aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2012-01-11Correct ordering of entries and spurious whitespace on the end of a line.ibm/2.12/masterRyan S. Arnold
2012-01-11Merge Adhemerval Zanella's pipe2 fix into ibm/2.12/master.Adhemerval Zanella
2012-01-05Fix for AIO stack size (backported from 'master' branch)Adhemerval Zanella
2011-12-29Add have_pipe2.Adhemerval Zanella
2011-12-29PowerPC - Optimization for str[n]casecmp functionsAdhemerval Zanella
This patch provides throughput boost for the strcasecmp function (25% on ppc32 and 40% on ppc64) and strncasecmp (15% on both ppc32 and ppc64) for POWER7. The optimization is done by manually (strcasecmp) or automatically (strncasecmp) unrolling the test loop to avoid CPU stalls caused by a test followed by a load.
2011-12-29PowerPC - Optimization for nearbyint[f]Adhemerval Zanella
This patch provides throughput boost for the nearbyint[f] functions for POWER. For POWER7, it improves performance for nearbyintf from 5 (ppc32) to 6 times (ppc64) and for nearbyint from 2.5 up to 5 times. For POWER6 it improves nearbyintf up to 2x (ppc64) and nearbyint up to 4x.
2011-12-29Fix compilation due to __nan definesAdhemerval Zanella
Backport of commit #ee2aafe08e68c88412b7dcfa4b1e0d11789348f8
2011-12-29Fix a wrong constant in powerpc hypot implementationAndreas Schwab
Backport of commit #850fb039cec802072f70ed9763927881bbbf639c
2011-12-29PowerPC: Arithmetic function optimizations for POWERAdhemerval Zanella
This patch creates inline assembly functions that use intrinsic PPC floating point instructions when the platform supports them but rely on the internal GLIBC functions when the instructions are not implemented (for instance, on POWER4).
2011-12-29Optimize libmUlrich Drepper
libm is now somewhat integrated with gcc's -ffinite-math-only option and lots of the wrapper functions have been optimized. Backport of commit #0ac5ae2335292908f39031b1ea9fe8edce433c0f
2011-12-29Fix PLT uses in libm on powerpcAndreas Schwab
Backport of commit #edc121be8b53da57950daf6f42a619b5826f30b6
2011-09-28Fix POWER4/POWER7 optimized strncmp to not read past differing bytesAndreas Schwab
2011-09-12Optimize the aligned copy for power7 with vector-scalar insns.Will Schmidt
2011-09-12Prevent Altivec and VSX insns on PowerPC64 when no FPRs or VRs are avail.Ryan S. Arnold
2011-09-12Fix profiling on powerpc32 secure-plt shared libs and PIEsAlan Modra
This patch moves the ppc32 _mcount to libc_shared.a, fixing a long-standing bug with profiling of secure-plt shared libraries and PIEs. The problem is that a ppc32 PIC PLT call stub uses r30 (GOT pointer) to load the function address from the PLT, r30 being set up in the function prologue, but _mcount is called before the function prologue. So chances are good that r30 will be pointing to the executable GOT when trying to call _mcount in a shared lib function. A similar problem can occur in a PIE if a shared lib calls a function in the executable.
2011-09-12Check for finite/infinity parms in IBM Long Double 128 fmal( )Dave Flaherty
This patch addresses some IBM Long Double 128 fmal () test-ldouble.out and test-ildoubl.out failures. If the ‘x’ and ‘y’ parameters are finite values and ‘z’ is infinity, the result of fmal () should be ‘z’ not NaN.
2011-09-12Provide a throughput boost to the 64-bit power7 strncmp code of approxWill Schmidt
15%. The 32-bit throughput is not notably affected by this change, so the change to the 32-bit code is done to keep the two files in sync with each other.
2011-09-12Trigonometric optimizations for POWER cpusAdhemerval Zanella
These POWER optimizations remove most of the FP->INT conversions in hypot/hypotf and sinf/cosf on POWER making the computation done with FP operations instead. This eliminates Load-Hit-Store (LHS) stalls, increasing performance of hypot/hypotf (by about 100% on POWER7 and 12% on POWER6) and sinf/cosf (by 80% on POWER7 and 30% on POWER6).
2011-09-12Complete revert of git commit ec11dbe1aabbb45d41f5e9c21e315e2496f3b473Ryan S. Arnold
Avoid deadcode and remove regressive optimization for POWER[5|6]. Complete revert of git commit 5939fc0867e1616a537c26d84ca1612b53b0303e Trigonometric optimizations for POWER cpus
2011-09-09Revert git commit ec11dbe1aabbb45d41f5e9c21e315e2496f3b473Ryan S. Arnold
Avoid deadcode and remove regressive optimization for POWER[5|6]. Revert git commit 5939fc0867e1616a537c26d84ca1612b53b0303e Trigonometric optimizations for POWER cpus
2011-07-01Remove pthread_attr_setstacksize from __aio_create_helper_thread.ryanarn
2011-07-01Revert git commit db90a4844fc2f0eb7d6a45a317d9e625cca370e8ryanarn
2011-07-01Static TLS memory leak fix (DTV case)Alan Modra
2011-07-01Set stacksize in __aio_create_helper_thread to default for AIO with TLS.Ryan S. Arnold
2011-06-28Avoid deadcode and remove regressive optimization for POWER[5|6].Adhemerval Zanella
2011-06-17Trigonometric optimizations for POWER cpusAdhemerval Zanella
These optimizations remove most of the FP->INT conversions making the computation done with FP operations instead. This eliminates Load-Hit-Store (LHS) stalls on POWER, increasing performance of hypot/hypotf (about 50% on POWER7, 25% on POWER6, and 30% on POWER5) and sinf/cosf (30% on POWER7, 15% on POWER6, and 10% on POWER5).
2011-05-25Correct "//" in search paths due to LD_AT_PLATFORM=""Ryan S. Arnold
2011-05-12Fix two printf handler issues.Jakub Jelinek
(cherry picked from commit c1d0e639a95d6d3d3f1c1f70cf97d596bd5a24ec)
2011-03-16Correct POWER6 END(__isnan) to END(__isnanf).Ryan S. Arnold
2011-02-17Add #include <stdint.h> to tst-writev.c for backport.Ryan S. Arnold
2011-02-17Fix INTERNAL_[V]SYSCALL_NCS macros to not cast return val to int.Ryan S. Arnold
(cherry picked from commit d55fd7a5573ea7c7c14af95bc37618469cde8a5f)
2011-02-16Prevent VSX type TOC ref in _dl_start before relocs are resolved.Ryan S. Arnold
Disable VSX instruction usage in rtld.c with -mno-vsx so that, under -O3 optimization, a TOC reference isn't used for a zero constant in a VSX register prior to resolution of relocations.
2011-02-16Reverse -mno-vsx commits in order to correct it later.Ryan S. Arnold
Reversing commits c3b77b811fc7106a4cc9d491c1cc9358fb266c42 and 4265a8e3556a90e6e2ee583388d8aaa0d93cfa29
2011-02-15Prevent VSX type TOC ref in _dl_start before relocs are resolved.Ryan S. Arnold
Disable VSX instruction usage in rtld.c with -mno-vsx so that, under -O3 optimization, a TOC reference isn't used for a zero constant in a VSX register prior to resolution of relocations.
2011-02-15Prevent VSX type TOC ref in _dl_start before relocs are resolved.Ryan S. Arnold
Disable VSX instruction usage in rtld.c with -mno-vsx so that, under -O3 optimization, a TOC reference isn't used for a zero constant in a VSX register prior to resolution of relocations.
2010-12-03Fix comparison in sqrtl for IBM long double 128.Luis Machado
(cherry picked from commit da93d21475878725c9e0cb2b6e650bd8d3628435)
2010-11-05power7-optimized mempcpyLuis Machado
(cherry picked from commit 344d0b545d0a0a0ab737ff333d807969721ce381)
2010-11-05This patch is not upstream. It is useful and non-invasive. It allows someRyan Arnold
recent versions of GCC to compile GLIBC when altivec is configured. 2009-11-06 Petr Baudis <pasky@suse.cz> * include/sys/uio.h: Change __vector to __iovec to avoid clash with altivec. (cherry picked from commit bcb75669959f740e04208b773e99d87c5342dc34)
2010-11-01PowerPC64 doesn't need an executable stack and therefore doesn't needRyan Arnold
PT_GNU_STACK to make the stack no-exec. This change abstracts the stack permissions settings into a macro defined in a header.
2010-11-01IEEE quad fmal fixesJakub Jelinek
(cherry picked from commit 7c08a05c5f4f86f9e249c2554b1628da0f812364)
2010-11-01Add C99 FP_FAST_FMA{,F,L} macros to math.hMichael Meissner
(cherry picked from commit d20501241ea423e477b357d93147530987e0d89e)
2010-11-01Missing CL.Jakub Jelinek
(cherry picked from commit 475facd1f553ad6d45de1137bd2a22e70307c7bf)
2010-11-01Implement fmal, some fma bugfixesJakub Jelinek
(cherry picked from commit 3e692e0518b4f4679352d25102bd47cf3f85c592)
2010-11-01Fix some more dbl-64/s_fma.c issueJakub Jelinek
(cherry picked from commit f3f7372de1401b99f0a318ce09caf73e42d6f022)
2010-11-01Implement accurate fma.Jakub Jelinek
(cherry picked from commit 5e908464b9603e9cd95047d2edd00de5f78fedaf)
2010-11-01Correct implementation of fmaf.Jakub Jelinek
2010-11-01powerpc: PPCA2/A2 optimized memcpy functionMichael B. Brutman
(cherry picked from commit d0b9e94faf18f2585e7d0ae0b10daefb12f93059)
2010-11-01PowerPC64 ABI fixesAlan Modra
(cherry picked from commit bebff237c522e4e8e23204ca1e5104896389158e)
2010-11-01Whitespace fixes.Ulrich Drepper
(cherry picked from commit 9b059f977477351b1660cf50f6925e1eb0ec498b)
2010-11-01POWER6/7 optimizations for copysignLuis Machado
(cherry picked from commit 77c1b06986d9ce8aacee8ea734ba42e635d79b2f)
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-03 04:22:51 +0000