| Message ID | 20160519193619.GB40563@msticlxl57.ims.intel.com |
|---|---|
| State | New |
| Headers | show |
On Thu, 19 May 2016, Ilya Enkovich wrote: > Hi, > > This patch introduces new options used for loop epilogues vectorization. Any patch adding a new option should update invoke.texi (both the summary list of options, and adding documentation for the new option).
On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: > Hi, > > This patch introduces new options used for loop epilogues vectorization. Why's that? This is a bit too much for the casual user and if it is really necessary to control this via options then it is not fine-grained enough. Why doesn't the vectorizer/backend have enough info to decide this itself? Richard. > Thanks, > Ilya > -- > gcc/ > > 2016-05-19 Ilya Enkovich <ilya.enkovich@intel.com> > > * common.opt (flag_tree_vectorize_epilogues): New. > (ftree-vectorize-short-loops): New. > (ftree-vectorize-epilogues=): New. > (fno-tree-vectorize-epilogues): New. > (fvect-epilogue-cost-model=): New. > * flag-types.h (enum vect_epilogue_mode): New. > * opts.c (parse_vectorizer_options): New. > (common_handle_option): Support -ftree-vectorize-epilogues= > and -fno-tree-vectorize-epilogues options. > > > diff --git a/gcc/common.opt b/gcc/common.opt > index 682cb41..6b83b79 100644 > --- a/gcc/common.opt > +++ b/gcc/common.opt > @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false > Variable > bool flag_disable_hsa = false > > +; Flag holding modes for loop epilogue vectorization > +Variable > +unsigned int flag_tree_vectorize_epilogues > + > ### > Driver > > @@ -2557,6 +2561,19 @@ ftree-vectorize > Common Report Var(flag_tree_vectorize) Optimization > Enable vectorization on trees. > > +ftree-vectorize-short-loops > +Common Report Var(flag_tree_vectorize_short_loops) Optimization > +Enable vectorization of loops with low trip count using masking. > + > +ftree-vectorize-epilogues= > +Common Report Joined Optimization > +Comma separated list of loop epilogue vectorization modes. > +Available modes: combine, mask, nomask. > + > +fno-tree-vectorize-epilogues > +Common RejectNegative Optimization > +Disable epilogues vectorization. > + > ftree-vectorizer-verbose= > Common Joined RejectNegative Ignore > Does nothing. Preserved for backward compatibility. > @@ -2577,6 +2594,10 @@ fsimd-cost-model= > Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization > Specifies the vectorization cost model for code marked with a simd directive. > > +fvect-epilogue-cost-model= > +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization > +Specifies the cost model for epilogue vectorization. > + > Enum > Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) > > diff --git a/gcc/flag-types.h b/gcc/flag-types.h > index dd57e16..24081b1 100644 > --- a/gcc/flag-types.h > +++ b/gcc/flag-types.h > @@ -200,6 +200,15 @@ enum vect_cost_model { > VECT_COST_MODEL_DEFAULT = 3 > }; > > +/* Epilogue vectorization modes. */ > +enum vect_epilogue_mode { > + VECT_EPILOGUE_COMBINE = 1 << 0, > + VECT_EPILOGUE_MASK = 1 << 1, > + VECT_EPILOGUE_NOMASK = 1 << 2, > + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK > + | VECT_EPILOGUE_NOMASK > +}; > + > /* Different instrumentation modes. */ > enum sanitize_code { > /* AddressSanitizer. */ > diff --git a/gcc/opts.c b/gcc/opts.c > index 0f9431a..a0c0987 100644 > --- a/gcc/opts.c > +++ b/gcc/opts.c > @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, > return flags; > } > > +/* Parse comma separated vectorizer suboptions from P for option SCODE, > + adjust previous FLAGS and return new ones. If COMPLAIN is false, > + don't issue diagnostics. */ > + > +unsigned int > +parse_vectorizer_options (const char *p, location_t loc, int scode, > + unsigned int flags, int value, bool complain) > +{ > + if (scode != OPT_ftree_vectorize_epilogues_) > + return flags; > + > + if (!p) > + return value; > + > + while (*p != 0) > + { > + size_t len; > + const char *comma = strchr (p, ','); > + unsigned int flag = 0; > + > + if (comma == NULL) > + len = strlen (p); > + else > + len = comma - p; > + if (len == 0) > + { > + p = comma + 1; > + continue; > + } > + > + /* Check to see if the string matches an option class name. */ > + if (len == strlen ("combine") > + && memcmp (p, "combine", len) == 0) > + flag = VECT_EPILOGUE_COMBINE; > + else if (len == strlen ("mask") > + && memcmp (p, "mask", len) == 0) > + flag = VECT_EPILOGUE_MASK; > + else if (len == strlen ("nomask") > + && memcmp (p, "nomask", len) == 0) > + flag = VECT_EPILOGUE_NOMASK; > + else if (complain) > + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " > + "option: %q.*s", (int) len, p); > + > + if (value) > + flags |= flag; > + else > + flags &= ~flag; > + > + if (comma == NULL) > + break; > + p = comma + 1; > + } > + > + return flags; > +} > + > /* Handle target- and language-independent options. Return zero to > generate an "unknown option" message. Only options that need > extra handling need to be listed here; if you simply want > @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, > if (!opts_set->x_flag_tree_slp_vectorize) > opts->x_flag_tree_slp_vectorize = value; > break; > + > + case OPT_ftree_vectorize_epilogues_: > + opts->x_flag_tree_vectorize_epilogues > + = parse_vectorizer_options (arg, loc, code, > + opts->x_flag_tree_vectorize_epilogues, > + value, true); > + break; > + > + case OPT_fno_tree_vectorize_epilogues: > + opts->x_flag_tree_vectorize_epilogues = 0; > + break; > + > case OPT_fshow_column: > dc->show_column = value; > break;
2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: > On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >> Hi, >> >> This patch introduces new options used for loop epilogues vectorization. > > Why's that? This is a bit too much for the casual user and if it is > really necessary > to control this via options then it is not fine-grained enough. > > Why doesn't the vectorizer/backend have enough info to decide this itself? I don't expect casual user to decide which modes to choose. These controls are added for debugging and performance measurement purposes. I see now I miss -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. Surely I expect epilogues and short loops vectorization be enabled by default on -O3 or by -ftree-vectorize-loops. Thanks, Ilya > > Richard. > >> Thanks, >> Ilya >> -- >> gcc/ >> >> 2016-05-19 Ilya Enkovich <ilya.enkovich@intel.com> >> >> * common.opt (flag_tree_vectorize_epilogues): New. >> (ftree-vectorize-short-loops): New. >> (ftree-vectorize-epilogues=): New. >> (fno-tree-vectorize-epilogues): New. >> (fvect-epilogue-cost-model=): New. >> * flag-types.h (enum vect_epilogue_mode): New. >> * opts.c (parse_vectorizer_options): New. >> (common_handle_option): Support -ftree-vectorize-epilogues= >> and -fno-tree-vectorize-epilogues options. >> >> >> diff --git a/gcc/common.opt b/gcc/common.opt >> index 682cb41..6b83b79 100644 >> --- a/gcc/common.opt >> +++ b/gcc/common.opt >> @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false >> Variable >> bool flag_disable_hsa = false >> >> +; Flag holding modes for loop epilogue vectorization >> +Variable >> +unsigned int flag_tree_vectorize_epilogues >> + >> ### >> Driver >> >> @@ -2557,6 +2561,19 @@ ftree-vectorize >> Common Report Var(flag_tree_vectorize) Optimization >> Enable vectorization on trees. >> >> +ftree-vectorize-short-loops >> +Common Report Var(flag_tree_vectorize_short_loops) Optimization >> +Enable vectorization of loops with low trip count using masking. >> + >> +ftree-vectorize-epilogues= >> +Common Report Joined Optimization >> +Comma separated list of loop epilogue vectorization modes. >> +Available modes: combine, mask, nomask. >> + >> +fno-tree-vectorize-epilogues >> +Common RejectNegative Optimization >> +Disable epilogues vectorization. >> + >> ftree-vectorizer-verbose= >> Common Joined RejectNegative Ignore >> Does nothing. Preserved for backward compatibility. >> @@ -2577,6 +2594,10 @@ fsimd-cost-model= >> Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization >> Specifies the vectorization cost model for code marked with a simd directive. >> >> +fvect-epilogue-cost-model= >> +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization >> +Specifies the cost model for epilogue vectorization. >> + >> Enum >> Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) >> >> diff --git a/gcc/flag-types.h b/gcc/flag-types.h >> index dd57e16..24081b1 100644 >> --- a/gcc/flag-types.h >> +++ b/gcc/flag-types.h >> @@ -200,6 +200,15 @@ enum vect_cost_model { >> VECT_COST_MODEL_DEFAULT = 3 >> }; >> >> +/* Epilogue vectorization modes. */ >> +enum vect_epilogue_mode { >> + VECT_EPILOGUE_COMBINE = 1 << 0, >> + VECT_EPILOGUE_MASK = 1 << 1, >> + VECT_EPILOGUE_NOMASK = 1 << 2, >> + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK >> + | VECT_EPILOGUE_NOMASK >> +}; >> + >> /* Different instrumentation modes. */ >> enum sanitize_code { >> /* AddressSanitizer. */ >> diff --git a/gcc/opts.c b/gcc/opts.c >> index 0f9431a..a0c0987 100644 >> --- a/gcc/opts.c >> +++ b/gcc/opts.c >> @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, >> return flags; >> } >> >> +/* Parse comma separated vectorizer suboptions from P for option SCODE, >> + adjust previous FLAGS and return new ones. If COMPLAIN is false, >> + don't issue diagnostics. */ >> + >> +unsigned int >> +parse_vectorizer_options (const char *p, location_t loc, int scode, >> + unsigned int flags, int value, bool complain) >> +{ >> + if (scode != OPT_ftree_vectorize_epilogues_) >> + return flags; >> + >> + if (!p) >> + return value; >> + >> + while (*p != 0) >> + { >> + size_t len; >> + const char *comma = strchr (p, ','); >> + unsigned int flag = 0; >> + >> + if (comma == NULL) >> + len = strlen (p); >> + else >> + len = comma - p; >> + if (len == 0) >> + { >> + p = comma + 1; >> + continue; >> + } >> + >> + /* Check to see if the string matches an option class name. */ >> + if (len == strlen ("combine") >> + && memcmp (p, "combine", len) == 0) >> + flag = VECT_EPILOGUE_COMBINE; >> + else if (len == strlen ("mask") >> + && memcmp (p, "mask", len) == 0) >> + flag = VECT_EPILOGUE_MASK; >> + else if (len == strlen ("nomask") >> + && memcmp (p, "nomask", len) == 0) >> + flag = VECT_EPILOGUE_NOMASK; >> + else if (complain) >> + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " >> + "option: %q.*s", (int) len, p); >> + >> + if (value) >> + flags |= flag; >> + else >> + flags &= ~flag; >> + >> + if (comma == NULL) >> + break; >> + p = comma + 1; >> + } >> + >> + return flags; >> +} >> + >> /* Handle target- and language-independent options. Return zero to >> generate an "unknown option" message. Only options that need >> extra handling need to be listed here; if you simply want >> @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, >> if (!opts_set->x_flag_tree_slp_vectorize) >> opts->x_flag_tree_slp_vectorize = value; >> break; >> + >> + case OPT_ftree_vectorize_epilogues_: >> + opts->x_flag_tree_vectorize_epilogues >> + = parse_vectorizer_options (arg, loc, code, >> + opts->x_flag_tree_vectorize_epilogues, >> + value, true); >> + break; >> + >> + case OPT_fno_tree_vectorize_epilogues: >> + opts->x_flag_tree_vectorize_epilogues = 0; >> + break; >> + >> case OPT_fshow_column: >> dc->show_column = value; >> break;
On Fri, May 20, 2016 at 11:50 AM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: > 2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >> On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>> Hi, >>> >>> This patch introduces new options used for loop epilogues vectorization. >> >> Why's that? This is a bit too much for the casual user and if it is >> really necessary >> to control this via options then it is not fine-grained enough. >> >> Why doesn't the vectorizer/backend have enough info to decide this itself? > > I don't expect casual user to decide which modes to choose. These controls are > added for debugging and performance measurement purposes. I see now I miss > -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. Surely > I expect epilogues and short loops vectorization be enabled by default on -O3 > or by -ftree-vectorize-loops. Can you make all these --params then? I think to be useful to users we'd want them to be loop pragmas rather than options. Richard. > Thanks, > Ilya > >> >> Richard. >> >>> Thanks, >>> Ilya >>> -- >>> gcc/ >>> >>> 2016-05-19 Ilya Enkovich <ilya.enkovich@intel.com> >>> >>> * common.opt (flag_tree_vectorize_epilogues): New. >>> (ftree-vectorize-short-loops): New. >>> (ftree-vectorize-epilogues=): New. >>> (fno-tree-vectorize-epilogues): New. >>> (fvect-epilogue-cost-model=): New. >>> * flag-types.h (enum vect_epilogue_mode): New. >>> * opts.c (parse_vectorizer_options): New. >>> (common_handle_option): Support -ftree-vectorize-epilogues= >>> and -fno-tree-vectorize-epilogues options. >>> >>> >>> diff --git a/gcc/common.opt b/gcc/common.opt >>> index 682cb41..6b83b79 100644 >>> --- a/gcc/common.opt >>> +++ b/gcc/common.opt >>> @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false >>> Variable >>> bool flag_disable_hsa = false >>> >>> +; Flag holding modes for loop epilogue vectorization >>> +Variable >>> +unsigned int flag_tree_vectorize_epilogues >>> + >>> ### >>> Driver >>> >>> @@ -2557,6 +2561,19 @@ ftree-vectorize >>> Common Report Var(flag_tree_vectorize) Optimization >>> Enable vectorization on trees. >>> >>> +ftree-vectorize-short-loops >>> +Common Report Var(flag_tree_vectorize_short_loops) Optimization >>> +Enable vectorization of loops with low trip count using masking. >>> + >>> +ftree-vectorize-epilogues= >>> +Common Report Joined Optimization >>> +Comma separated list of loop epilogue vectorization modes. >>> +Available modes: combine, mask, nomask. >>> + >>> +fno-tree-vectorize-epilogues >>> +Common RejectNegative Optimization >>> +Disable epilogues vectorization. >>> + >>> ftree-vectorizer-verbose= >>> Common Joined RejectNegative Ignore >>> Does nothing. Preserved for backward compatibility. >>> @@ -2577,6 +2594,10 @@ fsimd-cost-model= >>> Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization >>> Specifies the vectorization cost model for code marked with a simd directive. >>> >>> +fvect-epilogue-cost-model= >>> +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization >>> +Specifies the cost model for epilogue vectorization. >>> + >>> Enum >>> Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) >>> >>> diff --git a/gcc/flag-types.h b/gcc/flag-types.h >>> index dd57e16..24081b1 100644 >>> --- a/gcc/flag-types.h >>> +++ b/gcc/flag-types.h >>> @@ -200,6 +200,15 @@ enum vect_cost_model { >>> VECT_COST_MODEL_DEFAULT = 3 >>> }; >>> >>> +/* Epilogue vectorization modes. */ >>> +enum vect_epilogue_mode { >>> + VECT_EPILOGUE_COMBINE = 1 << 0, >>> + VECT_EPILOGUE_MASK = 1 << 1, >>> + VECT_EPILOGUE_NOMASK = 1 << 2, >>> + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK >>> + | VECT_EPILOGUE_NOMASK >>> +}; >>> + >>> /* Different instrumentation modes. */ >>> enum sanitize_code { >>> /* AddressSanitizer. */ >>> diff --git a/gcc/opts.c b/gcc/opts.c >>> index 0f9431a..a0c0987 100644 >>> --- a/gcc/opts.c >>> +++ b/gcc/opts.c >>> @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, >>> return flags; >>> } >>> >>> +/* Parse comma separated vectorizer suboptions from P for option SCODE, >>> + adjust previous FLAGS and return new ones. If COMPLAIN is false, >>> + don't issue diagnostics. */ >>> + >>> +unsigned int >>> +parse_vectorizer_options (const char *p, location_t loc, int scode, >>> + unsigned int flags, int value, bool complain) >>> +{ >>> + if (scode != OPT_ftree_vectorize_epilogues_) >>> + return flags; >>> + >>> + if (!p) >>> + return value; >>> + >>> + while (*p != 0) >>> + { >>> + size_t len; >>> + const char *comma = strchr (p, ','); >>> + unsigned int flag = 0; >>> + >>> + if (comma == NULL) >>> + len = strlen (p); >>> + else >>> + len = comma - p; >>> + if (len == 0) >>> + { >>> + p = comma + 1; >>> + continue; >>> + } >>> + >>> + /* Check to see if the string matches an option class name. */ >>> + if (len == strlen ("combine") >>> + && memcmp (p, "combine", len) == 0) >>> + flag = VECT_EPILOGUE_COMBINE; >>> + else if (len == strlen ("mask") >>> + && memcmp (p, "mask", len) == 0) >>> + flag = VECT_EPILOGUE_MASK; >>> + else if (len == strlen ("nomask") >>> + && memcmp (p, "nomask", len) == 0) >>> + flag = VECT_EPILOGUE_NOMASK; >>> + else if (complain) >>> + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " >>> + "option: %q.*s", (int) len, p); >>> + >>> + if (value) >>> + flags |= flag; >>> + else >>> + flags &= ~flag; >>> + >>> + if (comma == NULL) >>> + break; >>> + p = comma + 1; >>> + } >>> + >>> + return flags; >>> +} >>> + >>> /* Handle target- and language-independent options. Return zero to >>> generate an "unknown option" message. Only options that need >>> extra handling need to be listed here; if you simply want >>> @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, >>> if (!opts_set->x_flag_tree_slp_vectorize) >>> opts->x_flag_tree_slp_vectorize = value; >>> break; >>> + >>> + case OPT_ftree_vectorize_epilogues_: >>> + opts->x_flag_tree_vectorize_epilogues >>> + = parse_vectorizer_options (arg, loc, code, >>> + opts->x_flag_tree_vectorize_epilogues, >>> + value, true); >>> + break; >>> + >>> + case OPT_fno_tree_vectorize_epilogues: >>> + opts->x_flag_tree_vectorize_epilogues = 0; >>> + break; >>> + >>> case OPT_fshow_column: >>> dc->show_column = value; >>> break;
2016-05-20 14:17 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: > On Fri, May 20, 2016 at 11:50 AM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >> 2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>> On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>>> Hi, >>>> >>>> This patch introduces new options used for loop epilogues vectorization. >>> >>> Why's that? This is a bit too much for the casual user and if it is >>> really necessary >>> to control this via options then it is not fine-grained enough. >>> >>> Why doesn't the vectorizer/backend have enough info to decide this itself? >> >> I don't expect casual user to decide which modes to choose. These controls are >> added for debugging and performance measurement purposes. I see now I miss >> -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. Surely >> I expect epilogues and short loops vectorization be enabled by default on -O3 >> or by -ftree-vectorize-loops. > > Can you make all these --params then? I think to be useful to users we'd want > them to be loop pragmas rather than options. OK, I'll change it to params. I didn't think about control via pragmas but will do now. Thanks, Ilya > > Richard. > >> Thanks, >> Ilya >> >>> >>> Richard. >>> >>>> Thanks, >>>> Ilya >>>> -- >>>> gcc/ >>>> >>>> 2016-05-19 Ilya Enkovich <ilya.enkovich@intel.com> >>>> >>>> * common.opt (flag_tree_vectorize_epilogues): New. >>>> (ftree-vectorize-short-loops): New. >>>> (ftree-vectorize-epilogues=): New. >>>> (fno-tree-vectorize-epilogues): New. >>>> (fvect-epilogue-cost-model=): New. >>>> * flag-types.h (enum vect_epilogue_mode): New. >>>> * opts.c (parse_vectorizer_options): New. >>>> (common_handle_option): Support -ftree-vectorize-epilogues= >>>> and -fno-tree-vectorize-epilogues options. >>>> >>>> >>>> diff --git a/gcc/common.opt b/gcc/common.opt >>>> index 682cb41..6b83b79 100644 >>>> --- a/gcc/common.opt >>>> +++ b/gcc/common.opt >>>> @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false >>>> Variable >>>> bool flag_disable_hsa = false >>>> >>>> +; Flag holding modes for loop epilogue vectorization >>>> +Variable >>>> +unsigned int flag_tree_vectorize_epilogues >>>> + >>>> ### >>>> Driver >>>> >>>> @@ -2557,6 +2561,19 @@ ftree-vectorize >>>> Common Report Var(flag_tree_vectorize) Optimization >>>> Enable vectorization on trees. >>>> >>>> +ftree-vectorize-short-loops >>>> +Common Report Var(flag_tree_vectorize_short_loops) Optimization >>>> +Enable vectorization of loops with low trip count using masking. >>>> + >>>> +ftree-vectorize-epilogues= >>>> +Common Report Joined Optimization >>>> +Comma separated list of loop epilogue vectorization modes. >>>> +Available modes: combine, mask, nomask. >>>> + >>>> +fno-tree-vectorize-epilogues >>>> +Common RejectNegative Optimization >>>> +Disable epilogues vectorization. >>>> + >>>> ftree-vectorizer-verbose= >>>> Common Joined RejectNegative Ignore >>>> Does nothing. Preserved for backward compatibility. >>>> @@ -2577,6 +2594,10 @@ fsimd-cost-model= >>>> Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization >>>> Specifies the vectorization cost model for code marked with a simd directive. >>>> >>>> +fvect-epilogue-cost-model= >>>> +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization >>>> +Specifies the cost model for epilogue vectorization. >>>> + >>>> Enum >>>> Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) >>>> >>>> diff --git a/gcc/flag-types.h b/gcc/flag-types.h >>>> index dd57e16..24081b1 100644 >>>> --- a/gcc/flag-types.h >>>> +++ b/gcc/flag-types.h >>>> @@ -200,6 +200,15 @@ enum vect_cost_model { >>>> VECT_COST_MODEL_DEFAULT = 3 >>>> }; >>>> >>>> +/* Epilogue vectorization modes. */ >>>> +enum vect_epilogue_mode { >>>> + VECT_EPILOGUE_COMBINE = 1 << 0, >>>> + VECT_EPILOGUE_MASK = 1 << 1, >>>> + VECT_EPILOGUE_NOMASK = 1 << 2, >>>> + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK >>>> + | VECT_EPILOGUE_NOMASK >>>> +}; >>>> + >>>> /* Different instrumentation modes. */ >>>> enum sanitize_code { >>>> /* AddressSanitizer. */ >>>> diff --git a/gcc/opts.c b/gcc/opts.c >>>> index 0f9431a..a0c0987 100644 >>>> --- a/gcc/opts.c >>>> +++ b/gcc/opts.c >>>> @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, >>>> return flags; >>>> } >>>> >>>> +/* Parse comma separated vectorizer suboptions from P for option SCODE, >>>> + adjust previous FLAGS and return new ones. If COMPLAIN is false, >>>> + don't issue diagnostics. */ >>>> + >>>> +unsigned int >>>> +parse_vectorizer_options (const char *p, location_t loc, int scode, >>>> + unsigned int flags, int value, bool complain) >>>> +{ >>>> + if (scode != OPT_ftree_vectorize_epilogues_) >>>> + return flags; >>>> + >>>> + if (!p) >>>> + return value; >>>> + >>>> + while (*p != 0) >>>> + { >>>> + size_t len; >>>> + const char *comma = strchr (p, ','); >>>> + unsigned int flag = 0; >>>> + >>>> + if (comma == NULL) >>>> + len = strlen (p); >>>> + else >>>> + len = comma - p; >>>> + if (len == 0) >>>> + { >>>> + p = comma + 1; >>>> + continue; >>>> + } >>>> + >>>> + /* Check to see if the string matches an option class name. */ >>>> + if (len == strlen ("combine") >>>> + && memcmp (p, "combine", len) == 0) >>>> + flag = VECT_EPILOGUE_COMBINE; >>>> + else if (len == strlen ("mask") >>>> + && memcmp (p, "mask", len) == 0) >>>> + flag = VECT_EPILOGUE_MASK; >>>> + else if (len == strlen ("nomask") >>>> + && memcmp (p, "nomask", len) == 0) >>>> + flag = VECT_EPILOGUE_NOMASK; >>>> + else if (complain) >>>> + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " >>>> + "option: %q.*s", (int) len, p); >>>> + >>>> + if (value) >>>> + flags |= flag; >>>> + else >>>> + flags &= ~flag; >>>> + >>>> + if (comma == NULL) >>>> + break; >>>> + p = comma + 1; >>>> + } >>>> + >>>> + return flags; >>>> +} >>>> + >>>> /* Handle target- and language-independent options. Return zero to >>>> generate an "unknown option" message. Only options that need >>>> extra handling need to be listed here; if you simply want >>>> @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, >>>> if (!opts_set->x_flag_tree_slp_vectorize) >>>> opts->x_flag_tree_slp_vectorize = value; >>>> break; >>>> + >>>> + case OPT_ftree_vectorize_epilogues_: >>>> + opts->x_flag_tree_vectorize_epilogues >>>> + = parse_vectorizer_options (arg, loc, code, >>>> + opts->x_flag_tree_vectorize_epilogues, >>>> + value, true); >>>> + break; >>>> + >>>> + case OPT_fno_tree_vectorize_epilogues: >>>> + opts->x_flag_tree_vectorize_epilogues = 0; >>>> + break; >>>> + >>>> case OPT_fshow_column: >>>> dc->show_column = value; >>>> break;
Hi Richard, Do you have some comments on other parts of this series? Thanks, Ilya 2016-05-20 14:40 GMT+03:00 Ilya Enkovich <enkovich.gnu@gmail.com>: > 2016-05-20 14:17 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >> On Fri, May 20, 2016 at 11:50 AM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>> 2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>>> On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>>>> Hi, >>>>> >>>>> This patch introduces new options used for loop epilogues vectorization. >>>> >>>> Why's that? This is a bit too much for the casual user and if it is >>>> really necessary >>>> to control this via options then it is not fine-grained enough. >>>> >>>> Why doesn't the vectorizer/backend have enough info to decide this itself? >>> >>> I don't expect casual user to decide which modes to choose. These controls are >>> added for debugging and performance measurement purposes. I see now I miss >>> -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. Surely >>> I expect epilogues and short loops vectorization be enabled by default on -O3 >>> or by -ftree-vectorize-loops. >> >> Can you make all these --params then? I think to be useful to users we'd want >> them to be loop pragmas rather than options. > > OK, I'll change it to params. I didn't think about control via > pragmas but will do now. > > Thanks, > Ilya > >> >> Richard. >> >>> Thanks, >>> Ilya >>> >>>> >>>> Richard. >>>> >>>>> Thanks, >>>>> Ilya >>>>> -- >>>>> gcc/ >>>>> >>>>> 2016-05-19 Ilya Enkovich <ilya.enkovich@intel.com> >>>>> >>>>> * common.opt (flag_tree_vectorize_epilogues): New. >>>>> (ftree-vectorize-short-loops): New. >>>>> (ftree-vectorize-epilogues=): New. >>>>> (fno-tree-vectorize-epilogues): New. >>>>> (fvect-epilogue-cost-model=): New. >>>>> * flag-types.h (enum vect_epilogue_mode): New. >>>>> * opts.c (parse_vectorizer_options): New. >>>>> (common_handle_option): Support -ftree-vectorize-epilogues= >>>>> and -fno-tree-vectorize-epilogues options. >>>>> >>>>> >>>>> diff --git a/gcc/common.opt b/gcc/common.opt >>>>> index 682cb41..6b83b79 100644 >>>>> --- a/gcc/common.opt >>>>> +++ b/gcc/common.opt >>>>> @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false >>>>> Variable >>>>> bool flag_disable_hsa = false >>>>> >>>>> +; Flag holding modes for loop epilogue vectorization >>>>> +Variable >>>>> +unsigned int flag_tree_vectorize_epilogues >>>>> + >>>>> ### >>>>> Driver >>>>> >>>>> @@ -2557,6 +2561,19 @@ ftree-vectorize >>>>> Common Report Var(flag_tree_vectorize) Optimization >>>>> Enable vectorization on trees. >>>>> >>>>> +ftree-vectorize-short-loops >>>>> +Common Report Var(flag_tree_vectorize_short_loops) Optimization >>>>> +Enable vectorization of loops with low trip count using masking. >>>>> + >>>>> +ftree-vectorize-epilogues= >>>>> +Common Report Joined Optimization >>>>> +Comma separated list of loop epilogue vectorization modes. >>>>> +Available modes: combine, mask, nomask. >>>>> + >>>>> +fno-tree-vectorize-epilogues >>>>> +Common RejectNegative Optimization >>>>> +Disable epilogues vectorization. >>>>> + >>>>> ftree-vectorizer-verbose= >>>>> Common Joined RejectNegative Ignore >>>>> Does nothing. Preserved for backward compatibility. >>>>> @@ -2577,6 +2594,10 @@ fsimd-cost-model= >>>>> Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization >>>>> Specifies the vectorization cost model for code marked with a simd directive. >>>>> >>>>> +fvect-epilogue-cost-model= >>>>> +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization >>>>> +Specifies the cost model for epilogue vectorization. >>>>> + >>>>> Enum >>>>> Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) >>>>> >>>>> diff --git a/gcc/flag-types.h b/gcc/flag-types.h >>>>> index dd57e16..24081b1 100644 >>>>> --- a/gcc/flag-types.h >>>>> +++ b/gcc/flag-types.h >>>>> @@ -200,6 +200,15 @@ enum vect_cost_model { >>>>> VECT_COST_MODEL_DEFAULT = 3 >>>>> }; >>>>> >>>>> +/* Epilogue vectorization modes. */ >>>>> +enum vect_epilogue_mode { >>>>> + VECT_EPILOGUE_COMBINE = 1 << 0, >>>>> + VECT_EPILOGUE_MASK = 1 << 1, >>>>> + VECT_EPILOGUE_NOMASK = 1 << 2, >>>>> + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK >>>>> + | VECT_EPILOGUE_NOMASK >>>>> +}; >>>>> + >>>>> /* Different instrumentation modes. */ >>>>> enum sanitize_code { >>>>> /* AddressSanitizer. */ >>>>> diff --git a/gcc/opts.c b/gcc/opts.c >>>>> index 0f9431a..a0c0987 100644 >>>>> --- a/gcc/opts.c >>>>> +++ b/gcc/opts.c >>>>> @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, >>>>> return flags; >>>>> } >>>>> >>>>> +/* Parse comma separated vectorizer suboptions from P for option SCODE, >>>>> + adjust previous FLAGS and return new ones. If COMPLAIN is false, >>>>> + don't issue diagnostics. */ >>>>> + >>>>> +unsigned int >>>>> +parse_vectorizer_options (const char *p, location_t loc, int scode, >>>>> + unsigned int flags, int value, bool complain) >>>>> +{ >>>>> + if (scode != OPT_ftree_vectorize_epilogues_) >>>>> + return flags; >>>>> + >>>>> + if (!p) >>>>> + return value; >>>>> + >>>>> + while (*p != 0) >>>>> + { >>>>> + size_t len; >>>>> + const char *comma = strchr (p, ','); >>>>> + unsigned int flag = 0; >>>>> + >>>>> + if (comma == NULL) >>>>> + len = strlen (p); >>>>> + else >>>>> + len = comma - p; >>>>> + if (len == 0) >>>>> + { >>>>> + p = comma + 1; >>>>> + continue; >>>>> + } >>>>> + >>>>> + /* Check to see if the string matches an option class name. */ >>>>> + if (len == strlen ("combine") >>>>> + && memcmp (p, "combine", len) == 0) >>>>> + flag = VECT_EPILOGUE_COMBINE; >>>>> + else if (len == strlen ("mask") >>>>> + && memcmp (p, "mask", len) == 0) >>>>> + flag = VECT_EPILOGUE_MASK; >>>>> + else if (len == strlen ("nomask") >>>>> + && memcmp (p, "nomask", len) == 0) >>>>> + flag = VECT_EPILOGUE_NOMASK; >>>>> + else if (complain) >>>>> + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " >>>>> + "option: %q.*s", (int) len, p); >>>>> + >>>>> + if (value) >>>>> + flags |= flag; >>>>> + else >>>>> + flags &= ~flag; >>>>> + >>>>> + if (comma == NULL) >>>>> + break; >>>>> + p = comma + 1; >>>>> + } >>>>> + >>>>> + return flags; >>>>> +} >>>>> + >>>>> /* Handle target- and language-independent options. Return zero to >>>>> generate an "unknown option" message. Only options that need >>>>> extra handling need to be listed here; if you simply want >>>>> @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, >>>>> if (!opts_set->x_flag_tree_slp_vectorize) >>>>> opts->x_flag_tree_slp_vectorize = value; >>>>> break; >>>>> + >>>>> + case OPT_ftree_vectorize_epilogues_: >>>>> + opts->x_flag_tree_vectorize_epilogues >>>>> + = parse_vectorizer_options (arg, loc, code, >>>>> + opts->x_flag_tree_vectorize_epilogues, >>>>> + value, true); >>>>> + break; >>>>> + >>>>> + case OPT_fno_tree_vectorize_epilogues: >>>>> + opts->x_flag_tree_vectorize_epilogues = 0; >>>>> + break; >>>>> + >>>>> case OPT_fshow_column: >>>>> dc->show_column = value; >>>>> break;
On Thu, Jun 9, 2016 at 12:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: > Hi Richard, > > Do you have some comments on other parts of this series? It's queued for review ... I'll need a slot of some spare hours to go over it. Richard. > Thanks, > Ilya > > 2016-05-20 14:40 GMT+03:00 Ilya Enkovich <enkovich.gnu@gmail.com>: >> 2016-05-20 14:17 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>> On Fri, May 20, 2016 at 11:50 AM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>>> 2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>>>> On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>>>>> Hi, >>>>>> >>>>>> This patch introduces new options used for loop epilogues vectorization. >>>>> >>>>> Why's that? This is a bit too much for the casual user and if it is >>>>> really necessary >>>>> to control this via options then it is not fine-grained enough. >>>>> >>>>> Why doesn't the vectorizer/backend have enough info to decide this itself? >>>> >>>> I don't expect casual user to decide which modes to choose. These controls are >>>> added for debugging and performance measurement purposes. I see now I miss >>>> -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. Surely >>>> I expect epilogues and short loops vectorization be enabled by default on -O3 >>>> or by -ftree-vectorize-loops. >>> >>> Can you make all these --params then? I think to be useful to users we'd want >>> them to be loop pragmas rather than options. >> >> OK, I'll change it to params. I didn't think about control via >> pragmas but will do now. >> >> Thanks, >> Ilya >> >>> >>> Richard. >>> >>>> Thanks, >>>> Ilya >>>> >>>>> >>>>> Richard. >>>>> >>>>>> Thanks, >>>>>> Ilya >>>>>> -- >>>>>> gcc/ >>>>>> >>>>>> 2016-05-19 Ilya Enkovich <ilya.enkovich@intel.com> >>>>>> >>>>>> * common.opt (flag_tree_vectorize_epilogues): New. >>>>>> (ftree-vectorize-short-loops): New. >>>>>> (ftree-vectorize-epilogues=): New. >>>>>> (fno-tree-vectorize-epilogues): New. >>>>>> (fvect-epilogue-cost-model=): New. >>>>>> * flag-types.h (enum vect_epilogue_mode): New. >>>>>> * opts.c (parse_vectorizer_options): New. >>>>>> (common_handle_option): Support -ftree-vectorize-epilogues= >>>>>> and -fno-tree-vectorize-epilogues options. >>>>>> >>>>>> >>>>>> diff --git a/gcc/common.opt b/gcc/common.opt >>>>>> index 682cb41..6b83b79 100644 >>>>>> --- a/gcc/common.opt >>>>>> +++ b/gcc/common.opt >>>>>> @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false >>>>>> Variable >>>>>> bool flag_disable_hsa = false >>>>>> >>>>>> +; Flag holding modes for loop epilogue vectorization >>>>>> +Variable >>>>>> +unsigned int flag_tree_vectorize_epilogues >>>>>> + >>>>>> ### >>>>>> Driver >>>>>> >>>>>> @@ -2557,6 +2561,19 @@ ftree-vectorize >>>>>> Common Report Var(flag_tree_vectorize) Optimization >>>>>> Enable vectorization on trees. >>>>>> >>>>>> +ftree-vectorize-short-loops >>>>>> +Common Report Var(flag_tree_vectorize_short_loops) Optimization >>>>>> +Enable vectorization of loops with low trip count using masking. >>>>>> + >>>>>> +ftree-vectorize-epilogues= >>>>>> +Common Report Joined Optimization >>>>>> +Comma separated list of loop epilogue vectorization modes. >>>>>> +Available modes: combine, mask, nomask. >>>>>> + >>>>>> +fno-tree-vectorize-epilogues >>>>>> +Common RejectNegative Optimization >>>>>> +Disable epilogues vectorization. >>>>>> + >>>>>> ftree-vectorizer-verbose= >>>>>> Common Joined RejectNegative Ignore >>>>>> Does nothing. Preserved for backward compatibility. >>>>>> @@ -2577,6 +2594,10 @@ fsimd-cost-model= >>>>>> Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization >>>>>> Specifies the vectorization cost model for code marked with a simd directive. >>>>>> >>>>>> +fvect-epilogue-cost-model= >>>>>> +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization >>>>>> +Specifies the cost model for epilogue vectorization. >>>>>> + >>>>>> Enum >>>>>> Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) >>>>>> >>>>>> diff --git a/gcc/flag-types.h b/gcc/flag-types.h >>>>>> index dd57e16..24081b1 100644 >>>>>> --- a/gcc/flag-types.h >>>>>> +++ b/gcc/flag-types.h >>>>>> @@ -200,6 +200,15 @@ enum vect_cost_model { >>>>>> VECT_COST_MODEL_DEFAULT = 3 >>>>>> }; >>>>>> >>>>>> +/* Epilogue vectorization modes. */ >>>>>> +enum vect_epilogue_mode { >>>>>> + VECT_EPILOGUE_COMBINE = 1 << 0, >>>>>> + VECT_EPILOGUE_MASK = 1 << 1, >>>>>> + VECT_EPILOGUE_NOMASK = 1 << 2, >>>>>> + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK >>>>>> + | VECT_EPILOGUE_NOMASK >>>>>> +}; >>>>>> + >>>>>> /* Different instrumentation modes. */ >>>>>> enum sanitize_code { >>>>>> /* AddressSanitizer. */ >>>>>> diff --git a/gcc/opts.c b/gcc/opts.c >>>>>> index 0f9431a..a0c0987 100644 >>>>>> --- a/gcc/opts.c >>>>>> +++ b/gcc/opts.c >>>>>> @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, >>>>>> return flags; >>>>>> } >>>>>> >>>>>> +/* Parse comma separated vectorizer suboptions from P for option SCODE, >>>>>> + adjust previous FLAGS and return new ones. If COMPLAIN is false, >>>>>> + don't issue diagnostics. */ >>>>>> + >>>>>> +unsigned int >>>>>> +parse_vectorizer_options (const char *p, location_t loc, int scode, >>>>>> + unsigned int flags, int value, bool complain) >>>>>> +{ >>>>>> + if (scode != OPT_ftree_vectorize_epilogues_) >>>>>> + return flags; >>>>>> + >>>>>> + if (!p) >>>>>> + return value; >>>>>> + >>>>>> + while (*p != 0) >>>>>> + { >>>>>> + size_t len; >>>>>> + const char *comma = strchr (p, ','); >>>>>> + unsigned int flag = 0; >>>>>> + >>>>>> + if (comma == NULL) >>>>>> + len = strlen (p); >>>>>> + else >>>>>> + len = comma - p; >>>>>> + if (len == 0) >>>>>> + { >>>>>> + p = comma + 1; >>>>>> + continue; >>>>>> + } >>>>>> + >>>>>> + /* Check to see if the string matches an option class name. */ >>>>>> + if (len == strlen ("combine") >>>>>> + && memcmp (p, "combine", len) == 0) >>>>>> + flag = VECT_EPILOGUE_COMBINE; >>>>>> + else if (len == strlen ("mask") >>>>>> + && memcmp (p, "mask", len) == 0) >>>>>> + flag = VECT_EPILOGUE_MASK; >>>>>> + else if (len == strlen ("nomask") >>>>>> + && memcmp (p, "nomask", len) == 0) >>>>>> + flag = VECT_EPILOGUE_NOMASK; >>>>>> + else if (complain) >>>>>> + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " >>>>>> + "option: %q.*s", (int) len, p); >>>>>> + >>>>>> + if (value) >>>>>> + flags |= flag; >>>>>> + else >>>>>> + flags &= ~flag; >>>>>> + >>>>>> + if (comma == NULL) >>>>>> + break; >>>>>> + p = comma + 1; >>>>>> + } >>>>>> + >>>>>> + return flags; >>>>>> +} >>>>>> + >>>>>> /* Handle target- and language-independent options. Return zero to >>>>>> generate an "unknown option" message. Only options that need >>>>>> extra handling need to be listed here; if you simply want >>>>>> @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, >>>>>> if (!opts_set->x_flag_tree_slp_vectorize) >>>>>> opts->x_flag_tree_slp_vectorize = value; >>>>>> break; >>>>>> + >>>>>> + case OPT_ftree_vectorize_epilogues_: >>>>>> + opts->x_flag_tree_vectorize_epilogues >>>>>> + = parse_vectorizer_options (arg, loc, code, >>>>>> + opts->x_flag_tree_vectorize_epilogues, >>>>>> + value, true); >>>>>> + break; >>>>>> + >>>>>> + case OPT_fno_tree_vectorize_epilogues: >>>>>> + opts->x_flag_tree_vectorize_epilogues = 0; >>>>>> + break; >>>>>> + >>>>>> case OPT_fshow_column: >>>>>> dc->show_column = value; >>>>>> break;
On 05/20/2016 05:40 AM, Ilya Enkovich wrote: > 2016-05-20 14:17 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >> On Fri, May 20, 2016 at 11:50 AM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>> 2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>>> On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote: >>>>> Hi, >>>>> >>>>> This patch introduces new options used for loop epilogues vectorization. >>>> >>>> Why's that? This is a bit too much for the casual user and if it is >>>> really necessary >>>> to control this via options then it is not fine-grained enough. >>>> >>>> Why doesn't the vectorizer/backend have enough info to decide this itself? >>> >>> I don't expect casual user to decide which modes to choose. These controls are >>> added for debugging and performance measurement purposes. I see now I miss >>> -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. Surely >>> I expect epilogues and short loops vectorization be enabled by default on -O3 >>> or by -ftree-vectorize-loops. >> >> Can you make all these --params then? I think to be useful to users we'd want >> them to be loop pragmas rather than options. > > OK, I'll change it to params. I didn't think about control via > pragmas but will do now. So the questions I'd like to see answered: 1. You've got 3 modes for epilogue vectorization. Is this an artifact of not really having good heuristics yet for which mode to apply to a particular loop at this time? 2. Similarly for cost models. In the cover message you indicated you were getting expected gains of KNL, but not on Haswell. Do you have any sense yet why you're not getting good resuls on Haswell yet? For KNL are you getting those speedups with a generic set of options or are those with a custom set of options to set the mode & cost models? jeff
2016-06-16 8:06 GMT+03:00 Jeff Law <law@redhat.com>: > On 05/20/2016 05:40 AM, Ilya Enkovich wrote: >> >> 2016-05-20 14:17 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>> >>> On Fri, May 20, 2016 at 11:50 AM, Ilya Enkovich <enkovich.gnu@gmail.com> >>> wrote: >>>> >>>> 2016-05-20 12:26 GMT+03:00 Richard Biener <richard.guenther@gmail.com>: >>>>> >>>>> On Thu, May 19, 2016 at 9:36 PM, Ilya Enkovich <enkovich.gnu@gmail.com> >>>>> wrote: >>>>>> >>>>>> Hi, >>>>>> >>>>>> This patch introduces new options used for loop epilogues >>>>>> vectorization. >>>>> >>>>> >>>>> Why's that? This is a bit too much for the casual user and if it is >>>>> really necessary >>>>> to control this via options then it is not fine-grained enough. >>>>> >>>>> Why doesn't the vectorizer/backend have enough info to decide this >>>>> itself? >>>> >>>> >>>> I don't expect casual user to decide which modes to choose. These >>>> controls are >>>> added for debugging and performance measurement purposes. I see now I >>>> miss >>>> -ftree-vectorize-epilogues aliased to -ftree-vectorize-epilogues=all. >>>> Surely >>>> I expect epilogues and short loops vectorization be enabled by default >>>> on -O3 >>>> or by -ftree-vectorize-loops. >>> >>> >>> Can you make all these --params then? I think to be useful to users we'd >>> want >>> them to be loop pragmas rather than options. >> >> >> OK, I'll change it to params. I didn't think about control via >> pragmas but will do now. > > So the questions I'd like to see answered: > > 1. You've got 3 modes for epilogue vectorization. Is this an artifact of > not really having good heuristics yet for which mode to apply to a > particular loop at this time? > > 2. Similarly for cost models. All three modes are profitable in different situations. Profitable mode depends on a loop structure and target capabilities. Ultimate goal is to have all three modes enabled by default. I can't state current heuristics are good enough for all cases and targets and therefore don't enable epilogues vectorization by default for now. This is to be measured, analyzed and tuned in time for GCC 7.1. I add cost model simply to have an ability to force epilogue vectorization for stability testing (force some mode of epilogue vectorization and check nothing fails) and performance testing/tuning (try to find cases where we may benefit from epilogue vectorization but don't due to bad cost model). Also I don't want to force epilogue vectorization for all loops for which vectorization is forced using unlimited cost model because that may hurt performance for simd loops. > > In the cover message you indicated you were getting expected gains of KNL, > but not on Haswell. Do you have any sense yet why you're not getting good > resuls on Haswell yet? For KNL are you getting those speedups with a > generic set of options or are those with a custom set of options to set the > mode & cost models? Currently I have numbers collected on various suites for KNL machine. Masking mode (-ftree-vectorize-epilogues=mask) shows not bad results (dynamic cost model, -Ofast -flto -funroll-loops). I don't see significant losses and there are few significant gains. For combine and nomask modes the result is not good enough yet - there are several significant performance losses. My guess is that current threshold for combine is way too high and for nomask variant we better choose the smallest vector size for epilogues instead of the next available (use zmm for body and xmm for epilogue instead of zmmm for body and ymm for epilogue). ICC shows better results in these modes which makes me believe we can tune them as well. Overall nomask mode shows worse results comparing to options with masking which is quite expected for KNL. Unfortunately some big gains demonstrated by ICC are not reproducible using GCC because we originally can't vectorize required hot loops. E.g. on 200.sixtrack GCC has nothing and ICC has ~40% for all three modes. I don't have the whole statistics for Haswell but synthetic tests show the situation is really different from KNL. Even for the 'perfect' iterations count number (VF * 2 - 1) scalar version of epilogue shows the same result as a masked one. It means ratio of vector code performance vs. scalar code performance is not as high as for KNL (KNL is more vector oriented and has weaker scalar performance, double vector size also matters here) and masking cost is higher for Haswell. We still focus on AVX-512 targets more because of their rich masking capabilities and wider vector. Thanks, Ilya > > jeff
On 06/17/2016 04:41 AM, Ilya Enkovich wrote: >> >> 1. You've got 3 modes for epilogue vectorization. Is this an artifact of >> not really having good heuristics yet for which mode to apply to a >> particular loop at this time? >> >> 2. Similarly for cost models. > > All three modes are profitable in different situations. Profitable mode depends > on a loop structure and target capabilities. Ultimate goal is to have all three > modes enabled by default. I can't state current heuristics are good enough > for all cases and targets and therefore don't enable epilogues vectorization > by default for now. This is to be measured, analyzed and tuned in > time for GCC 7.1. > > I add cost model simply to have an ability to force epilogue vectorization for > stability testing (force some mode of epilogue vectorization and check nothing > fails) and performance testing/tuning (try to find cases where we may benefit > from epilogue vectorization but don't due to bad cost model). Also I don't > want to force epilogue vectorization for all loops for which vectorization is > forced using unlimited cost model because that may hurt performance for > simd loops. Thanks. That overview helps a lot. We've done something similar to what you're doing with cost models for testing in the scheduler and other places in the past. The costing models seem more geared towards us as developers rather than users. you might consider keep those as local changes and not documenting them. Understood completely on the modes. > > Currently I have numbers collected on various suites for KNL machine. Masking > mode (-ftree-vectorize-epilogues=mask) shows not bad results (dynamic > cost model, > -Ofast -flto -funroll-loops). I don't see significant losses and there are few > significant gains. For combine and nomask modes the result is not good enough > yet - there are several significant performance losses. My guess is that > current threshold for combine is way too high and for nomask variant we better > choose the smallest vector size for epilogues instead of the next available > (use zmm for body and xmm for epilogue instead of zmmm for body and ymm for > epilogue). > > ICC shows better results in these modes which makes me believe we can tune them > as well. Overall nomask mode shows worse results comparing to options with > masking which is quite expected for KNL. > > Unfortunately some big gains demonstrated by ICC are not reproducible > using GCC because we originally can't vectorize required hot loops. E.g. on > 200.sixtrack GCC has nothing and ICC has ~40% for all three modes. I hadn't pondered that case. Certainly if GCC isn't vectorizing as much, we're not going to have as many opportunities for optimizing the vec-tails. Given the results with ICC, we're probably best off keeping all 3 modes and working to get them tuned correctly. > > I don't have the whole statistics for Haswell but synthetic tests show the > situation is really different from KNL. Even for the 'perfect' iterations count > number (VF * 2 - 1) scalar version of epilogue shows the same result as a masked > one. It means ratio of vector code performance vs. scalar code performance is > not as high as for KNL (KNL is more vector oriented and has weaker > scalar performance, > double vector size also matters here) and masking cost is higher for Haswell. > We still focus on AVX-512 targets more because of their rich masking > capabilities and wider vector. Understood. Jeff
diff --git a/gcc/common.opt b/gcc/common.opt index 682cb41..6b83b79 100644 --- a/gcc/common.opt +++ b/gcc/common.opt @@ -243,6 +243,10 @@ bool dump_base_name_prefixed = false Variable bool flag_disable_hsa = false +; Flag holding modes for loop epilogue vectorization +Variable +unsigned int flag_tree_vectorize_epilogues + ### Driver @@ -2557,6 +2561,19 @@ ftree-vectorize Common Report Var(flag_tree_vectorize) Optimization Enable vectorization on trees. +ftree-vectorize-short-loops +Common Report Var(flag_tree_vectorize_short_loops) Optimization +Enable vectorization of loops with low trip count using masking. + +ftree-vectorize-epilogues= +Common Report Joined Optimization +Comma separated list of loop epilogue vectorization modes. +Available modes: combine, mask, nomask. + +fno-tree-vectorize-epilogues +Common RejectNegative Optimization +Disable epilogues vectorization. + ftree-vectorizer-verbose= Common Joined RejectNegative Ignore Does nothing. Preserved for backward compatibility. @@ -2577,6 +2594,10 @@ fsimd-cost-model= Common Joined RejectNegative Enum(vect_cost_model) Var(flag_simd_cost_model) Init(VECT_COST_MODEL_UNLIMITED) Optimization Specifies the vectorization cost model for code marked with a simd directive. +fvect-epilogue-cost-model= +Common Joined RejectNegative Enum(vect_cost_model) Var(flag_vect_epilogue_cost_model) Init(VECT_COST_MODEL_DEFAULT) Optimization +Specifies the cost model for epilogue vectorization. + Enum Name(vect_cost_model) Type(enum vect_cost_model) UnknownError(unknown vectorizer cost model %qs) diff --git a/gcc/flag-types.h b/gcc/flag-types.h index dd57e16..24081b1 100644 --- a/gcc/flag-types.h +++ b/gcc/flag-types.h @@ -200,6 +200,15 @@ enum vect_cost_model { VECT_COST_MODEL_DEFAULT = 3 }; +/* Epilogue vectorization modes. */ +enum vect_epilogue_mode { + VECT_EPILOGUE_COMBINE = 1 << 0, + VECT_EPILOGUE_MASK = 1 << 1, + VECT_EPILOGUE_NOMASK = 1 << 2, + VECT_EPILOGUE_ALL = VECT_EPILOGUE_COMBINE | VECT_EPILOGUE_MASK + | VECT_EPILOGUE_NOMASK +}; + /* Different instrumentation modes. */ enum sanitize_code { /* AddressSanitizer. */ diff --git a/gcc/opts.c b/gcc/opts.c index 0f9431a..a0c0987 100644 --- a/gcc/opts.c +++ b/gcc/opts.c @@ -1531,6 +1531,63 @@ parse_sanitizer_options (const char *p, location_t loc, int scode, return flags; } +/* Parse comma separated vectorizer suboptions from P for option SCODE, + adjust previous FLAGS and return new ones. If COMPLAIN is false, + don't issue diagnostics. */ + +unsigned int +parse_vectorizer_options (const char *p, location_t loc, int scode, + unsigned int flags, int value, bool complain) +{ + if (scode != OPT_ftree_vectorize_epilogues_) + return flags; + + if (!p) + return value; + + while (*p != 0) + { + size_t len; + const char *comma = strchr (p, ','); + unsigned int flag = 0; + + if (comma == NULL) + len = strlen (p); + else + len = comma - p; + if (len == 0) + { + p = comma + 1; + continue; + } + + /* Check to see if the string matches an option class name. */ + if (len == strlen ("combine") + && memcmp (p, "combine", len) == 0) + flag = VECT_EPILOGUE_COMBINE; + else if (len == strlen ("mask") + && memcmp (p, "mask", len) == 0) + flag = VECT_EPILOGUE_MASK; + else if (len == strlen ("nomask") + && memcmp (p, "nomask", len) == 0) + flag = VECT_EPILOGUE_NOMASK; + else if (complain) + error_at (loc, "unrecognized argument to -ftree-vectorize-epilogues= " + "option: %q.*s", (int) len, p); + + if (value) + flags |= flag; + else + flags &= ~flag; + + if (comma == NULL) + break; + p = comma + 1; + } + + return flags; +} + /* Handle target- and language-independent options. Return zero to generate an "unknown option" message. Only options that need extra handling need to be listed here; if you simply want @@ -2018,6 +2075,18 @@ common_handle_option (struct gcc_options *opts, if (!opts_set->x_flag_tree_slp_vectorize) opts->x_flag_tree_slp_vectorize = value; break; + + case OPT_ftree_vectorize_epilogues_: + opts->x_flag_tree_vectorize_epilogues + = parse_vectorizer_options (arg, loc, code, + opts->x_flag_tree_vectorize_epilogues, + value, true); + break; + + case OPT_fno_tree_vectorize_epilogues: + opts->x_flag_tree_vectorize_epilogues = 0; + break; + case OPT_fshow_column: dc->show_column = value; break;