[Xen-users] Xenoprof error and missing Xen-syms events

[Viren Kumar <virenk@xxxxxxx>]

Hello everyone,

I'm trying to run Xenoprof (Oprofile 0.9.4) in Gentoo Linux 2.6.21-xen (64 
bit) on an AMD 64 system and am facing the following two issues:

1. Xenoprof complains about a write error, but starts and runs anyway.

dom0 # 
opcontrol --start-daemon --active-domains=0 --event=CPU_CLK_UNHALTED:75000::1:1 
--vmlinux=/usr/src/linux-2.6.21-xen-dom0/vmlinux --xen=/boot/xen-syms-3.2.1 
--separate=library --verbose=all

/usr/bin/opcontrol: line 1056: echo: write error: Invalid argument

So, I checked line 1056 in /usr/bin/opcontrol and it turns out to be the line 
below

set_param()
{
        if test "$KERNEL_SUPPORT" = "yes"; then
                echo $2 >$MOUNT/$1      <--LINE 1056
        else
                $SYSCTL -w dev.oprofile.$1=$2
        fi
}

I'm not sure why that's happening, perhaps someone can shed some light on it 
for me.

2. My second issue might be linked to the error above, but when I run opreport 
after sampling, it shows me the samples for the application and the kernel, 
but not for xen-syms. As far as I can tell, I've started it the right way, 
and Xen-syms was installed by Gentoo for me. I've also tried to compile Xen 
on my own and used this Xen-syms, but to no avail. I gunzipped /boot/xen.gz 
and tried that, then tried /boot/xen.gz itself as a last resort, but nothing 
worked. 

In /var/lib/oprofile/samples/oprofiled.log, I see plenty of lines like the 
following:

XEN_ENTER_SWITCH to xen
CTX_SWITCH to tid 0, tgid 0, cookie 0(none)
Lost kernel sample ffff828c8011243f
Lost kernel sample ffff828c8011240e
Lost kernel sample ffff828c801124b2
Lost kernel sample ffff828c8011233b
KERNEL_ENTER_SWITCH to kernel
Sample 0x2aea2(0): kern (name /usr/src/linux-2.6.21-xen-dom0/vmlinux, 
0xffffffff80200000-0xffffffff805196ea), app invalid(ffffffffffffffff)
CTX_SWITCH to tid 28, tgid 28, cookie 0(none)
Sample 0x313054(0): kern (name /usr/src/linux-2.6.21-xen-dom0/vmlinux, 
0xffffffff80200000-0xffffffff805196ea), app invalid(ffffffffffffffff)


So, the vmlinux kernel samples are being mapped to their appropriate points, 
but the Xen samples are being lost. I'm not sure why this is happening 
either, has anyone face this before, or do they know what might be causing 
it?

Thanks in advance,
Viren

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users