fche changed the topic of #systemtap to: http://sourceware.org/systemtap; email systemtap@sourceware.org if answers here not timely, conversations may be logged
hpt has joined #systemtap
irker386 has quit [Quit: transmission timeout]
gregwork has joined #systemtap
khaled has quit [Remote host closed the connection]
orivej has quit [Ping timeout: 272 seconds]
orivej has joined #systemtap
orivej has quit [Ping timeout: 240 seconds]
orivej has joined #systemtap
orivej has quit [Ping timeout: 256 seconds]
hpt has quit [Ping timeout: 256 seconds]
yogananth has quit [Read error: Connection reset by peer]
sscox has quit [Ping timeout: 240 seconds]
khaled has joined #systemtap
sscox has joined #systemtap
<agentzh> i'm still bothered by the weird minor page faults on some of the Xeon CPU metals. seems like we're not alone: http://lkml.iu.edu/hypermail/linux/kernel/1103.0/01700.html
<agentzh> i wonder if it's possible for stap to allow minor page faults to work around such hardware issues.
<agentzh> i guess minor page faults do not sleep?
<agentzh> such minor faults are repeated even after being read successfully by another process via syscalls like process_vm_readv().
<agentzh> and the PTE is always perfect for such pages. and such pages are allocated by mmap MAP_PRIVATE in the target process.
<agentzh> the permissions are always good (rw)
<agentzh> and never changed since allocation.
<agentzh> such bogus minor read page faults are so annoying.
<agentzh> disabling swap completely on the system does not help (which has nothing to do with minor faults anyway).
<agentzh> any light on this?
<agentzh> thanks!
<agentzh> it's bare metal systems, not using any virtualization.
<agentzh> and it happens on more than 1 machine.
<fche> I believe the problem is that we cannot signal to the kernel that a minor page fault is okay but a full one isn't
derek088_ has quit [Remote host closed the connection]
<agentzh> yeah, that's tricky.
derek0883 has joined #systemtap
derek0883 has quit [Ping timeout: 244 seconds]
derek0883 has joined #systemtap
sscox has quit [Ping timeout: 260 seconds]