vDSO for sparc

Following patch is based on work done by Nick Alcock on 64-bit vDSO for sparc
in Oracle linux. I have extended it to include support for 32-bit vDSO for sparc
on 64-bit kernel.

vDSO for sparc is based on the X86 implementation. This patch
provides vDSO support for both 64-bit and 32-bit programs on 64-bit kernel.
vDSO will be disabled on 32-bit linux kernel on sparc.

*) vclock_gettime.c contains all the vdso functions. Since data page is mapped
   before the vdso code page, the pointer to data page is got by subracting offset
   from an address in the vdso code page. The return address stored in
   %i7 is used for this purpose.
*) During compilation, both 32-bit and 64-bit vdso images are compiled and are
   converted into raw bytes by vdso2c program to be ready for mapping into the
   process. 32-bit images are compiled only if CONFIG_COMPAT is enabled. vdso2c
   generates two files vdso-image-64.c and vdso-image-32.c which contains the
   respective vDSO image in C structure.
*) During vdso initialization, required number of vdso pages are allocated and
   raw bytes are copied into the pages.
*) During every exec, these pages are mapped into the process through
   arch_setup_additional_pages and the location of mapping is passed on to the
   process through aux vector AT_SYSINFO_EHDR which is used by glibc.
*) A new update_vsyscall routine for sparc is added to keep the data page in
   vdso updated.
*) As vDSO cannot contain dynamically relocatable references, a new version of
   cpu_relax is added for the use of vDSO.

This change also requires a putback to glibc to use vDSO. For testing,
programs planning to try vDSO can be compiled against the generated
vdso(64/32).so in the source.

Testing:

========
[root@localhost ~]# cat vdso_test.c
int main() {
        struct timespec tv_start, tv_end;
        struct timeval tv_tmp;
	int i;
        int count = 1 * 1000 * 10000;
	long long diff;

        clock_gettime(0, &tv_start);
        for (i = 0; i < count; i++)
              gettimeofday(&tv_tmp, NULL);
        clock_gettime(0, &tv_end);
        diff = (long long)(tv_end.tv_sec -
		tv_start.tv_sec)*(1*1000*1000*1000);
        diff += (tv_end.tv_nsec - tv_start.tv_nsec);
	printf("Start sec: %d\n", tv_start.tv_sec);
	printf("End sec  : %d\n", tv_end.tv_sec);
        printf("%d cycles in %lld ns = %f ns/cycle\n", count, diff,
		(double)diff / (double)count);
        return 0;
}

[root@localhost ~]# cc vdso_test.c -o t32_without_fix -m32 -lrt
[root@localhost ~]# ./t32_without_fix
Start sec: 1502396130
End sec  : 1502396140
10000000 cycles in 9565148528 ns = 956.514853 ns/cycle
[root@localhost ~]# cc vdso_test.c -o t32_with_fix -m32 ./vdso32.so.dbg
[root@localhost ~]# ./t32_with_fix
Start sec: 1502396168
End sec  : 1502396169
10000000 cycles in 798141262 ns = 79.814126 ns/cycle
[root@localhost ~]# cc vdso_test.c -o t64_without_fix -m64 -lrt
[root@localhost ~]# ./t64_without_fix
Start sec: 1502396208
End sec  : 1502396218
10000000 cycles in 9846091800 ns = 984.609180 ns/cycle
[root@localhost ~]# cc vdso_test.c -o t64_with_fix -m64 ./vdso64.so.dbg
[root@localhost ~]# ./t64_with_fix
Start sec: 1502396257
End sec  : 1502396257
10000000 cycles in 380984048 ns = 38.098405 ns/cycle

V1 to V2 Changes:
=================
	Added hot patching code to switch the read stick instruction to read
tick instruction based on the hardware.

V2 to V3 Changes:
=================
	Merged latest changes from sparc-next and moved the initialization
of clocksource_tick.archdata.vclock_mode to time_init_early. Disabled
queued spinlock and rwlock configuration when simulating 32-bit config
to compile 32-bit VDSO.

V3 to V4 Changes:
=================
	Hardcoded the page size as 8192 in linker script for both 64-bit and
32-bit binaries. Removed unused variables in vdso2c.h. Added -mv8plus flag to
Makefile to prevent the generation of relocation entries for __lshrdi3 in 32-bit
vdso binary.

Signed-off-by: Nick Alcock <nick.alcock@oracle.com>
Signed-off-by: Nagarathnam Muthusamy <nagarathnam.muthusamy@oracle.com>
Reviewed-by: Shannon Nelson <shannon.nelson@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 234 insertions, 0 deletions

diff --git a/arch/sparc/vdso/vdso2c.c b/arch/sparc/vdso/vdso2c.c
new file mode 100644
index 000000000000..9f5b1cd6d51d
--- /dev/null
+++ b/arch/sparc/vdso/vdso2c.c
@@ -0,0 +1,234 @@
+/*
+ * vdso2c - A vdso image preparation tool
+ * Copyright (c) 2014 Andy Lutomirski and others
+ * Licensed under the GPL v2
+ *
+ * vdso2c requires stripped and unstripped input.  It would be trivial
+ * to fully strip the input in here, but, for reasons described below,
+ * we need to write a section table.  Doing this is more or less
+ * equivalent to dropping all non-allocatable sections, but it's
+ * easier to let objcopy handle that instead of doing it ourselves.
+ * If we ever need to do something fancier than what objcopy provides,
+ * it would be straightforward to add here.
+ *
+ * We keep a section table for a few reasons:
+ *
+ * Binutils has issues debugging the vDSO: it reads the section table to
+ * find SHT_NOTE; it won't look at PT_NOTE for the in-memory vDSO, which
+ * would break build-id if we removed the section table.  Binutils
+ * also requires that shstrndx != 0.  See:
+ * https://sourceware.org/bugzilla/show_bug.cgi?id=17064
+ *
+ * elfutils might not look for PT_NOTE if there is a section table at
+ * all.  I don't know whether this matters for any practical purpose.
+ *
+ * For simplicity, rather than hacking up a partial section table, we
+ * just write a mostly complete one.  We omit non-dynamic symbols,
+ * though, since they're rather large.
+ *
+ * Once binutils gets fixed, we might be able to drop this for all but
+ * the 64-bit vdso, since build-id only works in kernel RPMs, and
+ * systems that update to new enough kernel RPMs will likely update
+ * binutils in sync.  build-id has never worked for home-built kernel
+ * RPMs without manual symlinking, and I suspect that no one ever does
+ * that.
+ */
+
+/*
+ * Copyright (c) 2017 Oracle and/or its affiliates. All rights reserved.
+ */
+
+#include <inttypes.h>
+#include <stdint.h>
+#include <unistd.h>
+#include <stdarg.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <fcntl.h>
+#include <err.h>
+
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <tools/be_byteshift.h>
+
+#include <linux/elf.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+
+const char *outfilename;
+
+/* Symbols that we need in vdso2c. */
+enum {
+	sym_vvar_start,
+	sym_VDSO_FAKE_SECTION_TABLE_START,
+	sym_VDSO_FAKE_SECTION_TABLE_END,
+	sym_vread_tick,
+	sym_vread_tick_patch_start,
+	sym_vread_tick_patch_end
+};
+
+struct vdso_sym {
+	const char *name;
+	int export;
+};
+
+struct vdso_sym required_syms[] = {
+	[sym_vvar_start] = {"vvar_start", 1},
+	[sym_VDSO_FAKE_SECTION_TABLE_START] = {
+		"VDSO_FAKE_SECTION_TABLE_START", 0
+	},
+	[sym_VDSO_FAKE_SECTION_TABLE_END] = {
+		"VDSO_FAKE_SECTION_TABLE_END", 0
+	},
+	[sym_vread_tick] = {"vread_tick", 1},
+	[sym_vread_tick_patch_start] = {"vread_tick_patch_start", 1},
+	[sym_vread_tick_patch_end] = {"vread_tick_patch_end", 1}
+};
+
+__attribute__((format(printf, 1, 2))) __attribute__((noreturn))
+static void fail(const char *format, ...)
+{
+	va_list ap;
+
+	va_start(ap, format);
+	fprintf(stderr, "Error: ");
+	vfprintf(stderr, format, ap);
+	if (outfilename)
+		unlink(outfilename);
+	exit(1);
+	va_end(ap);
+}
+
+/*
+ * Evil macros for big-endian reads and writes
+ */
+#define GBE(x, bits, ifnot)						\
+	__builtin_choose_expr(						\
+		(sizeof(*(x)) == bits/8),				\
+		(__typeof__(*(x)))get_unaligned_be##bits(x), ifnot)
+
+#define LAST_GBE(x)							\
+	__builtin_choose_expr(sizeof(*(x)) == 1, *(x), (void)(0))
+
+#define GET_BE(x)							\
+	GBE(x, 64, GBE(x, 32, GBE(x, 16, LAST_GBE(x))))
+
+#define PBE(x, val, bits, ifnot)					\
+	__builtin_choose_expr(						\
+		(sizeof(*(x)) == bits/8),				\
+		put_unaligned_be##bits((val), (x)), ifnot)
+
+#define LAST_PBE(x, val)						\
+	__builtin_choose_expr(sizeof(*(x)) == 1, *(x) = (val), (void)(0))
+
+#define PUT_BE(x, val)					\
+	PBE(x, val, 64, PBE(x, val, 32, PBE(x, val, 16, LAST_PBE(x, val))))
+
+#define NSYMS ARRAY_SIZE(required_syms)
+
+#define BITSFUNC3(name, bits, suffix) name##bits##suffix
+#define BITSFUNC2(name, bits, suffix) BITSFUNC3(name, bits, suffix)
+#define BITSFUNC(name) BITSFUNC2(name, ELF_BITS, )
+
+#define INT_BITS BITSFUNC2(int, ELF_BITS, _t)
+
+#define ELF_BITS_XFORM2(bits, x) Elf##bits##_##x
+#define ELF_BITS_XFORM(bits, x) ELF_BITS_XFORM2(bits, x)
+#define ELF(x) ELF_BITS_XFORM(ELF_BITS, x)
+
+#define ELF_BITS 64
+#include "vdso2c.h"
+#undef ELF_BITS
+
+#define ELF_BITS 32
+#include "vdso2c.h"
+#undef ELF_BITS
+
+static void go(void *raw_addr, size_t raw_len,
+	       void *stripped_addr, size_t stripped_len,
+	       FILE *outfile, const char *name)
+{
+	Elf64_Ehdr *hdr = (Elf64_Ehdr *)raw_addr;
+
+	if (hdr->e_ident[EI_CLASS] == ELFCLASS64) {
+		go64(raw_addr, raw_len, stripped_addr, stripped_len,
+		     outfile, name);
+	} else if (hdr->e_ident[EI_CLASS] == ELFCLASS32) {
+		go32(raw_addr, raw_len, stripped_addr, stripped_len,
+		     outfile, name);
+	} else {
+		fail("unknown ELF class\n");
+	}
+}
+
+static void map_input(const char *name, void **addr, size_t *len, int prot)
+{
+	off_t tmp_len;
+
+	int fd = open(name, O_RDONLY);
+
+	if (fd == -1)
+		err(1, "%s", name);
+
+	tmp_len = lseek(fd, 0, SEEK_END);
+	if (tmp_len == (off_t)-1)
+		err(1, "lseek");
+	*len = (size_t)tmp_len;
+
+	*addr = mmap(NULL, tmp_len, prot, MAP_PRIVATE, fd, 0);
+	if (*addr == MAP_FAILED)
+		err(1, "mmap");
+
+	close(fd);
+}
+
+int main(int argc, char **argv)
+{
+	size_t raw_len, stripped_len;
+	void *raw_addr, *stripped_addr;
+	FILE *outfile;
+	char *name, *tmp;
+	int namelen;
+
+	if (argc != 4) {
+		printf("Usage: vdso2c RAW_INPUT STRIPPED_INPUT OUTPUT\n");
+		return 1;
+	}
+
+	/*
+	 * Figure out the struct name.  If we're writing to a .so file,
+	 * generate raw output insted.
+	 */
+	name = strdup(argv[3]);
+	namelen = strlen(name);
+	if (namelen >= 3 && !strcmp(name + namelen - 3, ".so")) {
+		name = NULL;
+	} else {
+		tmp = strrchr(name, '/');
+		if (tmp)
+			name = tmp + 1;
+		tmp = strchr(name, '.');
+		if (tmp)
+			*tmp = '\0';
+		for (tmp = name; *tmp; tmp++)
+			if (*tmp == '-')
+				*tmp = '_';
+	}
+
+	map_input(argv[1], &raw_addr, &raw_len, PROT_READ);
+	map_input(argv[2], &stripped_addr, &stripped_len, PROT_READ);
+
+	outfilename = argv[3];
+	outfile = fopen(outfilename, "w");
+	if (!outfile)
+		err(1, "%s", argv[2]);
+
+	go(raw_addr, raw_len, stripped_addr, stripped_len, outfile, name);
+
+	munmap(raw_addr, raw_len);
+	munmap(stripped_addr, stripped_len);
+	fclose(outfile);
+
+	return 0;
+}