Andi Kleen <a...@firstfloor.org> writes: Ping!
> Andi Kleen <a...@firstfloor.org> writes: > > Ping! > >> From: Andi Kleen <a...@linux.intel.com> >> >> The Linux kernel dynamically patches in __fentry__ calls in and >> out at runtime. This allows using function tracing for debugging >> in production kernels without (significant) performance penalty. >> >> For this it needs a table pointing to each __fentry__ call. >> >> The way it is currently implemented is that a special >> perl script scans the object file, generates the table in a special >> section. When the kernel boots up it nops the calls, and >> then later patches in the calls again as needed. >> >> The recordmcount.pl script in the kernel works, but it seems >> cleaner and faster to support the code generation of the patch table >> directly in gcc. >> >> This also allows to nop the calls directly at code generation >> time, which allows to skip a patching step at kernel boot. >> I also expect that a patchable production tracing facility is also useful >> for other applications. >> >> For example it could be used in ftracer >> (https://github.com/andikleen/ftracer) >> >> Having a nop area at the beginning of each function can be also >> also useful for other things. For example it can be used to patch >> functions at runtime to point to different functions, to do >> binary updates without restarting the program (like ksplice or >> similar) >> >> This patch implements two new options for the i386 target: >> >> -mrecord-mcount >> Generate a __mcount_loc section entry for each __fentry__ or mcount >> call. The section is compatible with the kernel convention >> and the data is put into a section loaded at runtime. >> >> -mnop-mcount >> Generate the mcount/__fentry__ call as 5 byte nop that can be >> patched in later. The nop is generated as a single instruction, >> as the Linux kernel run time patching relies on this. >> >> Limitations: >> - I didn't implement -mnop-mcount for -fPIC. This >> would need a good single instruction 6 byte NOP and it seems a >> bit pointless, as the patching would prevent text sharing. >> - I didn't implement noping for targets that pass a variable >> to mcount. >> - The facility could be useful on architectures too. Currently >> the mcount code is target specific, so I made it a i386 option. >> >> Passes bootstrap and testing on x86_64-linux. >> >> Cc: rost...@goodmis.org >> >> gcc/: >> >> 2014-09-01 Andi Kleen <a...@linux.intel.com> >> >> * config/i386/i386.c (x86_print_call_or_nop): New function. >> (x86_function_profiler): Support -mnop-mcount and >> -mrecord-mcount. >> * config/i386/i386.opt (-mnop-mcount, -mrecord-mcount): Add >> * doc/invoke.texi: Document -mnop-mcount, -mrecord-mcount >> * testsuite/gcc/gcc.target/i386/nop-mcount.c: New file. >> * testsuite/gcc/gcc.target/i386/record-mcount.c: New file. >> --- >> gcc/config/i386/i386.c | 34 >> +++++++++++++++++++++++---- >> gcc/config/i386/i386.opt | 9 +++++++ >> gcc/doc/invoke.texi | 17 +++++++++++++- >> gcc/testsuite/gcc.target/i386/nop-mcount.c | 24 +++++++++++++++++++ >> gcc/testsuite/gcc.target/i386/record-mcount.c | 24 +++++++++++++++++++ >> 5 files changed, 102 insertions(+), 6 deletions(-) >> create mode 100644 gcc/testsuite/gcc.target/i386/nop-mcount.c >> create mode 100644 gcc/testsuite/gcc.target/i386/record-mcount.c >> >> diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c >> index 61b33782..a651aa1 100644 >> --- a/gcc/config/i386/i386.c >> +++ b/gcc/config/i386/i386.c >> @@ -3974,6 +3974,13 @@ ix86_option_override_internal (bool main_args_p, >> } >> } >> >> +#ifndef NO_PROFILE_COUNTERS >> + if (flag_nop_mcount) >> + error ("-mnop-mcount is not compatible with this target"); >> +#endif >> + if (flag_nop_mcount && flag_pic) >> + error ("-mnop-mcount is not implemented for -fPIC"); >> + >> /* Accept -msseregparm only if at least SSE support is enabled. */ >> if (TARGET_SSEREGPARM_P (opts->x_target_flags) >> && ! TARGET_SSE_P (opts->x_ix86_isa_flags)) >> @@ -39042,6 +39049,17 @@ x86_field_alignment (tree field, int computed) >> return computed; >> } >> >> +/* Print call to TARGET to FILE. */ >> + >> +static void >> +x86_print_call_or_nop (FILE *file, const char *target) >> +{ >> + if (flag_nop_mcount) >> + fprintf (file, "1:\tnopl 0x00(%%eax,%%eax,1)\n"); /* 5 byte nop. */ >> + else >> + fprintf (file, "1:\tcall\t%s\n", target); >> +} >> + >> /* Output assembler code to FILE to increment profiler label # LABELNO >> for profiling a function entry. */ >> void >> @@ -39049,7 +39067,6 @@ x86_function_profiler (FILE *file, int labelno >> ATTRIBUTE_UNUSED) >> { >> const char *mcount_name = (flag_fentry ? MCOUNT_NAME_BEFORE_PROLOGUE >> : MCOUNT_NAME); >> - >> if (TARGET_64BIT) >> { >> #ifndef NO_PROFILE_COUNTERS >> @@ -39057,9 +39074,9 @@ x86_function_profiler (FILE *file, int labelno >> ATTRIBUTE_UNUSED) >> #endif >> >> if (!TARGET_PECOFF && flag_pic) >> - fprintf (file, "\tcall\t*%s@GOTPCREL(%%rip)\n", mcount_name); >> + fprintf (file, "1:\tcall\t*%s@GOTPCREL(%%rip)\n", mcount_name); >> else >> - fprintf (file, "\tcall\t%s\n", mcount_name); >> + x86_print_call_or_nop (file, mcount_name); >> } >> else if (flag_pic) >> { >> @@ -39067,7 +39084,7 @@ x86_function_profiler (FILE *file, int labelno >> ATTRIBUTE_UNUSED) >> fprintf (file, "\tleal\t%sP%d@GOTOFF(%%ebx),%%" >> PROFILE_COUNT_REGISTER "\n", >> LPREFIX, labelno); >> #endif >> - fprintf (file, "\tcall\t*%s@GOT(%%ebx)\n", mcount_name); >> + fprintf (file, "1:\tcall\t*%s@GOT(%%ebx)\n", mcount_name); >> } >> else >> { >> @@ -39075,7 +39092,14 @@ x86_function_profiler (FILE *file, int labelno >> ATTRIBUTE_UNUSED) >> fprintf (file, "\tmovl\t$%sP%d,%%" PROFILE_COUNT_REGISTER "\n", >> LPREFIX, labelno); >> #endif >> - fprintf (file, "\tcall\t%s\n", mcount_name); >> + x86_print_call_or_nop (file, mcount_name); >> + } >> + >> + if (flag_record_mcount) >> + { >> + fprintf (file, "\t.section __mcount_loc, \"r\"\n"); >> + fprintf (file, "\t.%s 1b\n", TARGET_64BIT ? "quad" : "long"); >> + fprintf (file, "\t.previous\n"); >> } >> } >> >> diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt >> index 9208b76..acf6b37 100644 >> --- a/gcc/config/i386/i386.opt >> +++ b/gcc/config/i386/i386.opt >> @@ -789,6 +789,15 @@ mfentry >> Target Report Var(flag_fentry) Init(-1) >> Emit profiling counter call at function entry before prologue. >> >> +mrecord-mcount >> +Target Report Var(flag_record_mcount) Init(0) >> +Generate __mcount_loc section with all mcount or __fentry__ calls. >> + >> +mnop-mcount >> +Target Report Var(flag_nop_mcount) Init(0) >> +Generate mcount/__fentry__ calls as nops. To activate they need to be >> +patched in. >> + >> m8bit-idiv >> Target Report Mask(USE_8BIT_IDIV) Save >> Expand 32bit/64bit integer divide into 8bit unsigned integer divide with >> run-time check >> diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi >> index d15d4a9..43fd6b7 100644 >> --- a/gcc/doc/invoke.texi >> +++ b/gcc/doc/invoke.texi >> @@ -691,7 +691,7 @@ Objective-C and Objective-C++ Dialects}. >> -momit-leaf-frame-pointer -mno-red-zone -mno-tls-direct-seg-refs @gol >> -mcmodel=@var{code-model} -mabi=@var{name} -maddress-mode=@var{mode} @gol >> -m32 -m64 -mx32 -m16 -mlarge-data-threshold=@var{num} @gol >> --msse2avx -mfentry -m8bit-idiv @gol >> +-msse2avx -mfentry -mrecord-mcount -mnop-mcount -m8bit-idiv @gol >> -mavx256-split-unaligned-load -mavx256-split-unaligned-store @gol >> -mstack-protector-guard=@var{guard}} >> >> @@ -15954,6 +15954,21 @@ counter call before the prologue. >> Note: On x86 architectures the attribute @code{ms_hook_prologue} >> isn't possible at the moment for @option{-mfentry} and @option{-pg}. >> >> +@item -mrecord-mcount >> +@itemx -mno-record-mcount >> +@opindex mrecord-mcount >> +If profiling is active (@option{-pg}), generate a __mcount_loc section >> +that contains pointers to each profiling call. This is useful for >> +automatically patching and out calls. >> + >> +@item -mnop-mcount >> +@itemx -mno-nop-mcount >> +@opindex mnop-mcount >> +If profiling is active (@option{-pg}), generate the calls to >> +the profiling functions as nops. This is useful when they >> +should be patched in later dynamically. This is likely only >> +useful together with @option{-mrecord-mcount}. >> + >> @item -m8bit-idiv >> @itemx -mno-8bit-idiv >> @opindex 8bit-idiv >> diff --git a/gcc/testsuite/gcc.target/i386/nop-mcount.c >> b/gcc/testsuite/gcc.target/i386/nop-mcount.c >> new file mode 100644 >> index 0000000..2592231 >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/i386/nop-mcount.c >> @@ -0,0 +1,24 @@ >> +/* Test -mnop-mcount */ >> +/* { dg-do compile } */ >> +/* { dg-options "-pg -mfentry -mrecord-mcount -mnop-mcount" } */ >> +/* { dg-final { scan-assembler-not "__fentry__" } } */ >> +/* Origin: Andi Kleen */ >> +extern void foobar(char *); >> + >> +void func(void) >> +{ >> + foobar ("Hello world\n"); >> +} >> + >> +void func2(void) >> +{ >> + int i; >> + for (i = 0; i < 10; i++) >> + foobar ("Hello world"); >> +} >> + >> +void func3(a) >> +char *a; >> +{ >> + foobar("Hello world"); >> +} >> diff --git a/gcc/testsuite/gcc.target/i386/record-mcount.c >> b/gcc/testsuite/gcc.target/i386/record-mcount.c >> new file mode 100644 >> index 0000000..dae413e >> --- /dev/null >> +++ b/gcc/testsuite/gcc.target/i386/record-mcount.c >> @@ -0,0 +1,24 @@ >> +/* Test -mrecord-mcount */ >> +/* { dg-do compile } */ >> +/* { dg-options "-pg -mrecord-mcount" } */ >> +/* { dg-final { scan-assembler "mcount_loc" } } */ >> +/* Origin: Andi Kleen */ >> +extern void foobar(char *); >> + >> +void func(void) >> +{ >> + foobar ("Hello world\n"); >> +} >> + >> +void func2(void) >> +{ >> + int i; >> + for (i = 0; i < 10; i++) >> + foobar ("Hello world"); >> +} >> + >> +void func3(a) >> +char *a; >> +{ >> + foobar("Hello world"); >> +} -- a...@linux.intel.com -- Speaking for myself only
