Hello Dmitry, I remember I removed "Linux-kernel" because I wanted to reduce modification as little as possible and it does not affect my debug operations. I am glad if someone tells me what it is for. -caz
-----Original Message----- From: Dmitry Mikhin [mailto:[email protected]] Sent: Wednesday, April 29, 2009 12:31 AM To: Caz Yokoyama Cc: [email protected] Subject: Re: [Kgdb-bugreport] debugging kernel modules with kgdb-light in 2.6.29.1 Yes, now kgdb-light brannch compiles without error. But I've got another problem: ***** (kgdb) set osabi Linux-kernel Undefined item: "Linux-kernel". ***** Caz Yokoyama wrote: > The latest code invalidates those code on other than x86_64. > -caz > > -----Original Message----- > From: Dmitry Mikhin [mailto:[email protected]] > Sent: Tuesday, April 28, 2009 12:24 PM > To: Caz Yokoyama > Cc: 'Dmitrij Mihin'; [email protected] > Subject: Re: [Kgdb-bugreport] debugging kernel modules with kgdb-light > in2.6.29.1 > > Thanks, I've got it. I'll do more research tommorow. > > Caz Yokoyama wrote: >> Hello Dmitry, >> >> Yes, it is useless on i386 architecture, but it is very important on >> x86_64. I spent several weeks to find this bug on gdb. >> >> -caz >> >> >> >> ------------------------------------------------------------------------ >> >> *From:* Dmitrij Mihin [mailto:[email protected]] >> *Sent:* Tuesday, April 28, 2009 11:26 AM >> *To:* Caz Yokoyama; [email protected] >> *Cc:* [email protected] >> *Subject:* Re: [Kgdb-bugreport] debugging kernel modules with >> kgdb-light in2.6.29.1 >> >> >> >> Hello Caz, >> >> the problem is in constant in string 1781 in file dwarf2read.c: >> ****** >> 1778: pdi = pdi->die_sibling; >> 1779: >> 1780: if (debugkernel) { >> 1781: if (pdi != NULL && pdi->highpc >= */0xffffffffff000000/*) { >> 1782: if (pdi->die_sibling != NULL) >> 1783: pdi = pdi->die_sibling; >> 1784: } >> 1785: } >> ****** >> >> I think code in strings 1780-1785 is useless because of 1778 and it >> can be safely removed. >> >> Regards, >> Dmitry. >> >> On Tue, Apr 28, 2009 at 9:32 PM, Caz Yokoyama <[email protected] >> <mailto:[email protected]>> wrote: >> >> Hello Dmitry, >> My build on F8 x86_64 finished successfully. So, the problem you saw >> is the >> one on your environment. My gcc is red hat 4.1.2-33. >> Are you able to figure out which integer constant produces the >> warning? The >> warning message does not show which. >> >> -caz >> >> -----Original Message----- >> From: Dmitry Mikhin [mailto:[email protected] <mailto:[email protected]>] >> >> Sent: Tuesday, April 28, 2009 10:00 AM >> To: Caz Yokoyama >> >> Cc: 'Dmitry Mikhin'; [email protected] >> <mailto:[email protected]> >> Subject: Re: [Kgdb-bugreport] debugging kernel modules with kgdb-light >> in2.6.29.1 >> >> Hello Caz, >> >> first of all, thank you very much for your work. >> Actually I'm Gentoo user, so I prefer to build everything from source >> code :) >> I'm using gcc 4.1.2 on x86. If you could share you local generalized >> copy, it will be great. >> In any case, now I'm working in this field, so if you need any >> assistance, I can try to help. >> >> Best Regards, >> Dmitry. >> >> Caz Yokoyama wrote: >>> Hello Dmitry, >>> I am an owner of kgdb-light. I generalized to support 2.6.27 and later >>> locally. Let me build the code. BTW, which environment do you build? I > am >>> using Fedora 8 on x86_64. >>> -caz >>> >>> -----Original Message----- >>> From: Dmitry Mikhin [mailto:[email protected] <mailto:[email protected]>] >>> Sent: Tuesday, April 28, 2009 7:57 AM >>> To: [email protected] >> <mailto:[email protected]> >>> Subject: [Kgdb-bugreport] debugging kernel modules with kgdb-light >>> in2.6.29.1 >>> >>> Hello all, >>> >>> i've managed to debug kernel 2.6.29.1 with kgdb-light and standart gdb >>> 6.8 (no modifications). >>> Next step is debugging of kernel modules. Unfortunatly standart gdb >>> doesn't support it. >>> I've found gdbmod-2.4, but it is pretty old (gdb 6.4), and looks like it >>> is not supported any more. >>> Another project is http://sourceforge.net/projects/kgdb-light/, it is >>> based on gdb 6.8. >>> But it gives errors during compilation: >>> >>> cc1: warnings being treated as errors >>> dwarf2read.c: In function 'scan_partial_symbols': >>> dwarf2read.c:1781: warning: integer constant is too large for 'long' > type >>> dwarf2read.c:1781: warning: comparison is always false due to limited >>> range of data type >>> make[2]: *** [dwarf2read.o] Error 1 >>> >>> and looks like it supports only kernels 2.6.18 and 2.6.27 (look at >>> string 1821 in solid-srv4.c) >>> >>> What version of gdb do you use for debugging kernel modules? >>> >>> Any help will be greatly appreciated. >>> >>> Kind Regards, >>> Dmitry >>> >>> > ---------------------------------------------------------------------------- >>> -- >>> Register Now & Save for Velocity, the Web Performance & Operations >>> Conference from O'Reilly Media. Velocity features a full day of >>> expert-led, hands-on workshops and two days of sessions from industry >>> leaders in dedicated Performance & Operations tracks. Use code vel09scf >>> and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf >>> _______________________________________________ >>> Kgdb-bugreport mailing list >>> [email protected] >> <mailto:[email protected]> >>> https://lists.sourceforge.net/lists/listinfo/kgdb-bugreport >>> >>> >>> >> >> > > ------------------------------------------------------------------------------ Register Now & Save for Velocity, the Web Performance & Operations Conference from O'Reilly Media. Velocity features a full day of expert-led, hands-on workshops and two days of sessions from industry leaders in dedicated Performance & Operations tracks. Use code vel09scf and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf _______________________________________________ Kgdb-bugreport mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/kgdb-bugreport
