2010-02-19 08:13:04 +08:00
|
|
|
/*
|
2015-10-21 16:54:38 +08:00
|
|
|
* Copyright (C) 2015 Imagination Technologies
|
|
|
|
* Author: Alex Smith <alex.smith@imgtec.com>
|
2010-02-19 08:13:04 +08:00
|
|
|
*
|
2015-10-21 16:54:38 +08:00
|
|
|
* This program is free software; you can redistribute it and/or modify it
|
|
|
|
* under the terms of the GNU General Public License as published by the
|
|
|
|
* Free Software Foundation; either version 2 of the License, or (at your
|
|
|
|
* option) any later version.
|
2010-02-19 08:13:04 +08:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/binfmts.h>
|
|
|
|
#include <linux/elf.h>
|
2015-10-21 16:54:38 +08:00
|
|
|
#include <linux/err.h>
|
|
|
|
#include <linux/init.h>
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
#include <linux/ioport.h>
|
|
|
|
#include <linux/irqchip/mips-gic.h>
|
2015-10-21 16:54:38 +08:00
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/slab.h>
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
#include <linux/timekeeper_internal.h>
|
2010-02-19 08:13:04 +08:00
|
|
|
|
2015-10-21 16:54:38 +08:00
|
|
|
#include <asm/abi.h>
|
2010-02-19 08:13:04 +08:00
|
|
|
#include <asm/vdso.h>
|
2015-10-21 16:54:38 +08:00
|
|
|
|
|
|
|
/* Kernel-provided data used by the VDSO. */
|
|
|
|
static union mips_vdso_data vdso_data __page_aligned_data;
|
2010-02-19 08:13:04 +08:00
|
|
|
|
|
|
|
/*
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
* Mapping for the VDSO data/GIC pages. The real pages are mapped manually, as
|
2015-10-21 16:54:38 +08:00
|
|
|
* what we map and where within the area they are mapped is determined at
|
|
|
|
* runtime.
|
2010-02-19 08:13:04 +08:00
|
|
|
*/
|
2015-10-21 16:54:38 +08:00
|
|
|
static struct page *no_pages[] = { NULL };
|
|
|
|
static struct vm_special_mapping vdso_vvar_mapping = {
|
|
|
|
.name = "[vvar]",
|
|
|
|
.pages = no_pages,
|
|
|
|
};
|
2010-02-19 08:13:04 +08:00
|
|
|
|
2015-10-21 16:54:38 +08:00
|
|
|
static void __init init_vdso_image(struct mips_vdso_image *image)
|
2010-02-19 08:13:04 +08:00
|
|
|
{
|
2015-10-21 16:54:38 +08:00
|
|
|
unsigned long num_pages, i;
|
|
|
|
|
|
|
|
BUG_ON(!PAGE_ALIGNED(image->data));
|
|
|
|
BUG_ON(!PAGE_ALIGNED(image->size));
|
|
|
|
|
|
|
|
num_pages = image->size / PAGE_SIZE;
|
|
|
|
|
|
|
|
for (i = 0; i < num_pages; i++) {
|
|
|
|
image->mapping.pages[i] =
|
|
|
|
virt_to_page(image->data + (i * PAGE_SIZE));
|
|
|
|
}
|
2010-02-19 08:13:04 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static int __init init_vdso(void)
|
|
|
|
{
|
2015-10-21 16:54:38 +08:00
|
|
|
init_vdso_image(&vdso_image);
|
|
|
|
|
|
|
|
#ifdef CONFIG_MIPS32_O32
|
|
|
|
init_vdso_image(&vdso_image_o32);
|
2010-02-19 08:13:04 +08:00
|
|
|
#endif
|
|
|
|
|
2015-10-21 16:54:38 +08:00
|
|
|
#ifdef CONFIG_MIPS32_N32
|
|
|
|
init_vdso_image(&vdso_image_n32);
|
|
|
|
#endif
|
2010-02-19 08:13:04 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
2010-06-17 06:00:28 +08:00
|
|
|
subsys_initcall(init_vdso);
|
2010-02-19 08:13:04 +08:00
|
|
|
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
void update_vsyscall(struct timekeeper *tk)
|
|
|
|
{
|
|
|
|
vdso_data_write_begin(&vdso_data);
|
|
|
|
|
|
|
|
vdso_data.xtime_sec = tk->xtime_sec;
|
|
|
|
vdso_data.xtime_nsec = tk->tkr_mono.xtime_nsec;
|
|
|
|
vdso_data.wall_to_mono_sec = tk->wall_to_monotonic.tv_sec;
|
|
|
|
vdso_data.wall_to_mono_nsec = tk->wall_to_monotonic.tv_nsec;
|
|
|
|
vdso_data.cs_shift = tk->tkr_mono.shift;
|
|
|
|
|
|
|
|
vdso_data.clock_mode = tk->tkr_mono.clock->archdata.vdso_clock_mode;
|
|
|
|
if (vdso_data.clock_mode != VDSO_CLOCK_NONE) {
|
|
|
|
vdso_data.cs_mult = tk->tkr_mono.mult;
|
|
|
|
vdso_data.cs_cycle_last = tk->tkr_mono.cycle_last;
|
|
|
|
vdso_data.cs_mask = tk->tkr_mono.mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
vdso_data_write_end(&vdso_data);
|
|
|
|
}
|
|
|
|
|
|
|
|
void update_vsyscall_tz(void)
|
|
|
|
{
|
|
|
|
if (vdso_data.clock_mode != VDSO_CLOCK_NONE) {
|
|
|
|
vdso_data.tz_minuteswest = sys_tz.tz_minuteswest;
|
|
|
|
vdso_data.tz_dsttime = sys_tz.tz_dsttime;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-02-19 08:13:04 +08:00
|
|
|
int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp)
|
|
|
|
{
|
2015-10-21 16:54:38 +08:00
|
|
|
struct mips_vdso_image *image = current->thread.abi->vdso;
|
2010-02-19 08:13:04 +08:00
|
|
|
struct mm_struct *mm = current->mm;
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
unsigned long gic_size, vvar_size, size, base, data_addr, vdso_addr;
|
2015-10-21 16:54:38 +08:00
|
|
|
struct vm_area_struct *vma;
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
struct resource gic_res;
|
2015-10-21 16:54:38 +08:00
|
|
|
int ret;
|
2010-02-19 08:13:04 +08:00
|
|
|
|
|
|
|
down_write(&mm->mmap_sem);
|
|
|
|
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
/*
|
|
|
|
* Determine total area size. This includes the VDSO data itself, the
|
|
|
|
* data page, and the GIC user page if present. Always create a mapping
|
|
|
|
* for the GIC user area if the GIC is present regardless of whether it
|
|
|
|
* is the current clocksource, in case it comes into use later on. We
|
|
|
|
* only map a page even though the total area is 64K, as we only need
|
|
|
|
* the counter registers at the start.
|
|
|
|
*/
|
|
|
|
gic_size = gic_present ? PAGE_SIZE : 0;
|
|
|
|
vvar_size = gic_size + PAGE_SIZE;
|
|
|
|
size = vvar_size + image->size;
|
|
|
|
|
|
|
|
base = get_unmapped_area(NULL, 0, size, 0, 0);
|
2015-10-21 16:54:38 +08:00
|
|
|
if (IS_ERR_VALUE(base)) {
|
|
|
|
ret = base;
|
|
|
|
goto out;
|
2010-02-19 08:13:04 +08:00
|
|
|
}
|
|
|
|
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
data_addr = base + gic_size;
|
|
|
|
vdso_addr = data_addr + PAGE_SIZE;
|
2015-10-21 16:54:38 +08:00
|
|
|
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
vma = _install_special_mapping(mm, base, vvar_size,
|
2015-10-21 16:54:38 +08:00
|
|
|
VM_READ | VM_MAYREAD,
|
|
|
|
&vdso_vvar_mapping);
|
|
|
|
if (IS_ERR(vma)) {
|
|
|
|
ret = PTR_ERR(vma);
|
|
|
|
goto out;
|
|
|
|
}
|
2010-02-19 08:13:04 +08:00
|
|
|
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
/* Map GIC user page. */
|
|
|
|
if (gic_size) {
|
|
|
|
ret = gic_get_usm_range(&gic_res);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
ret = io_remap_pfn_range(vma, base,
|
|
|
|
gic_res.start >> PAGE_SHIFT,
|
|
|
|
gic_size,
|
|
|
|
pgprot_noncached(PAGE_READONLY));
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
2015-10-21 16:54:38 +08:00
|
|
|
/* Map data page. */
|
MIPS: VDSO: Add implementations of gettimeofday() and clock_gettime()
Add user-mode implementations of gettimeofday() and clock_gettime() to
the VDSO. This is currently usable with 2 clocksources: the CP0 count
register, which is accessible to user-mode via RDHWR on R2 and later
cores, or the MIPS Global Interrupt Controller (GIC) timer, which
provides a "user-mode visible" section containing a mirror of its
counter registers. This section must be mapped into user memory, which
is done below the VDSO data page.
When a supported clocksource is not in use, the VDSO functions will
return -ENOSYS, which causes libc to fall back on the standard syscall
path.
When support for neither of these clocksources is compiled into the
kernel at all, the VDSO still provides clock_gettime(), as the coarse
realtime/monotonic clocks can still be implemented. However,
gettimeofday() is not provided in this case as nothing can be done
without a suitable clocksource. This causes the symbol lookup to fail
in libc and it will then always use the standard syscall path.
This patch includes a workaround for a bug in QEMU which results in
RDHWR on the CP0 count register always returning a constant (incorrect)
value. A fix for this has been submitted, and the workaround can be
removed after the fix has been in stable releases for a reasonable
amount of time.
A simple performance test which calls gettimeofday() 1000 times in a
loop and calculates the average execution time gives the following
results on a Malta + I6400 (running at 20MHz):
- Syscall: ~31000 ns
- VDSO (GIC): ~15000 ns
- VDSO (CP0): ~9500 ns
[markos.chandras@imgtec.com:
- Minor code re-arrangements in order for mappings to be made
in the order they appear to the process' address space.
- Move do_{monotonic, realtime} outside of the MIPS_CLOCK_VSYSCALL ifdef
- Use gic_get_usm_range so we can do the GIC mapping in the
arch/mips/kernel/vdso instead of the GIC irqchip driver]
Signed-off-by: Alex Smith <alex.smith@imgtec.com>
Signed-off-by: Markos Chandras <markos.chandras@imgtec.com>
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/11338/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2015-10-21 16:57:44 +08:00
|
|
|
ret = remap_pfn_range(vma, data_addr,
|
2015-10-21 16:54:38 +08:00
|
|
|
virt_to_phys(&vdso_data) >> PAGE_SHIFT,
|
|
|
|
PAGE_SIZE, PAGE_READONLY);
|
2010-02-19 08:13:04 +08:00
|
|
|
if (ret)
|
2015-10-21 16:54:38 +08:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
/* Map VDSO image. */
|
|
|
|
vma = _install_special_mapping(mm, vdso_addr, image->size,
|
|
|
|
VM_READ | VM_EXEC |
|
|
|
|
VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC,
|
|
|
|
&image->mapping);
|
|
|
|
if (IS_ERR(vma)) {
|
|
|
|
ret = PTR_ERR(vma);
|
|
|
|
goto out;
|
|
|
|
}
|
2010-02-19 08:13:04 +08:00
|
|
|
|
2015-10-21 16:54:38 +08:00
|
|
|
mm->context.vdso = (void *)vdso_addr;
|
|
|
|
ret = 0;
|
2010-02-19 08:13:04 +08:00
|
|
|
|
2015-10-21 16:54:38 +08:00
|
|
|
out:
|
2010-02-19 08:13:04 +08:00
|
|
|
up_write(&mm->mmap_sem);
|
|
|
|
return ret;
|
|
|
|
}
|