Rather than buffering 16 MiB of entropy in userspace (by way of
chacha20), simply call getrandom() every time.
This approach is doubtlessly slower, for now, but trying to prematurely
optimize arc4random appears to be leading toward all sorts of nasty
properties and gotchas. Instead, this patch takes a much more
conservative approach. The interface is added as a basic loop wrapper
around getrandom(), and then later, the kernel and libc together can
work together on optimizing that.
This prevents numerous issues in which userspace is unaware of when it
really must throw away its buffer, since we avoid buffering all
together. Future improvements may include userspace learning more from
the kernel about when to do that, which might make these sorts of
chacha20-based optimizations more possible. The current heuristic of 16
MiB is meaningless garbage that doesn't correspond to anything the
kernel might know about. So for now, let's just do something
conservative that we know is correct and won't lead to cryptographic
issues for users of this function.
This patch might be considered along the lines of, "optimization is the
root of all evil," in that the much more complex implementation it
replaces moves too fast without considering security implications,
whereas the incremental approach done here is a much safer way of going
about things. Once this lands, we can take our time in optimizing this
properly using new interplay between the kernel and userspace.
getrandom(0) is used, since that's the one that ensures the bytes
returned are cryptographically secure. But on systems without it, we
fallback to using /dev/urandom. This is unfortunate because it means
opening a file descriptor, but there's not much of a choice. Secondly,
as part of the fallback, in order to get more or less the same
properties of getrandom(0), we poll on /dev/random, and if the poll
succeeds at least once, then we assume the RNG is initialized. This is a
rough approximation, as the ancient "non-blocking pool" initialized
after the "blocking pool", not before, and it may not port back to all
ancient kernels, though it does to all kernels supported by glibc
(≥3.2), so generally it's the best approximation we can do.
The motivation for including arc4random, in the first place, is to have
source-level compatibility with existing code. That means this patch
doesn't attempt to litigate the interface itself. It does, however,
choose a conservative approach for implementing it.
Cc: Adhemerval Zanella Netto <adhemerval.zanella@linaro.org>
Cc: Florian Weimer <fweimer@redhat.com>
Cc: Cristian Rodríguez <crrodriguez@opensuse.org>
Cc: Paul Eggert <eggert@cs.ucla.edu>
Cc: Mark Harris <mark.hsj@gmail.com>
Cc: Eric Biggers <ebiggers@kernel.org>
Cc: linux-crypto@vger.kernel.org
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
It adds vectorized ChaCha20 implementation based on libgcrypt
cipher/chacha20-aarch64.S. It is used as default and only
little-endian is supported (BE uses generic code).
As for generic implementation, the last step that XOR with the
input is omited. The final state register clearing is also
omitted.
On a virtualized Linux on Apple M1 it shows the following
improvements (using formatted bench-arc4random data):
GENERIC MB/s
-----------------------------------------------
arc4random [single-thread] 380.89
arc4random_buf(16) [single-thread] 500.73
arc4random_buf(32) [single-thread] 552.61
arc4random_buf(48) [single-thread] 566.82
arc4random_buf(64) [single-thread] 574.01
arc4random_buf(80) [single-thread] 581.02
arc4random_buf(96) [single-thread] 591.19
arc4random_buf(112) [single-thread] 592.29
arc4random_buf(128) [single-thread] 596.43
-----------------------------------------------
OPTIMIZED MB/s
-----------------------------------------------
arc4random [single-thread] 569.60
arc4random_buf(16) [single-thread] 825.78
arc4random_buf(32) [single-thread] 987.03
arc4random_buf(48) [single-thread] 1042.39
arc4random_buf(64) [single-thread] 1075.50
arc4random_buf(80) [single-thread] 1094.68
arc4random_buf(96) [single-thread] 1130.16
arc4random_buf(112) [single-thread] 1129.58
arc4random_buf(128) [single-thread] 1137.91
-----------------------------------------------
Checked on aarch64-linux-gnu.
We found that string functions were using AND+ADDP
to find the nibble/syndrome mask but there is an easier
opportunity through `SHRN dst.8b, src.8h, 4` (shift
right every 2 bytes by 4 and narrow to 1 byte) and has
same latency on all SIMD ARMv8 targets as ADDP. There
are also possible gaps for memcmp but that's for
another patch.
We see 10-20% savings for small-mid size cases (<=128)
which are primary cases for general workloads.
The AFP feature (Alternate floating-point behavior) was added in armv8.7 and
introduced new FPCR bits.
Currently, HWCAP2_AFP bits (bit 0, 1, 2) in FPCR are preserved when fenv is
set to default environment. This is a deviation from standard behaviour.
Clear these bits when setting the fenv to default.
There is no libc API to modify the new FPCR bits. Restoring those bits matters
if the user changed them directly.
The RTLD_BOOTSTRAP branch is used to relocate ld.so itself. It only
needs to handle RELATIVE, GLOB_DAT, and JUMP_SLOT.
TLSDESC/TLS_DTPMOD/TLS_DTPREL handling can be removed. Remove
`case AARCH64_R(RELATIVE)` as well as elf_machine_rela has checked it.
Tested on aarch64-linux-gnu.
Add a proper bounds check to __libc_ifunc_impl_list. This makes MAX_IFUNC
redundant and fixes several targets that will write outside the array.
To avoid unnecessary large diffs, pass the maximum in the argument 'i' to
IFUNC_IMPL_ADD - 'max' can be used in new ifunc definitions and existing
ones can be updated if desired.
Passes buildmanyglibc.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Add an initial SVE memcpy implementation. Copies up to 32 bytes use SVE
vectors which improves the random memcpy benchmark significantly.
Cleanup the memcpy and memmove ifunc selectors.
This reverts commit 0910702c4d.
Say both a.so and b.so define protected data symbol `var` and the executable
copy relocates var. ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA has strange
semantics: a.so accesses the copy in the executable while b.so accesses its
own. This behavior requires that (a) the compiler emits GOT-generating
relocations (b) the linker produces GLOB_DAT instead of RELATIVE.
Without the ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA code, b.so's GLOB_DAT
will bind to the executable (normal behavior).
For aarch64 it makes sense to restore the original behavior and don't
pay the ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA cost. The behavior is very
unlikely used by anyone.
* Clang code generator treats STV_PROTECTED the same way as STV_HIDDEN:
no GOT-generating relocation in the first place.
* gold and lld reject copy relocation on a STV_PROTECTED symbol.
* Nowadays -fpie/-fpic modes are popular. GCC/Clang's codegen uses
GOT-generating relocation when accessing an default visibility
external symbol which avoids copy relocation.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
A separate asm file is easier to maintain than a macro that expands to
inline asm.
The RTLD_START macro is only needed now because _dl_start is local in
rtld.c, but _start has to call it, if _dl_start was made hidden then it
could be empty.
_dl_skip_args is no longer needed.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
_dl_skip_args is always 0, so the target specific code that modifies
argv after relro protection is applied is no longer used.
After the patch relro protection is applied to _dl_argv consistently
on all targets.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
PI_STATIC_AND_HIDDEN indicates whether accesses to internal linkage
variables and hidden visibility variables in a shared object (ld.so)
need dynamic relocations (usually R_*_RELATIVE). PI (position
independent) in the macro name is a misnomer: a code sequence using GOT
is typically position-independent as well, but using dynamic relocations
does not meet the requirement.
Not defining PI_STATIC_AND_HIDDEN is legacy and we expect that all new
ports will define PI_STATIC_AND_HIDDEN. Current ports defining
PI_STATIC_AND_HIDDEN are more than the opposite. Change the configure
default.
No functional change.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
The rtld audit support show two problems on aarch64:
1. _dl_runtime_resolve does not preserve x8, the indirect result
location register, which might generate wrong result calls
depending of the function signature.
2. The NEON Q registers pushed onto the stack by _dl_runtime_resolve
were twice the size of D registers extracted from the stack frame by
_dl_runtime_profile.
While 2. might result in wrong information passed on the PLT tracing,
1. generates wrong runtime behaviour.
The aarch64 rtld audit support is changed to:
* Both La_aarch64_regs and La_aarch64_retval are expanded to include
both x8 and the full sized NEON V registers, as defined by the
ABI.
* dl_runtime_profile needed to extract registers saved by
_dl_runtime_resolve and put them into the new correctly sized
La_aarch64_regs structure.
* The LAV_CURRENT check is change to only accept new audit modules
to avoid the undefined behavior of not save/restore x8.
* Different than other architectures, audit modules older than
LAV_CURRENT are rejected (both La_aarch64_regs and La_aarch64_retval
changed their layout and there are no requirements to support multiple
audit interface with the inherent aarch64 issues).
* A new field is also reserved on both La_aarch64_regs and
La_aarch64_retval to support variant pcs symbols.
Similar to x86, a new La_aarch64_vector type to represent the NEON
register is added on the La_aarch64_regs (so each type can be accessed
directly).
Since LAV_CURRENT was already bumped to support bind-now, there is
no need to increase it again.
Checked on aarch64-linux-gnu.
Co-authored-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
I used these shell commands:
../glibc/scripts/update-copyrights $PWD/../gnulib/build-aux/update-copyright
(cd ../glibc && git commit -am"[this commit message]")
and then ignored the output, which consisted lines saying "FOO: warning:
copyright statement not found" for each of 7061 files FOO.
I then removed trailing white space from math/tgmath.h,
support/tst-support-open-dev-null-range.c, and
sysdeps/x86_64/multiarch/strlen-vec.S, to work around the following
obscure pre-commit check failure diagnostics from Savannah. I don't
know why I run into these diagnostics whereas others evidently do not.
remote: *** 912-#endif
remote: *** 913:
remote: *** 914-
remote: *** error: lines with trailing whitespace found
...
remote: *** error: sysdeps/unix/sysv/linux/statx_cp.c: trailing lines
And use machine-sp.h instead. The Linux implementation is based on
already provided CURRENT_STACK_FRAME (used on nptl code) and
STACK_GROWS_UPWARD is replaced with _STACK_GROWS_UP.
It consolidates the code required to call la_pltexit audit
callback.
Checked on x86_64-linux-gnu, i686-linux-gnu, and aarch64-linux-gnu.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
TLS_INIT_TCB_ALIGN is not actually used. TLS_TCB_ALIGN was likely
introduced to support a configuration where the thread pointer
has not the same alignment as THREAD_SELF. Only ia64 seems to use
that, but for the stack/pointer guard, not for storing tcbhead_t.
Some ports use TLS_TCB_OFFSET and TLS_PRE_TCB_SIZE to shift
the thread pointer, potentially landing in a different residue class
modulo the alignment, but the changes should not impact that.
In general, given that TLS variables have their own alignment
requirements, having different alignment for the (unshifted) thread
pointer and struct pthread would potentially result in dynamic
offsets, leading to more complexity.
hppa had different values before: __alignof__ (tcbhead_t), which
seems to be 4, and __alignof__ (struct pthread), which was 8
(old default) and is now 32. However, it defines THREAD_SELF as:
/* Return the thread descriptor for the current thread. */
# define THREAD_SELF \
({ struct pthread *__self; \
__self = __get_cr27(); \
__self - 1; \
})
So the thread pointer points after struct pthread (hence __self - 1),
and they have to have the same alignment on hppa as well.
Similarly, on ia64, the definitions were different. We have:
# define TLS_PRE_TCB_SIZE \
(sizeof (struct pthread) \
+ (PTHREAD_STRUCT_END_PADDING < 2 * sizeof (uintptr_t) \
? ((2 * sizeof (uintptr_t) + __alignof__ (struct pthread) - 1) \
& ~(__alignof__ (struct pthread) - 1)) \
: 0))
# define THREAD_SELF \
((struct pthread *) ((char *) __thread_self - TLS_PRE_TCB_SIZE))
And TLS_PRE_TCB_SIZE is a multiple of the struct pthread alignment
(confirmed by the new _Static_assert in sysdeps/ia64/libc-tls.c).
On m68k, we have a larger gap between tcbhead_t and struct pthread.
But as far as I can tell, the port is fine with that. The definition
of TCB_OFFSET is sufficient to handle the shifted TCB scenario.
This fixes commit 23c77f6018
("nptl: Increase default TCB alignment to 32").
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
rseq support will use a 32-byte aligned field in struct pthread,
so the whole struct needs to have at least that alignment.
nptl/tst-tls3mod.c uses TCB_ALIGNMENT, therefore include <descr.h>
to obtain the fallback definition.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
v2 is a complete rewrite of the A64FX memcpy. Performance is improved
by streamlining the code, aligning all large copies and using a single
unrolled loop for all sizes. The code size for memcpy and memmove goes
down from 1796 bytes to 868 bytes. Performance is better in all cases:
bench-memcpy-random is 2.3% faster overall, bench-memcpy-large is ~33%
faster for large sizes, bench-memcpy-walk is 25% faster for small sizes
and 20% for the largest sizes. The geomean of all tests in bench-memcpy
is 5.1% faster, and total time is reduced by 4%.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Rewrite memcmp to improve performance. On small and medium inputs performance
is 10-20% better. Large inputs use a SIMD loop processing 64 bytes per
iteration, which is 30-50% faster depending on the size.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
No bug.
This commit adds support for __memcmpeq() as a new ABI for all
targets. In this commit __memcmpeq() is implemented only as an alias
to the corresponding targets memcmp() implementation. __memcmpeq() is
added as a new symbol starting with GLIBC_2.35 and defined in string.h
with comments explaining its behavior. Basic tests that it is callable
and works where added in string/tester.c
As discussed in the proposal "Add new ABI '__memcmpeq()' to libc"
__memcmpeq() is essentially a reserved namespace for bcmp(). The means
is shares the same specifications as memcmp() except the return value
for non-equal byte sequences is any non-zero value. This is less
strict than memcmp()'s return value specification and can be better
optimized when a boolean return is all that is needed.
__memcmpeq() is meant to only be called by compilers if they can prove
that the return value of a memcmp() call is only used for its boolean
value.
All tests in string/tester.c passed. As well build succeeds on
x86_64-linux-gnu target.
The 4af6982e4c fix does not fully handle RTLD_BOOTSTRAP usage on
rtld.c due two issues:
1. RTLD_BOOTSTRAP is also used on dl-machine.h on various
architectures and it changes the semantics of various machine
relocation functions.
2. The elf_get_dynamic_info() change was done sideways, previously
to 490e6c62aa get-dynamic-info.h was included by the first
dynamic-link.h include *without* RTLD_BOOTSTRAP being defined.
It means that the code within elf_get_dynamic_info() that uses
RTLD_BOOTSTRAP is in fact unused.
To fix 1. this patch now includes dynamic-link.h only once with
RTLD_BOOTSTRAP defined. The ELF_DYNAMIC_RELOCATE call will now have
the relocation fnctions with the expected semantics for the loader.
And to fix 2. part of 4af6982e4c is reverted (the check argument
elf_get_dynamic_info() is not required) and the RTLD_BOOTSTRAP
pieces are removed.
To reorganize the includes the static TLS definition is moved to
its own header to avoid a circular dependency (it is defined on
dynamic-link.h and dl-machine.h requires it at same time other
dynamic-link.h definition requires dl-machine.h defitions).
Also ELF_MACHINE_NO_REL, ELF_MACHINE_NO_RELA, and ELF_MACHINE_PLT_REL
are moved to its own header. Only ancient ABIs need special values
(arm, i386, and mips), so a generic one is used as default.
The powerpc Elf64_FuncDesc is also moved to its own header, since
csu code required its definition (which would require either include
elf/ folder or add a full path with elf/).
Checked on x86_64, i686, aarch64, armhf, powerpc64, powerpc32,
and powerpc64le.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
dynamic-link.h is included more than once in some elf/ files (rtld.c,
dl-conflict.c, dl-reloc.c, dl-reloc-static-pie.c) and uses GCC nested
functions. This harms readability and the nested functions usage
is the biggest obstacle prevents Clang build (Clang doesn't support GCC
nested functions).
The key idea for unnesting is to add extra parameters (struct link_map
*and struct r_scope_elm *[]) to RESOLVE_MAP,
ELF_MACHINE_BEFORE_RTLD_RELOC, ELF_DYNAMIC_RELOCATE, elf_machine_rel[a],
elf_machine_lazy_rel, and elf_machine_runtime_setup. (This is inspired
by Stan Shebs' ppc64/x86-64 implementation in the
google/grte/v5-2.27/master which uses mixed extra parameters and static
variables.)
Future simplification:
* If mips elf_machine_runtime_setup no longer needs RESOLVE_GOTSYM,
elf_machine_runtime_setup can drop the `scope` parameter.
* If TLSDESC no longer need to be in elf_machine_lazy_rel,
elf_machine_lazy_rel can drop the `scope` parameter.
Tested on aarch64, i386, x86-64, powerpc64le, powerpc64, powerpc32,
sparc64, sparcv9, s390x, s390, hppa, ia64, armhf, alpha, and mips64.
In addition, tested build-many-glibcs.py with {arc,csky,microblaze,nios2}-linux-gnu
and riscv64-linux-gnu-rv64imafdc-lp64d.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This patch disables A64FX memcpy/memmove BTI instruction insertion
unconditionally such as A64FX memset patch [1] for performance.
[1] commit 07b427296b
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
All the ports now have THREAD_GSCOPE_IN_TCB set to 1. Remove all
support for !THREAD_GSCOPE_IN_TCB, along with the definition itself.
Signed-off-by: Sergey Bugaev <bugaevc@gmail.com>
Message-Id: <20210915171110.226187-4-bugaevc@gmail.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
This patch updates unroll8 code so as not to degrade at the peak
performance 16KB for both FX1000 and FX700.
Inserted 2 instructions at the beginning of the unroll8 loop,
cmp and branch, are a workaround that is found heuristically.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
We stopped adding "Contributed by" or similar lines in sources in 2012
in favour of git logs and keeping the Contributors section of the
glibc manual up to date. Removing these lines makes the license
header a bit more consistent across files and also removes the
possibility of error in attribution when license blocks or files are
copied across since the contributed-by lines don't actually reflect
reality in those cases.
Move all "Contributed by" and similar lines (Written by, Test by,
etc.) into a new file CONTRIBUTED-BY to retain record of these
contributions. These contributors are also mentioned in
manual/contrib.texi, so we just maintain this additional record as a
courtesy to the earlier developers.
The following scripts were used to filter a list of files to edit in
place and to clean up the CONTRIBUTED-BY file respectively. These
were not added to the glibc sources because they're not expected to be
of any use in future given that this is a one time task:
https://gist.github.com/siddhesh/b5ecac94eabfd72ed2916d6d8157e7dchttps://gist.github.com/siddhesh/15ea1f5e435ace9774f485030695ee02
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
This patch updates unroll8 code so as not to degrade at the peak
performance 16KB for both FX1000 and FX700.
Inserted 2 instructions at the beginning of the unroll8 loop,
cmp and branch, are a workaround that is found heuristically.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
They provide TLS_GD/TLS_LD/TLS_IE/TLS_IE macros for TLS testing. Now
that we have migrated to __thread and tls_model attributes, these macros
are unused and the tls-macros.h files can retire.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
The AArch64 ABI is largely platform agnostic and does not specify
_GLOBAL_OFFSET_TABLE_[0] ([1]). glibc ld.so turns out to be probably the
only user of _GLOBAL_OFFSET_TABLE_[0] and GNU ld defines the value
to the link-time address _DYNAMIC. [2]
In 2012, __ehdr_start was implemented in GNU ld and gold in binutils
2.23. Using adrp+add / (-mcmodel=tiny) adr to access
__ehdr_start/_DYNAMIC gives us a robust way to get the load address and
the link-time address of _DYNAMIC.
[1]: From a psABI maintainer, https://bugs.llvm.org/show_bug.cgi?id=49672#c2
[2]: LLD's aarch64 port does not set _GLOBAL_OFFSET_TABLE_[0] to the
link-time address _DYNAMIC.
LLD is widely used on aarch64 Android and ChromeOS devices. Software
just works without the need for _GLOBAL_OFFSET_TABLE_[0].
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Simplify handling of remaining bytes. Avoid lots of taken branches and complex
whilelo computations, instead unconditionally write vectors from the end.
Reviewed-by: Naohiro Tamura <naohirot@fujitsu.com>
Improve performance of large memsets. Simplify alignment code. For zero memset
use DC ZVA, which almost doubles performance. For non-zero memsets use the
unroll8 loop which is about 10% faster.
Reviewed-by: Naohiro Tamura <naohirot@fujitsu.com>
Improve performance of small memsets by reducing instruction counts and
improving code alignment. Bench-memset shows 35-45% performance gain for
small sizes.
Reviewed-by: Naohiro Tamura <naohirot@fujitsu.com>
The malloc-check debugging feature is tightly integrated into glibc
malloc, so thanks to an idea from Florian Weimer, much of the malloc
implementation has been moved into libc_malloc_debug.so to support
malloc-check. Due to this, glibc malloc and malloc-check can no
longer work together; they use altogether different (but identical)
structures for heap management. This should not make a difference
though since the malloc check hook is not disabled anywhere.
malloc_set_state does, but it does so early enough that it shouldn't
cause any problems.
The malloc check tunable is now in the debug DSO and has no effect
when the DSO is not preloaded.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Add hp-timing.h using the cntvct_el0 counter. Return timing in nanoseconds
so it is fully compatible with generic hp-timing. Don't set HP_TIMING_INLINE
in the dynamic linker since it adds unnecessary overheads and some ancient
kernels may not handle emulating cntcvt correctly. Currently cntvct_el0 is
only used for timing in the benchtests.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Optimize strnlen by avoiding UMINV which is slow on most cores. On Neoverse N1
large strings are 1.8x faster than the current version, and bench-strnlen is
50% faster overall. This version is MTE compatible.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
Add inline assembler for the roundeven functions.
Passes GLIBC regression. Note GCC does not inline the builtin (PR100966),
so this cannot be used for now.
This patch optimizes the performance of memset for A64FX [1] which
implements ARMv8-A SVE and has L1 64KB cache per core and L2 8MB cache
per NUMA node.
The performance optimization makes use of Scalable Vector Register
with several techniques such as loop unrolling, memory access
alignment, cache zero fill and prefetch.
SVE assembler code for memset is implemented as Vector Length Agnostic
code so theoretically it can be run on any SOC which supports ARMv8-A
SVE standard.
We confirmed that all testcases have been passed by running 'make
check' and 'make xcheck' not only on A64FX but also on ThunderX2.
And also we confirmed that the SVE 512 bit vector register performance
is roughly 4 times better than Advanced SIMD 128 bit register and 8
times better than scalar 64 bit register by running 'make bench'.
[1] https://github.com/fujitsu/A64FX
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
Reviewed-by: Szabolcs Nagy <Szabolcs.Nagy@arm.com>
