- User Since
- Nov 13 2017, 4:43 AM (171 w, 4 d)
Dec 16 2020
Fix when the number of swap devise is 1.
Dec 7 2020
Fix off-by-one error over 60% and stop extra +1 for percentage.
Add a , to .Xr ls
Update getbsize man page.
Oct 19 2020
Remove extra spaces around a function argument.
Use bitmask to indicate which protocol versions to try.
Also remove duplicate check.
Oct 18 2020
I make multiple read-only connection for installworld/installkernel from different machines.
Otherwise, I use ggate to export some ufs file systems.
I could use nfs for these as well.
Address some of problems pointed by code review.
Sep 28 2020
The slowness issue existed back in a decade ago around 6.1-RELEASE and some years/releases thereafter.
Sep 24 2020
Sep 20 2020
Sep 9 2020
Thank you for checking this again.
It was also my bad that the problem wasn't described well enough.
Sep 7 2020
Stop using PR* formatter.
Tested on i386 and amd64.
Sep 6 2020
ggate defines its offset in unsigned int 64 as below.
So, printing negative numbers from printf indicates using a wrong formatter.
Aug 29 2020
Thank you for your support, Mark.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=132845 is the corresponding bugzilla entry.
Aug 28 2020
Verify paths are the same between handshakes for a single session.
Aug 27 2020
Moved close() to connection_remove() and aligned with send/recv fds.
Aug 26 2020
First of all, thank you for your quick response.
Aug 25 2020
Aug 24 2020
Aug 2 2020
Jun 12 2020
Added static to implementatoin; follow style of other existing static functions.
Drop unnecessary NULL check. This is a field on a struct and won't be NULL.
Jun 11 2020
May 29 2020
I think I addressed all of feed backs so far.
I'm wondering if someone can take a look.
May 14 2020
Address other review comments.
Mar 20 2020
Fixed style, moved 0 assignment to *before and *after when SWWAPBLK_NONE,
and adjusted if/else statement for *after case.
Mar 15 2020
Split backward and forward search into separate functions for ease of reading.
Mar 5 2020
Thank you for quick response, Mark.
Feb 25 2020
Jan 20 2020
Updated function name in comment.
Changed its prefix to swdevt.
Jan 18 2020
Jan 17 2020
Nov 22 2019
Check NULL pointers explicitly.
Nov 21 2019
Nov 20 2019
Update and merge D22437.
Nov 18 2019
It looks we better drop swp_pager_meta_ctl() changes from this, D22324, and let D22409 take over the lookup implementation while dropping SWM_POP for simpler diff.
D22409 change is to optimize swap_pager_haspage() with additional searches after successful pindex lookup.
Nov 16 2019
Nov 15 2019
D22280 has been committed.
Nov 11 2019
Nov 9 2019
Correct swblk index and exclude swp_pager_swblk_free_empty() changes.
I don't think this code duplication in swap_pager_copy() is a good idea.
I updated D22256 to avoid the double lookup cost from the first revision.
I'm still testing the 2nd version though.
Created swp_pager_meta_find_least private API to find and delete in a single
operation. swap_pager_find_least calls it for look up.
Nov 6 2019
Nov 5 2019
Suspend for now.
Oct 25 2019
I've been using the comparison based optimization for 2 weeks on current
following daily, buildworld/buildkernel daily and haven't seen an issue
with this approach.
Oct 21 2019
I see. I just added zfs/zvol swap space temporary from command line within the last a couple of weeks for other reasons. I see few of too early OOM on few of machines although it is very rare and spontaneous such that it hasn't been easy to find a trigger.
Oct 20 2019
swap_pager_full != 0 was indeed too aggressive. I'm wondering if we can feed in swap space usage such that we can delay OOM electing at low usage and push for OOM at high usage.
Oct 19 2019
Oct 17 2019
Oct 14 2019
That's another idea.
I will check the code and test if that is also an option.
Thanks for follow up, markj.
Oct 13 2019
Oct 12 2019
Use "== SWPG_LOW" comparison for almost full warning.
Drop swp_gauge > SWPG_LOW comparison for high watermark check.
Oct 3 2019
The only reason I created a patch was to proceed removal of sys/dir.h in the base system.
Given the upstream source hasn't been updated for more than a couple of decade.
It looks the code still assume 20 year old standard.
Sep 29 2019
But sysctl describes totally different:
Sep 18 2019
destLen -> destlen.
Sep 7 2019
Sep 6 2019
Sep 5 2019
Aug 27 2019
Remove uninteded debug print.
Aug 26 2019
Also print length with PRIu32.