Bug 1131498
| Summary: | E.2.1. /proc/buddyinfo contains and error | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 6 | Reporter: | Elie De Brauwer <eliedebrauwer> |
| Component: | doc-Deployment_Guide | Assignee: | Jaromir Hradilek <jhradile> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | ecs-bugs |
| Severity: | low | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 6.6 | Keywords: | Documentation, EasyFix |
| Target Milestone: | rc | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2014-10-17 12:29:31 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Elie De Brauwer
2014-08-19 12:29:13 UTC
Thank you very much for taking the time to report this issue. Hi Elie,
I have corrected the formula and edited the description a little to make it easier to read and understand. As a consequence, the updated section now reads:
E.2.1. /proc/buddyinfo
This file is used primarily for diagnosing memory fragmentation
issues. The following is an example of a typical /proc/buddyinfo
file:
Node 0, zone DMA 90 6 2 1 1 ...
Node 0, zone Normal 1650 310 5 0 0 ...
Node 0, zone HighMem 2 0 0 1 1 ...
Using the buddy algorithm, each column represents the number of
pages of a certain order (a certain size) that are available at any
given time. For example, for zone direct memory access (DMA), there
are 90 of (2^0)*PAGE_SIZE chunks of memory. Similarly, there are 6
of (2^1)*PAGE_SIZE chunks, and 2 of (2^2)*PAGE_SIZE chunks of memory
available.
The DMA row references the first 16 MB of memory on the system, the
HighMem row references all memory greater than 4 GB on the system,
and the Normal row references the memory in between.
When you have a moment, could you please take a look at this updated section and let me know if this resolves your issue or if there is anything else you would like me to change? I can see that the description of available rows needs to be updated too (on my 64-bit system, I also have DMA32 and don't have HighMem at all), but I will open a separate bug for that.
Thank you,
Jaromir Hradilek
> I can see that the description of available rows needs to be updated too > (on my 64-bit system, I also have DMA32 and don't have HighMem at all), > but I will open a separate bug for that. Created bug 1131631. Jaromir, Thanks for looking so quickly at this. The only thing I would add is an indication of the units, since you are essentialy using the order to calculate the size of these chunks. The reason why I bumped in this page was to figure out that 'order 11' implied chunks of 8 megabyte large. So I would change (imagine you replace the formula by the actual size) ... are 90 of 4096 chunks of memory. And replace it by something like ... are 90 of (2^0)*PAGE_SIZE _bytes large_ chunks of memory. What do you think ? my 2 cents E. That is a very good point. How about something like this: Using the buddy algorithm, each column represents the number of pages of a certain order (a certain size) that are available at any given time. In the example above, for zone DMA, there are 90 of (2^0)*PAGE_SIZE bytes large chunks of memory. Similarly, there are 6 of (2^1)*PAGE_SIZE chunks, and 2 of (2^2)*PAGE_SIZE chunks of memory available. I do not want to repeat the "bytes large chunks of memory" part every time, because it makes the sentence a little too complicated, I think. Thank you, Jaromir Fine for me ! Hi Elie, I have published a new version of the book on the Customer Portal so that you can find the corrected section here: https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6-Beta/html/Deployment_Guide/s1-proc-topfiles.html#s2-proc-buddyinfo Thank you, Jaromir |