Tcwg-commits July 2015

tcwg-commits@lists.linaro.org

2 participants
1355 discussions

[gcc] 04/08: Turn on X86_TUNE_USE_BT for IA MCU

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit 69eee0f57e384755a5c4c2753365d5ce82db6fcc Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Thu Jul 9 18:51:01 2015 -0700 Turn on X86_TUNE_USE_BT for IA MCU --- gcc/config/i386/x86-tune.def | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def index eb271b8..6d65ca8 100644 --- a/gcc/config/i386/x86-tune.def +++ b/gcc/config/i386/x86-tune.def @@ -292,7 +292,7 @@ DEF_TUNE (X86_TUNE_USE_CLTD, "use_cltd", /* X86_TUNE_USE_BT: Enable use of BT (bit test) instructions. */ DEF_TUNE (X86_TUNE_USE_BT, "use_bt", m_CORE_ALL | m_BONNELL | m_SILVERMONT | m_KNL | m_INTEL - | m_AMD_MULTIPLE | m_GENERIC) + | m_IAMCU | m_AMD_MULTIPLE | m_GENERIC) /*****************************************************************************/ /* 387 instruction selection tuning */ -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] branch hjl/iamcu/improve updated (9334ac2 -> 985b5b7)

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a change to branch hjl/iamcu/improve in repository gcc. discards 9334ac2 IA MCU run-time doesn't support TLS discards b980e72 Skip incompatible tests on IA MCU target discards fe133bc Skip error_mark_node type decls discards c452977 Turn off loop and funcion alignments for IA MCU discards c53e8f9 Turn on X86_TUNE_USE_BT for IA MCU discards 5ccb678 Turn off X86_TUNE_ZERO_EXTEND_WITH_AND for IA MCU discards db8fdf5 Allow indirect sibcall with register arguments discards ad2477a Allow SFmode for IA MCU in FP splitter new acf8a75 Allow software FP SFmode in FP splitter new c851071 Allow indirect sibcall with register arguments new 31cdcd0 Turn off X86_TUNE_ZERO_EXTEND_WITH_AND for IA MCU new 69eee0f Turn on X86_TUNE_USE_BT for IA MCU new e0e3eb8 Turn off loop and funcion alignments for IA MCU new 577a3c8 Skip error_mark_node type decls new f954767 Skip incompatible tests on IA MCU target new 985b5b7 IA MCU run-time doesn't support TLS This update added new revisions after undoing existing revisions. That is to say, some revisions that were in the old version of the branch are not in the new version. This situation occurs when a user --force pushes a change and generates a repository containing something like this: * -- * -- B -- O -- O -- O (9334ac2) \ N -- N -- N refs/heads/hjl/iamcu/improve (985b5b7) You should already have received notification emails for all of the O revisions, and so the following emails describe only the N revisions from the common base, B. Any revisions marked "omits" are not gone; other references still refer to them. Any revisions marked "discards" are gone forever. The 8 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference. Summary of changes: gcc/config/i386/i386.c | 8 ++++++-- gcc/config/i386/i386.md | 5 ++++- gcc/testsuite/gcc.target/i386/pr66824.c | 8 +++++++- 3 files changed, 17 insertions(+), 4 deletions(-) -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 01/08: Allow software FP SFmode in FP splitter

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit acf8a75eb6fda26dc23318e5aef6aee89431f776 Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Thu Jul 9 12:06:40 2015 -0700 Allow software FP SFmode in FP splitter ix86_split_long_move can optimize floating point constant move, which can be used to optimize SFmode move with software floating point. gcc/ PR target/66824 * config/i386/i386.c (ix86_split_to_parts): Allow SFmode move without 387, MMX nor SSE. (ix86_split_long_move): Support single move. * config/i386/i386.md (FP splitter): Allow SFmode without 387, MMX nor SSE. gcc/testsuite/ PR target/66824 * gcc.target/i386/pr66824.c: New test. --- gcc/config/i386/i386.c | 18 +++++++++++++++++- gcc/config/i386/i386.md | 6 +++++- gcc/testsuite/gcc.target/i386/pr66824.c | 29 +++++++++++++++++++++++++++++ 3 files changed, 51 insertions(+), 2 deletions(-) diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c index a18c733..a9a67a9 100644 --- a/gcc/config/i386/i386.c +++ b/gcc/config/i386/i386.c @@ -22744,7 +22744,13 @@ ix86_split_to_parts (rtx operand, rtx *parts, ma [...] size = (GET_MODE_SIZE (mode) + 4) / 8; gcc_assert (!REG_P (operand) || !MMX_REGNO_P (REGNO (operand))); - gcc_assert (size >= 2 && size <= 4); + /* For software floating point, we also optimize SFmode move. */ + gcc_assert ((size >= 2 + || (mode == SFmode + && !TARGET_80387 + && !TARGET_MMX + && !TARGET_SSE)) + && size <= 4); /* Optimize constant pool reference to immediates. This is used by fp moves, that force all constants to memory to allow combining. */ @@ -22822,10 +22828,14 @@ ix86_split_to_parts (rtx operand, rtx *parts, m [...] case DFmode: REAL_VALUE_TO_TARGET_DOUBLE (r, l); break; + case SFmode: + REAL_VALUE_TO_TARGET_SINGLE (r, l[0]); + goto part0; default: gcc_unreachable (); } parts[1] = gen_int_mode (l[1], SImode); +part0: parts[0] = gen_int_mode (l[0], SImode); } else @@ -22932,6 +22942,12 @@ ix86_split_long_move (rtx operands[]) nparts = ix86_split_to_parts (operands[1], part[1], GET_MODE (operands[0])); ix86_split_to_parts (operands[0], part[0], GET_MODE (operands[0])); + if (nparts == 1) + { + emit_move_insn (part[0][0], part[1][0]); + return; + } + /* When emitting push, take care for source operands on the stack. */ if (push && MEM_P (operands[1]) && reg_overlap_mentioned_p (stack_pointer_rtx, operands[1])) diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md index 1d43aaf..11e3a81 100644 --- a/gcc/config/i386/i386.md +++ b/gcc/config/i386/i386.md @@ -3507,7 +3507,11 @@ "reload_completed && (GET_MODE (operands[0]) == TFmode || GET_MODE (operands[0]) == XFmode - || GET_MODE (operands[0]) == DFmode) + || GET_MODE (operands[0]) == DFmode + || (GET_MODE (operands[0]) == SFmode + && !TARGET_80387 + && !TARGET_MMX + && !TARGET_SSE)) && !(ANY_FP_REG_P (operands[0]) || ANY_FP_REG_P (operands[1]))" [(const_int 0)] "ix86_split_long_move (operands); DONE;") diff --git a/gcc/testsuite/gcc.target/i386/pr66824.c b/gcc/testsuite/gcc. [...] new file mode 100644 index 0000000..3511e4c --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66824.c @@ -0,0 +1,29 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-O2 -mno-sse -mno-mmx -mno-80387" } */ +/* { dg-final { scan-assembler-not "\.LC\[0-9\]" } } */ + +double foo (float); + +double +f1 (void) +{ + return foo (1.0); +} + +double +f2 (void) +{ + return foo (0.0); +} + +void +f3 (float *x, float t) +{ + *x = 0.0 + t; +} + +float +f4 (void) +{ + return 1.0; +} -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] branch hjl/iamcu/improve updated (97cf0cb -> 9334ac2)

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a change to branch hjl/iamcu/improve in repository gcc. discards 97cf0cb IA MCU run-time doesn't support TLS discards aad652e Skip incompatible tests on IA MCU target discards 60d8da9 Skip error_mark_node type decls discards e6f431f Allow indirect sibcall with register arguments discards ee6e3eb Allow SFmode for IA MCU in FP splitter adds 68fea6c Daily bump. adds 4c8d20e 2015-07-09 Paolo Carlini <paolo.carlini(a)oracle.com> adds cb791e7 2015-07-10 Richard Biener <rguenther(a)suse.de> adds 8f84077 ipa-icf.c: Fix typo in dump message adds 9cda83a 2015-07-10 Richard Biener <rguenther(a)suse.de> adds 3c31a6c Insert new bound in try_transform_to_exit_first_loop_alt adds 2e3a49a 2015-07-10 Paolo Carlini <paolo.carlini(a)oracle.com> adds a3e1e71 [obvious] Fix typos above expand_cond_expr_using_cmove adds c3a8197 PR middle-end/66820 * gimplify.c (maybe_fold_stmt [...] new ad2477a Allow SFmode for IA MCU in FP splitter new db8fdf5 Allow indirect sibcall with register arguments new 5ccb678 Turn off X86_TUNE_ZERO_EXTEND_WITH_AND for IA MCU new c53e8f9 Turn on X86_TUNE_USE_BT for IA MCU new c452977 Turn off loop and funcion alignments for IA MCU new fe133bc Skip error_mark_node type decls new b980e72 Skip incompatible tests on IA MCU target new 9334ac2 IA MCU run-time doesn't support TLS This update added new revisions after undoing existing revisions. That is to say, some revisions that were in the old version of the branch are not in the new version. This situation occurs when a user --force pushes a change and generates a repository containing something like this: * -- * -- B -- O -- O -- O (97cf0cb) \ N -- N -- N refs/heads/hjl/iamcu/improve (9334ac2) You should already have received notification emails for all of the O revisions, and so the following emails describe only the N revisions from the common base, B. Any revisions marked "omits" are not gone; other references still refer to them. Any revisions marked "discards" are gone forever. The 8 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference. Summary of changes: gcc/ChangeLog | 36 +++++++++++++++++++ gcc/DATESTAMP | 2 +- gcc/config/i386/i386.c | 2 +- gcc/config/i386/x86-tune.def | 4 +-- gcc/expr.c | 6 ++-- gcc/gimplify.c | 9 ++--- gcc/ipa-icf.c | 2 +- gcc/omp-low.c | 4 +-- gcc/testsuite/ChangeLog | 26 ++++++++++++++ gcc/testsuite/g++.dg/cpp0x/lambda/lambda-ice14.C | 32 +++++++++++++++++ gcc/testsuite/g++.dg/cpp0x/scoped_enum6.C | 19 ++++++++++ gcc/testsuite/gcc.dg/gomp/pr66820.c | 18 ++++++++++ gcc/testsuite/gcc.dg/vect/pr61194.c | 3 +- .../gfortran.dg/parloops-exit-first-loop-alt-2.f95 | 24 +++++++++++++ .../gfortran.dg/parloops-exit-first-loop-alt.f95 | 25 +++++++++++++ gcc/tree-if-conv.c | 38 ++++++++------------ gcc/tree-parloops.c | 12 ++++++- libgomp/ChangeLog | 5 +++ .../parloops-exit-first-loop-alt-2.f95 | 40 +++++++++++++++++++++ .../parloops-exit-first-loop-alt.f95 | 41 ++++++++++++++++++++++ 20 files changed, 308 insertions(+), 40 deletions(-) create mode 100644 gcc/testsuite/g++.dg/cpp0x/lambda/lambda-ice14.C create mode 100644 gcc/testsuite/g++.dg/cpp0x/scoped_enum6.C create mode 100644 gcc/testsuite/gcc.dg/gomp/pr66820.c create mode 100644 gcc/testsuite/gfortran.dg/parloops-exit-first-loop-alt-2.f95 create mode 100644 gcc/testsuite/gfortran.dg/parloops-exit-first-loop-alt.f95 create mode 100644 libgomp/testsuite/libgomp.fortran/parloops-exit-first [...] create mode 100644 libgomp/testsuite/libgomp.fortran/parloops-exit-first [...] -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 07/08: Skip incompatible tests on IA MCU target

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit b980e727e5f06cd9b2dc4ff7a7b936c436328389 Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Wed Jul 8 04:55:05 2015 -0700 Skip incompatible tests on IA MCU target --- gcc/testsuite/gcc.target/i386/i386.exp | 92 +++++++++++++++++++++++++++++++++- 1 file changed, 90 insertions(+), 2 deletions(-) diff --git a/gcc/testsuite/gcc.target/i386/i386.exp b/gcc/testsuite/gcc.t [...] index 8ca6c51..1b4657d 100644 --- a/gcc/testsuite/gcc.target/i386/i386.exp +++ b/gcc/testsuite/gcc.target/i386/i386.exp @@ -394,7 +394,12 @@ global runtests # Special case compilation of vect-args.c so we don't have to # replicate it 16 times. if [runtest_file_p $runtests $srcdir/$subdir/vect-args.c] { - foreach type { "" -mmmx -m3dnow -msse -msse2 -mavx -mavx2 -mavx512f } { + if { [istarget *-*-elfiamcu] } then { + set types { "" } + } else { + set types { "" -mmmx -m3dnow -msse -msse2 -mavx -mavx2 -mavx512f } + } + foreach type $types { foreach level { "" -O } { set flags "$type $level" verbose -log "Testing vect-args, $flags" 1 @@ -405,7 +410,90 @@ if [runtest_file_p $runtests $srcdir/$subdir/vect-args.c] { # Everything else. set tests [lsort [glob -nocomplain $srcdir/$subdir/*.\[cS\]]] -set tests [prune $tests $srcdir/$subdir/vect-args.c] +set all_tests [prune $tests $srcdir/$subdir/vect-args.c] +if { [istarget *-*-elfiamcu] } then { + # Skip tests which are incompatible with Intel MCU. + set tests {} + foreach t $all_tests { + if { [string match "$srcdir/$subdir/20001127-1.c" $t] \ + || [string match "$srcdir/$subdir/20080723-1.c" $t] \ + || [string match "$srcdir/$subdir/980226-1.c" $t] \ + || [string match "$srcdir/$subdir/980414-1.c" $t] \ + || [string match "$srcdir/$subdir/addr-sel-1.c" $t] \ + || [string match "$srcdir/$subdir/incoming-5.c" $t] \ + || [string match "$srcdir/$subdir/mcount_pic.c" $t] \ + || [string match "$srcdir/$subdir/pause-2.c" $t] \ + || [string match "$srcdir/$subdir/pow-1.c" $t] \ + || [string match "$srcdir/$subdir/pr30848.c" $t] \ + || [string match "$srcdir/$subdir/pr45296.c" $t] \ + || [string match "$srcdir/$subdir/pr57736.c" $t] \ + || [string match "$srcdir/$subdir/pr57848.c" $t] \ + || [string match "$srcdir/$subdir/pr58048.c" $t] \ + || [string match "$srcdir/$subdir/pr66047.c" $t] \ + || [string match "$srcdir/$subdir/sibcall-6.c" $t] \ + || [string match "$srcdir/$subdir/sse-5.c" $t] \ + || [string match "$srcdir/$subdir/aggregate-ret*.c" $t] \ + || [string match "$srcdir/$subdir/avx*.c" $t] \ + || [string match "$srcdir/$subdir/funcspec-*.c" $t] \ + || [string match "$srcdir/$subdir/long-double-*.c" $t] \ + || [string match "$srcdir/$subdir/pr57756*.c" $t] \ + || [string match "$srcdir/$subdir/pr59390*.c" $t] \ + || [string match "$srcdir/$subdir/pr59794-*.c" $t] \ + || [string match "$srcdir/$subdir/pr60205-*.c" $t] \ + || [string match "$srcdir/$subdir/pr61925-*.c" $t] } { + continue + } + set options [dg-get-options $t] + if { ! [string match "*-mmmx*" $options] \ + && ! [string match "*-m3dnow*" $options] \ + && ! [string match "*-msse*" $options] \ + && ! [string match "*-mssse*" $options] \ + && ! [string match "*-mavx*" $options] \ + && ! [string match "*-madx*" $options] \ + && ! [string match "*-maes*" $options] \ + && ! [string match "*-mbmi*" $options] \ + && ! [string match "*-mclflushopt*" $options] \ + && ! [string match "*-mclwb*" $options] \ + && ! [string match "*-mf16c*" $options] \ + && ! [string match "*-mfma*" $options] \ + && ! [string match "*-mfxsr*" $options] \ + && ! [string match "*-mhle*" $options] \ + && ! [string match "*-mlzcnt*" $options] \ + && ! [string match "*-mmwaitx*" $options] \ + && ! [string match "*-mpclmul*" $options] \ + && ! [string match "*-mpcommit*" $options] \ + && ! [string match "*-mprefetchwt1*" $options] \ + && ! [string match "*-mprfchw*" $options] \ + && ! [string match "*-mrdrnd*" $options] \ + && ! [string match "*-mrdseed*" $options] \ + && ! [string match "*-mrtm*" $options] \ + && ! [string match "*-msha*" $options] \ + && ! [string match "*-mtbm*" $options] \ + && ! [string match "*-mxop*" $options] \ + && ! [string match "*-mxsave*" $options] \ + && ! [string match "*-mvzeroupper*" $options] \ + && ! [string match "*-march=i686*" $options] \ + && ! [string match "*-march=atom*" $options] \ + && ! [string match "*-march=core*" $options] \ + && ! [string match "*-march=nocona*" $options] \ + && ! [string match "*-march=pentium*" $options] \ + && ! [string match "*-march=amdfam10*" $options] \ + && ! [string match "*-march=athlon*" $options] \ + && ! [string match "*-march=barcelona*" $options] \ + && ! [string match "*-march=bdver*" $options] \ + && ! [string match "*-march=k6*" $options] \ + && ! [string match "*-march=k8*" $options] \ + && ! [string match "*-march=opteron*" $options] \ + && ! [string match "*-march=x86-64*" $options] \ + && ! [string match "*-march=native*" $options] \ + && ! [string match "*-mfpmath=*" $options] \ + && ! [string match "*-mregparm=*" $options] } { + lappend tests $t + } + } +} else { + set tests $all_tests +} # Main loop. dg-runtest $tests "" $DEFAULT_CFLAGS -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 08/08: IA MCU run-time doesn't support TLS

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit 9334ac23a324dbe8ec1b9e4067d41ddfad2390ef Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Wed Jul 8 19:46:43 2015 -0700 IA MCU run-time doesn't support TLS Return 0 in check_effective_target_tls_native and check_effective_target_tls_emulated for IA MCU target. --- gcc/testsuite/lib/target-supports.exp | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/ta [...] index 1988301..c97fa08 100644 --- a/gcc/testsuite/lib/target-supports.exp +++ b/gcc/testsuite/lib/target-supports.exp @@ -701,8 +701,9 @@ proc check_effective_target_tls {} { proc check_effective_target_tls_native {} { # VxWorks uses emulated TLS machinery, but with non-standard helper - # functions, so we fail to automatically detect it. - if { [istarget *-*-vxworks*] } { + # functions, so we fail to automatically detect it. IA MCU run-time + # doesn't support TLS. + if { [istarget *-*-vxworks*] || [istarget *-*-elfiamcu] } { return 0 } @@ -716,6 +717,11 @@ proc check_effective_target_tls_native {} { # Return 1 if *emulated* thread local storage (TLS) is supported, 0 otherwise. proc check_effective_target_tls_emulated {} { + # IA MCU run-time doesn't support TLS. + if { [istarget *-*-elfiamcu] } { + return 0 + } + # VxWorks uses emulated TLS machinery, but with non-standard helper # functions, so we fail to automatically detect it. if { [istarget *-*-vxworks*] } { -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 02/08: Allow indirect sibcall with register arguments

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit db8fdf5c03419ed3c573797b0f4173179c88b6cb Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Wed Jul 8 21:10:25 2015 -0700 Allow indirect sibcall with register arguments Indirect sibcall with register arguments is OK when there is register available for argument passing. gcc/ PR target/66819 * config/i386/i386.c (ix86_function_ok_for_sibcall): Allow indirect sibcall with register arguments if register available for argument passing. (init_cumulative_args): Set cfun->machine->arg_reg_available_p to cum->nregs != 0. (function_arg_advance_32): Set cfun->machine->arg_reg_available_p to 0 when setting cum->nregs = 0. * config/i386/i386.h (machine_function): Add arg_reg_available_p. gcc/testsuite/ PR target/66819 * gcc.target/i386/pr66819-1.c: New test. * gcc.target/i386/pr66819-2.c: Likewise. * gcc.target/i386/pr66819-3.c: Likewise. * gcc.target/i386/pr66819-4.c: Likewise. * gcc.target/i386/pr66819-5.c: Likewise. --- gcc/config/i386/i386.c | 15 +++++++++------ gcc/config/i386/i386.h | 3 +++ gcc/testsuite/gcc.target/i386/pr66819-1.c | 8 ++++++++ gcc/testsuite/gcc.target/i386/pr66819-2.c | 8 ++++++++ gcc/testsuite/gcc.target/i386/pr66819-3.c | 10 ++++++++++ gcc/testsuite/gcc.target/i386/pr66819-4.c | 12 ++++++++++++ gcc/testsuite/gcc.target/i386/pr66819-5.c | 10 ++++++++++ 7 files changed, 60 insertions(+), 6 deletions(-) diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c index d2925bc..bca1fee 100644 --- a/gcc/config/i386/i386.c +++ b/gcc/config/i386/i386.c @@ -5628,12 +5628,12 @@ ix86_function_ok_for_sibcall (tree decl, tree exp) if (!decl || (TARGET_DLLIMPORT_DECL_ATTRIBUTES && DECL_DLLIMPORT_P (decl))) { - if (ix86_function_regparm (type, NULL) >= 3) - { - /* ??? Need to count the actual number of registers to be used, - not the possible number of registers. Fix later. */ - return false; - } + /* FIXME: The symbol indirect call doesn't need a + call-clobbered register. But we don't know if + this is a symbol indirect call or not here. */ + if (ix86_function_regparm (type, NULL) >= 3 + && !cfun->machine->arg_reg_available_p) + return false; } } @@ -6567,6 +6567,7 @@ init_cumulative_args (CUMULATIVE_ARGS *cum, /* Arg [...] ? X86_64_REGPARM_MAX : X86_64_MS_REGPARM_MAX); } + cfun->machine->arg_reg_available_p = cum->nregs != 0; if (TARGET_SSE) { cum->sse_nregs = SSE_REGPARM_MAX; @@ -6636,6 +6637,7 @@ init_cumulative_args (CUMULATIVE_ARGS *cum, /* Arg [...] else cum->nregs = ix86_function_regparm (fntype, fndecl); } + cfun->machine->arg_reg_available_p = cum->nregs != 0; /* Set up the number of SSE registers used for passing SFmode and DFmode arguments. Warn for mismatching ABI. */ @@ -7584,6 +7586,7 @@ pass_in_reg: { cum->nregs = 0; cum->regno = 0; + cfun->machine->arg_reg_available_p = 0; } break; diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h index 74334ff..0b6e304 100644 --- a/gcc/config/i386/i386.h +++ b/gcc/config/i386/i386.h @@ -2479,6 +2479,9 @@ struct GTY(()) machine_function { /* If true, it is safe to not save/restore DRAP register. */ BOOL_BITFIELD no_drap_save_restore : 1; + /* If true, there is register available for argument passing. */ + BOOL_BITFIELD arg_reg_available_p : 1; + /* During prologue/epilogue generation, the current frame state. Otherwise, the frame state at the end of the prologue. */ struct machine_frame_state fs; diff --git a/gcc/testsuite/gcc.target/i386/pr66819-1.c b/gcc/testsuite/gc [...] new file mode 100644 index 0000000..7c8a1ab --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66819-1.c @@ -0,0 +1,8 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-O2 -mregparm=3" } */ +/* { dg-final { scan-assembler-not "call" } } */ + +void foo(void (*bar)(void)) +{ + bar(); +} diff --git a/gcc/testsuite/gcc.target/i386/pr66819-2.c b/gcc/testsuite/gc [...] new file mode 100644 index 0000000..9de4f97 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66819-2.c @@ -0,0 +1,8 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-fPIC -O2 -mregparm=3" } */ +/* { dg-final { scan-assembler-not "call" } } */ + +void foo(void (*bar)(void)) +{ + bar(); +} diff --git a/gcc/testsuite/gcc.target/i386/pr66819-3.c b/gcc/testsuite/gc [...] new file mode 100644 index 0000000..3bc5a34 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66819-3.c @@ -0,0 +1,10 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-O2 -mregparm=3" } */ +/* { dg-final { scan-assembler-not "call" } } */ + +void (*bar)(int, int); + +void foo(int i, int j) +{ + bar(i, j); +} diff --git a/gcc/testsuite/gcc.target/i386/pr66819-4.c b/gcc/testsuite/gc [...] new file mode 100644 index 0000000..18b2ccf --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66819-4.c @@ -0,0 +1,12 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-O2 -mregparm=3" } */ +/* { dg-final { scan-assembler-not "call" } } */ + +#include <stdarg.h> + +void (*bar)(int, va_list); + +void foo(int i, va_list args) +{ + bar(i, args); +} diff --git a/gcc/testsuite/gcc.target/i386/pr66819-5.c b/gcc/testsuite/gc [...] new file mode 100644 index 0000000..6b019d1 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66819-5.c @@ -0,0 +1,10 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-O2 -mregparm=3" } */ +/* { dg-final { scan-assembler "call" } } */ + +void (*bar)(int, int, int); + +void foo(int i, int j, int k) +{ + bar(i, j, k); +} -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 06/08: Skip error_mark_node type decls

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit fe133bc2eebc892c498b0d26f264a8a0479c2f04 Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Wed Jul 8 08:28:03 2015 -0700 Skip error_mark_node type decls There is no need to analyze error_mark_node type decls. gcc/ PR target/66810 * varpool.cvarpool.c (varpool_node::analyze): Skip error_mark_node type decls. gcc/testsuite/ PR target/66810 * gcc.target/i386/pr66810.c: New test. --- gcc/testsuite/gcc.target/i386/pr66810.c | 10 ++++++++++ gcc/varpool.c | 29 +++++++++++++++++------------ 2 files changed, 27 insertions(+), 12 deletions(-) diff --git a/gcc/testsuite/gcc.target/i386/pr66810.c b/gcc/testsuite/gcc. [...] new file mode 100644 index 0000000..4778b37 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66810.c @@ -0,0 +1,10 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-mno-sse -mno-mmx -miamcu" } */ + +int vv; + +void +i (void) +{ + static int a[vv]; /* { dg-error "storage size" } */ +} diff --git a/gcc/varpool.c b/gcc/varpool.c index 10fa93c..f7c4d46 100644 --- a/gcc/varpool.c +++ b/gcc/varpool.c @@ -514,20 +514,25 @@ varpool_node::get_availability (void) void varpool_node::analyze (void) { - /* When reading back varpool at LTO time, we re-construct the queue in order - to have "needed" list right by inserting all needed nodes into varpool. - We however don't want to re-analyze already analyzed nodes. */ - if (!analyzed) + /* Skip error_mark_node type decls. */ + if (TREE_TYPE (decl) != error_mark_node) { - gcc_assert (!in_lto_p || symtab->function_flags_ready); - /* Compute the alignment early so function body expanders are - already informed about increased alignment. */ - align_variable (decl, 0); + /* When reading back varpool at LTO time, we re-construct the + queue in order to have "needed" list right by inserting all + needed nodes into varpool. We however don't want to re-analyze + already analyzed nodes. */ + if (!analyzed) + { + gcc_assert (!in_lto_p || symtab->function_flags_ready); + /* Compute the alignment early so function body expanders are + already informed about increased alignment. */ + align_variable (decl, 0); + } + if (alias) + resolve_alias (varpool_node::get (alias_target)); + else if (DECL_INITIAL (decl)) + record_references_in_initializer (decl, analyzed); } - if (alias) - resolve_alias (varpool_node::get (alias_target)); - else if (DECL_INITIAL (decl)) - record_references_in_initializer (decl, analyzed); analyzed = true; } -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 01/08: Allow SFmode for IA MCU in FP splitter

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit ad2477a45a8ffb6e6cc5beaa52a505d41222d858 Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Thu Jul 9 12:06:40 2015 -0700 Allow SFmode for IA MCU in FP splitter ix86_split_long_move can optimize floating point constant move, which can be used to optimize SFmode move for IA MCU. gcc/ PR target/66824 * config/i386/i386.c (ix86_split_to_parts): Allow SFmode move for IA MCU. (ix86_split_long_move): Support single move. * config/i386/i386.md (FP splitter): Allow SFmode for IA MCU. gcc/testsuite/ PR target/66824 * gcc.target/i386/pr66824.c: New test. --- gcc/config/i386/i386.c | 14 +++++++++++++- gcc/config/i386/i386.md | 3 ++- gcc/testsuite/gcc.target/i386/pr66824.c | 23 +++++++++++++++++++++++ 3 files changed, 38 insertions(+), 2 deletions(-) diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c index a18c733..d2925bc 100644 --- a/gcc/config/i386/i386.c +++ b/gcc/config/i386/i386.c @@ -22744,7 +22744,9 @@ ix86_split_to_parts (rtx operand, rtx *parts, mac [...] size = (GET_MODE_SIZE (mode) + 4) / 8; gcc_assert (!REG_P (operand) || !MMX_REGNO_P (REGNO (operand))); - gcc_assert (size >= 2 && size <= 4); + /* For IA MCU, we can optimize SFmode move. */ + gcc_assert ((size >= 2 || (TARGET_IAMCU && mode == SFmode)) + && size <= 4); /* Optimize constant pool reference to immediates. This is used by fp moves, that force all constants to memory to allow combining. */ @@ -22822,10 +22824,14 @@ ix86_split_to_parts (rtx operand, rtx *parts, m [...] case DFmode: REAL_VALUE_TO_TARGET_DOUBLE (r, l); break; + case SFmode: + REAL_VALUE_TO_TARGET_SINGLE (r, l[0]); + goto part0; default: gcc_unreachable (); } parts[1] = gen_int_mode (l[1], SImode); +part0: parts[0] = gen_int_mode (l[0], SImode); } else @@ -22932,6 +22938,12 @@ ix86_split_long_move (rtx operands[]) nparts = ix86_split_to_parts (operands[1], part[1], GET_MODE (operands[0])); ix86_split_to_parts (operands[0], part[0], GET_MODE (operands[0])); + if (nparts == 1) + { + emit_move_insn (part[0][0], part[1][0]); + return; + } + /* When emitting push, take care for source operands on the stack. */ if (push && MEM_P (operands[1]) && reg_overlap_mentioned_p (stack_pointer_rtx, operands[1])) diff --git a/gcc/config/i386/i386.md b/gcc/config/i386/i386.md index 1d43aaf..e723eab 100644 --- a/gcc/config/i386/i386.md +++ b/gcc/config/i386/i386.md @@ -3507,7 +3507,8 @@ "reload_completed && (GET_MODE (operands[0]) == TFmode || GET_MODE (operands[0]) == XFmode - || GET_MODE (operands[0]) == DFmode) + || GET_MODE (operands[0]) == DFmode + || (TARGET_IAMCU && GET_MODE (operands[0]) == SFmode)) && !(ANY_FP_REG_P (operands[0]) || ANY_FP_REG_P (operands[1]))" [(const_int 0)] "ix86_split_long_move (operands); DONE;") diff --git a/gcc/testsuite/gcc.target/i386/pr66824.c b/gcc/testsuite/gcc. [...] new file mode 100644 index 0000000..0966736 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr66824.c @@ -0,0 +1,23 @@ +/* { dg-do compile { target ia32 } } */ +/* { dg-options "-O2 -miamcu" } */ +/* { dg-final { scan-assembler-not "\.LC\[0-9\]" } } */ + +double foo (float); + +double +f1 (void) +{ + return foo (1.0); +} + +double +f2 (void) +{ + return foo (0.0); +} + +void +f3 (float *x, float t) +{ + *x = 0.0 + t; +} -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

[gcc] 05/08: Turn off loop and funcion alignments for IA MCU

by git

This is an automated email from the git hooks/post-receive script. unknown user pushed a commit to branch hjl/iamcu/improve in repository gcc. commit c45297796b7c2d50d97d2acfe1a1340f9ab2ab9f Author: H.J. Lu <hjl.tools(a)gmail.com> Date: Fri Jul 10 04:05:53 2015 -0700 Turn off loop and funcion alignments for IA MCU --- gcc/config/i386/i386.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c index bca1fee..4f08d0a 100644 --- a/gcc/config/i386/i386.c +++ b/gcc/config/i386/i386.c @@ -2556,7 +2556,7 @@ static const struct ptt processor_target_table[PROC [...] {"i386", &i386_cost, 4, 3, 4, 3, 4}, {"i486", &i486_cost, 16, 15, 16, 15, 16}, {"pentium", &pentium_cost, 16, 7, 16, 7, 16}, - {"iamcu", &iamcu_cost, 16, 7, 16, 7, 16}, + {"iamcu", &iamcu_cost, 0, 0, 0, 0, 0}, {"pentiumpro", &pentiumpro_cost, 16, 15, 16, 10, 16}, {"pentium4", &pentium4_cost, 0, 0, 0, 0, 0}, {"nocona", &nocona_cost, 0, 0, 0, 0, 0}, -- To stop receiving notification emails like this one, please contact the administrator of this repository.

9 years, 4 months

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

Tcwg-commits July 2015