I'm not sure how to test it - but it looks somewhat suspect. The difficulty is in the way that ia64 tools report program addresses within bundles. ia64 crams 3 41-bit instructions (and a 5-bit template) into a 16-byte "bundle". Some tools report consecutive addresses like this: 0x000 0x001 0x002 0x010 0x011 0x012 where the low nibble is the instruction number within the bundle. Others go for: 0x000 0x006 0x00c 0x010 0x016 0x01c with the illusion that instructions take 6-byte, 6bytes, 4-bytes. This patch does some fast and loose translation from hex (0x12) in the kernel to decimal (18) in the user tools ... which doesn't look like it fits with either of these conventions. -Tony -----Original Message----- From: linux-arch-owner@xxxxxxxxxxxxxxx [mailto:linux-arch-owner@xxxxxxxxxxxxxxx] On Behalf Of Steven Rostedt Sent: Monday, May 16, 2011 11:58 AM To: Luck, Tony Cc: Martin Schwidefsky; linux-arch@xxxxxxxxxxxxxxx Subject: Re: [patch 3/4] ia64 mcount offset calculation Tony, Can you give me an Acked-by for this patch? Thanks, -- Steve On Tue, 2011-05-10 at 10:10 +0200, Martin Schwidefsky wrote: > plain text document attachment (302-mcount-adjust-ia64.diff) > From: Martin Schwidefsky <schwidefsky@xxxxxxxxxx> > > Do the mcount offset adjustment in the recordmcount.pl/recordmcount.[ch] > at compile time and not in ftrace_call_adjust at run time. > > Signed-off-by: Martin Schwidefsky <schwidefsky@xxxxxxxxxx> > --- > > > arch/ia64/include/asm/ftrace.h | 7 +++++-- > scripts/recordmcount.c | 7 ++++++- > scripts/recordmcount.pl | 1 + > 3 files changed, 12 insertions(+), 3 deletions(-) > Index: test-2.6/arch/ia64/include/asm/ftrace.h > =================================================================== > --- test-2.6.orig/arch/ia64/include/asm/ftrace.h 2010-01-11 10:58:31.000000000 +0100 > +++ test-2.6/arch/ia64/include/asm/ftrace.h 2011-05-03 09:39:04.929486388 +0200 > @@ -14,8 +14,11 @@ > > static inline unsigned long ftrace_call_adjust(unsigned long addr) > { > - /* second bundle, insn 2 */ > - return addr - 0x12; > + /* > + * addr is the address of the mcount call instruction. > + * recordmcount does the necessary offset calculation. > + */ > + return addr; > } > > struct dyn_arch_ftrace { > Index: test-2.6/scripts/recordmcount.c > =================================================================== > --- test-2.6.orig/scripts/recordmcount.c 2011-05-03 09:39:02.145472579 +0200 > +++ test-2.6/scripts/recordmcount.c 2011-05-03 09:39:04.929486388 +0200 > @@ -307,7 +307,12 @@ > case EM_ARM: reltype = R_ARM_ABS32; > altmcount = "__gnu_mcount_nc"; > break; > - case EM_IA_64: reltype = R_IA64_IMM64; gpfx = '_'; break; > + case EM_IA_64: > + reltype = R_IA64_IMM64; > + /* Adjust relocation to second bundle, insn 2 */ > + mcount_adjust_32 = -18; > + gpfx = '_'; > + break; > case EM_MIPS: /* reltype: e_class */ gpfx = '_'; break; > case EM_PPC: reltype = R_PPC_ADDR32; gpfx = '_'; break; > case EM_PPC64: reltype = R_PPC64_ADDR64; gpfx = '_'; break; > Index: test-2.6/scripts/recordmcount.pl > =================================================================== > --- test-2.6.orig/scripts/recordmcount.pl 2011-05-03 09:39:02.145472579 +0200 > +++ test-2.6/scripts/recordmcount.pl 2011-05-03 09:39:04.929486388 +0200 > @@ -278,6 +278,7 @@ > > } elsif ($arch eq "ia64") { > $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s_mcount\$"; > + $mcount_adjust = -18; > $type = "data8"; > > if ($is_module eq "0") { -- To unsubscribe from this list: send the line "unsubscribe linux-arch" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-arch" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
