Do the DirectMap?? entries in /proc/meminfo account for huge page compaction?
up vote
0
down vote
favorite
I've been tracking down some Transparent Huge Pages performance issues--nothing unusual there--in x86_64 CentOS 6.7, and I've noticed something odd. The values in /proc/meminfo for DirectMap?? are supposed to keep track of the number of pages mapped for each of the three page sizes. But once DirectMap1G has dropped to 0, it appears that DirectMap2M will only decrease. No amount of huge page compaction ever seems to increase the value back up.
When I look at the relevant kernel source in arch/x86/mm/pageattr.c, it's pretty clear that the DirectMap?? counters get set at initialization via update_page_count() and updated when pages are split via split_page_count(). But there's nothing in the kernel that does the reverse split adjustment to the counters when pages are compacted.
I've looked at kernel source from 2.6 up to 4.18.20, and it seems that this behavior hasn't changed.
So, am I missing something, or is this a bug?
linux-kernel memory-management huge-pages
asked Dec 6 at 1:02
Harold Zable
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
add a comment |
up vote
0
down vote
favorite
I've been tracking down some Transparent Huge Pages performance issues--nothing unusual there--in x86_64 CentOS 6.7, and I've noticed something odd. The values in /proc/meminfo for DirectMap?? are supposed to keep track of the number of pages mapped for each of the three page sizes. But once DirectMap1G has dropped to 0, it appears that DirectMap2M will only decrease. No amount of huge page compaction ever seems to increase the value back up.
When I look at the relevant kernel source in arch/x86/mm/pageattr.c, it's pretty clear that the DirectMap?? counters get set at initialization via update_page_count() and updated when pages are split via split_page_count(). But there's nothing in the kernel that does the reverse split adjustment to the counters when pages are compacted.
I've looked at kernel source from 2.6 up to 4.18.20, and it seems that this behavior hasn't changed.
So, am I missing something, or is this a bug?
linux-kernel memory-management huge-pages
asked Dec 6 at 1:02
Harold Zable
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
add a comment |
up vote
0
down vote
favorite
up vote
0
down vote
favorite
I've been tracking down some Transparent Huge Pages performance issues--nothing unusual there--in x86_64 CentOS 6.7, and I've noticed something odd. The values in /proc/meminfo for DirectMap?? are supposed to keep track of the number of pages mapped for each of the three page sizes. But once DirectMap1G has dropped to 0, it appears that DirectMap2M will only decrease. No amount of huge page compaction ever seems to increase the value back up.
When I look at the relevant kernel source in arch/x86/mm/pageattr.c, it's pretty clear that the DirectMap?? counters get set at initialization via update_page_count() and updated when pages are split via split_page_count(). But there's nothing in the kernel that does the reverse split adjustment to the counters when pages are compacted.
I've looked at kernel source from 2.6 up to 4.18.20, and it seems that this behavior hasn't changed.
So, am I missing something, or is this a bug?
linux-kernel memory-management huge-pages
asked Dec 6 at 1:02
Harold Zable
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
I've been tracking down some Transparent Huge Pages performance issues--nothing unusual there--in x86_64 CentOS 6.7, and I've noticed something odd. The values in /proc/meminfo for DirectMap?? are supposed to keep track of the number of pages mapped for each of the three page sizes. But once DirectMap1G has dropped to 0, it appears that DirectMap2M will only decrease. No amount of huge page compaction ever seems to increase the value back up.
When I look at the relevant kernel source in arch/x86/mm/pageattr.c, it's pretty clear that the DirectMap?? counters get set at initialization via update_page_count() and updated when pages are split via split_page_count(). But there's nothing in the kernel that does the reverse split adjustment to the counters when pages are compacted.
I've looked at kernel source from 2.6 up to 4.18.20, and it seems that this behavior hasn't changed.
So, am I missing something, or is this a bug?
linux-kernel memory-management huge-pages
linux-kernel memory-management huge-pages
asked Dec 6 at 1:02
Harold Zable
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
asked Dec 6 at 1:02
Harold Zable
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
asked Dec 6 at 1:02
Harold Zable
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
asked Dec 6 at 1:02
Harold Zable
11
asked Dec 6 at 1:02
Harold Zable
11
11
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
New contributor
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
Harold Zable is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.
add a comment |
add a comment |
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
Harold Zable is a new contributor. Be nice, and check out our Code of Conduct.
Harold Zable is a new contributor. Be nice, and check out our Code of Conduct.
Harold Zable is a new contributor. Be nice, and check out our Code of Conduct.
Harold Zable is a new contributor. Be nice, and check out our Code of Conduct.
Thanks for contributing an answer to Unix & Linux Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Some of your past answers have not been well-received, and you're in danger of being blocked from answering.
Please pay close attention to the following guidance:
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2funix.stackexchange.com%2fquestions%2f486265%2fdo-the-directmap-entries-in-proc-meminfo-account-for-huge-page-compaction%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
