| 2017-11-08 02:42:57 |
Alice Liu |
bug |
|
|
added bug |
| 2017-11-08 02:43:06 |
Alice Liu |
information type |
Proprietary |
Private |
|
| 2018-02-24 00:26:58 |
Leann Ogasawara |
tags |
|
kernel |
|
| 2018-02-24 00:27:01 |
Leann Ogasawara |
intel: status |
New |
Incomplete |
|
| 2018-05-01 21:45:13 |
quanxian |
description |
Description
Huge page helps to reduce TLB miss rate, but it has higher cache footprint, sometimes this may cause some issue. For example, when clearing huge page on x86_64 platform, the cache footprint is 2M. But on a Xeon E5 v3 2699 CPU, there are 18 cores, 36 threads, and only 45M LLC (last level cache). That is, in average, there are 2.5M LLC for each core and 1.25M LLC for each thread. If the cache pressure is heavy when clearing the huge page, and we clear the huge page from the begin to the end, it is possible that the begin of huge page is evicted from the cache after we finishing clearing the end of the huge page. And it is possible for the application to access the begin of the huge page after clearing the huge page.
To help the above situation, when we clear or copy a huge page, the order to clear or copy the sub-pages can be changed to clear or copy the sub-page to access last.
Target Kernel: 4.15
Target Release: 18.04 |
Description
Huge page helps to reduce TLB miss rate, but it has higher cache footprint, sometimes this may cause some issue. For example, when clearing huge page on x86_64 platform, the cache footprint is 2M. But on a Xeon E5 v3 2699 CPU, there are 18 cores, 36 threads, and only 45M LLC (last level cache). That is, in average, there are 2.5M LLC for each core and 1.25M LLC for each thread. If the cache pressure is heavy when clearing the huge page, and we clear the huge page from the begin to the end, it is possible that the begin of huge page is evicted from the cache after we finishing clearing the end of the huge page. And it is possible for the application to access the begin of the huge page after clearing the huge page.
To help the above situation, when we clear or copy a huge page, the order to clear or copy the sub-pages can be changed to clear or copy the sub-page to access last.
Target Kernel: TBD
Target Release: 18.10 |
|
| 2018-05-01 21:45:26 |
quanxian |
tags |
kernel |
intel-kernel-18.10 kernel |
|
| 2018-07-25 23:06:24 |
Joseph Salisbury |
intel: importance |
Undecided |
Medium |
|
| 2018-07-25 23:06:27 |
Joseph Salisbury |
intel: status |
Incomplete |
Triaged |
|
| 2018-08-07 01:24:42 |
quanxian |
description |
Description
Huge page helps to reduce TLB miss rate, but it has higher cache footprint, sometimes this may cause some issue. For example, when clearing huge page on x86_64 platform, the cache footprint is 2M. But on a Xeon E5 v3 2699 CPU, there are 18 cores, 36 threads, and only 45M LLC (last level cache). That is, in average, there are 2.5M LLC for each core and 1.25M LLC for each thread. If the cache pressure is heavy when clearing the huge page, and we clear the huge page from the begin to the end, it is possible that the begin of huge page is evicted from the cache after we finishing clearing the end of the huge page. And it is possible for the application to access the begin of the huge page after clearing the huge page.
To help the above situation, when we clear or copy a huge page, the order to clear or copy the sub-pages can be changed to clear or copy the sub-page to access last.
Target Kernel: TBD
Target Release: 18.10 |
Description
Huge page helps to reduce TLB miss rate, but it has higher cache footprint, sometimes this may cause some issue. For example, when clearing huge page on x86_64 platform, the cache footprint is 2M. But on a Xeon E5 v3 2699 CPU, there are 18 cores, 36 threads, and only 45M LLC (last level cache). That is, in average, there are 2.5M LLC for each core and 1.25M LLC for each thread. If the cache pressure is heavy when clearing the huge page, and we clear the huge page from the begin to the end, it is possible that the begin of huge page is evicted from the cache after we finishing clearing the end of the huge page. And it is possible for the application to access the begin of the huge page after clearing the huge page.
To help the above situation, when we clear or copy a huge page, the order to clear or copy the sub-pages can be changed to clear or copy the sub-page to access last.
Target Kernel: 4,19
Target Release: 18.10 |
|
| 2018-08-31 16:15:14 |
Joseph Salisbury |
bug task added |
|
linux (Ubuntu) |
|
| 2018-08-31 16:15:21 |
Joseph Salisbury |
linux (Ubuntu): status |
New |
Triaged |
|
| 2018-08-31 16:15:24 |
Joseph Salisbury |
linux (Ubuntu): importance |
Undecided |
Medium |
|
| 2018-09-07 19:43:04 |
Joseph Salisbury |
linux (Ubuntu): status |
Triaged |
In Progress |
|
| 2018-09-07 19:43:07 |
Joseph Salisbury |
linux (Ubuntu): assignee |
|
Joseph Salisbury (jsalisbury) |
|
| 2018-09-10 17:04:45 |
Joseph Salisbury |
information type |
Private |
Public |
|
| 2018-09-11 12:09:30 |
Thadeu Lima de Souza Cascardo |
linux (Ubuntu): status |
In Progress |
Fix Committed |
|
| 2018-10-12 13:48:54 |
Launchpad Janitor |
linux (Ubuntu): status |
Fix Committed |
Fix Released |
|
| 2018-10-12 13:48:54 |
Launchpad Janitor |
cve linked |
|
2018-14633 |
|
| 2018-10-12 13:48:54 |
Launchpad Janitor |
cve linked |
|
2018-5391 |
|
| 2018-10-26 07:31:28 |
quanxian |
intel: status |
Triaged |
Fix Released |
|
| 2019-02-14 12:11:51 |
Brad Figg |
tags |
intel-kernel-18.10 kernel |
intel-kernel-18.10 kernel verification-needed-bionic |
|
| 2019-02-14 15:52:59 |
Andy Whitcroft |
tags |
intel-kernel-18.10 kernel verification-needed-bionic |
intel-kernel-18.10 kernel kernel-fixup-verification-needed-bionic |
|
| 2019-02-14 16:18:28 |
Andy Whitcroft |
tags |
intel-kernel-18.10 kernel kernel-fixup-verification-needed-bionic |
intel-kernel-18.10 kernel kernel-fixup-verification-needed-bionic verification-done-bionic |
|
| 2019-07-24 21:07:58 |
Brad Figg |
tags |
intel-kernel-18.10 kernel kernel-fixup-verification-needed-bionic verification-done-bionic |
cscc intel-kernel-18.10 kernel kernel-fixup-verification-needed-bionic verification-done-bionic |
|
