[PATCH v9 0/6] ARM: VDSO
Nathan Lynch
nathan_lynch at mentor.com
Fri Aug 22 14:52:25 PDT 2014
Provide fast userspace implementations of gettimeofday and
clock_gettime on systems that implement the generic timers extension
defined in ARMv7. This follows the example of arm64 in conception but
significantly differs in some aspects of the implementation (C vs
assembly, mainly).
Clocks supported:
- CLOCK_REALTIME
- CLOCK_MONOTONIC
- CLOCK_REALTIME_COARSE
- CLOCK_MONOTONIC_COARSE
This also provides clock_getres (as arm64 does).
getcpu support is planned but not included at this time.
For applications to transparently benefit from this change,
ARM-specific support code needs to be added to glibc. I have such a
patch, and have verified that glibc's self tests do not detect any
regressions. I hope to have that code added to glibc for the 2.21
release.
The VDSO symbols are available for lookup via dlsym even with an
unpatched glibc.
Note that while the high-precision realtime and monotonic clock
support depends on the generic timers extension, support for
clock_getres and coarse clocks is independent of the timer
implementation and is provided unconditionally. High-resolution clock
support requires changes to the arch timer code, posted here:
http://lists.infradead.org/pipermail/linux-arm-kernel/2014-August/281280.html
The VDSO will function correctly without those changes, but
gettimeofday and clock_gettime with CLOCK_REALTIME/CLOCK_MONOTONIC
will not be accelerated.
Tested on OMAP5 and i.MX6, verifying that results obtained with the
vdso are consistent with those obtained from the kernel. On OMAP5 I
observe a 3- to 4-fold speedup for gettimeofday / CLOCK_REALTIME, with
even better (if less interesting) speedups for the coarse clock ids
and clock_getres.
I've been testing and benchmarking this with some custom test code
which I have hosted here:
https://github.com/nlynch-mentor/vdsotest
Unpatched FSF GDB may complain "warning: Could not load shared library
symbols for linux-vdso.so.1." This is not an ARM-specific issue.
Current Fedora and Debian patch their GDB packages to prevent this
warning.
Changes since v8:
- Update to 3.17-rc1.
- Split out arch timer changes into separate series.
- Ensure that VDSO will not attempt to read the counter if access is
not enabled; this check can be removed after the arch timer changes
are merged. See update_vsyscall and vdso_can_use_arch_timer in
patch #5.
Changes since v7:
- Update to next-20140801.
- In arch_setup_additional_pages, fix call to get_unmapped_area - use
bytes (not pages) for length argument.
- As x86 does, separate data and text into two VMAs, [vvar] and [vdso]
respectively. These have different permissions; [vdso] will allow
debuggers to set breakpoints, but [vvar] is read-only and cannot be
modified even via ptrace.
- Use _install_special_mapping for signal page, vvar, and vdso
mappings.
- Add -DDISABLE_BRANCH_PROFILING.
- Add --no-undefined -Bsymbolic to link options to cause linker to
error out on unresolved references, making checkundef script
unnecessary.
- Specify max-page-size, common-page-size in linker options so the
true alignment is reflected in program header; otherwise gdb gets
confused.
- Fix incremental build vs. generated auxvec.h.
- Use appropriate unwind directives in __get_datapage.
- Added vdso_install target and help text. Install build-id symlinks
as x86 does.
- Adjust update_vsyscall for changes in struct timekeeper.
Changes since v6:
- Update to 3.16-rc1.
- Remove -lgcc from link step - need to support GCC installations
without libgcc.
- Force -O2 compilation to prevent GCC from emitting calls to libgcc
math routines.
- Use custom post-processing to clear the EF_ARM_ABI_FLOAT_SOFT flag
if set in the ELF header to produce a shared object which is
architecturally allowed to be used by both soft- and hard-float
code.
- Consolidate common arch timer code instead of duplicating it.
- Prevent the VDSO from attempting CP15 access on memory-only
architected timer implementations by renaming the clocksource.
Changes since v5:
- Update to 3.15-rc1.
- Place vdso at a randomized offset above the stack along with the
sigpage.
- Properly export asm/auxvec.h.
- Split patch into series for ease of review.
Changes since v4:
- Map data page at the beginning of the VMA to prevent orphan
sections at the end of output invalidating the calculated offset.
- Move checkundef into cmd_vdsold to avoid spurious rebuilds.
- Change vdso_init message to pr_debug.
- Add -fno-stack-protector to cflags.
Changes since v3:
- Update to 3.14-rc6.
- Record vdso base in mm context before installing mapping (for the
sake of perf_mmap_event).
- Use a more seqcount-like API for critical sections. Using seqcount
API directly, however, would leak kernel pointers to userspace when
lockdep is enabled.
- Trap instead of looping forever in division-by-zero stubs.
Changes since v2:
- Update to 3.14-rc4.
- Make vDSO configurable, depending on AEABI and MMU.
- Defer shifting of nanosecond component of timespec: fixes observed
1ns inconsistencies for CLOCK_REALTIME, CLOCK_MONOTONIC (see
45a7905fc48f for arm64 equivalent).
- Force reload of seq_count when spinning: without a memory clobber
after the load of vdata->seq_count, GCC can generate code like this:
2f8: e59c9020 ldr r9, [ip, #32]
2fc: e3190001 tst r9, #1
300: 1a000033 bne 3d4 <do_realtime+0x104>
304: f57ff05b dmb ish
308: e59c3034 ldr r3, [ip, #52] ; 0x34
...
3d4: eafffffe b 3d4 <do_realtime+0x104>
- Build vdso.so with -lgcc: calls to __lshrdi3, __divsi3 sometimes
emitted (especially with -Os). Override certain libgcc functions to
prevent undefined symbols.
- Do not clear PG_reserved on vdso pages.
- Remove unnecessary get_page calls.
- Simplify ELF signature check during init.
- Use volatile for asm syscall fallbacks.
- Check whether vdso_pagelist is initialized in arm_install_vdso.
- Record clocksource mask in data page.
- Reduce code duplication in do_realtime, do_monotonic.
- Reduce calculations performed in critical sections.
- Simplify coarse clock handling.
- Move datapage load to its own assembly routine.
- Tune vdso_data layout and tweak field names.
- Check vdso shared object for undefined symbols during build.
Changes since v1:
- update to 3.14-rc1
- ensure cache coherency for data page
- Document the kernel-to-userspace protocol for vdso data page updates,
and note that the timekeeping core prevents concurrent updates.
- update wall-to-monotonic fields unconditionally
- move vdso_start, vdso_end declarations to vdso.h
- correctly build and run when CONFIG_ARM_ARCH_TIMER=n
- rearrange linker script to avoid overlapping sections when CONFIG_DEBUGINFO=n
- remove use_syscall checks from coarse clock paths
- crib BUG_INSTR (0xe7f001f2) from asm/bug.h for text fill
Nathan Lynch (6):
ARM: use _install_special_mapping for sigpage
ARM: place sigpage at a random offset above stack
ARM: miscellaneous vdso infrastructure, preparation
ARM: add vdso user-space code
ARM: vdso initialization, mapping, and synchronization
ARM: add CONFIG_VDSO Kconfig and Makefile bits
arch/arm/Makefile | 8 +
arch/arm/include/asm/Kbuild | 1 -
arch/arm/include/asm/auxvec.h | 1 +
arch/arm/include/asm/elf.h | 9 +
arch/arm/include/asm/mmu.h | 3 +
arch/arm/include/asm/vdso.h | 34 ++++
arch/arm/include/asm/vdso_datapage.h | 60 +++++++
arch/arm/include/uapi/asm/Kbuild | 1 +
arch/arm/include/uapi/asm/auxvec.h | 7 +
arch/arm/kernel/Makefile | 1 +
arch/arm/kernel/asm-offsets.c | 5 +
arch/arm/kernel/process.c | 71 +++++++-
arch/arm/kernel/vdso.c | 207 ++++++++++++++++++++++
arch/arm/mm/Kconfig | 15 ++
arch/arm/vdso/.gitignore | 1 +
arch/arm/vdso/Makefile | 74 ++++++++
arch/arm/vdso/datapage.S | 15 ++
arch/arm/vdso/vdso.S | 35 ++++
arch/arm/vdso/vdso.lds.S | 88 ++++++++++
arch/arm/vdso/vdsomunge.c | 208 +++++++++++++++++++++++
arch/arm/vdso/vgettimeofday.c | 320 +++++++++++++++++++++++++++++++++++
21 files changed, 1154 insertions(+), 10 deletions(-)
create mode 100644 arch/arm/include/asm/auxvec.h
create mode 100644 arch/arm/include/asm/vdso.h
create mode 100644 arch/arm/include/asm/vdso_datapage.h
create mode 100644 arch/arm/include/uapi/asm/auxvec.h
create mode 100644 arch/arm/kernel/vdso.c
create mode 100644 arch/arm/vdso/.gitignore
create mode 100644 arch/arm/vdso/Makefile
create mode 100644 arch/arm/vdso/datapage.S
create mode 100644 arch/arm/vdso/vdso.S
create mode 100644 arch/arm/vdso/vdso.lds.S
create mode 100644 arch/arm/vdso/vdsomunge.c
create mode 100644 arch/arm/vdso/vgettimeofday.c
--
1.9.3
More information about the linux-arm-kernel
mailing list