[S390] introduce vdso on s390

Add a vdso to speed up gettimeofday and clock_getres/clock_gettime for
CLOCK_REALTIME/CLOCK_MONOTONIC.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

7 files changed, 467 insertions, 0 deletions

diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile
new file mode 100644
index 0000000..ca78ad6
--- /dev/null
+++ b/arch/s390/kernel/vdso32/Makefile
@@ -0,0 +1,55 @@
+# List of files in the vdso, has to be asm only for now
+
+obj-vdso32 = gettimeofday.o clock_getres.o clock_gettime.o note.o
+
+# Build rules
+
+targets := $(obj-vdso32) vdso32.so vdso32.so.dbg
+obj-vdso32 := $(addprefix $(obj)/, $(obj-vdso32))
+
+KBUILD_AFLAGS_31 := $(filter-out -m64,$(KBUILD_AFLAGS))
+KBUILD_AFLAGS_31 += -m31 -s
+
+KBUILD_CFLAGS_31 := $(filter-out -m64,$(KBUILD_CFLAGS))
+KBUILD_CFLAGS_31 += -m31 -fPIC -shared -fno-common -fno-builtin
+KBUILD_CFLAGS_31 += -nostdlib -Wl,-soname=linux-vdso32.so.1 \
+			$(call ld-option, -Wl$(comma)--hash-style=sysv)
+
+$(targets:%=$(obj)/%.dbg): KBUILD_CFLAGS = $(KBUILD_CFLAGS_31)
+$(targets:%=$(obj)/%.dbg): KBUILD_AFLAGS = $(KBUILD_AFLAGS_31)
+
+obj-y += vdso32_wrapper.o
+extra-y += vdso32.lds
+CPPFLAGS_vdso32.lds += -P -C -U$(ARCH)
+
+# Force dependency (incbin is bad)
+$(obj)/vdso32_wrapper.o : $(obj)/vdso32.so
+
+# link rule for the .so file, .lds has to be first
+$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32)
+	$(call if_changed,vdso32ld)
+
+# strip rule for the .so file
+$(obj)/%.so: OBJCOPYFLAGS := -S
+$(obj)/%.so: $(obj)/%.so.dbg FORCE
+	$(call if_changed,objcopy)
+
+# assembly rules for the .S files
+$(obj-vdso32): %.o: %.S
+	$(call if_changed_dep,vdso32as)
+
+# actual build commands
+quiet_cmd_vdso32ld = VDSO32L $@
+      cmd_vdso32ld = $(CC) $(c_flags) -Wl,-T $^ -o $@
+quiet_cmd_vdso32as = VDSO32A $@
+      cmd_vdso32as = $(CC) $(a_flags) -c -o $@ $<
+
+# install commands for the unstripped file
+quiet_cmd_vdso_install = INSTALL $@
+      cmd_vdso_install = cp $(obj)/$@.dbg $(MODLIB)/vdso/$@
+
+vdso32.so: $(obj)/vdso32.so.dbg
+	@mkdir -p $(MODLIB)/vdso
+	$(call cmd,vdso_install)
+
+vdso_install: vdso32.so
diff --git a/arch/s390/kernel/vdso32/clock_getres.S b/arch/s390/kernel/vdso32/clock_getres.S
new file mode 100644
index 0000000..9532c4e
--- /dev/null
+++ b/arch/s390/kernel/vdso32/clock_getres.S
@@ -0,0 +1,39 @@
+/*
+ * Userland implementation of clock_getres() for 32 bits processes in a
+ * s390 kernel for use in the vDSO
+ *
+ *  Copyright IBM Corp. 2008
+ *  Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+#include <asm/vdso.h>
+#include <asm/asm-offsets.h>
+#include <asm/unistd.h>
+
+	.text
+	.align 4
+	.globl __kernel_clock_getres
+	.type  __kernel_clock_getres,@function
+__kernel_clock_getres:
+	.cfi_startproc
+	chi	%r2,CLOCK_REALTIME
+	je	0f
+	chi	%r2,CLOCK_MONOTONIC
+	jne	3f
+0:	ltr	%r3,%r3
+	jz	2f				/* res == NULL */
+	basr	%r1,0
+1:	l	%r0,4f-1b(%r1)
+	xc	0(4,%r3),0(%r3)			/* set tp->tv_sec to zero */
+	st	%r0,4(%r3)			/* store tp->tv_usec */
+2:	lhi	%r2,0
+	br	%r14
+3:	lhi	%r1,__NR_clock_getres		/* fallback to svc */
+	svc	0
+	br	%r14
+4:	.long	CLOCK_REALTIME_RES
+	.cfi_endproc
+	.size	__kernel_clock_getres,.-__kernel_clock_getres
diff --git a/arch/s390/kernel/vdso32/clock_gettime.S b/arch/s390/kernel/vdso32/clock_gettime.S
new file mode 100644
index 0000000..4a98909
--- /dev/null
+++ b/arch/s390/kernel/vdso32/clock_gettime.S
@@ -0,0 +1,128 @@
+/*
+ * Userland implementation of clock_gettime() for 32 bits processes in a
+ * s390 kernel for use in the vDSO
+ *
+ *  Copyright IBM Corp. 2008
+ *  Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+#include <asm/vdso.h>
+#include <asm/asm-offsets.h>
+#include <asm/unistd.h>
+
+	.text
+	.align 4
+	.globl __kernel_clock_gettime
+	.type  __kernel_clock_gettime,@function
+__kernel_clock_gettime:
+	.cfi_startproc
+	basr	%r5,0
+0:	al	%r5,21f-0b(%r5)			/* get &_vdso_data */
+	chi	%r2,CLOCK_REALTIME
+	je	10f
+	chi	%r2,CLOCK_MONOTONIC
+	jne	19f
+
+	/* CLOCK_MONOTONIC */
+	ltr	%r3,%r3
+	jz	9f				/* tp == NULL */
+1:	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
+	tml	%r4,0x0001			/* pending update ? loop */
+	jnz	1b
+	stck	24(%r15)			/* Store TOD clock */
+	lm	%r0,%r1,24(%r15)
+	s	%r0,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
+	sl	%r1,__VDSO_XTIME_STAMP+4(%r5)
+	brc	3,2f
+	ahi	%r0,-1
+2:	mhi	%r0,1000			/* cyc2ns(clock,cycle_delta) */
+	lr	%r2,%r0
+	lhi	%r0,1000
+	ltr	%r1,%r1
+	mr	%r0,%r0
+	jnm	3f
+	ahi	%r0,1000
+3:	alr	%r0,%r2
+	srdl	%r0,12
+	al	%r0,__VDSO_XTIME_NSEC(%r5)	/*  + xtime */
+	al	%r1,__VDSO_XTIME_NSEC+4(%r5)
+	brc	12,4f
+	ahi	%r0,1
+4:	l	%r2,__VDSO_XTIME_SEC+4(%r5)
+	al	%r0,__VDSO_WTOM_NSEC(%r5)	/*  + wall_to_monotonic */
+	al	%r1,__VDSO_WTOM_NSEC+4(%r5)
+	brc	12,5f
+	ahi	%r0,1
+5:	al	%r2,__VDSO_WTOM_SEC+4(%r5)
+	cl	%r4,__VDSO_UPD_COUNT+4(%r5)	/* check update counter */
+	jne	1b
+	basr	%r5,0
+6:	ltr	%r0,%r0
+	jnz	7f
+	cl	%r1,20f-6b(%r5)
+	jl	8f
+7:	ahi	%r2,1
+	sl	%r1,20f-6b(%r5)
+	brc	3,6b
+	ahi	%r0,-1
+	j	6b
+8:	st	%r2,0(%r3)			/* store tp->tv_sec */
+	st	%r1,4(%r3)			/* store tp->tv_nsec */
+9:	lhi	%r2,0
+	br	%r14
+
+	/* CLOCK_REALTIME */
+10:	ltr	%r3,%r3				/* tp == NULL */
+	jz	18f
+11:	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
+	tml	%r4,0x0001			/* pending update ? loop */
+	jnz	11b
+	stck	24(%r15)			/* Store TOD clock */
+	lm	%r0,%r1,24(%r15)
+	s	%r0,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
+	sl	%r1,__VDSO_XTIME_STAMP+4(%r5)
+	brc	3,12f
+	ahi	%r0,-1
+12:	mhi	%r0,1000			/* cyc2ns(clock,cycle_delta) */
+	lr	%r2,%r0
+	lhi	%r0,1000
+	ltr	%r1,%r1
+	mr	%r0,%r0
+	jnm	13f
+	ahi	%r0,1000
+13:	alr	%r0,%r2
+	srdl	%r0,12
+	al	%r0,__VDSO_XTIME_NSEC(%r5)	/*  + xtime */
+	al	%r1,__VDSO_XTIME_NSEC+4(%r5)
+	brc	12,14f
+	ahi	%r0,1
+14:	l	%r2,__VDSO_XTIME_SEC+4(%r5)
+	cl	%r4,__VDSO_UPD_COUNT+4(%r5)	/* check update counter */
+	jne	11b
+	basr	%r5,0
+15:	ltr	%r0,%r0
+	jnz	16f
+	cl	%r1,20f-15b(%r5)
+	jl	17f
+16:	ahi	%r2,1
+	sl	%r1,20f-15b(%r5)
+	brc	3,15b
+	ahi	%r0,-1
+	j	15b
+17:	st	%r2,0(%r3)			/* store tp->tv_sec */
+	st	%r1,4(%r3)			/* store tp->tv_nsec */
+18:	lhi	%r2,0
+	br	%r14
+
+	/* Fallback to system call */
+19:	lhi	%r1,__NR_clock_gettime
+	svc	0
+	br	%r14
+
+20:	.long	1000000000
+21:	.long	_vdso_data - 0b
+	.cfi_endproc
+	.size	__kernel_clock_gettime,.-__kernel_clock_gettime
diff --git a/arch/s390/kernel/vdso32/gettimeofday.S b/arch/s390/kernel/vdso32/gettimeofday.S
new file mode 100644
index 0000000..c32f29c
--- /dev/null
+++ b/arch/s390/kernel/vdso32/gettimeofday.S
@@ -0,0 +1,82 @@
+/*
+ * Userland implementation of gettimeofday() for 32 bits processes in a
+ * s390 kernel for use in the vDSO
+ *
+ *  Copyright IBM Corp. 2008
+ *  Author(s): Martin Schwidefsky (schwidefsky@de.ibm.com)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+#include <asm/vdso.h>
+#include <asm/asm-offsets.h>
+#include <asm/unistd.h>
+
+#include <asm/vdso.h>
+#include <asm/asm-offsets.h>
+#include <asm/unistd.h>
+
+	.text
+	.align 4
+	.globl __kernel_gettimeofday
+	.type  __kernel_gettimeofday,@function
+__kernel_gettimeofday:
+	.cfi_startproc
+	basr	%r5,0
+0:	al	%r5,13f-0b(%r5)			/* get &_vdso_data */
+1:	ltr	%r3,%r3				/* check if tz is NULL */
+	je	2f
+	mvc	0(8,%r3),__VDSO_TIMEZONE(%r5)
+2:	ltr	%r2,%r2				/* check if tv is NULL */
+	je	10f
+	l	%r4,__VDSO_UPD_COUNT+4(%r5)	/* load update counter */
+	tml	%r4,0x0001			/* pending update ? loop */
+	jnz	1b
+	stck	24(%r15)			/* Store TOD clock */
+	lm	%r0,%r1,24(%r15)
+	s	%r0,__VDSO_XTIME_STAMP(%r5)	/* TOD - cycle_last */
+	sl	%r1,__VDSO_XTIME_STAMP+4(%r5)
+	brc	3,3f
+	ahi	%r0,-1
+3:	mhi	%r0,1000			/* cyc2ns(clock,cycle_delta) */
+	st	%r0,24(%r15)
+	lhi	%r0,1000
+	ltr	%r1,%r1
+	mr	%r0,%r0
+	jnm	4f
+	ahi	%r0,1000
+4:	al	%r0,24(%r15)
+	srdl	%r0,12
+	al	%r0,__VDSO_XTIME_NSEC(%r5)	/*  + xtime */
+	al	%r1,__VDSO_XTIME_NSEC+4(%r5)
+	brc	12,5f
+	ahi	%r0,1
+5:	mvc	24(4,%r15),__VDSO_XTIME_SEC+4(%r5)
+	cl	%r4,__VDSO_UPD_COUNT+4(%r5)	/* check update counter */
+	jne	1b
+	l	%r4,24(%r15)			/* get tv_sec from stack */
+	basr	%r5,0
+6:	ltr	%r0,%r0
+	jnz	7f
+	cl	%r1,11f-6b(%r5)
+	jl	8f
+7:	ahi	%r4,1
+	sl	%r1,11f-6b(%r5)
+	brc	3,6b
+	ahi	%r0,-1
+	j	6b
+8:	st	%r4,0(%r2)			/* store tv->tv_sec */
+	ltr	%r1,%r1
+	m	%r0,12f-6b(%r5)
+	jnm	9f
+	al	%r0,12f-6b(%r5)
+9:	srl	%r0,6
+	st	%r0,4(%r2)			/* store tv->tv_usec */
+10:	slr	%r2,%r2
+	br	%r14
+11:	.long	1000000000
+12:	.long	274877907
+13:	.long	_vdso_data - 0b
+	.cfi_endproc
+	.size	__kernel_gettimeofday,.-__kernel_gettimeofday
diff --git a/arch/s390/kernel/vdso32/note.S b/arch/s390/kernel/vdso32/note.S
new file mode 100644
index 0000000..79a071e
--- /dev/null
+++ b/arch/s390/kernel/vdso32/note.S
@@ -0,0 +1,12 @@
+/*
+ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
+ * Here we can supply some information useful to userland.
+ */
+
+#include <linux/uts.h>
+#include <linux/version.h>
+#include <linux/elfnote.h>
+
+ELFNOTE_START(Linux, 0, "a")
+	.long LINUX_VERSION_CODE
+ELFNOTE_END
diff --git a/arch/s390/kernel/vdso32/vdso32.lds.S b/arch/s390/kernel/vdso32/vdso32.lds.S
new file mode 100644
index 0000000..a8c379f
--- /dev/null
+++ b/arch/s390/kernel/vdso32/vdso32.lds.S
@@ -0,0 +1,138 @@
+/*
+ * This is the infamous ld script for the 32 bits vdso
+ * library
+ */
+#include <asm/vdso.h>
+
+OUTPUT_FORMAT("elf32-s390", "elf32-s390", "elf32-s390")
+OUTPUT_ARCH(s390:31-bit)
+ENTRY(_start)
+
+SECTIONS
+{
+	. = VDSO32_LBASE + SIZEOF_HEADERS;
+
+	.hash		: { *(.hash) }			:text
+	.gnu.hash	: { *(.gnu.hash) }
+	.dynsym		: { *(.dynsym) }
+	.dynstr		: { *(.dynstr) }
+	.gnu.version	: { *(.gnu.version) }
+	.gnu.version_d	: { *(.gnu.version_d) }
+	.gnu.version_r	: { *(.gnu.version_r) }
+
+	.note		: { *(.note.*) }		:text	:note
+
+	. = ALIGN(16);
+	.text		: {
+		*(.text .stub .text.* .gnu.linkonce.t.*)
+	} :text
+	PROVIDE(__etext = .);
+	PROVIDE(_etext = .);
+	PROVIDE(etext = .);
+
+	/*
+	 * Other stuff is appended to the text segment:
+	 */
+	.rodata		: { *(.rodata .rodata.* .gnu.linkonce.r.*) }
+	.rodata1	: { *(.rodata1) }
+
+	.dynamic	: { *(.dynamic) }		:text	:dynamic
+
+	.eh_frame_hdr	: { *(.eh_frame_hdr) }		:text	:eh_frame_hdr
+	.eh_frame	: { KEEP (*(.eh_frame)) }	:text
+	.gcc_except_table : { *(.gcc_except_table .gcc_except_table.*) }
+
+	.rela.dyn ALIGN(8) : { *(.rela.dyn) }
+	.got ALIGN(8)	: { *(.got .toc) }
+
+	_end = .;
+	PROVIDE(end = .);
+
+	/*
+	 * Stabs debugging sections are here too.
+	 */
+	.stab	       0 : { *(.stab) }
+	.stabstr       0 : { *(.stabstr) }
+	.stab.excl     0 : { *(.stab.excl) }
+	.stab.exclstr  0 : { *(.stab.exclstr) }
+	.stab.index    0 : { *(.stab.index) }
+	.stab.indexstr 0 : { *(.stab.indexstr) }
+	.comment       0 : { *(.comment) }
+
+	/*
+	 * DWARF debug sections.
+	 * Symbols in the DWARF debugging sections are relative to the
+	 * beginning of the section so we begin them at 0.
+	 */
+	/* DWARF 1 */
+	.debug		0 : { *(.debug) }
+	.line		0 : { *(.line) }
+	/* GNU DWARF 1 extensions */
+	.debug_srcinfo	0 : { *(.debug_srcinfo) }
+	.debug_sfnames	0 : { *(.debug_sfnames) }
+	/* DWARF 1.1 and DWARF 2 */
+	.debug_aranges	0 : { *(.debug_aranges) }
+	.debug_pubnames 0 : { *(.debug_pubnames) }
+	/* DWARF 2 */
+	.debug_info	0 : { *(.debug_info .gnu.linkonce.wi.*) }
+	.debug_abbrev	0 : { *(.debug_abbrev) }
+	.debug_line	0 : { *(.debug_line) }
+	.debug_frame	0 : { *(.debug_frame) }
+	.debug_str	0 : { *(.debug_str) }
+	.debug_loc	0 : { *(.debug_loc) }
+	.debug_macinfo	0 : { *(.debug_macinfo) }
+	/* SGI/MIPS DWARF 2 extensions */
+	.debug_weaknames 0 : { *(.debug_weaknames) }
+	.debug_funcnames 0 : { *(.debug_funcnames) }
+	.debug_typenames 0 : { *(.debug_typenames) }
+	.debug_varnames  0 : { *(.debug_varnames) }
+	/* DWARF 3 */
+	.debug_pubtypes 0 : { *(.debug_pubtypes) }
+	.debug_ranges	0 : { *(.debug_ranges) }
+	.gnu.attributes 0 : { KEEP (*(.gnu.attributes)) }
+
+	. = ALIGN(4096);
+	PROVIDE(_vdso_data = .);
+
+	/DISCARD/	: {
+		*(.note.GNU-stack)
+		*(.branch_lt)
+		*(.data .data.* .gnu.linkonce.d.* .sdata*)
+		*(.bss .sbss .dynbss .dynsbss)
+	}
+}
+
+/*
+ * Very old versions of ld do not recognize this name token; use the constant.
+ */
+#define PT_GNU_EH_FRAME	0x6474e550
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+	text		PT_LOAD FILEHDR PHDRS FLAGS(5);	/* PF_R|PF_X */
+	dynamic		PT_DYNAMIC FLAGS(4);		/* PF_R */
+	note		PT_NOTE FLAGS(4);		/* PF_R */
+	eh_frame_hdr	PT_GNU_EH_FRAME;
+}
+
+/*
+ * This controls what symbols we export from the DSO.
+ */
+VERSION
+{
+	VDSO_VERSION_STRING {
+	global:
+		/*
+		 * Has to be there for the kernel to find
+		 */
+		__kernel_gettimeofday;
+		__kernel_clock_gettime;
+		__kernel_clock_getres;
+
+	local: *;
+	};
+}
diff --git a/arch/s390/kernel/vdso32/vdso32_wrapper.S b/arch/s390/kernel/vdso32/vdso32_wrapper.S
new file mode 100644
index 0000000..61639a8
--- /dev/null
+++ b/arch/s390/kernel/vdso32/vdso32_wrapper.S
@@ -0,0 +1,13 @@
+#include <linux/init.h>
+#include <asm/page.h>
+
+	.section ".data.page_aligned"
+
+	.globl vdso32_start, vdso32_end
+	.balign PAGE_SIZE
+vdso32_start:
+	.incbin "arch/s390/kernel/vdso32/vdso32.so"
+	.balign PAGE_SIZE
+vdso32_end:
+
+	.previous