Ubuntu
linux package

libhugetlbfs tests fail on powerpc/ppc64el

Bug #1328251 reported by Tim Gardner on 2014年06月09日

This bug affects 2 people

Affects		Status	Importance
linux (Ubuntu)	Fix Released	Undecided	Unassigned
Trusty	Fix Released	Undecided	Tim Gardner
Utopic	Fix Released	Undecided	Unassigned

Bug Description

The test provided for libhugetlbfs along with ubuntu_beta release fail.

Steps to recreate:
============

1. apt-get install libhugetlb
2. cd /usr/lib/libhugetlbs
3. ./run_tests.py

./run_tests.py
Unable to find available page sizes, are you sure hugetlbfs
is mounted and there are available huge pages?

We see the above message.

Tags:

Related branches

lp:ubuntu/precise-proposed/linux-lts-trusty

lp:ubuntu/trusty-proposed/linux-keystone

Revision history for this message

Tim Gardner (timg-tpi) wrote on 2014年06月09日: #1

hugetlb: ensure hugepage access is denied if hugepages are not supported Edit (4.9 KiB, text/plain)

Commit 457c1b27ed56ec472d202731b12417bff023594a fixes the issue.

Changed in linux (Ubuntu Trusty):

assignee: nobody → Tim Gardner (timg-tpi)

status: New → In Progress

Changed in linux (Ubuntu Utopic):

status: New → Fix Released

summary: - libhugetlbfs tests on powerpc/ppc64el
+ libhugetlbfs tests fail on powerpc/ppc64el

Revision history for this message

Nish Aravamudan (nacc-p) wrote on 2014年06月09日: #2

Hi Tim,

Note that the referenced commit only makes it so that if you don't back a PowerKVM guest with hugepages, hugepages do not (falsely) appear to be supported. In such an environment the message referenced in this bug report will still be displayed as it should -- hugepages aren't supported in such a guest.

To run the libhugetlbfs test suite in a PowerKVM guest, you must back the guest with hugepages in the host and then allocate hugepages in the guest as appropriate.

Thanks,
Nish

Revision history for this message

Brad Figg (brad-figg) wrote on 2014年06月12日: #3

This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-trusty' to 'verification-done-trusty'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

tags: added: verification-needed-trusty

Revision history for this message

Anasuya M Shankar (anasuyams) wrote on 2014年06月17日: #4

Working on Enabling the proposed patch and testing the same.

Revision history for this message

Anasuya M Shankar (anasuyams) wrote on 2014年06月17日: #5

Download full text (7.0 KiB)

Applied the proposed patch as follows:

1.Added below entry in /etc/apt/source.list
deb http://ports.ubuntu.com/ubuntu-ports/ trusty-proposed restricted main multiverse universe

2. Did apt-get update

3. Check for libhugetlbfs

root@ubuntu:~# apt-cache policy libhugetlbfs*
libhugetlbfs-dev:
Installed: (none)
Candidate: 2.17-0ubuntu2
Version table:
2.17-0ubuntu2 0
500 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe ppc64el Packages
libhugetlbfs0:
Installed: (none)
Candidate: 2.17-0ubuntu2
Version table:
2.17-0ubuntu2 0
500 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe ppc64el Packages
libhugetlbfs-tests:
Installed: (none)
Candidate: 2.17-0ubuntu2
Version table:
2.17-0ubuntu2 0
500 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe ppc64el Packages
root@ubuntu:~#

4. Install libhugetlbfs
root@ubuntu:~# apt-get install libhugetlbfs*
Reading package lists... Done
Building dependency tree
Reading state information... Done
Note, selecting 'libhugetlbfs-dev' for regex 'libhugetlbfs*'
Note, selecting 'libhugetlbfs0' for regex 'libhugetlbfs*'
Note, selecting 'libhugetlbfs-tests' for regex 'libhugetlbfs*'
The following extra packages will be installed:
hugepages
Suggested packages:
oprofile
The following NEW packages will be installed:
hugepages libhugetlbfs-dev libhugetlbfs-tests libhugetlbfs0
0 upgraded, 4 newly installed, 0 to remove and 56 not upgraded.
Need to get 526 kB of archives.
After this operation, 5,555 kB of additional disk space will be used.
Do you want to continue? [Y/n] y
Get:1 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe libhugetlbfs0 ppc64el 2.17-0ubuntu2 [52.1 kB]
Get:2 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe hugepages ppc64el 2.17-0ubuntu2 [45.0 kB]
Get:3 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe libhugetlbfs-dev ppc64el 2.17-0ubuntu2 [48.6 kB]
Get:4 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe libhugetlbfs-tests ppc64el 2.17-0ubuntu2 [381 kB]
Fetched 526 kB in 3s (171 kB/s)
Selecting previously unselected package libhugetlbfs0.
(Reading database ... 45325 files and directories currently installed.)
Preparing to unpack .../libhugetlbfs0_2.17-0ubuntu2_ppc64el.deb ...
Unpacking libhugetlbfs0 (2.17-0ubuntu2) ...
Selecting previously unselected package hugepages.
Preparing to unpack .../hugepages_2.17-0ubuntu2_ppc64el.deb ...
Unpacking hugepages (2.17-0ubuntu2) ...
Selecting previously unselected package libhugetlbfs-dev.
Preparing to unpack .../libhugetlbfs-dev_2.17-0ubuntu2_ppc64el.deb ...
Unpacking libhugetlbfs-dev (2.17-0ubuntu2) ...
Selecting previously unselected package libhugetlbfs-tests.
Preparing to unpack .../libhugetlbfs-tests_2.17-0ubuntu2_ppc64el.deb ...
Unpacking libhugetlbfs-tests (2.17-0ubuntu2) ...
Processing triggers for man-db (2.6.7.1-1) ...
Setting up libhugetlbfs0 (2.17-0ubuntu2) ...
Setting up hugepages (2.17-0ubuntu2) ...
Setting up libhugetlbfs-dev (2.17-0ubuntu2) ...
Setting up libhugetlbfs-tests (2.17-0ubuntu2) ...
Processing triggers for libc-bin (2.19-0ubuntu6) ...
root@ubuntu:~#
root@ubuntu:~#

5. performed 2 scenrio tests

Tests Performed:
===============
...

Applied the proposed patch as follows:
1.Added below entry in /etc/apt/source.list
deb http://ports.ubuntu.com/ubuntu-ports/ trusty-proposed restricted main multiverse universe
2. Did apt-get update
3. Check for libhugetlbfs
root@ubuntu:~# apt-cache policy libhugetlbfs*
libhugetlbfs-dev:
 Installed: (none)
 Candidate: 2.17-0ubuntu2
 Version table:
 2.17-0ubuntu2 0
 500 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe ppc64el Packages
libhugetlbfs0:
 Installed: (none)
 Candidate: 2.17-0ubuntu2
 Version table:
 2.17-0ubuntu2 0
 500 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe ppc64el Packages
libhugetlbfs-tests:
 Installed: (none)
 Candidate: 2.17-0ubuntu2
 Version table:
 2.17-0ubuntu2 0
 500 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe ppc64el Packages
root@ubuntu:~# 
4. Install libhugetlbfs
root@ubuntu:~# apt-get install libhugetlbfs*
Reading package lists... Done
Building dependency tree 
Reading state information... Done
Note, selecting 'libhugetlbfs-dev' for regex 'libhugetlbfs*'
Note, selecting 'libhugetlbfs0' for regex 'libhugetlbfs*'
Note, selecting 'libhugetlbfs-tests' for regex 'libhugetlbfs*'
The following extra packages will be installed:
 hugepages
Suggested packages:
 oprofile
The following NEW packages will be installed:
 hugepages libhugetlbfs-dev libhugetlbfs-tests libhugetlbfs0
0 upgraded, 4 newly installed, 0 to remove and 56 not upgraded.
Need to get 526 kB of archives.
After this operation, 5,555 kB of additional disk space will be used.
Do you want to continue? [Y/n] y
Get:1 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe libhugetlbfs0 ppc64el 2.17-0ubuntu2 [52.1 kB]
Get:2 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe hugepages ppc64el 2.17-0ubuntu2 [45.0 kB]
Get:3 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe libhugetlbfs-dev ppc64el 2.17-0ubuntu2 [48.6 kB]
Get:4 http://ports.ubuntu.com/ubuntu-ports/ trusty/universe libhugetlbfs-tests ppc64el 2.17-0ubuntu2 [381 kB]
Fetched 526 kB in 3s (171 kB/s) 
Selecting previously unselected package libhugetlbfs0.
(Reading database ... 45325 files and directories currently installed.)
Preparing to unpack .../libhugetlbfs0_2.17-0ubuntu2_ppc64el.deb ...
Unpacking libhugetlbfs0 (2.17-0ubuntu2) ...
Selecting previously unselected package hugepages.
Preparing to unpack .../hugepages_2.17-0ubuntu2_ppc64el.deb ...
Unpacking hugepages (2.17-0ubuntu2) ...
Selecting previously unselected package libhugetlbfs-dev.
Preparing to unpack .../libhugetlbfs-dev_2.17-0ubuntu2_ppc64el.deb ...
Unpacking libhugetlbfs-dev (2.17-0ubuntu2) ...
Selecting previously unselected package libhugetlbfs-tests.
Preparing to unpack .../libhugetlbfs-tests_2.17-0ubuntu2_ppc64el.deb ...
Unpacking libhugetlbfs-tests (2.17-0ubuntu2) ...
Processing triggers for man-db (2.6.7.1-1) ...
Setting up libhugetlbfs0 (2.17-0ubuntu2) ...
Setting up hugepages (2.17-0ubuntu2) ...
Setting up libhugetlbfs-dev (2.17-0ubuntu2) ...
Setting up libhugetlbfs-tests (2.17-0ubuntu2) ...
Processing triggers for libc-bin (2.19-0ubuntu6) ...
root@ubuntu:~# 
root@ubuntu:~# 
5. performed 2 scenrio tests
Tests Performed:
===============
Scenario 1:
=========
Guests backed with hugepages
-------------------------------------------
1. On the Host:
[root@lep8b ubuntu_ga]# sysctl vm.nr_hugepages=1000
vm.nr_hugepages = 1000
[root@lep8b ubuntu_ga]# cat /proc/meminfo | grep -i Huge
HugePages_Total: 1000
HugePages_Free: 1000
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 16384 kB
2. make an entry for Hugepages in guest XML
Add the lines 
<memoryBacking>
 <hugepages/>
 </memoryBacking>
3. Spawn the guest
4. root@ubuntu:~# cat /proc/meminfo | grep -i Huge
AnonHugePages: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 16384 kB
root@ubuntu:~# 
5. root@ubuntu:/usr/lib/libhugetlbfs/tests# mkdir -p /mnt/hugetlbfs
root@ubuntu:/usr/lib/libhugetlbfs/tests# mount -t hugetlbfs none /mnt/hugetlbfs
root@ubuntu:/usr/lib/libhugetlbfs/tests# 
root@ubuntu:/usr/lib/libhugetlbfs/tests# mount | grep -i Huge
none on /mnt/hugetlbfs type hugetlbfs (rw)
6. root@ubuntu:/usr/lib/libhugetlbfs/tests# sysctl vm.nr_hugepages=120
vm.nr_hugepages = 120
root@ubuntu:/usr/lib/libhugetlbfs/tests# cat /proc/meminfo | grep -i huge
AnonHugePages: 0 kB
HugePages_Total: 120
HugePages_Free: 120
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 16384 kB
root@ubuntu:/usr/lib/libhugetlbfs/tests# ./run_tests.py 
run_tests.py: The 32 bit word size is not compatible with 16M pages
zero_filesize_segment (16M: 64):	PASS
test_root (16M: 64):	PASS
meminfo_nohuge (16M: 64):	PASS
gethugepagesize (16M: 64):	PASS
gethugepagesizes (16M: 64):	PASS
HUGETLB_VERBOSE=1 empty_mounts (16M: 64):	PASS
HUGETLB_VERBOSE=1 large_mounts (16M: 64):	PASS
find_path (16M: 64):	PASS
unlinked_fd (16M: 64):	PASS
readback (16M: 64):	PASS
truncate (16M: 64):	PASS
shared (16M: 64):	PASS
mprotect (16M: 64):	PASS
This Sceanrio is PASS.
Sceanrio 2:
============
Guests not backed with hugepages
-------------------------------------------------
1. On the host
[root@lep8b ubuntu_ga]# cat /proc/meminfo | grep -i Huge
HugePages_Total: 1000
HugePages_Free: 1000
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 16384 kB
[root@lep8b ubuntu_ga]# 
2. Do not provide any information about backing guests in the XML.
3. Spawn the guest.
4. root@ubuntu:~# cat /proc/meminfo | grep -i huge
AnonHugePages: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 64 kB
root@ubuntu:/proc/sys/vm# cat nr_hugepages
0
root@ubuntu:/proc/sys/vm# ls
admin_reserve_kbytes laptop_mode oom_dump_tasks
block_dump legacy_va_layout oom_kill_allocating_task
compact_memory lowmem_reserve_ratio overcommit_memory
dirty_background_bytes max_map_count overcommit_ratio
dirty_background_ratio min_free_kbytes page-cluster
dirty_bytes min_slab_ratio panic_on_oom
dirty_expire_centisecs min_unmapped_ratio percpu_pagelist_fraction
dirty_ratio mmap_min_addr scan_unevictable_pages
dirty_writeback_centisecs nr_hugepages stat_interval
drop_caches nr_hugepages_mempolicy swappiness
extfrag_threshold nr_overcommit_hugepages user_reserve_kbytes
hugepages_treat_as_movable nr_pdflush_threads vfs_cache_pressure
hugetlb_shm_group numa_zonelist_order zone_reclaim_mode
root@ubuntu:/proc/sys/vm# cat nr_hugepages
All information about hugepages is still available. As per the patch provided in defect https://bugs.launchpad.net/tasty-taco/+bug/1295705, the hugepage data should not be available. However it doesn't seem to be the same.
This sceanrio as a result will be a FAIL.
Kindly do run through my patch application steps and let me know if the right steps have been followed or should I be choosing a different repo altogether.

Tim Gardner (timg-tpi) on 2014年06月17日

Changed in linux (Ubuntu Trusty):

status: In Progress → Fix Committed

Brad Figg (brad-figg) on 2014年06月20日

tags: added: verification-done-trusty
removed: verification-needed-trusty

Revision history for this message

Launchpad Janitor (janitor) wrote on 2014年06月26日: #6

Download full text (19.3 KiB)

This bug was fixed in the package linux - 3.13.0-30.54

---------------
linux (3.13.0-30.54) trusty; urgency=low

[ Adam Conrad ]

* [Config] Enable building the sata-modules udeb on ppc64el.
- LP: #1323980

[ Anton Blanchard ]

* SAUCE: (no-up) powerpc: 64bit sendfile is capped at 2GB
- LP: #1328230

[ Ben Widawsky ]

* SAUCE: i915_bdw: drm/i915: Fix PSR programming
- LP: #1321729
* SAUCE: i915_bdw: drm/i915: Correct PPGTT total size
- LP: #1321729

[ Chris Wilson ]

* SAUCE: i915_bdw: drm/i915: Broadwell expands ACTHD to 64bit
- LP: #1321729

[ Damien Lespiau ]

* SAUCE: i915_bdw: drm/i915/bdw: Implement
Wa4x4STCOptimizationDisable:bdw
- LP: #1321729

[ Dave Chiluk ]

* [Config] Enable CONFIG_IP_VS_IPV6=y
- LP: #1300739

[ Kamal Mostafa ]

* [Config] add debian/gbp.conf
* Release Tracking Bug
- LP: #1328286

[ Mika Kuoppala ]

* SAUCE: i915_bdw: drm/i915: add render state initialization
- LP: #1321729

[ Paulo Zanoni ]

* SAUCE: i915_bdw: drm/i915: fix assert_cursor on BDW
- LP: #1321729

[ Steven Rostedt ]

* SAUCE: i915_bdw: drm/i915: Do not dereference pointers from ring buffer
in evict event
- LP: #1321729

[ Tim Gardner ]

* [Config] CONFIG_POWERNV_CPUFREQ=y for ppc64el
- LP: #1324571
* [Debian] Treat vdso install as an environment variable
* [Config] Treat vdso install as an environment variable
* [Config] CONFIG_MLX4_DEBUG=y
- LP: #1328256
* [Config] CONFIG_I40EVF=m, CONFIG_I40E_DCB=y, CONFIG_I40E_VXLAN=y
- LP: #1328037

[ Timo Aaltonen ]

* SAUCE: i915_bdw: Rebase to drm-intel-next-2014-03-07 + fixes
- LP: #1321729
* SAUCE: i915_bdw: Add BDW specific power well calls
- LP: #1317865

[ Upstream Kernel Changes ]

* drm: expose subpixel order name routine v3
* drm: dp helper: Add DP test sink CRC definition.
* drm: export cmdline and preferred mode functions from fb helper
* hugetlb: ensure hugepage access is denied if hugepages are not
supported
- LP: #1328251
* powerpc/powernv: Move SG list structure to header file
- LP: #1326015
* powerpc/powernv: Read OPAL error log and export it through sysfs
- LP: #1326015
* powerpc/powernv Platform dump interface
- LP: #1326015
* pci_regs.h: Add PCI bus link speed and width defines
- LP: #1328037
* net_tstamp: Add SIOCGHWTSTAMP ioctl to match SIOCSHWTSTAMP
- LP: #1328037
* PCI/MSI: Add pci_enable_msi_range() and pci_enable_msix_range()
- LP: #1328037
* net: Change skb_get_rxhash to skb_get_hash
- LP: #1328037
* net: Add utility functions to clear rxhash
- LP: #1328037
* net: Add function to set the rxhash
- LP: #1328037
* i40e: set pf_id based on device and function numbers
- LP: #1328037
* i40e: register file updates
- LP: #1328037
* i40e: clear AQ head and tail registers
- LP: #1328037
* i40e: simplify aq head-tail-len setups
- LP: #1328037
* i40e: firmware version fields offsets update
- LP: #1328037
* i40e: allow one more vector for VFs
- LP: #1328037
* i40e: select reset counters correctly
- LP: #1328037
* i40e: retry call on timeout
- LP: #...

This bug was fixed in the package linux - 3.13.0-30.54
---------------
linux (3.13.0-30.54) trusty; urgency=low
 [ Adam Conrad ]
 * [Config] Enable building the sata-modules udeb on ppc64el.
 - LP: #1323980
 [ Anton Blanchard ]
 * SAUCE: (no-up) powerpc: 64bit sendfile is capped at 2GB
 - LP: #1328230
 [ Ben Widawsky ]
 * SAUCE: i915_bdw: drm/i915: Fix PSR programming
 - LP: #1321729
 * SAUCE: i915_bdw: drm/i915: Correct PPGTT total size
 - LP: #1321729
 [ Chris Wilson ]
 * SAUCE: i915_bdw: drm/i915: Broadwell expands ACTHD to 64bit
 - LP: #1321729
 [ Damien Lespiau ]
 * SAUCE: i915_bdw: drm/i915/bdw: Implement
 Wa4x4STCOptimizationDisable:bdw
 - LP: #1321729
 [ Dave Chiluk ]
 * [Config] Enable CONFIG_IP_VS_IPV6=y
 - LP: #1300739
 [ Kamal Mostafa ]
 * [Config] add debian/gbp.conf
 * Release Tracking Bug
 - LP: #1328286
 [ Mika Kuoppala ]
 * SAUCE: i915_bdw: drm/i915: add render state initialization
 - LP: #1321729
 [ Paulo Zanoni ]
 * SAUCE: i915_bdw: drm/i915: fix assert_cursor on BDW
 - LP: #1321729
 [ Steven Rostedt ]
 * SAUCE: i915_bdw: drm/i915: Do not dereference pointers from ring buffer
 in evict event
 - LP: #1321729
 [ Tim Gardner ]
 * [Config] CONFIG_POWERNV_CPUFREQ=y for ppc64el
 - LP: #1324571
 * [Debian] Treat vdso install as an environment variable
 * [Config] Treat vdso install as an environment variable
 * [Config] CONFIG_MLX4_DEBUG=y
 - LP: #1328256
 * [Config] CONFIG_I40EVF=m, CONFIG_I40E_DCB=y, CONFIG_I40E_VXLAN=y
 - LP: #1328037
 [ Timo Aaltonen ]
 * SAUCE: i915_bdw: Rebase to drm-intel-next-2014年03月07日 + fixes
 - LP: #1321729
 * SAUCE: i915_bdw: Add BDW specific power well calls
 - LP: #1317865
 [ Upstream Kernel Changes ]
 * drm: expose subpixel order name routine v3
 * drm: dp helper: Add DP test sink CRC definition.
 * drm: export cmdline and preferred mode functions from fb helper
 * hugetlb: ensure hugepage access is denied if hugepages are not
 supported
 - LP: #1328251
 * powerpc/powernv: Move SG list structure to header file
 - LP: #1326015
 * powerpc/powernv: Read OPAL error log and export it through sysfs
 - LP: #1326015
 * powerpc/powernv Platform dump interface
 - LP: #1326015
 * pci_regs.h: Add PCI bus link speed and width defines
 - LP: #1328037
 * net_tstamp: Add SIOCGHWTSTAMP ioctl to match SIOCSHWTSTAMP
 - LP: #1328037
 * PCI/MSI: Add pci_enable_msi_range() and pci_enable_msix_range()
 - LP: #1328037
 * net: Change skb_get_rxhash to skb_get_hash
 - LP: #1328037
 * net: Add utility functions to clear rxhash
 - LP: #1328037
 * net: Add function to set the rxhash
 - LP: #1328037
 * i40e: set pf_id based on device and function numbers
 - LP: #1328037
 * i40e: register file updates
 - LP: #1328037
 * i40e: clear AQ head and tail registers
 - LP: #1328037
 * i40e: simplify aq head-tail-len setups
 - LP: #1328037
 * i40e: firmware version fields offsets update
 - LP: #1328037
 * i40e: allow one more vector for VFs
 - LP: #1328037
 * i40e: select reset counters correctly
 - LP: #1328037
 * i40e: retry call on timeout
 - LP: #1328037
 * i40e: properly add VF MAC addresses
 - LP: #1328037
 * i40e: fix debugging messages
 - LP: #1328037
 * i40e: default debug mask setting
 - LP: #1328037
 * i40e: add interrupt test
 - LP: #1328037
 * i40e: add support for triggering EMPR
 - LP: #1328037
 * i40e: restrict diag test length
 - LP: #1328037
 * i40e: sync header files with hardware
 - LP: #1328037
 * i40e: separate TSYNVALID and TSYNINDX fields in Rx descriptor
 - LP: #1328037
 * i40e: check multi-bit state correctly
 - LP: #1328037
 * i40e: get media type during link info
 - LP: #1328037
 * i40e: Add flag for L2 VEB filtering
 - LP: #1328037
 * i40e: enable early hardware support
 - LP: #1328037
 * i40e: whitespace
 - LP: #1328037
 * i40e: Bump version
 - LP: #1328037
 * i40e: refactor reset code
 - LP: #1328037
 * i40e: Enable all PCTYPEs except FCOE for RSS.
 - LP: #1328037
 * i40e: only set up the rings to be used
 - LP: #1328037
 * i40e: clear test state bit after all ethtool tests
 - LP: #1328037
 * i40e: refactor ethtool tests
 - LP: #1328037
 * i40e: add num_VFs message
 - LP: #1328037
 * i40e: Add a new variable to track number of pf instances
 - LP: #1328037
 * i40e: restrict diag test messages
 - LP: #1328037
 * i40e: loopback info and set loopback fix
 - LP: #1328037
 * i40e: complain about out-of-range descriptor request
 - LP: #1328037
 * i40e: remove and fix confusing define name
 - LP: #1328037
 * i40e: Bump version number
 - LP: #1328037
 * i40e: fix up some of the ethtool connection reporting
 - LP: #1328037
 * i40e: fix pf reset after offline test
 - LP: #1328037
 * i40e: Tell the stack about our actual number of queues
 - LP: #1328037
 * i40e: init flow control settings to disabled
 - LP: #1328037
 * i40e: trivial fixes
 - LP: #1328037
 * i40e: use same number of queues as CPUs
 - LP: #1328037
 * i40e: reinit flow for the main VSI
 - LP: #1328037
 * i40e: function to reconfigure RSS queues and rebuild
 - LP: #1328037
 * i40e: Add basic support for get/set channels for RSS
 - LP: #1328037
 * i40e: rtnl_lock in reset path fixes
 - LP: #1328037
 * i40e: support for suspend and resume
 - LP: #1328037
 * i40e: Remove FCoE in i40e_virtchnl_pf.c code
 - LP: #1328037
 * i40e: Fix dump output from debugfs calls
 - LP: #1328037
 * i40e: prevent null pointer exception in dump descriptor
 - LP: #1328037
 * i40e: simplify error messages for dump descriptor
 - LP: #1328037
 * i40e: fix up scanf decoders
 - LP: #1328037
 * i40e: more print_hex_dump use
 - LP: #1328037
 * i40e: Fix wrong mask bits being used in misc interrupt
 - LP: #1328037
 * i40e: Bump version number
 - LP: #1328037
 * i40e: Fix off by one in i40e_dbg_command_write
 - LP: #1328037
 * i40e: make functions static and remove dead code
 - LP: #1328037
 * i40evf: main driver core
 - LP: #1328037
 * i40evf: transmit and receive functionality
 - LP: #1328037
 * i40evf: core ethtool functionality
 - LP: #1328037
 * i40evf: virtual channel interface
 - LP: #1328037
 * i40evf: driver core headers
 - LP: #1328037
 * i40evf: init code and hardware support
 - LP: #1328037
 * i40evf: add driver to kernel build system
 - LP: #1328037
 * i40evf: A0 silicon specific
 - LP: #1328037
 * i40e: using for_each_set_bit to simplify the code
 - LP: #1328037
 * i40e: Suppress HMC error to Interrupt message level
 - LP: #1328037
 * i40e: Populate and check pci bus speed and width
 - LP: #1328037
 * i40e: add wake-on-lan support
 - LP: #1328037
 * i40e: fix curly brace use and return type
 - LP: #1328037
 * i40e: Implementation of VXLAN ndo's
 - LP: #1328037
 * i40e: Rx checksum offload for VXLAN
 - LP: #1328037
 * i40e: move i40e_reset_vf
 - LP: #1328037
 * i40e: refactor VF reset flow
 - LP: #1328037
 * i40e: remove redundant code
 - LP: #1328037
 * i40e: remove chatty log messages
 - LP: #1328037
 * i40e: fix error return
 - LP: #1328037
 * i40e: be more informative
 - LP: #1328037
 * i40e: make a define from a large constant
 - LP: #1328037
 * i40e: update led set args
 - LP: #1328037
 * i40e: report VF MAC addresses correctly
 - LP: #1328037
 * i40e: Dump the whole NVM, not half
 - LP: #1328037
 * i40e: fix mac address checking
 - LP: #1328037
 * i40e: Change the ethtool NVM read method to use AQ
 - LP: #1328037
 * i40e: fix constant cast issues
 - LP: #1328037
 * i40e: guard against vf message races
 - LP: #1328037
 * i40e: add header file flag _I40E_TXRX_H_
 - LP: #1328037
 * i40e: use functions to enable and disable icr 0
 - LP: #1328037
 * i40e: reinit buffer size each time
 - LP: #1328037
 * i40e: fix error handling when alloc of vsi array fails
 - LP: #1328037
 * i40e: keep allocated memory in structs
 - LP: #1328037
 * i40e: catch unset q_vector
 - LP: #1328037
 * i40e: Fix ring allocation
 - LP: #1328037
 * i40e: I40E_FLAG_MQ_ENABLED is not used
 - LP: #1328037
 * i40e: Remove unnecessary prototypes
 - LP: #1328037
 * i40e: remove un-necessary io-write
 - LP: #1328037
 * i40e: Record dma buffer info for dummy packets
 - LP: #1328037
 * i40e: Fix SR-IOV VF port VLAN
 - LP: #1328037
 * i40e: fix whitespace
 - LP: #1328037
 * i40e: avoid unnecessary register read
 - LP: #1328037
 * i40e: Do not enable default port on the VEB
 - LP: #1328037
 * i40e: use struct assign instead of memcpy
 - LP: #1328037
 * i40e: don't allocate zero size
 - LP: #1328037
 * i40e: acknowledge VFLR when disabling SR-IOV
 - LP: #1328037
 * i40e: support VFs on PFs other than 0
 - LP: #1328037
 * i40e: Fix VF driver MAC address configuration
 - LP: #1328037
 * i40e: use correct struct for get and update vsi params
 - LP: #1328037
 * i40e: Hide the Port VLAN VLAN ID
 - LP: #1328037
 * i40e: Admin queue shutdown fixes
 - LP: #1328037
 * i40e: check asq alive before notify
 - LP: #1328037
 * i40e: Do not allow AQ calls from ndo-ops
 - LP: #1328037
 * i40e: Expose AQ debugfs hooks
 - LP: #1328037
 * i40e: Do not enable broadcast promiscuous by default
 - LP: #1328037
 * i40e: Stop accepting any VLAN tag on VLAN 0 filter set
 - LP: #1328037
 * i40e: Allow VF to set already assigned MAC address
 - LP: #1328037
 * i40e: Bump version
 - LP: #1328037
 * i40e: Add code to wait for FW to complete in reset path
 - LP: #1328037
 * i40e: update firmware api to 1.1
 - LP: #1328037
 * i40e: Reduce range of interrupt reg in reg test
 - LP: #1328037
 * i40e: move PF ID init from PF reset to SC init
 - LP: #1328037
 * i40e: check MAC type before any REG access
 - LP: #1328037
 * i40e: rework shadow ram read functions
 - LP: #1328037
 * i40e: whitespace paren and comment tweaks
 - LP: #1328037
 * i40e: Enable/Disable PF switch LB on SR-IOV configure changes
 - LP: #1328037
 * i40e: remove redundant AQ enable
 - LP: #1328037
 * i40e: correctly setup ARQ descriptors
 - LP: #1328037
 * i40e: Re-enable interrupt on ICR0
 - LP: #1328037
 * i40e: use kernel specific defines
 - LP: #1328037
 * i40e: Fix GPL header
 - LP: #1328037
 * i40e: Fix MAC format in Write MAC address AQ cmd
 - LP: #1328037
 * i40e: add a comment on barrier and fix panic on reset
 - LP: #1328037
 * i40e: disable packet split
 - LP: #1328037
 * i40e: Cleanup reconfig rss path
 - LP: #1328037
 * i40e: release NVM resource reservation on startup
 - LP: #1328037
 * i40e: remove interrupt on AQ error
 - LP: #1328037
 * i40e: accept pf to pf adminq messages
 - LP: #1328037
 * i40e: shorten wordy fields
 - LP: #1328037
 * i40e: trivial: formatting and checkpatch fixes
 - LP: #1328037
 * i40e: fix spelling errors
 - LP: #1328037
 * i40e: Add a dummy packet template
 - LP: #1328037
 * i40e: Turn flow director off in MFP mode
 - LP: #1328037
 * i40e: use assignment instead of memcpy
 - LP: #1328037
 * i40e: drop unused macros
 - LP: #1328037
 * i40e: Update the Current NVM version Low value
 - LP: #1328037
 * i40e: Bump version
 - LP: #1328037
 * i40e: fix long lines
 - LP: #1328037
 * i40e: Cleanup Doxygen warnings
 - LP: #1328037
 * i40e: Setting queue count to 1 using ethtool is valid
 - LP: #1328037
 * i40e: do not bail when disabling if Tx queue disable fails
 - LP: #1328037
 * i40e: allow VF to remove any MAC filter
 - LP: #1328037
 * i40e: check for possible incorrect ipv6 checksum
 - LP: #1328037
 * i40e: adjust ITR max and min values
 - LP: #1328037
 * i40e: clear qtx_head before enabling Tx queue
 - LP: #1328037
 * i40e: call clear_pxe after adminq is initialized
 - LP: #1328037
 * i40e: enable PTP
 - LP: #1328037
 * i40e: fix log message wording
 - LP: #1328037
 * i40e: Bump version
 - LP: #1328037
 * i40evf: fix s390 build failure due to implicit prefetch.h
 - LP: #1328037
 * i40e: remove extra register write
 - LP: #1328037
 * i40e: associate VMDq queue with VM type
 - LP: #1328037
 * i40e: make message meaningful
 - LP: #1328037
 * i40e: whitespace fixes
 - LP: #1328037
 * i40e: trivial cleanup
 - LP: #1328037
 * i40e: Bump version number
 - LP: #1328037
 * i40e: Warn admin to reload VF driver on port VLAN configuration
 - LP: #1328037
 * i40e: Retain MAC filters on port VLAN deletion
 - LP: #1328037
 * i40e: Remove autogenerated Module.symvers file.
 - LP: #1328037
 * i40e: check desc pointer before printing
 - LP: #1328037
 * i40e: updates to AdminQ interface
 - LP: #1328037
 * i40e: fix compile warning on checksum_local
 - LP: #1328037
 * i40e: Change firmware workaround
 - LP: #1328037
 * i40e: whitespace fixes
 - LP: #1328037
 * i40e: rename defines
 - LP: #1328037
 * i40e: refactor flow director
 - LP: #1328037
 * i40e: implement DCB support infastructure
 - LP: #1328037
 * i40e: add DCB and DCBNL support
 - LP: #1328037
 * i40e: add DCB option to Kconfig
 - LP: #1328037
 * i40e: Fix device ID define names to align to standard
 - LP: #1328037
 * i40e: Add missing braces to i40e_dcb_need_reconfig()
 - LP: #1328037
 * i40e: spelling error
 - LP: #1328037
 * i40e: bump driver version
 - LP: #1328037
 * i40evf: trivial fixes
 - LP: #1328037
 * i40evf: clean up memsets
 - LP: #1328037
 * i40e: Setting i40e_down bit for tx_timeout
 - LP: #1328037
 * i40e: remove dead code
 - LP: #1328037
 * i40e: set VF state to active when reset is complete
 - LP: #1328037
 * i40e: reset VFs after PF reset
 - LP: #1328037
 * i40e: enable extant VFs
 - LP: #1328037
 * i40e: don't handle VF reset on unload
 - LP: #1328037
 * i40evf: clean up adapter struct
 - LP: #1328037
 * i40evf: fix bogus comment
 - LP: #1328037
 * i40evf: don't store unnecessary array of strings
 - LP: #1328037
 * i40evf: change type of flags variable
 - LP: #1328037
 * i40evf: refactor reset handling
 - LP: #1328037
 * net: i40evf: Remove duplicate include
 - LP: #1328037
 * i40e: Use pci_enable_msix_range() instead of pci_enable_msix()
 - LP: #1328037
 * i40evf: request reset on tx hang
 - LP: #1328037
 * i40evf: remove VLAN filters on close
 - LP: #1328037
 * i40evf: fix multiple crashes on remove
 - LP: #1328037
 * i40evf: get rid of pci_using_dac
 - LP: #1328037
 * i40evf: fix up strings in init task
 - LP: #1328037
 * i40evf: remove bogus comment
 - LP: #1328037
 * i40evf: don't guess device name
 - LP: #1328037
 * i40evf: store ring size in ring structs
 - LP: #1328037
 * i40evf: update version and copyright date
 - LP: #1328037
 * i40evf: remove errant space
 - LP: #1328037
 * i40e: remove unnecessary delay
 - LP: #1328037
 * i40e: tighten up ring enable/disable flow
 - LP: #1328037
 * i40e: Change MSIX to MSI-X
 - LP: #1328037
 * i40e and i40evf: Bump driver versions
 - LP: #1328037
 * i40evf: Enable the ndo_set_features netdev op
 - LP: #1328037
 * i40e: Flow Director sideband accounting
 - LP: #1328037
 * i40e: Prevent overflow due to kzalloc
 - LP: #1328037
 * i40e/i40evf: i40e implementation for skb_set_hash
 - LP: #1328037
 * i40e: clean up comment style
 - LP: #1328037
 * i40e: Remove a FW workaround for Number of MSIX vectors
 - LP: #1328037
 * i40e: count timeout events
 - LP: #1328037
 * i40e: Remove a redundant filter addition
 - LP: #1328037
 * i40e: Fix static checker warning
 - LP: #1328037
 * i40e: fix nvm version and remove firmware report
 - LP: #1328037
 * i40e/i40evf: carefully fill tx ring
 - LP: #1328037
 * i40e/i40evf: Bump pf&vf build versions
 - LP: #1328037
 * i40e: delete netdev after deleting napi and vectors
 - LP: #1328037
 * i40e: Fix a bug in the update logic for FDIR SB filter.
 - LP: #1328037
 * i40e/i40evf: Some flow director HW definition fixes
 - LP: #1328037
 * i40e: make string references to q be queue
 - LP: #1328037
 * i40e: cleanup strings
 - LP: #1328037
 * i40e: simplified init string
 - LP: #1328037
 * i40e: Fix function comments
 - LP: #1328037
 * i40e: Define a new state variable to keep track of feature auto disable
 - LP: #1328037
 * i40e: Add code to handle FD table full condition
 - LP: #1328037
 * i40e: Bug fix for FDIR replay logic
 - LP: #1328037
 * i40e: Let MDD events be handled by MDD handler
 - LP: #1328037
 * i40e/i40evf: Use correct number of VF vectors
 - LP: #1328037
 * i40e/i40evf: Use dma_set_mask_and_coherent
 - LP: #1328037
 * net: Replace u64_stats_fetch_begin_bh to u64_stats_fetch_begin_irq
 - LP: #1328037
 * i40e: Don't receive packets when the napi budget == 0
 - LP: #1328037
 * i40evf: Rename i40e_ptype_lookup i40evf_ptype_lookup
 - LP: #1328037
 * net/i40e: Avoid double setting of NETIF_F_SG for the HW encapsulation
 feature mask
 - LP: #1328037
 * i40e: support VF link state ndo
 - LP: #1328037
 * i40evf: correctly program RSS HLUT table
 - LP: #1328037
 * i40evf: use min_t
 - LP: #1328037
 * i40e: Patch to enable Ethtool/netdev feature flag for NTUPLE control
 - LP: #1328037
 * i40e: Refactor and cleanup i40e_open(), adding i40e_vsi_open()
 - LP: #1328037
 * i40e/i40evf: enable hardware feature head write back
 - LP: #1328037
 * i40e/i40evf: reduce context descriptors
 - LP: #1328037
 * i40e: potential array underflow in i40e_vc_process_vf_msg()
 - LP: #1328037
 * i40e/i40evf: Bump build versions
 - LP: #1328037
 * i40e/i40evf: Add EEE LPI stats
 - LP: #1328037
 * i40e: Fix a message string
 - LP: #1328037
 * i40evf: don't shut down admin queue on error
 - LP: #1328037
 * i40evf: clean up init error messages
 - LP: #1328037
 * i40e: Delete ATR filter on RST
 - LP: #1328037
 * i40evf: fix oops in watchdog handler
 - LP: #1328037
 * i40e: Make the alloc and free queue vector calls orthogonal
 - LP: #1328037
 * i40e: eeprom integrity check on load and empr
 - LP: #1328037
 * i40e: Cleanup in FDIR SB ethtool code
 - LP: #1328037
 * i40e: Add functionality for FD SB to drop packets
 - LP: #1328037
 * i40evf: remove double space after return
 - LP: #1328037
 * i40e: check for netdev before debugfs use
 - LP: #1328037
 * i40e/i40evf: Add an FD message level
 - LP: #1328037
 * i40e: Use DEBUG_FD message level for an FD message
 - LP: #1328037
 * i40e: fix function kernel doc description
 - LP: #1328037
 * i40e/i40evf: fix error checking path
 - LP: #1328037
 * i40e/i40evf: Remove addressof casts to same type
 - LP: #1328037
 * i40e: Remove casts of pointer to same type
 - LP: #1328037
 * i40evf: remove open-coded skb_cow_head
 - LP: #1328037
 * i40evf: program RSS LUT correctly
 - LP: #1328037
 * i40e: remove open-coded skb_cow_head
 - LP: #1328037
 * i40e: fix TCP flag replication for hardware offload
 - LP: #1328037
 * e1000e/igb/ixgbe/i40e: fix message terminations
 - LP: #1328037
 * i40e: fix Timesync Tx interrupt handler code
 - LP: #1328037
 * mm: use paravirt friendly ops for NUMA hinting ptes
 - LP: #1313450
 [ Ville Syrjälä ]
 * SAUCE: i915_bdw: drm/i915: Fix scanline counter fixup on BDW
 - LP: #1321729
 -- Kamal Mostafa <kamal@canonical.com> 2014年6月09日 15:09:43 -0700

Changed in linux (Ubuntu Trusty):

status: Fix Committed → Fix Released

See full activity log