Fix a Linux kernel TSC calibration issue in the presence of SMI
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
intel |
Fix Released
|
Undecided
|
Unassigned | ||
linux (Ubuntu) |
Fix Released
|
Medium
|
Leann Ogasawara | ||
Oneiric |
Fix Released
|
Medium
|
Tim Gardner | ||
Precise |
Fix Released
|
Medium
|
Leann Ogasawara |
Bug Description
Linux TSC calibration using PIT was written to take potential SMI's in to account. But it had one corner case where the SMI was resulting in a variation of approximately +/- 20MHz in the calibrated values.
NTP can fix the variations with in 500ppm and this was resulting in time sync issue on the large cluster using Romley platforms. Our debug showed that it was indeed related to the SMI. Kernel fix is pushed into upstream (included in 3.3-rc1):
http://
This NTP failure is reported by the Lawrence Livermore National Labs on their cluster using Jaketown cpu's. And Intel developer suggested all the enterprise distributions to backport this patch to their kernels.
So please consider it for Precise.
Changed in linux (Ubuntu Precise): | |
assignee: | nobody → Leann Ogasawara (leannogasawara) |
importance: | Undecided → Medium |
status: | New → In Progress |
visibility: | private → public |
Changed in linux (Ubuntu Oneiric): | |
assignee: | nobody → Tim Gardner (timg-tpi) |
status: | New → In Progress |
Changed in linux (Ubuntu Oneiric): | |
importance: | Undecided → Medium |
Changed in linux (Ubuntu Precise): | |
status: | In Progress → Fix Committed |
Changed in linux (Ubuntu Oneiric): | |
status: | In Progress → Fix Committed |
tags: | added: verification-needed-oneiric |
Changed in intel: | |
status: | New → Fix Released |
This bug was fixed in the package linux - 3.2.0-21.34
---------------
linux (3.2.0-21.34) precise; urgency=low
[ AceLan Kao ]
* SAUCE: (drop after 3.4) asus-wmi: add scalar board brightness adj.
support
- LP: #961879
* SAUCE: (drop after 3.4) asus-wmi: store backlight power status for AIO
machine
- LP: #961880
[ Andy Whitcroft ]
* [Config]: enable CONFIG_ DRM_LOAD_ EDID_FIRMWARE
[ John Johansen ]
* SAUCE: AppArmor: Add the ability to mediate mount
* SAUCE: AppArmor: Add profile introspection file to interface
* SAUCE: AppArmor: basic networking rules
* SAUCE: Update aufs for build failure caused by apparmor backport
* SAUCE: AppArmor: Fix change_onexec when called from a confined task
- LP: #963756
[ Leann Ogasawara ]
* Revert "SAUCE: (drop after 3.3) platform/x86: Add driver for Apple gmux
device"
* Rebase to v3.2.13
* d-i: Fix iwl3945 driver name typo in nic-modules
* d-i: Add missing iwl* drivers to nic-modules
- LP: #965116
* d-i: Remove obsolete bcm43xx* drivers from nic-modules
* d-i: Add brcm*mac drivers to nic-modules
[ Seth Forshee ]
* SAUCE: (drop after 3.3) apple_bl: Add register/unregister functions
* SAUCE: (drop after 3.3) platform/x86: Add driver for Apple gmux device
[ Tim Gardner ]
* Add isci to debian scsi-modules udeb
- LP: #963306
* SAUCE: fsam7400: use UMH_WAIT_PROC consistently
- LP: #963685
* SAUCE: disable_nx should not be in __cpuinitdata section for X86_32
- LP: #968233
[ Upstream Kernel Changes ]
* Bluetooth: Adding USB device 13d3:3375 as an Atheros AR3012. FMA/AVX2) to guest path_chmod( ) to umode_t create_ file() to umode_t path_chmod( ) to struct path *
- LP: #916319
* x86: Report cpb and eff_freq_ro flags correctly
- LP: #960461
* x86/amd: Add missing feature flag for fam15h models 10h-1fh processors
- LP: #960461
* KVM: Move cpuid code to new file
- LP: #960466
* KVM: expose latest Intel cpu new features (BMI1/BMI2/
- LP: #960466
* apparmor: add missing rcu_dereference()
* apparmor: fix module parameter handling
* btrfs, nfs, apparmor: don't pull mnt_namespace.h for no reason...
* switch security_
* switch securityfs_
* switch ->path_mkdir() to umode_t
* switch ->path_mknod() to umode_t
* capabilities: remove the task from capable LSM hook entirely
* switch security_
* AppArmor: refactor securityfs to use structures
* AppArmor: add initial "features" directory to securityfs
* AppArmor: add "file" details to securityfs
* AppArmor: export known rlimit names/value mappings in securityfs
* AppArmor: Add mising end of structure test to caps unpacking
* AppArmor: Fix dropping of allowed operations that are force audited
* AppArmor: Fix underflow in xindex calculation
* AppArmor: fix mapping of META_READ to audit and quiet flags
* AppArmor: Fix the error case for chroot relative path name lookup
* AppArmor: Fix error returned when a path lookup is disconnected
* AppArmor: Fix oops in policy unpack auditing
* AppArmor: Add const qualifiers to generated string tables
* AppArmor: Retrieve the dentry_path for error reporting when path lookup
fails
...