When fixing namespace issues for <fenv.h> functions I missed one call
to fesetenv for powerpc-nofpu. This patch changes this to a call to
__fesetenv.
Tested for powerpc-nofpu; it fixes the previously observed math.h
linknamespace test failures.
[BZ #17748]
* sysdeps/powerpc/nofpu/feholdexcpt.c (__feholdexcept): Call
__fesetenv instead of fesetenv.
commit 050f7298e1 added an extern
declaration for __tls_get_addr that conflicts with the one in s390
dl-tls.h, based on whether __tls_get_addr is defined as a macro. The
rationale seems to be based on the assumption that __tls_get_addr is
exported for every architecture and hence an internal non-plt alias is
needed. This is not true for s390 though, since it exports
__tls_get_offset and not __tls_get_addr. This results in tst-audit9
being stuck in an infinite loop.
This patch fixes this by defining a __tls_get_addr macro to itself so
as to not use the conflicting declaration.
This patch fixes a performance regression on the POWER7/PPC64 memcmp
porting for Little Endian. The LE code uses 'ldbrx' instruction to read
the memory on byte reversed form, however ISA 2.06 just provide the indexed
form which uses a register value as additional index, instead of a fixed value
enconded in the instruction.
And the port strategy for LE uses r0 index value and update the address
value on each compare loop interation. For large compare size values,
it adds 8 more instructions plus some more depending of trailing
size. This patch fixes it by adding pre-calculate indexes to remove the
address update on loops and tailing sizes.
For large sizes it shows a considerable gain, with double performance
pairing with BE.
This patch adds an optimized POWER8 strncmp. The implementation focus
on speeding up unaligned cases follwing the ideas of power8 strcmp.
The algorithm first check the initial 16 bytes, then align the first
function source and uses unaligned loads on second argument only.
Aditional checks for page boundaries are done for unaligned cases
(where sources alignment are different).
This patch optimized the POWER7 trailing check by avoiding using byte
read operations and instead use the doubleword already readed with
bitwise operations.
This patch adds an optimized POWER8 strcmp using unaligned accesses.
The algorithm first check the initial 16 bytes, then align the first
function source and uses unaligned loads on second argument only.
Aditional checks for page boundaries are done for unaligned cases
This patch adds an optimized POWER8 st{r,p}ncpy using unaligned accesses.
It shows 10%-80% improvement over the optimized POWER7 one that uses
only aligned accesses, specially on unaligned inputs.
The algorithm first read and check 16 bytes (if inputs do not cross a 4K
page size). The it realign source to 16-bytes and issue a 16 bytes read
and compare loop to speedup null byte checks for large strings. Also,
different from POWER7 optimization, the null pad is done inline in the
implementation using possible unaligned accesses, instead of realying on
a memset call. Special case is added for page cross reads.
With 3eb38795db (Simplify strncat) the generic algorithms uses
strlen, strnlen, and memcpy. This is faster than POWER7 current
implementation, especially for unaligned strings (where POWER7 code
uses byte-byte operations).
This patch removes the assembly implementation and uses a multiarch
specialization based on default algorithm calling optimized POWER7
symbols.
This patch adds an optimized POWER8 strcpy using unaligned accesses.
For strings up to 16 bytes the implementation first calculate the
string size, like strlen, and issues a memcpy. For larger strings,
source is first aligned to 16 bytes and then tested over a loop that
reads 16 bytes am combine the cmpb results for speedup. Special case is
added for page cross reads.
It shows 30%-60% improvement over the optimized POWER7 one that uses
only aligned accesses.
[Modified from the original email by Siddhesh Poyarekar]
This patch solves bug #16009 by implementing an additional path in
strxfrm that does not depend on caching the weight and rule indices.
In detail the following changed:
* The old main loop was factored out of strxfrm_l into the function
do_xfrm_cached to be able to alternativly use the non-caching version
do_xfrm.
* strxfrm_l allocates a a fixed size array on the stack. If this is not
sufficiant to store the weight and rule indices, the non-caching path is
taken. As the cache size is not dependent on the input there can be no
problems with integer overflows or stack allocations greater than
__MAX_ALLOCA_CUTOFF. Note that malloc-ing is not possible because the
definition of strxfrm does not allow an oom errorhandling.
* The uncached path determines the weight and rule index for every char
and for every pass again.
* Passing all the locale data array by array resulted in very long
parameter lists, so I introduced a structure that holds them.
* Checking for zero src string has been moved a bit upwards, it is
before the locale data initialization now.
* To verify that the non-caching path works correct I added a test run
to localedata/sort-test.sh & localedata/xfrm-test.c where all strings
are patched up with spaces so that they are too large for the caching path.
The ldbl-96 implementation of scalblnl (used for x86_64 and ia64) uses
a condition k <= -63 to determine when a standard underflowing result
tiny*__copysignl(tiny,x) should be returned. However, that condition
corresponds to values with exponent -16446 or less, and in the case of
-16446, the correct result for round-to-nearest depends on whether the
value is exactly 0x1p-16446 (half the least subnormal) or more than
that. This patch fixes the bug by changing the condition to k <= -64
and accordingly adjusting the exponent by 64 not 63 when converting to
a normal value.
Tested for x86_64.
[BZ #17803]
* sysdeps/ieee754/ldbl-96/s_scalblnl.c (twom63): Rename to
twom64. Adjust value to 0x1p-64L.
(__scalblnl): Only return standard underflowing result for K <=
-64 not K <= -63; adjust exponent for underflowing result by 64
not 63.
* math/libm-test.inc (scalbn_test_data): Add more tests.
(scalbln_test_data): Likewise.
The ldbl-96 implementation of scalblnl (used for x86_64 and ia64) is
incorrect for subnormal arguments (this is a separate bug from bug
17803, which is about underflowing results). There are two problems
with the adjustments of subnormal arguments: the "two63" variable
multiplied by is actually 0x1p52L not 0x1p63L, so is insufficient to
make values normal, and then GET_LDOUBLE_EXP(es,x), used to extract
the new exponent, extracts it into a variable that isn't used, while
the value taken to by the new exponent is wrongly taken from the high
part of the mantissa before the adjustment (hx). This patch fixes
both those problems and adds appropriate tests.
Tested for x86_64.
[BZ #17834]
* sysdeps/ieee754/ldbl-96/s_scalblnl.c (two63): Change value to
0x1p63L.
(__scalblnl): Get new exponent of adjusted subnormal value from ES
not HX.
* math/libm-test.inc (scalbn_test_data): Add more tests.
(scalbln_test_data): Likewise.
Linux 3.15 adds support for clock_gettime, gettimeofday, and time vDSO
(commit id 37c975545ec63320789962bf307f000f08fabd48). This patch adds
GLIBC supports to use such symbol when they are avaiable.
Along with x86 vDSO support, this patch cleanup x86_64 code by moving
all common code to x86 common folder. Only init-first.c is different
between implementations.
Linux kernel powerpc documentation states issuing a syscall inside a
transaction is not recommended and may lead to undefined behavior. It
also states syscalls does not abort transactoin neither they run in
transactional state.
To avoid side-effects being visible outside transactions, GLIBC with
lock elision enabled will issue a transaction abort instruction just
before all syscalls if hardware supports hardware transactions.
This patch adds support for lock elision using ISA 2.07 hardware
transactional memory for rwlocks. The logic is similar to the
one presented in pthread_mutex lock elision.
This patch adds support for lock elision using ISA 2.07 hardware
transactional memory instructions for pthread_mutex primitives.
Similar to s390 version, the for elision logic defined in
'force-elision.h' is only enabled if ENABLE_LOCK_ELISION is defined.
Also, the lock elision code should be able to be built even with
a compiler that does not provide HTM support with builtins.
However I have noted the performance is sub-optimal due scheduling
pressures.
Microblaze apparently has a variable page size (see thread below) and
should not hard-code any page-size related macros.
Also remove macros that are only used for BFD's trad-core support
which is not relavant for microblaze also according to the thread
starting here:
https://sourceware.org/ml/libc-ports/2013-11/msg00028.html
This patch is neither built nor tested but mirrors a MIPS patch that
fixes the same issue.
Thanks,
Matthew
* sysdepsysdeps/unix/sysv/linux/microblaze/sys/user.h
(PAGE_SHIFT, PAGE_SIZE, PAGE_MASK, NBPG, UPAGES): Remove.
(HOST_TEXT_START_ADDR, HOST_STACK_END_ADDR): Remove.
Signed-off-by: David Holsgrove <david.holsgrove@xilinx.com>
2015-01-06 David Holsgrove <david.holsgrove@xilinx.com>
* sysdeps/microblaze/jmpbuf-unwind.h (_jmpbuf_sp): Declare SP as void
pointer and cast to uintptr_t.
Signed-off-by: David Holsgrove <david.holsgrove@xilinx.com>
2015-01-06 David Holsgrove <david.holsgrove@xilinx.com>
* sysdeps/microblaze/nptl/tls.h (TLS_INIT_TP): Use NULL instead
of 0.
Signed-off-by: David Holsgrove <david.holsgrove@xilinx.com>
GCC 5.0 emits an warning when using sizeof on array function parameters
and powerpc internal syscall macros add a check for such cases. More
specifically, on powerpc64 and powerpc32 sysdep.h:
if (__builtin_classify_type (__arg3) != 5 && sizeof (__arg3) > 8) \
__illegally_sized_syscall_arg3 (); \
And for sysdeps/unix/sysv/linux/utimensat.c build GCC emits:
error: ‘sizeof’ on array function parameter ‘tsp’ will return size of
‘const struct timespec *’
This patch uses the address of first struct member instead of the struct
itself in syscall macro.
Concluding the fixes for C90 libm functions calling C99 fe* functions,
this patch fixes the case of feupdateenv by making it a weak alias for
__feupdateenv and making the affected code call __feupdateenv.
Tested for x86_64 (testsuite, and that installed stripped shared
libraries are unchanged by the patch). Also tested for ARM
(soft-float) that the math.h linknamespace tests now pass.
[BZ #17748]
* include/fenv.h (__feupdateenv): Use libm_hidden_proto.
* math/feupdateenv.c (__feupdateenv): Use libm_hidden_def.
* sysdeps/aarch64/fpu/feupdateenv.c (feupdateenv): Rename to
__feupdateenv and define as weak alias of __feupdateenv. Use
libm_hidden_weak.
* sysdeps/alpha/fpu/feupdateenv.c (__feupdateenv): Use
libm_hidden_def.
* sysdeps/arm/feupdateenv.c (feupdateenv): Rename to __feupdateenv
and define as weak alias of __feupdateenv. Use libm_hidden_weak.
* sysdeps/hppa/fpu/feupdateenv.c (feupdateenv): Likewise.
* sysdeps/i386/fpu/feupdateenv.c (__feupdateenv): Use
libm_hidden_def.
* sysdeps/ia64/fpu/feupdateenv.c (feupdateenv): Rename to
__feupdateenv and define as weak alias of __feupdateenv. Use
libm_hidden_weak.
* sysdeps/m68k/fpu/feupdateenv.c (__feupdateenv): Use
libm_hidden_def.
* sysdeps/mips/fpu/feupdateenv.c (feupdateenv): Rename to
__feupdateenv and define as weak alias of __feupdateenv. Use
libm_hidden_weak.
* sysdeps/powerpc/fpu/feupdateenv.c (__feupdateenv): Use
libm_hidden_def.
* sysdeps/powerpc/nofpu/feupdateenv.c (__feupdateenv): Likewise.
* sysdeps/powerpc/powerpc32/e500/nofpu/feupdateenv.c
(__feupdateenv): Likewise.
* sysdeps/s390/fpu/feupdateenv.c (feupdateenv): Rename to
__feupdateenv and define as weak alias of __feupdateenv. Use
libm_hidden_weak.
* sysdeps/sh/sh4/fpu/feupdateenv.c (feupdateenv): Likewise.
* sysdeps/sparc/fpu/feupdateenv.c (__feupdateenv): Use
libm_hidden_def.
* sysdeps/tile/math_private.h (__feupdateenv): New inline
function.
* sysdeps/x86_64/fpu/feupdateenv.c (__feupdateenv): Use
libm_hidden_def.
* sysdeps/generic/math_private.h (default_libc_feupdateenv): Call
__feupdateenv instead of feupdateenv.
(default_libc_feupdateenv_test): Likewise.
(libc_feresetround_ctx): Likewise.
We see some surprising warnings on tilegx with gcc 4.8.2:
In file included from regex.c:66:0:
regcomp.c: In function ‘parse_expression’:
regcomp.c:2849:15: error: ‘end_elem’ may be used uninitialized in this
function [-Werror=maybe-uninitialized]
else if (br_elem->type == COLL_SYM)
^
regcomp.c:3109:34: note: ‘end_elem’ was declared here
bracket_elem_t start_elem, end_elem;
^
regcomp.c:3109:22: error: ‘start_elem’ may be used uninitialized in
this function [-Werror=maybe-uninitialized]
bracket_elem_t start_elem, end_elem;
^
These warnings are not seen on x86, and in fact if I compile the
preprocessed tile sources with the x86 gcc 4.8.2, I don't see the
warnings. I do see eqiuvalent warnings if I compile the
x86-preprocessed source code with tilegx gcc 4.8.2.
The fix here is to initialize the union type field appropriately in
a couple of places where we pass a union pointer to a subroutine that
"knows" what type the union is. Setting the type explicitly seems like
a more robust way to manage such a data structure in any case.
Chung-Lin Tang
Stefan Liebler
Joseph Myers
Siddhesh Poyarekar
Adhemerval Zanella
Rajalakshmi Srinivasaraghavan
David S. Miller
Roland McGrath
Leonhard Holz
Torvald Riegel
Andreas Schwab
Matthew Fortune
David Holsgrove
Richard Henderson
Chris Metcalf
Richard Earnshaw