* [PATCH v6] x86-64: Find a scratch register for large model profiling @ 2024-02-05 16:43 H.J. Lu 2024-02-05 18:01 ` Uros Bizjak 0 siblings, 1 reply; 3+ messages in thread From: H.J. Lu @ 2024-02-05 16:43 UTC (permalink / raw) To: gcc-patches; +Cc: ubizjak, jakub, rep.dot.nop Changes in v6: 1. Use ix86_save_reg and accessible_reg_set in x86_64_select_profile_regnum. 2. Construct a complete reg name in x86_function_profiler. Changes in v5: 1. Add pr113689-3.c. 2. Use %r10 if ix86_profile_before_prologue () return true. 3. Try a callee-saved register which has been saved on stack in the prologue. Changes in v4: 1. Remove pr113689-3.c. 2. Use df_get_live_out. Changes in v3: 1. Remove r10_ok. Changes in v2: 1. Add int_parameter_registers to machine_function to track integer registers used for parameter passing. 2. Update x86_64_select_profile_regnum to try %r10 first and use an caller-saved register, which isn't used for parameter passing. --- 2 scratch registers, %r10 and %r11, are available at function entry for large model profiling. But %r10 may be used by stack realignment and we can't use %r10 in this case. Add x86_64_select_profile_regnum to find a caller-saved register which isn't live or a callee-saved register which has been saved on stack in the prologue at entry for large model profiling and sorry if we can't find one. gcc/ PR target/113689 * config/i386/i386.cc (x86_64_select_profile_regnum): New. (x86_function_profiler): Call x86_64_select_profile_regnum to get a scratch register for large model profiling. gcc/testsuite/ PR target/113689 * gcc.target/i386/pr113689-1.c: New file. * gcc.target/i386/pr113689-2.c: Likewise. * gcc.target/i386/pr113689-3.c: Likewise. --- gcc/config/i386/i386.cc | 91 ++++++++++++++++++---- gcc/testsuite/gcc.target/i386/pr113689-1.c | 49 ++++++++++++ gcc/testsuite/gcc.target/i386/pr113689-2.c | 41 ++++++++++ gcc/testsuite/gcc.target/i386/pr113689-3.c | 48 ++++++++++++ 4 files changed, 214 insertions(+), 15 deletions(-) create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-1.c create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-2.c create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-3.c diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc index b3e7c74846e..08aad32af85 100644 --- a/gcc/config/i386/i386.cc +++ b/gcc/config/i386/i386.cc @@ -22749,6 +22749,48 @@ current_fentry_section (const char **name) return true; } +/* Return a caller-saved register which isn't live or a callee-saved + register which has been saved on stack in the prologue at entry for + profile. */ + +static int +x86_64_select_profile_regnum (bool r11_ok ATTRIBUTE_UNUSED) +{ + /* Use %r10 if the profiler is emitted before the prologue or it isn't + used by DRAP. */ + if (ix86_profile_before_prologue () + || !crtl->drap_reg + || REGNO (crtl->drap_reg) != R10_REG) + return R10_REG; + + /* The profiler is emitted after the prologue. If there is a + caller-saved register which isn't live or a callee-saved + register saved on stack in the prologue, use it. */ + + bitmap reg_live = df_get_live_out (ENTRY_BLOCK_PTR_FOR_FN (cfun)); + + int i; + for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) + if (GENERAL_REGNO_P (i) + && i != R10_REG +#ifdef NO_PROFILE_COUNTERS + && (r11_ok || i != R11_REG) +#else + && i != R11_REG +#endif + && TEST_HARD_REG_BIT (accessible_reg_set, i) + && !fixed_regs[i] + && (ix86_save_reg (i, true, true) + || (call_used_regs[i] + && !REGNO_REG_SET_P (reg_live, i)))) + return i; + + sorry ("no register available for profiling %<-mcmodel=large%s%>", + ix86_cmodel == CM_LARGE_PIC ? " -fPIC" : ""); + + return INVALID_REGNUM; +} + /* Output assembler code to FILE to increment profiler label # LABELNO for profiling a function entry. */ void @@ -22783,42 +22825,61 @@ x86_function_profiler (FILE *file, int labelno ATTRIBUTE_UNUSED) fprintf (file, "\tleaq\t%sP%d(%%rip), %%r11\n", LPREFIX, labelno); #endif + int scratch; + const char *reg; + char legacy_reg[4] = { 0 }; + if (!TARGET_PECOFF) { switch (ix86_cmodel) { case CM_LARGE: - /* NB: R10 is caller-saved. Although it can be used as a - static chain register, it is preserved when calling - mcount for nested functions. */ + scratch = x86_64_select_profile_regnum (true); + reg = hi_reg_name[scratch]; + if (LEGACY_INT_REGNO_P (scratch)) + { + legacy_reg[0] = 'r'; + legacy_reg[1] = reg[0]; + legacy_reg[2] = reg[1]; + reg = legacy_reg; + } if (ASSEMBLER_DIALECT == ASM_INTEL) - fprintf (file, "1:\tmovabs\tr10, OFFSET FLAT:%s\n" - "\tcall\tr10\n", mcount_name); + fprintf (file, "1:\tmovabs\t%s, OFFSET FLAT:%s\n" + "\tcall\t%s\n", reg, mcount_name, reg); else - fprintf (file, "1:\tmovabsq\t$%s, %%r10\n\tcall\t*%%r10\n", - mcount_name); + fprintf (file, "1:\tmovabsq\t$%s, %%%s\n\tcall\t*%%%s\n", + mcount_name, reg, reg); break; case CM_LARGE_PIC: #ifdef NO_PROFILE_COUNTERS + scratch = x86_64_select_profile_regnum (false); + reg = hi_reg_name[scratch]; + if (LEGACY_INT_REGNO_P (scratch)) + { + legacy_reg[0] = 'r'; + legacy_reg[1] = reg[0]; + legacy_reg[2] = reg[1]; + reg = legacy_reg; + } if (ASSEMBLER_DIALECT == ASM_INTEL) { fprintf (file, "1:movabs\tr11, " "OFFSET FLAT:_GLOBAL_OFFSET_TABLE_-1b\n"); - fprintf (file, "\tlea\tr10, 1b[rip]\n"); - fprintf (file, "\tadd\tr10, r11\n"); + fprintf (file, "\tlea\t%s, 1b[rip]\n", reg); + fprintf (file, "\tadd\t%s, r11\n", reg); fprintf (file, "\tmovabs\tr11, OFFSET FLAT:%s@PLTOFF\n", mcount_name); - fprintf (file, "\tadd\tr10, r11\n"); - fprintf (file, "\tcall\tr10\n"); + fprintf (file, "\tadd\t%s, r11\n", reg); + fprintf (file, "\tcall\t%s\n", reg); break; } fprintf (file, "1:\tmovabsq\t$_GLOBAL_OFFSET_TABLE_-1b, %%r11\n"); - fprintf (file, "\tleaq\t1b(%%rip), %%r10\n"); - fprintf (file, "\taddq\t%%r11, %%r10\n"); + fprintf (file, "\tleaq\t1b(%%rip), %%%s\n", reg); + fprintf (file, "\taddq\t%%r11, %%%s\n", reg); fprintf (file, "\tmovabsq\t$%s@PLTOFF, %%r11\n", mcount_name); - fprintf (file, "\taddq\t%%r11, %%r10\n"); - fprintf (file, "\tcall\t*%%r10\n"); + fprintf (file, "\taddq\t%%r11, %%%s\n", reg); + fprintf (file, "\tcall\t*%%%s\n", reg); #else sorry ("profiling %<-mcmodel=large%> with PIC is not supported"); #endif diff --git a/gcc/testsuite/gcc.target/i386/pr113689-1.c b/gcc/testsuite/gcc.target/i386/pr113689-1.c new file mode 100644 index 00000000000..8285c0a07b7 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr113689-1.c @@ -0,0 +1,49 @@ +/* { dg-do run { target { lp64 && fpic } } } */ +/* { dg-options "-O2 -fno-pic -fprofile -mcmodel=large" } */ + +#include <stdarg.h> + +__attribute__((noipa)) +void +bar (int a1, int a2, int a3, int a4, int a5, int a6, + char *x, char *y, int *z) +{ + if (a1 != 1) + __builtin_abort (); + if (a2 != 2) + __builtin_abort (); + if (a3 != 3) + __builtin_abort (); + if (a4 != 4) + __builtin_abort (); + if (a5 != 5) + __builtin_abort (); + if (a6 != 6) + __builtin_abort (); + x[0] = 42; + y[0] = 42; + if (z[0] != 16) + __builtin_abort (); +} + +__attribute__((noipa)) +void +foo (int c, int d, int e, int f, int g, int h, int z, ...) +{ + typedef char B[32]; + B b __attribute__((aligned (32))); + va_list ap; + va_start (ap, z); + double x = va_arg (ap, double); + if (x > 40.0) + __builtin_abort (); + bar (c, d, e, f, g, h, &b[0], __builtin_alloca (z), &z); + va_end (ap); +} + +int +main () +{ + foo (1, 2, 3, 4, 5, 6, 16, 38.0); + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr113689-2.c b/gcc/testsuite/gcc.target/i386/pr113689-2.c new file mode 100644 index 00000000000..2e5579ac546 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr113689-2.c @@ -0,0 +1,41 @@ +/* { dg-do run { target { lp64 && fpic } } } */ +/* { dg-options "-O2 -fpic -fprofile -mcmodel=large" } */ + +__attribute__((noipa)) +void +bar (int a1, int a2, int a3, int a4, int a5, int a6, + char *x, char *y, int *z) +{ + if (a1 != 1) + __builtin_abort (); + if (a2 != 2) + __builtin_abort (); + if (a3 != 3) + __builtin_abort (); + if (a4 != 4) + __builtin_abort (); + if (a5 != 5) + __builtin_abort (); + if (a6 != 6) + __builtin_abort (); + x[0] = 42; + y[0] = 42; + if (z[0] != 16) + __builtin_abort (); +} + +__attribute__((noipa)) +void +foo (int c, int d, int e, int f, int g, int h, int z) +{ + typedef char B[32]; + B b __attribute__((aligned (32))); + bar (c, d, e, f, g, h, &b[0], __builtin_alloca (z), &z); +} + +int +main () +{ + foo (1, 2, 3, 4, 5, 6, 16); + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr113689-3.c b/gcc/testsuite/gcc.target/i386/pr113689-3.c new file mode 100644 index 00000000000..dab75190635 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr113689-3.c @@ -0,0 +1,48 @@ +/* { dg-do run { target { lp64 && fpic } } } */ +/* { dg-options "-O2 -fpic -fprofile -mcmodel=large" } */ + +#include <stdarg.h> + +__attribute__((noipa)) +void +bar (char *x, char *y, int *z) +{ + x[0] = 42; + y[0] = 42; + if (z[0] != 16) + __builtin_abort (); +} + +__attribute__((noipa)) +void +foo (int a1, int a2, int a3, int a4, int a5, int a6, int z, ...) +{ + typedef char B[32]; + B b __attribute__((aligned (32))); + va_list ap; + va_start (ap, z); + double x = va_arg (ap, double); + if (x > 40.0) + __builtin_abort (); + if (a1 != 1) + __builtin_abort (); + if (a2 != 2) + __builtin_abort (); + if (a3 != 3) + __builtin_abort (); + if (a4 != 4) + __builtin_abort (); + if (a5 != 5) + __builtin_abort (); + if (a6 != 6) + __builtin_abort (); + bar (&b[0], __builtin_alloca (z), &z); + va_end (ap); +} + +int +main () +{ + foo (1, 2, 3, 4, 5, 6, 16, 38.0); + return 0; +} -- 2.43.0 ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH v6] x86-64: Find a scratch register for large model profiling 2024-02-05 16:43 [PATCH v6] x86-64: Find a scratch register for large model profiling H.J. Lu @ 2024-02-05 18:01 ` Uros Bizjak 2024-02-05 19:07 ` H.J. Lu 0 siblings, 1 reply; 3+ messages in thread From: Uros Bizjak @ 2024-02-05 18:01 UTC (permalink / raw) To: H.J. Lu; +Cc: gcc-patches, jakub, rep.dot.nop On Mon, Feb 5, 2024 at 5:43 PM H.J. Lu <hjl.tools@gmail.com> wrote: > > Changes in v6: > > 1. Use ix86_save_reg and accessible_reg_set in > x86_64_select_profile_regnum. > 2. Construct a complete reg name in x86_function_profiler. > > Changes in v5: > > 1. Add pr113689-3.c. > 2. Use %r10 if ix86_profile_before_prologue () return true. > 3. Try a callee-saved register which has been saved on stack in the > prologue. > > Changes in v4: > > 1. Remove pr113689-3.c. > 2. Use df_get_live_out. > > Changes in v3: > > 1. Remove r10_ok. > > Changes in v2: > > 1. Add int_parameter_registers to machine_function to track integer > registers used for parameter passing. > 2. Update x86_64_select_profile_regnum to try %r10 first and use an > caller-saved register, which isn't used for parameter passing. > > --- > 2 scratch registers, %r10 and %r11, are available at function entry for > large model profiling. But %r10 may be used by stack realignment and we > can't use %r10 in this case. Add x86_64_select_profile_regnum to find > a caller-saved register which isn't live or a callee-saved register > which has been saved on stack in the prologue at entry for large model > profiling and sorry if we can't find one. > > gcc/ > > PR target/113689 > * config/i386/i386.cc (x86_64_select_profile_regnum): New. > (x86_function_profiler): Call x86_64_select_profile_regnum to > get a scratch register for large model profiling. > > gcc/testsuite/ > > PR target/113689 > * gcc.target/i386/pr113689-1.c: New file. > * gcc.target/i386/pr113689-2.c: Likewise. > * gcc.target/i386/pr113689-3.c: Likewise. > --- > gcc/config/i386/i386.cc | 91 ++++++++++++++++++---- > gcc/testsuite/gcc.target/i386/pr113689-1.c | 49 ++++++++++++ > gcc/testsuite/gcc.target/i386/pr113689-2.c | 41 ++++++++++ > gcc/testsuite/gcc.target/i386/pr113689-3.c | 48 ++++++++++++ > 4 files changed, 214 insertions(+), 15 deletions(-) > create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-1.c > create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-2.c > create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-3.c > > diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc > index b3e7c74846e..08aad32af85 100644 > --- a/gcc/config/i386/i386.cc > +++ b/gcc/config/i386/i386.cc > @@ -22749,6 +22749,48 @@ current_fentry_section (const char **name) > return true; > } > > +/* Return a caller-saved register which isn't live or a callee-saved > + register which has been saved on stack in the prologue at entry for > + profile. */ > + > +static int > +x86_64_select_profile_regnum (bool r11_ok ATTRIBUTE_UNUSED) > +{ > + /* Use %r10 if the profiler is emitted before the prologue or it isn't > + used by DRAP. */ > + if (ix86_profile_before_prologue () > + || !crtl->drap_reg > + || REGNO (crtl->drap_reg) != R10_REG) > + return R10_REG; > + > + /* The profiler is emitted after the prologue. If there is a > + caller-saved register which isn't live or a callee-saved > + register saved on stack in the prologue, use it. */ > + > + bitmap reg_live = df_get_live_out (ENTRY_BLOCK_PTR_FOR_FN (cfun)); > + > + int i; > + for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) > + if (GENERAL_REGNO_P (i) > + && i != R10_REG > +#ifdef NO_PROFILE_COUNTERS > + && (r11_ok || i != R11_REG) > +#else > + && i != R11_REG > +#endif > + && TEST_HARD_REG_BIT (accessible_reg_set, i) > + && !fixed_regs[i] > + && (ix86_save_reg (i, true, true) > + || (call_used_regs[i] > + && !REGNO_REG_SET_P (reg_live, i)))) > + return i; ix86_save_reg will never save fixed regs, so the above can be optimized a bit: && TEST_HARD_REG_BIT (accessible_reg_set, i) && (ix86_save_reg (i, true, true) || (call_used_regs[i] && !fixed_regs[i] && !REGNO_REG_SET_P (reg_live, i)))) OK with the above change. Thanks, Uros. > + sorry ("no register available for profiling %<-mcmodel=large%s%>", > + ix86_cmodel == CM_LARGE_PIC ? " -fPIC" : ""); > + > + return INVALID_REGNUM; > +} > + > /* Output assembler code to FILE to increment profiler label # LABELNO > for profiling a function entry. */ > void > @@ -22783,42 +22825,61 @@ x86_function_profiler (FILE *file, int labelno ATTRIBUTE_UNUSED) > fprintf (file, "\tleaq\t%sP%d(%%rip), %%r11\n", LPREFIX, labelno); > #endif > > + int scratch; > + const char *reg; > + char legacy_reg[4] = { 0 }; > + > if (!TARGET_PECOFF) > { > switch (ix86_cmodel) > { > case CM_LARGE: > - /* NB: R10 is caller-saved. Although it can be used as a > - static chain register, it is preserved when calling > - mcount for nested functions. */ > + scratch = x86_64_select_profile_regnum (true); > + reg = hi_reg_name[scratch]; > + if (LEGACY_INT_REGNO_P (scratch)) > + { > + legacy_reg[0] = 'r'; > + legacy_reg[1] = reg[0]; > + legacy_reg[2] = reg[1]; > + reg = legacy_reg; > + } > if (ASSEMBLER_DIALECT == ASM_INTEL) > - fprintf (file, "1:\tmovabs\tr10, OFFSET FLAT:%s\n" > - "\tcall\tr10\n", mcount_name); > + fprintf (file, "1:\tmovabs\t%s, OFFSET FLAT:%s\n" > + "\tcall\t%s\n", reg, mcount_name, reg); > else > - fprintf (file, "1:\tmovabsq\t$%s, %%r10\n\tcall\t*%%r10\n", > - mcount_name); > + fprintf (file, "1:\tmovabsq\t$%s, %%%s\n\tcall\t*%%%s\n", > + mcount_name, reg, reg); > break; > case CM_LARGE_PIC: > #ifdef NO_PROFILE_COUNTERS > + scratch = x86_64_select_profile_regnum (false); > + reg = hi_reg_name[scratch]; > + if (LEGACY_INT_REGNO_P (scratch)) > + { > + legacy_reg[0] = 'r'; > + legacy_reg[1] = reg[0]; > + legacy_reg[2] = reg[1]; > + reg = legacy_reg; > + } > if (ASSEMBLER_DIALECT == ASM_INTEL) > { > fprintf (file, "1:movabs\tr11, " > "OFFSET FLAT:_GLOBAL_OFFSET_TABLE_-1b\n"); > - fprintf (file, "\tlea\tr10, 1b[rip]\n"); > - fprintf (file, "\tadd\tr10, r11\n"); > + fprintf (file, "\tlea\t%s, 1b[rip]\n", reg); > + fprintf (file, "\tadd\t%s, r11\n", reg); > fprintf (file, "\tmovabs\tr11, OFFSET FLAT:%s@PLTOFF\n", > mcount_name); > - fprintf (file, "\tadd\tr10, r11\n"); > - fprintf (file, "\tcall\tr10\n"); > + fprintf (file, "\tadd\t%s, r11\n", reg); > + fprintf (file, "\tcall\t%s\n", reg); > break; > } > fprintf (file, > "1:\tmovabsq\t$_GLOBAL_OFFSET_TABLE_-1b, %%r11\n"); > - fprintf (file, "\tleaq\t1b(%%rip), %%r10\n"); > - fprintf (file, "\taddq\t%%r11, %%r10\n"); > + fprintf (file, "\tleaq\t1b(%%rip), %%%s\n", reg); > + fprintf (file, "\taddq\t%%r11, %%%s\n", reg); > fprintf (file, "\tmovabsq\t$%s@PLTOFF, %%r11\n", mcount_name); > - fprintf (file, "\taddq\t%%r11, %%r10\n"); > - fprintf (file, "\tcall\t*%%r10\n"); > + fprintf (file, "\taddq\t%%r11, %%%s\n", reg); > + fprintf (file, "\tcall\t*%%%s\n", reg); > #else > sorry ("profiling %<-mcmodel=large%> with PIC is not supported"); > #endif > diff --git a/gcc/testsuite/gcc.target/i386/pr113689-1.c b/gcc/testsuite/gcc.target/i386/pr113689-1.c > new file mode 100644 > index 00000000000..8285c0a07b7 > --- /dev/null > +++ b/gcc/testsuite/gcc.target/i386/pr113689-1.c > @@ -0,0 +1,49 @@ > +/* { dg-do run { target { lp64 && fpic } } } */ > +/* { dg-options "-O2 -fno-pic -fprofile -mcmodel=large" } */ > + > +#include <stdarg.h> > + > +__attribute__((noipa)) > +void > +bar (int a1, int a2, int a3, int a4, int a5, int a6, > + char *x, char *y, int *z) > +{ > + if (a1 != 1) > + __builtin_abort (); > + if (a2 != 2) > + __builtin_abort (); > + if (a3 != 3) > + __builtin_abort (); > + if (a4 != 4) > + __builtin_abort (); > + if (a5 != 5) > + __builtin_abort (); > + if (a6 != 6) > + __builtin_abort (); > + x[0] = 42; > + y[0] = 42; > + if (z[0] != 16) > + __builtin_abort (); > +} > + > +__attribute__((noipa)) > +void > +foo (int c, int d, int e, int f, int g, int h, int z, ...) > +{ > + typedef char B[32]; > + B b __attribute__((aligned (32))); > + va_list ap; > + va_start (ap, z); > + double x = va_arg (ap, double); > + if (x > 40.0) > + __builtin_abort (); > + bar (c, d, e, f, g, h, &b[0], __builtin_alloca (z), &z); > + va_end (ap); > +} > + > +int > +main () > +{ > + foo (1, 2, 3, 4, 5, 6, 16, 38.0); > + return 0; > +} > diff --git a/gcc/testsuite/gcc.target/i386/pr113689-2.c b/gcc/testsuite/gcc.target/i386/pr113689-2.c > new file mode 100644 > index 00000000000..2e5579ac546 > --- /dev/null > +++ b/gcc/testsuite/gcc.target/i386/pr113689-2.c > @@ -0,0 +1,41 @@ > +/* { dg-do run { target { lp64 && fpic } } } */ > +/* { dg-options "-O2 -fpic -fprofile -mcmodel=large" } */ > + > +__attribute__((noipa)) > +void > +bar (int a1, int a2, int a3, int a4, int a5, int a6, > + char *x, char *y, int *z) > +{ > + if (a1 != 1) > + __builtin_abort (); > + if (a2 != 2) > + __builtin_abort (); > + if (a3 != 3) > + __builtin_abort (); > + if (a4 != 4) > + __builtin_abort (); > + if (a5 != 5) > + __builtin_abort (); > + if (a6 != 6) > + __builtin_abort (); > + x[0] = 42; > + y[0] = 42; > + if (z[0] != 16) > + __builtin_abort (); > +} > + > +__attribute__((noipa)) > +void > +foo (int c, int d, int e, int f, int g, int h, int z) > +{ > + typedef char B[32]; > + B b __attribute__((aligned (32))); > + bar (c, d, e, f, g, h, &b[0], __builtin_alloca (z), &z); > +} > + > +int > +main () > +{ > + foo (1, 2, 3, 4, 5, 6, 16); > + return 0; > +} > diff --git a/gcc/testsuite/gcc.target/i386/pr113689-3.c b/gcc/testsuite/gcc.target/i386/pr113689-3.c > new file mode 100644 > index 00000000000..dab75190635 > --- /dev/null > +++ b/gcc/testsuite/gcc.target/i386/pr113689-3.c > @@ -0,0 +1,48 @@ > +/* { dg-do run { target { lp64 && fpic } } } */ > +/* { dg-options "-O2 -fpic -fprofile -mcmodel=large" } */ > + > +#include <stdarg.h> > + > +__attribute__((noipa)) > +void > +bar (char *x, char *y, int *z) > +{ > + x[0] = 42; > + y[0] = 42; > + if (z[0] != 16) > + __builtin_abort (); > +} > + > +__attribute__((noipa)) > +void > +foo (int a1, int a2, int a3, int a4, int a5, int a6, int z, ...) > +{ > + typedef char B[32]; > + B b __attribute__((aligned (32))); > + va_list ap; > + va_start (ap, z); > + double x = va_arg (ap, double); > + if (x > 40.0) > + __builtin_abort (); > + if (a1 != 1) > + __builtin_abort (); > + if (a2 != 2) > + __builtin_abort (); > + if (a3 != 3) > + __builtin_abort (); > + if (a4 != 4) > + __builtin_abort (); > + if (a5 != 5) > + __builtin_abort (); > + if (a6 != 6) > + __builtin_abort (); > + bar (&b[0], __builtin_alloca (z), &z); > + va_end (ap); > +} > + > +int > +main () > +{ > + foo (1, 2, 3, 4, 5, 6, 16, 38.0); > + return 0; > +} > -- > 2.43.0 > ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH v6] x86-64: Find a scratch register for large model profiling 2024-02-05 18:01 ` Uros Bizjak @ 2024-02-05 19:07 ` H.J. Lu 0 siblings, 0 replies; 3+ messages in thread From: H.J. Lu @ 2024-02-05 19:07 UTC (permalink / raw) To: Uros Bizjak; +Cc: gcc-patches, jakub, rep.dot.nop [-- Attachment #1: Type: text/plain, Size: 4502 bytes --] On Mon, Feb 5, 2024 at 10:01 AM Uros Bizjak <ubizjak@gmail.com> wrote: > > On Mon, Feb 5, 2024 at 5:43 PM H.J. Lu <hjl.tools@gmail.com> wrote: > > > > Changes in v6: > > > > 1. Use ix86_save_reg and accessible_reg_set in > > x86_64_select_profile_regnum. > > 2. Construct a complete reg name in x86_function_profiler. > > > > Changes in v5: > > > > 1. Add pr113689-3.c. > > 2. Use %r10 if ix86_profile_before_prologue () return true. > > 3. Try a callee-saved register which has been saved on stack in the > > prologue. > > > > Changes in v4: > > > > 1. Remove pr113689-3.c. > > 2. Use df_get_live_out. > > > > Changes in v3: > > > > 1. Remove r10_ok. > > > > Changes in v2: > > > > 1. Add int_parameter_registers to machine_function to track integer > > registers used for parameter passing. > > 2. Update x86_64_select_profile_regnum to try %r10 first and use an > > caller-saved register, which isn't used for parameter passing. > > > > --- > > 2 scratch registers, %r10 and %r11, are available at function entry for > > large model profiling. But %r10 may be used by stack realignment and we > > can't use %r10 in this case. Add x86_64_select_profile_regnum to find > > a caller-saved register which isn't live or a callee-saved register > > which has been saved on stack in the prologue at entry for large model > > profiling and sorry if we can't find one. > > > > gcc/ > > > > PR target/113689 > > * config/i386/i386.cc (x86_64_select_profile_regnum): New. > > (x86_function_profiler): Call x86_64_select_profile_regnum to > > get a scratch register for large model profiling. > > > > gcc/testsuite/ > > > > PR target/113689 > > * gcc.target/i386/pr113689-1.c: New file. > > * gcc.target/i386/pr113689-2.c: Likewise. > > * gcc.target/i386/pr113689-3.c: Likewise. > > --- > > gcc/config/i386/i386.cc | 91 ++++++++++++++++++---- > > gcc/testsuite/gcc.target/i386/pr113689-1.c | 49 ++++++++++++ > > gcc/testsuite/gcc.target/i386/pr113689-2.c | 41 ++++++++++ > > gcc/testsuite/gcc.target/i386/pr113689-3.c | 48 ++++++++++++ > > 4 files changed, 214 insertions(+), 15 deletions(-) > > create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-1.c > > create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-2.c > > create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-3.c > > > > diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc > > index b3e7c74846e..08aad32af85 100644 > > --- a/gcc/config/i386/i386.cc > > +++ b/gcc/config/i386/i386.cc > > @@ -22749,6 +22749,48 @@ current_fentry_section (const char **name) > > return true; > > } > > > > +/* Return a caller-saved register which isn't live or a callee-saved > > + register which has been saved on stack in the prologue at entry for > > + profile. */ > > + > > +static int > > +x86_64_select_profile_regnum (bool r11_ok ATTRIBUTE_UNUSED) > > +{ > > + /* Use %r10 if the profiler is emitted before the prologue or it isn't > > + used by DRAP. */ > > + if (ix86_profile_before_prologue () > > + || !crtl->drap_reg > > + || REGNO (crtl->drap_reg) != R10_REG) > > + return R10_REG; > > + > > + /* The profiler is emitted after the prologue. If there is a > > + caller-saved register which isn't live or a callee-saved > > + register saved on stack in the prologue, use it. */ > > + > > + bitmap reg_live = df_get_live_out (ENTRY_BLOCK_PTR_FOR_FN (cfun)); > > + > > + int i; > > + for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) > > + if (GENERAL_REGNO_P (i) > > + && i != R10_REG > > +#ifdef NO_PROFILE_COUNTERS > > + && (r11_ok || i != R11_REG) > > +#else > > + && i != R11_REG > > +#endif > > + && TEST_HARD_REG_BIT (accessible_reg_set, i) > > + && !fixed_regs[i] > > + && (ix86_save_reg (i, true, true) > > + || (call_used_regs[i] > > + && !REGNO_REG_SET_P (reg_live, i)))) > > + return i; > > ix86_save_reg will never save fixed regs, so the above can be optimized a bit: > > && TEST_HARD_REG_BIT (accessible_reg_set, i) > && (ix86_save_reg (i, true, true) > || (call_used_regs[i] && !fixed_regs[i] > && !REGNO_REG_SET_P (reg_live, i)))) > > OK with the above change. Fixed. This is the patch I am checking in. Thanks. > Thanks, > Uros. > -- H.J. [-- Attachment #2: 0001-x86-64-Find-a-scratch-register-for-large-model-profi.patch --] [-- Type: text/x-patch, Size: 9325 bytes --] From dc3fe511a13bccf510793c22e6ba7a0cc0b9c1f6 Mon Sep 17 00:00:00 2001 From: "H.J. Lu" <hjl.tools@gmail.com> Date: Thu, 1 Feb 2024 08:02:27 -0800 Subject: [PATCH] x86-64: Find a scratch register for large model profiling 2 scratch registers, %r10 and %r11, are available at function entry for large model profiling. But %r10 may be used by stack realignment and we can't use %r10 in this case. Add x86_64_select_profile_regnum to find a caller-saved register which isn't live or a callee-saved register which has been saved on stack in the prologue at entry for large model profiling and sorry if we can't find one. gcc/ PR target/113689 * config/i386/i386.cc (x86_64_select_profile_regnum): New. (x86_function_profiler): Call x86_64_select_profile_regnum to get a scratch register for large model profiling. gcc/testsuite/ PR target/113689 * gcc.target/i386/pr113689-1.c: New file. * gcc.target/i386/pr113689-2.c: Likewise. * gcc.target/i386/pr113689-3.c: Likewise. --- gcc/config/i386/i386.cc | 91 ++++++++++++++++++---- gcc/testsuite/gcc.target/i386/pr113689-1.c | 49 ++++++++++++ gcc/testsuite/gcc.target/i386/pr113689-2.c | 41 ++++++++++ gcc/testsuite/gcc.target/i386/pr113689-3.c | 48 ++++++++++++ 4 files changed, 214 insertions(+), 15 deletions(-) create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-1.c create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-2.c create mode 100644 gcc/testsuite/gcc.target/i386/pr113689-3.c diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc index b3e7c74846e..f02c6c02ac6 100644 --- a/gcc/config/i386/i386.cc +++ b/gcc/config/i386/i386.cc @@ -22749,6 +22749,48 @@ current_fentry_section (const char **name) return true; } +/* Return a caller-saved register which isn't live or a callee-saved + register which has been saved on stack in the prologue at entry for + profile. */ + +static int +x86_64_select_profile_regnum (bool r11_ok ATTRIBUTE_UNUSED) +{ + /* Use %r10 if the profiler is emitted before the prologue or it isn't + used by DRAP. */ + if (ix86_profile_before_prologue () + || !crtl->drap_reg + || REGNO (crtl->drap_reg) != R10_REG) + return R10_REG; + + /* The profiler is emitted after the prologue. If there is a + caller-saved register which isn't live or a callee-saved + register saved on stack in the prologue, use it. */ + + bitmap reg_live = df_get_live_out (ENTRY_BLOCK_PTR_FOR_FN (cfun)); + + int i; + for (i = 0; i < FIRST_PSEUDO_REGISTER; i++) + if (GENERAL_REGNO_P (i) + && i != R10_REG +#ifdef NO_PROFILE_COUNTERS + && (r11_ok || i != R11_REG) +#else + && i != R11_REG +#endif + && TEST_HARD_REG_BIT (accessible_reg_set, i) + && (ix86_save_reg (i, true, true) + || (call_used_regs[i] + && !fixed_regs[i] + && !REGNO_REG_SET_P (reg_live, i)))) + return i; + + sorry ("no register available for profiling %<-mcmodel=large%s%>", + ix86_cmodel == CM_LARGE_PIC ? " -fPIC" : ""); + + return INVALID_REGNUM; +} + /* Output assembler code to FILE to increment profiler label # LABELNO for profiling a function entry. */ void @@ -22783,42 +22825,61 @@ x86_function_profiler (FILE *file, int labelno ATTRIBUTE_UNUSED) fprintf (file, "\tleaq\t%sP%d(%%rip), %%r11\n", LPREFIX, labelno); #endif + int scratch; + const char *reg; + char legacy_reg[4] = { 0 }; + if (!TARGET_PECOFF) { switch (ix86_cmodel) { case CM_LARGE: - /* NB: R10 is caller-saved. Although it can be used as a - static chain register, it is preserved when calling - mcount for nested functions. */ + scratch = x86_64_select_profile_regnum (true); + reg = hi_reg_name[scratch]; + if (LEGACY_INT_REGNO_P (scratch)) + { + legacy_reg[0] = 'r'; + legacy_reg[1] = reg[0]; + legacy_reg[2] = reg[1]; + reg = legacy_reg; + } if (ASSEMBLER_DIALECT == ASM_INTEL) - fprintf (file, "1:\tmovabs\tr10, OFFSET FLAT:%s\n" - "\tcall\tr10\n", mcount_name); + fprintf (file, "1:\tmovabs\t%s, OFFSET FLAT:%s\n" + "\tcall\t%s\n", reg, mcount_name, reg); else - fprintf (file, "1:\tmovabsq\t$%s, %%r10\n\tcall\t*%%r10\n", - mcount_name); + fprintf (file, "1:\tmovabsq\t$%s, %%%s\n\tcall\t*%%%s\n", + mcount_name, reg, reg); break; case CM_LARGE_PIC: #ifdef NO_PROFILE_COUNTERS + scratch = x86_64_select_profile_regnum (false); + reg = hi_reg_name[scratch]; + if (LEGACY_INT_REGNO_P (scratch)) + { + legacy_reg[0] = 'r'; + legacy_reg[1] = reg[0]; + legacy_reg[2] = reg[1]; + reg = legacy_reg; + } if (ASSEMBLER_DIALECT == ASM_INTEL) { fprintf (file, "1:movabs\tr11, " "OFFSET FLAT:_GLOBAL_OFFSET_TABLE_-1b\n"); - fprintf (file, "\tlea\tr10, 1b[rip]\n"); - fprintf (file, "\tadd\tr10, r11\n"); + fprintf (file, "\tlea\t%s, 1b[rip]\n", reg); + fprintf (file, "\tadd\t%s, r11\n", reg); fprintf (file, "\tmovabs\tr11, OFFSET FLAT:%s@PLTOFF\n", mcount_name); - fprintf (file, "\tadd\tr10, r11\n"); - fprintf (file, "\tcall\tr10\n"); + fprintf (file, "\tadd\t%s, r11\n", reg); + fprintf (file, "\tcall\t%s\n", reg); break; } fprintf (file, "1:\tmovabsq\t$_GLOBAL_OFFSET_TABLE_-1b, %%r11\n"); - fprintf (file, "\tleaq\t1b(%%rip), %%r10\n"); - fprintf (file, "\taddq\t%%r11, %%r10\n"); + fprintf (file, "\tleaq\t1b(%%rip), %%%s\n", reg); + fprintf (file, "\taddq\t%%r11, %%%s\n", reg); fprintf (file, "\tmovabsq\t$%s@PLTOFF, %%r11\n", mcount_name); - fprintf (file, "\taddq\t%%r11, %%r10\n"); - fprintf (file, "\tcall\t*%%r10\n"); + fprintf (file, "\taddq\t%%r11, %%%s\n", reg); + fprintf (file, "\tcall\t*%%%s\n", reg); #else sorry ("profiling %<-mcmodel=large%> with PIC is not supported"); #endif diff --git a/gcc/testsuite/gcc.target/i386/pr113689-1.c b/gcc/testsuite/gcc.target/i386/pr113689-1.c new file mode 100644 index 00000000000..8285c0a07b7 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr113689-1.c @@ -0,0 +1,49 @@ +/* { dg-do run { target { lp64 && fpic } } } */ +/* { dg-options "-O2 -fno-pic -fprofile -mcmodel=large" } */ + +#include <stdarg.h> + +__attribute__((noipa)) +void +bar (int a1, int a2, int a3, int a4, int a5, int a6, + char *x, char *y, int *z) +{ + if (a1 != 1) + __builtin_abort (); + if (a2 != 2) + __builtin_abort (); + if (a3 != 3) + __builtin_abort (); + if (a4 != 4) + __builtin_abort (); + if (a5 != 5) + __builtin_abort (); + if (a6 != 6) + __builtin_abort (); + x[0] = 42; + y[0] = 42; + if (z[0] != 16) + __builtin_abort (); +} + +__attribute__((noipa)) +void +foo (int c, int d, int e, int f, int g, int h, int z, ...) +{ + typedef char B[32]; + B b __attribute__((aligned (32))); + va_list ap; + va_start (ap, z); + double x = va_arg (ap, double); + if (x > 40.0) + __builtin_abort (); + bar (c, d, e, f, g, h, &b[0], __builtin_alloca (z), &z); + va_end (ap); +} + +int +main () +{ + foo (1, 2, 3, 4, 5, 6, 16, 38.0); + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr113689-2.c b/gcc/testsuite/gcc.target/i386/pr113689-2.c new file mode 100644 index 00000000000..2e5579ac546 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr113689-2.c @@ -0,0 +1,41 @@ +/* { dg-do run { target { lp64 && fpic } } } */ +/* { dg-options "-O2 -fpic -fprofile -mcmodel=large" } */ + +__attribute__((noipa)) +void +bar (int a1, int a2, int a3, int a4, int a5, int a6, + char *x, char *y, int *z) +{ + if (a1 != 1) + __builtin_abort (); + if (a2 != 2) + __builtin_abort (); + if (a3 != 3) + __builtin_abort (); + if (a4 != 4) + __builtin_abort (); + if (a5 != 5) + __builtin_abort (); + if (a6 != 6) + __builtin_abort (); + x[0] = 42; + y[0] = 42; + if (z[0] != 16) + __builtin_abort (); +} + +__attribute__((noipa)) +void +foo (int c, int d, int e, int f, int g, int h, int z) +{ + typedef char B[32]; + B b __attribute__((aligned (32))); + bar (c, d, e, f, g, h, &b[0], __builtin_alloca (z), &z); +} + +int +main () +{ + foo (1, 2, 3, 4, 5, 6, 16); + return 0; +} diff --git a/gcc/testsuite/gcc.target/i386/pr113689-3.c b/gcc/testsuite/gcc.target/i386/pr113689-3.c new file mode 100644 index 00000000000..dab75190635 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr113689-3.c @@ -0,0 +1,48 @@ +/* { dg-do run { target { lp64 && fpic } } } */ +/* { dg-options "-O2 -fpic -fprofile -mcmodel=large" } */ + +#include <stdarg.h> + +__attribute__((noipa)) +void +bar (char *x, char *y, int *z) +{ + x[0] = 42; + y[0] = 42; + if (z[0] != 16) + __builtin_abort (); +} + +__attribute__((noipa)) +void +foo (int a1, int a2, int a3, int a4, int a5, int a6, int z, ...) +{ + typedef char B[32]; + B b __attribute__((aligned (32))); + va_list ap; + va_start (ap, z); + double x = va_arg (ap, double); + if (x > 40.0) + __builtin_abort (); + if (a1 != 1) + __builtin_abort (); + if (a2 != 2) + __builtin_abort (); + if (a3 != 3) + __builtin_abort (); + if (a4 != 4) + __builtin_abort (); + if (a5 != 5) + __builtin_abort (); + if (a6 != 6) + __builtin_abort (); + bar (&b[0], __builtin_alloca (z), &z); + va_end (ap); +} + +int +main () +{ + foo (1, 2, 3, 4, 5, 6, 16, 38.0); + return 0; +} -- 2.43.0 ^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2024-02-05 19:08 UTC | newest] Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2024-02-05 16:43 [PATCH v6] x86-64: Find a scratch register for large model profiling H.J. Lu 2024-02-05 18:01 ` Uros Bizjak 2024-02-05 19:07 ` H.J. Lu
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).