Date: Tue, 22 Mar 2011 23:15:25 +0800 From: Xingxing Pan <panxingxing@mprc.pku.edu.cn> To: Chagin Dmitry <dchagin@freebsd.org> Cc: freebsd-hackers@freebsd.org Subject: Re: GSoC'11: DWARF2 call frame information Message-ID: <AANLkTindWYpEKEYXVOC5YTEM2ySTpSTC9c9jPSGPe=Yq@mail.gmail.com> In-Reply-To: <20110321173204.GA7575@dchagin.static.corbina.ru> References: <AANLkTinz6B7fsG7Cjc3ZTZ0OAmsOG8x1P25EfOE7Bt4J@mail.gmail.com> <20110319174115.GA33282@dchagin.static.corbina.ru> <AANLkTimp9qoOFr0VfK1F_r1nE3n_5d7W2-q_vSK_UzgW@mail.gmail.com> <20110320071847.GA10579@dchagin.static.corbina.ru> <AANLkTinxVYZcfvpj4mn7kd%2BfiJ83p7430JSsqxatHCeZ@mail.gmail.com> <20110320181911.GA79862@dchagin.static.corbina.ru> <AANLkTinx%2Bv5ZRr7zq%2BP8gW=umX3PNvhfXpgW8bmUZdtw@mail.gmail.com> <20110321173204.GA7575@dchagin.static.corbina.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
2011/3/22 Chagin Dmitry <dchagin@freebsd.org>: > On Mon, Mar 21, 2011 at 05:36:13PM +0800, Xingxing Pan wrote: >> 2011/3/21 Chagin Dmitry <dchagin@freebsd.org>: >> >> powerfull script. >> >> >> >> Xingxing Pan >> > >> > hmm, which script? I think enough amd64, i386 and amd64/ia32. >> > >> > I suggest to write a example before continuing the conversation >> > about the GSoC. For example (bcopy || bzero) && cpu_switch. >> > Is it ok for you? >> > >> > -- >> > Have fun! >> > chd >> > >> >> Hi Chargin, >> >> Thank you for your reply. >> The followings shows how I try to add DWARF for bcopy. >> >> --- ../8.2.0/sys/i386/include/asm.h =A0 =A0 2011-03-21 14:35:56.11197372= 2 +0800 >> +++ asm.h =A0 =A0 =A0 2011-03-21 15:25:31.564636162 +0800 >> @@ -71,7 +71,7 @@ >> >> =A0#define _ENTRY(x) =A0 =A0 =A0_START_ENTRY; \ >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 .globl CNAME(x); .type C= NAME(x),@function; CNAME(x): >> -#define =A0 =A0 =A0 =A0END(x) =A0 =A0 =A0 =A0 =A0.size x, . - x >> +#define =A0 =A0 =A0 =A0END(x) =A0 =A0 =A0 =A0 =A0.cfi_endproc; .size x,= . - x >> >> =A0#ifdef PROF >> =A0#define =A0 =A0 =A0 =A0ALTENTRY(x) =A0 =A0 _ENTRY(x); \ >> @@ -80,9 +80,10 @@ >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 popl %ebp; \ >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 jmp 9f >> =A0#define =A0 =A0 =A0 =A0ENTRY(x) =A0 =A0 =A0 =A0_ENTRY(x); \ >> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 pushl %ebp; movl %esp,%ebp= ; \ >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 .cfi_startproc; \ >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 pushl %ebp; .cfi_adjust_cf= a_offset 4; movl >> %esp,%ebp; .cfi_def_cfa_register %ebp; \ >> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 call PIC_PLT(HIDENAME(mc= ount)); \ >> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 popl %ebp; \ >> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 popl %ebp; .cfi_def_cfa %e= sp, 4; \ >> >> --- bcopy.S =A0 =A0 2011-03-21 15:51:26.804203809 +0800 >> +++ ../8.2.0/lib/libc/i386/string/bcopy.S =A0 =A0 =A0 2011-03-21 >> 14:28:15.023069890 +0800 >> @@ -51,9 +51,7 @@ ENTRY(bcopy) >> =A0#endif >> =A0#endif >> =A0 =A0 =A0 =A0 pushl =A0 %esi >> - =A0 =A0 =A0 .cfi_adjust_cfa_offset 4; >> =A0 =A0 =A0 =A0 pushl =A0 %edi >> - =A0 =A0 =A0 .cfi_adjust_cfa_offset 4; >> =A0#if defined(MEMCOPY) || defined(MEMMOVE) >> =A0 =A0 =A0 =A0 movl =A0 =A012(%esp),%edi >> =A0 =A0 =A0 =A0 movl =A0 =A016(%esp),%esi >> @@ -77,9 +75,7 @@ ENTRY(bcopy) >> =A0 =A0 =A0 =A0 rep >> =A0 =A0 =A0 =A0 movsb >> =A0 =A0 =A0 =A0 popl =A0 =A0%edi >> - =A0 =A0 =A0 .cfi_adjust_cfa_offset -4; >> =A0 =A0 =A0 =A0 popl =A0 =A0%esi >> - =A0 =A0 =A0 .cfi_adjust_cfa_offset -4; >> =A0 =A0 =A0 =A0 ret >> =A01: >> =A0 =A0 =A0 =A0 addl =A0 =A0%ecx,%edi =A0 =A0 =A0 /* copy backwards. */ >> @@ -98,9 +94,7 @@ ENTRY(bcopy) >> =A0 =A0 =A0 =A0 rep >> =A0 =A0 =A0 =A0 movsl >> =A0 =A0 =A0 =A0 popl =A0 =A0%edi >> - =A0 =A0 =A0 .cfi_adjust_cfa_offset -4; >> =A0 =A0 =A0 =A0 popl =A0 =A0%esi >> - =A0 =A0 =A0 .cfi_adjust_cfa_offset -4; >> =A0 =A0 =A0 =A0 cld >> =A0 =A0 =A0 =A0 ret >> =A0#ifdef MEMCOPY >> >> But I don't know how to add DWARF for cpu_switch, because I have no >> idea about the circumstance when we need to backtrace through this >> function. Suppose there's a cpu switch like this, >> threadA->kernel->threadB. Then should the expected backtrace has the >> following result? >> >> threadB's stack >> kernel's stack >> threadA's stack > > > hmm, ok. good, avoid cpu_switch. > First of all, please, read style(9) man page. > In the second, evaluate the proposed plan (discussed with kib@): > > 1) Annotate libc, msun, rtld, libthr (you) > 2) vdso (I'm) > 3) Annotate signal trampolines (you, after vdso) > > And i'm going to understand what I need to do to start GSoC for you. > Thanks! > > > -- > Have fun! > chd > I think the plan is ok for me. Thank you. Xingxing Pan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AANLkTindWYpEKEYXVOC5YTEM2ySTpSTC9c9jPSGPe=Yq>