remaining xsec and convert to nsec. we scale
-	 * up remaining xsec by 12 bits and get the top 32 bits
-	 * of the multiplication, then we multiply by 1000
-	 */
-	rlwinm	r7,r4,12,0,19
-	lis	r5,1000000@h
-	ori	r5,r5,1000000@l
-	mulhwu	r7,r7,r5
-	mulli	r7,r7,1000
-
 	/* now we must fixup using wall to monotonic. We need to snapshot
 	 * that value and do the counter trick again. Fortunately, we still
 	 * have the counter value in r8 that was returned by __do_get_xsec.
-	 * At this point, r6,r7 contain our sec/nsec values, r3,r4 and r5
-	 * can be used
+	 * At this point, r3,r4 contain our sec/nsec values, r5 and r6
+	 * can be used, r7 contains NSEC_PER_SEC.
 	 */
 
-	lwz	r3,WTOM_CLOCK_SEC(r9)
-	lwz	r4,WTOM_CLOCK_NSEC(r9)
+	lwz	r5,WTOM_CLOCK_SEC(r9)
+	lwz	r6,WTOM_CLOCK_NSEC(r9)
 
-	/* We now have our result in r3,r4. We create a fake dependency
-	 * on that result and re-check the counter
+	/* We now have our offset in r5,r6. We create a fake dependency
+	 * on that value and re-check the counter
 	 */
-	or	r5,r4,r3
-	xor	r0,r5,r5
+	or	r0,r6,r5
+	xor	r0,r0,r0
 	add	r9,r9,r0
-#ifdef CONFIG_PPC64
-	lwz	r0,(CFG_TB_UPDATE_COUNT+4)(r9)
-#else
-	lwz	r0,(CFG_TB_UPDATE_COUNT)(r9)
-#endif
+	lwz	r0,(CFG_TB_UPDATE_COUNT+LOPART)(r9)
         cmpl    cr0,r8,r0		/* check if updated */
 	bne-	50b
 
-	/* Calculate and store result. Note that this mimmics the C code,
+	/* Calculate and store result. Note that this mimics the C code,
 	 * which may cause funny results if nsec goes negative... is that
 	 * possible at all ?
 	 */
-	add	r3,r3,r6
-	add	r4,r4,r7
-	lis	r5,NSEC_PER_SEC@h
-	ori	r5,r5,NSEC_PER_SEC@l
-	cmpl	cr0,r4,r5
-	cmpli	cr1,r4,0
+	add	r3,r3,r5
+	add	r4,r4,r6
+	cmpw	cr0,r4,r7
+	cmpwi	cr1,r4,0
 	blt	1f
-	subf	r4,r5,r4
+	subf	r4,r7,r4
 	addi	r3,r3,1
-1:	bge	cr1,1f
+1:	bge	cr1,80f
 	addi	r3,r3,-1
-	add	r4,r4,r5
-1:	stw	r3,TSPC32_TV_SEC(r11)
+	add	r4,r4,r7
+
+80:	stw	r3,TSPC32_TV_SEC(r11)
 	stw	r4,TSPC32_TV_NSEC(r11)
 
 	mtlr	r12
@@ -193,10 +152,6 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	/*
 	 * syscall fallback
 	 */
-98:
-	mtlr	r12
-	mr	r3,r10
-	mr	r4,r11
 99:
 	li	r0,__NR_clock_gettime
 	sc
@@ -320,3 +275,98 @@ __do_get_xsec:
 	 */
 3:	blr
   .cfi_endproc
+
+/*
+ * This is the core of clock_gettime(), it returns the current
+ * time in seconds and nanoseconds in r3 and r4.
+ * It expects the datapage ptr in r9 and doesn't clobber it.
+ * It clobbers r0, r5, r6, r10 and returns NSEC_PER_SEC in r7.
+ * On return, r8 contains the counter value that can be reused.
+ * This clobbers cr0 but not any other cr field.
+ */
+__do_get_tspec:
+  .cfi_startproc
+	/* Check for update count & load values. We use the low
+	 * order 32 bits of the update count
+	 */
+1:	lwz	r8,(CFG_TB_UPDATE_COUNT+LOPART)(r9)
+	andi.	r0,r8,1			/* pending update ? loop */
+	bne-	1b
+	xor	r0,r8,r8		/* create dependency */
+	add	r9,r9,r0
+
+	/* Load orig stamp (offset to TB) */
+	lwz	r5,CFG_TB_ORIG_STAMP(r9)
+	lwz	r6,(CFG_TB_ORIG_STAMP+4)(r9)
+
+	/* Get a stable TB value */
+2:	mftbu	r3
+	mftbl	r4
+	mftbu	r0
+	cmpl	cr0,r3,r0
+	bne-	2b
+
+	/* Subtract tb orig stamp and shift left 12 bits.
+	 */
+	subfc	r7,r6,r4
+	subfe	r0,r5,r3
+	slwi	r0,r0,12
+	rlwimi.	r0,r7,12,20,31
+	slwi	r7,r7,12
+
+	/* Load scale factor & do multiplication */
+	lwz	r5,CFG_TB_TO_XS(r9)	/* load values */
+	lwz	r6,(CFG_TB_TO_XS+4)(r9)
+	mulhwu	r3,r7,r6
+	mullw	r10,r7,r5
+	mulhwu	r4,r7,r5
+	addc	r10,r3,r10
+	li	r3,0
+
+	beq+	4f			/* skip high part computation if 0 */
+	mulhwu	r3,r0,r5
+	mullw	r7,r0,r5
+	mulhwu	r5,r0,r6
+	mullw	r6,r0,r6
+	adde	r4,r4,r7
+	addze	r3,r3
+	addc	r4,r4,r5
+	addze	r3,r3
+	addc	r10,r10,r6
+
+4:	addze	r4,r4			/* add in carry */
+	lis	r7,NSEC_PER_SEC@h
+	ori	r7,r7,NSEC_PER_SEC@l
+	mulhwu	r4,r4,r7		/* convert to nanoseconds */
+
+	/* At this point, we have seconds & nanoseconds since the xtime
+	 * stamp in r3+CA and r4.  Load & add the xtime stamp.
+	 */
+#ifdef CONFIG_PPC64
+	lwz	r5,STAMP_XTIME+TSPC64_TV_SEC+LOPART(r9)
+	lwz	r6,STAMP_XTIME+TSPC64_TV_NSEC+LOPART(r9)
+#else
+	lwz	r5,STAMP_XTIME+TSPC32_TV_SEC(r9)
+	lwz	r6,STAMP_XTIME+TSPC32_TV_NSEC(r9)
+#endif
+	add	r4,r4,r6
+	adde	r3,r3,r5
+
+	/* We now have our result in r3,r4. We create a fake dependency
+	 * on that result and re-check the counter
+	 */
+	or	r6,r4,r3
+	xor	r0,r6,r6
+	add	r9,r9,r0
+	lwz	r0,(CFG_TB_UPDATE_COUNT+LOPART)(r9)
+        cmpl    cr0,r8,r0		/* check if updated */
+	bne-	1b
+
+	/* check for nanosecond overflow and adjust if necessary */
+	cmpw	r4,r7
+	bltlr				/* all done if no overflow */
+	subf	r4,r7,r4		/* adjust if overflow */
+	addi	r3,r3,1
+
+	blr
+  .cfi_endproc
diff --git a/arch/powerpc/kernel/vdso64/gettimeofday.S b/arch/powerpc/kernel/vdso64/gettimeofday.S
index 40ffd9b..bc3e9bf 100644
--- a/arch/powerpc/kernel/vdso64/gettimeofday.S
+++ b/arch/powerpc/kernel/vdso64/gettimeofday.S
@@ -74,90 +74,49 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 
 	mflr	r12			/* r12 saves lr */
   .cfi_register lr,r12
-	mr	r10,r3			/* r10 saves id */
 	mr	r11,r4			/* r11 saves tp */
 	bl	V_LOCAL_FUNC(__get_datapage)	/* get data page */
-	beq	cr1,50f			/* if monotonic -> jump there */
-
-	/*
-	 * CLOCK_REALTIME
-	 */
-
-	bl	V_LOCAL_FUNC(__do_get_xsec)	/* get xsec from tb & kernel */
-
-	lis     r7,15			/* r7 = 1000000 = USEC_PER_SEC */
-	ori     r7,r7,16960
-	rldicl  r5,r4,44,20		/* r5 = sec = xsec / XSEC_PER_SEC */
-	rldicr  r6,r5,20,43		/* r6 = sec * XSEC_PER_SEC */
-	std	r5,TSPC64_TV_SEC(r11)	/* store sec in tv */
-	subf	r0,r6,r4		/* r0 = xsec = (xsec - r6) */
-	mulld   r0,r0,r7		/* usec = (xsec * USEC_PER_SEC) /
-					 * XSEC_PER_SEC
-					 */
-	rldicl  r0,r0,44,20
-	mulli	r0,r0,1000		/* nsec = usec * 1000 */
-	std	r0,TSPC64_TV_NSEC(r11)	/* store nsec in tp */
-
-	mtlr	r12
-	crclr	cr0*4+so
-	li	r3,0
-	blr
+50:	bl	V_LOCAL_FUNC(__do_get_tspec)	/* get time from tb & kernel */
+	bne	cr1,80f			/* if not monotonic, all done */
 
 	/*
 	 * CLOCK_MONOTONIC
 	 */
 
-50:	bl	V_LOCAL_FUNC(__do_get_xsec)	/* get xsec from tb & kernel */
-
-	lis     r7,15			/* r7 = 1000000 = USEC_PER_SEC */
-	ori     r7,r7,16960
-	rldicl  r5,r4,44,20		/* r5 = sec = xsec / XSEC_PER_SEC */
-	rldicr  r6,r5,20,43		/* r6 = sec * XSEC_PER_SEC */
-	subf	r0,r6,r4		/* r0 = xsec = (xsec - r6) */
-	mulld   r0,r0,r7		/* usec = (xsec * USEC_PER_SEC) /
-					 * XSEC_PER_SEC
-					 */
-	rldicl  r6,r0,44,20
-	mulli	r6,r6,1000		/* nsec = usec * 1000 */
-
 	/* now we must fixup using wall to monotonic. We need to snapshot
 	 * that value and do the counter trick again. Fortunately, we still
-	 * have the counter value in r8 that was returned by __do_get_xsec.
-	 * At this point, r5,r6 contain our sec/nsec values.
-	 * can be used
+	 * have the counter value in r8 that was returned by __do_get_tspec.
+	 * At this point, r4,r5 contain our sec/nsec values.
 	 */
 
-	lwa	r4,WTOM_CLOCK_SEC(r3)
-	lwa	r7,WTOM_CLOCK_NSEC(r3)
+	lwa	r6,WTOM_CLOCK_SEC(r3)
+	lwa	r9,WTOM_CLOCK_NSEC(r3)
 
-	/* We now have our result in r4,r7. We create a fake dependency
+	/* We now have our result in r6,r9. We create a fake dependency
 	 * on that result and re-check the counter
 	 */
-	or	r9,r4,r7
-	xor	r0,r9,r9
+	or	r0,r6,r9
+	xor	r0,r0,r0
 	add	r3,r3,r0
 	ld	r0,CFG_TB_UPDATE_COUNT(r3)
         cmpld   cr0,r0,r8		/* check if updated */
 	bne-	50b
 
-	/* Calculate and store result. Note that this mimmics the C code,
-	 * which may cause funny results if nsec goes negative... is that
-	 * possible at all ?
+	/* Add wall->monotonic offset and check for overflow or underflow.
 	 */
-	add	r4,r4,r5
-	add	r7,r7,r6
-	lis	r9,NSEC_PER_SEC@h
-	ori	r9,r9,NSEC_PER_SEC@l
-	cmpl	cr0,r7,r9
-	cmpli	cr1,r7,0
+	add	r4,r4,r6
+	add	r5,r5,r9
+	cmpd	cr0,r5,r7
+	cmpdi	cr1,r5,0
 	blt	1f
-	subf	r7,r9,r7
+	subf	r5,r7,r5
 	addi	r4,r4,1
-1:	bge	cr1,1f
+1:	bge	cr1,80f
 	addi	r4,r4,-1
-	add	r7,r7,r9
-1:	std	r4,TSPC64_TV_SEC(r11)
-	std	r7,TSPC64_TV_NSEC(r11)
+	add	r5,r5,r7
+
+80:	std	r4,TSPC64_TV_SEC(r11)
+	std	r5,TSPC64_TV_NSEC(r11)
 
 	mtlr	r12
 	crclr	cr0*4+so
@@ -167,10 +126,6 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	/*
 	 * syscall fallback
 	 */
-98:
-	mtlr	r12
-	mr	r3,r10
-	mr	r4,r11
 99:
 	li	r0,__NR_clock_gettime
 	sc
@@ -252,3 +207,59 @@ V_FUNCTION_BEGIN(__do_get_xsec)
 	blr
   .cfi_endproc
 V_FUNCTION_END(__do_get_xsec)
+
+/*
+ * This is the core of clock_gettime(), it returns the current
+ * time in seconds and nanoseconds in r4 and r5.
+ * It expects the datapage ptr in r3 and doesn't clobber it.
+ * It clobbers r0 and r6 and returns NSEC_PER_SEC in r7.
+ * On return, r8 contains the counter value that can be reused.
+ * This clobbers cr0 but not any other cr field.
+ */
+V_FUNCTION_BEGIN(__do_get_tspec)
+  .cfi_startproc
+	/* check for update count & load values */
+1:	ld	r8,CFG_TB_UPDATE_COUNT(r3)
+	andi.	r0,r8,1			/* pending update ? loop */
+	bne-	1b
+	xor	r0,r8,r8		/* create dependency */
+	add	r3,r3,r0
+
+	/* Get TB & offset it. We use the MFTB macro which will generate
+	 * workaround code for Cell.
+	 */
+	MFTB(r7)
+	ld	r9,CFG_TB_ORIG_STAMP(r3)
+	subf	r7,r9,r7
+
+	/* Scale result */
+	ld	r5,CFG_TB_TO_XS(r3)
+	sldi	r7,r7,12		/* compute time since stamp_xtime */
+	mulhdu	r6,r7,r5		/* in units of 2^-32 seconds */
+
+	/* Add stamp since epoch */
+	ld	r4,STAMP_XTIME+TSPC64_TV_SEC(r3)
+	ld	r5,STAMP_XTIME+TSPC64_TV_NSEC(r3)
+	or	r0,r4,r5
+	or	r0,r0,r6
+	xor	r0,r0,r0
+	add	r3,r3,r0
+	ld	r0,CFG_TB_UPDATE_COUNT(r3)
+	cmpld   r0,r8			/* check if updated */
+	bne-	1b			/* reload if so */
+
+	/* convert to seconds & nanoseconds and add to stamp */
+	lis	r7,NSEC_PER_SEC@h
+	ori	r7,r7,NSEC_PER_SEC@l
+	mulhwu	r0,r6,r7		/* compute nanoseconds and */
+	srdi	r6,r6,32		/* seconds since stamp_xtime */
+	clrldi	r0,r0,32
+	add	r5,r5,r0		/* add nanoseconds together */
+	cmpd	r5,r7			/* overflow? */
+	add	r4,r4,r6
+	bltlr				/* all done if no overflow */
+	subf	r5,r7,r5		/* if overflow, adjust */
+	addi	r4,r4,1
+	blr
+  .cfi_endproc
+V_FUNCTION_END(__do_get_ysec)
diff --git a/include/asm-powerpc/vdso_datapage.h b/include/asm-powerpc/vdso_datapage.h
index 8a94f0e..3bf61d7 100644
--- a/include/asm-powerpc/vdso_datapage.h
+++ b/include/asm-powerpc/vdso_datapage.h
@@ -39,6 +39,7 @@
 #ifndef __ASSEMBLY__
 
 #include <linux/unistd.h>
+#include <linux/time.h>
 
 #define SYSCALL_MAP_SIZE      ((__NR_syscalls + 31) / 32)
 
@@ -79,6 +80,7 @@ struct vdso_data {
 	 */
 	__s32 wtom_clock_sec;			/* Wall to monotonic clock */
 	__s32 wtom_clock_nsec;
+	struct timespec stamp_xtime;	/* xtime value for tb_orig_stamp */
    	__u32 syscall_map_64[SYSCALL_MAP_SIZE]; /* map of syscalls  */
    	__u32 syscall_map_32[SYSCALL_MAP_SIZE]; /* map of syscalls */
 };
@@ -98,6 +100,7 @@ struct vdso_data {
 	__u32 tz_dsttime;		/* Type of dst correction	0x5C */
 	__s32 wtom_clock_sec;			/* Wall to monotonic clock */
 	__s32 wtom_clock_nsec;
+	struct timespec stamp_xtime;	/* xtime value for tb_orig_stamp */
    	__u32 syscall_map_32[SYSCALL_MAP_SIZE]; /* map of syscalls */
 };