The systemtap folks are interested in having System.map installed by
kernel-devel, not just by the kernel rpm itself. This makes it possible to
do more preparations off-line for a kernel other than an installed one.
Without objection, I''ll check this in on all branches.
Thanks,
Roland
Index: kernel.spec
==================================================================RCS file:
/cvs/pkgs/rpms/kernel/F-8/kernel.spec,v
retrieving revision 1.239
diff -u -b -p -r1.239 kernel.spec
--- kernel.spec 22 Oct 2007 20:49:43 -0000 1.239
+++ kernel.spec 23 Oct 2007 13:02:39 -0000
@@ -1446,6 +1446,7 @@ BuildKernel() {
# first copy everything
cp --parents `find -type f -name "Makefile*" -o -name
"Kconfig*"` $RPM_BUILD_ROOT/lib/modules/$KernelVer/build
cp Module.symvers $RPM_BUILD_ROOT/lib/modules/$KernelVer/build
+ cp System.map $RPM_BUILD_ROOT/lib/modules/$KernelVer/build
# then drop all but the needed Makefiles/Kconfig files
rm -rf $RPM_BUILD_ROOT/lib/modules/$KernelVer/build/Documentation
rm -rf $RPM_BUILD_ROOT/lib/modules/$KernelVer/build/scripts
On 10/23/2007 09:05 AM, Roland McGrath wrote:> The systemtap folks are interested in having System.map installed by > kernel-devel, not just by the kernel rpm itself. This makes it possible to > do more preparations off-line for a kernel other than an installed one. > > Without objection, I''ll check this in on all branches. >Are the addresses in System.map accurate? On the F7 2.6.23 kernel, I had to subtract 0x400000 and add 0x1000000 to the address in an oops message to get an address to use with eu-addr2line.
On Tue, Oct 23, 2007 at 06:26:21PM -0400, Chuck Ebbert wrote: > On 10/23/2007 09:05 AM, Roland McGrath wrote: > > The systemtap folks are interested in having System.map installed by > > kernel-devel, not just by the kernel rpm itself. This makes it possible to > > do more preparations off-line for a kernel other than an installed one. > > > > Without objection, I''ll check this in on all branches. > > > > Are the addresses in System.map accurate? On the F7 2.6.23 kernel, > I had to subtract 0x400000 and add 0x1000000 to the address in an > oops message to get an address to use with eu-addr2line. Relocatable kernel is another thing that really screws with this. Dave -- http://www.codemonkey.org.uk
On 10/23/2007 06:28 PM, Dave Jones wrote:> > Are the addresses in System.map accurate? On the F7 2.6.23 kernel, > > I had to subtract 0x400000 and add 0x1000000 to the address in an > > oops message to get an address to use with eu-addr2line. > > Relocatable kernel is another thing that really screws with this. >CONFIG_RELOCATABLE=y CONFIG_PHYSICAL_ALIGN=0x400000 CONFIG_PHYSICAL_START=0x1000000 Address in oops was c04622db. I had to use "eu-addr2line -e vmlinux 0xc10622db" And objdump just flat refuses to show line number information from vmlinux containing debug info. (But on Fedora 8 it will.)
Chuck Ebbert wrote:> On 10/23/2007 06:28 PM, Dave Jones wrote: >> > Are the addresses in System.map accurate? On the F7 2.6.23 kernel, >> > I had to subtract 0x400000 and add 0x1000000 to the address in an >> > oops message to get an address to use with eu-addr2line. >> >> Relocatable kernel is another thing that really screws with this. >> > > CONFIG_RELOCATABLE=y > CONFIG_PHYSICAL_ALIGN=0x400000 > CONFIG_PHYSICAL_START=0x1000000 > > Address in oops was c04622db. > > I had to use "eu-addr2line -e vmlinux 0xc10622db" > > And objdump just flat refuses to show line number information > from vmlinux containing debug info. (But on Fedora 8 it will.)I vaguely recall seeing a bug about this one, and I thought the solution was to set _ALIGH and _START to the same value, but these are only vague recollections... -- Jarod Wilson jwilson@redhat.com
On Tue, Oct 23, 2007 at 08:50:23PM -0400, Jarod Wilson wrote: > Chuck Ebbert wrote: > > On 10/23/2007 06:28 PM, Dave Jones wrote: > >> > Are the addresses in System.map accurate? On the F7 2.6.23 kernel, > >> > I had to subtract 0x400000 and add 0x1000000 to the address in an > >> > oops message to get an address to use with eu-addr2line. > >> > >> Relocatable kernel is another thing that really screws with this. > >> > > > > CONFIG_RELOCATABLE=y > > CONFIG_PHYSICAL_ALIGN=0x400000 > > CONFIG_PHYSICAL_START=0x1000000 > > > > Address in oops was c04622db. > > > > I had to use "eu-addr2line -e vmlinux 0xc10622db" > > > > And objdump just flat refuses to show line number information > > from vmlinux containing debug info. (But on Fedora 8 it will.) > > I vaguely recall seeing a bug about this one, and I thought the solution > was to set _ALIGH and _START to the same value, but these are only vague > recollections... That sounds a bit.. odd. to say the least. Though we current should have 4MB alignment and 4MB start. Dave -- http://www.codemonkey.org.uk
On Tue, Oct 23, 2007 at 06:35:31PM -0400, Chuck Ebbert wrote: > On 10/23/2007 06:28 PM, Dave Jones wrote: > > > Are the addresses in System.map accurate? On the F7 2.6.23 kernel, > > > I had to subtract 0x400000 and add 0x1000000 to the address in an > > > oops message to get an address to use with eu-addr2line. > > > > Relocatable kernel is another thing that really screws with this. > > > > CONFIG_RELOCATABLE=y > CONFIG_PHYSICAL_ALIGN=0x400000 > CONFIG_PHYSICAL_START=0x1000000 > > Address in oops was c04622db. > > I had to use "eu-addr2line -e vmlinux 0xc10622db" > > And objdump just flat refuses to show line number information > from vmlinux containing debug info. (But on Fedora 8 it will.) Hmm, they should both be @4mb. I''ll poke at this tomorrow. Dave -- http://www.codemonkey.org.uk
Ok, well all that is not a reason not to install the file, just reasons to fix it. I''ll go ahead and check in the .spec change. Is there really a problem, or just the relocatable kernel issue? i.e. are System.map addresses "wrong" or just biased? Thanks, Roland