Page MenuHomeFreeBSD
Feed Advanced Search

Tue, Oct 12

ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.
  • systat: Stop displaying total if there is one swap devise.
  • systat: Implemented per-process swap display on -swap.
  • Avoid double counting and address review comments.
  • systat.1 updates
  • D31163 - Add sysctl vm.objects_swap.
  • Use new vm.objects_swap instead of vm.objects.
Tue, Oct 12, 3:54 AM

Jul 24 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.

I started looking into https://reviews.freebsd.org/D31163.
I've been releng/13.0 based system (with cherrypick).
This time, the patch doesn't apply to 13 and I need to prepare main branch based environment.
I will need a bit time to build the latest.

Jul 24 2021, 4:01 AM

May 31 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.

I tested the 2 kernel changes both applied. In short, it helps since the 2nd refresh.

May 31 2021, 3:30 PM
ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

Update systat.1 based on suggestions.

May 31 2021, 2:49 PM

May 18 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.
In D29754#680269, @kib wrote:

Ok. Could you add the following to the kernel, please, and see if it makes any difference?

diff --git a/sys/vm/vm_object.c b/sys/vm/vm_object.c
index dbb09f67b4c1..cf8b3fcb8f9c 100644
--- a/sys/vm/vm_object.c
+++ b/sys/vm/vm_object.c
@@ -2478,7 +2478,7 @@ sysctl_vm_object_list(SYSCTL_HANDLER_ARGS)
 	struct vnode *vp;
 	struct vattr va;
 	vm_object_t obj;
-	vm_page_t m;
+//	vm_page_t m;
 	u_long sp;
 	int count, error;
 
@@ -2524,6 +2524,7 @@ sysctl_vm_object_list(SYSCTL_HANDLER_ARGS)
 		kvo->kvo_memattr = obj->memattr;
 		kvo->kvo_active = 0;
 		kvo->kvo_inactive = 0;
+#if 0
May 18 2021, 3:13 AM

May 15 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.
In D29754#677726, @kib wrote:

I mean, when asking whether my one-line patch helped, did interactivity of the system during vm.objects retrieval improved? Of course it cannot optimize the running time of the sysctl, but it should allow for other threads to run some more while sysctl is gathering data and copying it out.

May 15 2021, 3:54 AM

May 11 2021

ota_j.email.ne.jp accepted D29354: systat: Add SPDX-License-Identifier.
May 11 2021, 9:55 PM
ota_j.email.ne.jp updated the diff for D29354: systat: Add SPDX-License-Identifier.

Fix once more.

May 11 2021, 1:12 AM

May 9 2021

ota_j.email.ne.jp added a reviewer for D29754: Augment systat(1) -swap to display large swap space processes: bcr.
May 9 2021, 3:10 PM
ota_j.email.ne.jp abandoned D29695: systat: Implemented per-process swap display on -swap.

We have a duplicate review for this and are tracking changes in https://reviews.freebsd.org/D29754.
The comments are addressed in https://reviews.freebsd.org/D29754.

May 9 2021, 3:09 PM
ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

Adjust systat.1 from the other review.

May 9 2021, 3:09 PM
ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.

Most likely it was sysctl vm.objects and not per-proc vmmap sysctl that caused the intermediate hang and CPU usage. As I and Mark discussed it earlier, some more fine-grained API to retrieve e.g. just the shadow chain for given object handle probably would help there.

But before I implement this sysctl, can you recheck, please, that the following patch solves the transient hang issue?

 commit fd2da9e1515bb50fb1c8fb747d9987ee9b8a314d
Author: Konstantin Belousov <kib@FreeBSD.org>
Date:   Sat May 8 01:13:29 2021 +0300

    sysctl vm.objects: yield if hog

diff --git a/sys/vm/vm_object.c b/sys/vm/vm_object.c
index 1c4e879d82ea..b1fa3ebdf9ab 100644
--- a/sys/vm/vm_object.c
+++ b/sys/vm/vm_object.c
@@ -2614,6 +2614,7 @@ sysctl_vm_object_list(SYSCTL_HANDLER_ARGS)
 		kvo->kvo_structsize = roundup(kvo->kvo_structsize,
 		    sizeof(uint64_t));
 		error = SYSCTL_OUT(req, kvo, kvo->kvo_structsize);
+		maybe_yield();
 		mtx_lock(&vm_object_list_mtx);
 		if (error)
 			break;
May 9 2021, 2:43 PM

May 7 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.

Thank you for reviews.

May 7 2021, 6:50 AM
ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

Remove extra comments.

May 7 2021, 6:39 AM
ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

another style fix.

May 7 2021, 2:14 AM

May 6 2021

ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

Remove #include <devstat.h>

May 6 2021, 5:21 AM
ota_j.email.ne.jp added inline comments to D29754: Augment systat(1) -swap to display large swap space processes.
May 6 2021, 5:16 AM
ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

Style fixes based on review comments and recstored Kirk's systat.1 updates.

May 6 2021, 5:07 AM

May 5 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.

Thanks for the program. Indeed, I saw lots of duplicates from the java program.

May 5 2021, 4:25 AM
ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.
  • Avoid double counting and address review comments.
May 5 2021, 4:10 AM

May 4 2021

ota_j.email.ne.jp updated the diff for D29754: Augment systat(1) -swap to display large swap space processes.

The algorithm has been adjusted proposed approach as below by Konstantin:

May 4 2021, 6:01 AM
ota_j.email.ne.jp commandeered D29754: Augment systat(1) -swap to display large swap space processes.

Kirk and I have been looking into this together.
Given the history of comments, this is a better one for tracking objective of this change.

May 4 2021, 5:56 AM

Apr 23 2021

ota_j.email.ne.jp added a comment to D29354: systat: Add SPDX-License-Identifier.

I had mixed with another changeset and caused failures.

Apr 23 2021, 4:16 AM
ota_j.email.ne.jp updated the diff for D29354: systat: Add SPDX-License-Identifier.

systat: Update to SPDX BSD-2-Clause - I've written zarc.c and sysput.c and University of California hasn't been involved

Apr 23 2021, 4:13 AM

Apr 15 2021

ota_j.email.ne.jp added a comment to D29754: Augment systat(1) -swap to display large swap space processes.
In D29754#667588, @kib wrote:

I saw this patch on a list, and I decided to avoid it.

It does not do what is advertised, i.e. the numbers displayed has nothing related to the 'process swap usage'. First, the swap usage per process is quite hard to define: due to the shadow chaining and sharing of the shadowing objects between siblings, a lot of swap objects is commonly used but simply not visible to the view from the process map level. Second, (and to be fair, it is mentioned in the man page update of this patch), swap object type itself does not imply that a swap is used, at all. Third, even if the top-level object is default, below it, the objects might have swap types.

Apr 15 2021, 5:38 AM

Apr 10 2021

ota_j.email.ne.jp updated the summary of D29695: systat: Implemented per-process swap display on -swap.
Apr 10 2021, 3:21 PM
ota_j.email.ne.jp requested review of D29695: systat: Implemented per-process swap display on -swap.
Apr 10 2021, 3:18 PM

Mar 20 2021

ota_j.email.ne.jp requested review of D29354: systat: Add SPDX-License-Identifier.
Mar 20 2021, 6:22 PM

Mar 19 2021

ota_j.email.ne.jp added a reviewer for D29337: systat: Handle SIGWINCH to properly window resizing and adjust -swap disk stat based on new size.: mr.
Mar 19 2021, 2:40 AM
ota_j.email.ne.jp requested review of D29337: systat: Handle SIGWINCH to properly window resizing and adjust -swap disk stat based on new size..
Mar 19 2021, 2:38 AM

Mar 17 2021

ota_j.email.ne.jp added a reviewer for D29303: systat: Avoid incorrect reallocation in pigs.c: mr.
Mar 17 2021, 12:44 AM
ota_j.email.ne.jp requested review of D29303: systat: Avoid incorrect reallocation in pigs.c.
Mar 17 2021, 12:42 AM

Dec 16 2020

ota_j.email.ne.jp updated the diff for D27496: Improve systat -swap readability and fix display issues.

Fix when the number of swap devise is 1.

Dec 16 2020, 5:22 AM

Dec 7 2020

ota_j.email.ne.jp updated the diff for D27496: Improve systat -swap readability and fix display issues.

Fix off-by-one error over 60% and stop extra +1 for percentage.

Dec 7 2020, 11:22 PM
ota_j.email.ne.jp updated the diff for D27496: Improve systat -swap readability and fix display issues.

Fix style.

Dec 7 2020, 1:11 AM
ota_j.email.ne.jp updated the diff for D27496: Improve systat -swap readability and fix display issues.

Add a , to .Xr ls

Dec 7 2020, 12:55 AM
ota_j.email.ne.jp updated the diff for D27496: Improve systat -swap readability and fix display issues.

Update getbsize man page.

Dec 7 2020, 12:53 AM
ota_j.email.ne.jp updated the summary of D27496: Improve systat -swap readability and fix display issues.
Dec 7 2020, 12:25 AM
ota_j.email.ne.jp requested review of D27496: Improve systat -swap readability and fix display issues.
Dec 7 2020, 12:24 AM

Oct 19 2020

ota_j.email.ne.jp updated the diff for D26542: [ggated/ggatec] Simplify Gate Handshake..

Remove extra spaces around a function argument.

Oct 19 2020, 4:09 AM
ota_j.email.ne.jp added inline comments to D26542: [ggated/ggatec] Simplify Gate Handshake..
Oct 19 2020, 4:07 AM
ota_j.email.ne.jp updated the diff for D26542: [ggated/ggatec] Simplify Gate Handshake..

Use bitmask to indicate which protocol versions to try.
Also remove duplicate check.

Oct 19 2020, 2:26 AM

Oct 18 2020

ota_j.email.ne.jp planned changes to D26542: [ggated/ggatec] Simplify Gate Handshake..

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.

Oct 18 2020, 4:57 AM
ota_j.email.ne.jp updated the diff for D26542: [ggated/ggatec] Simplify Gate Handshake..

Address some of problems pointed by code review.

Oct 18 2020, 3:17 AM

Sep 28 2020

ota_j.email.ne.jp added a comment to D26542: [ggated/ggatec] Simplify Gate Handshake..

The slowness issue existed back in a decade ago around 6.1-RELEASE and some years/releases thereafter.

Sep 28 2020, 4:48 AM

Sep 24 2020

ota_j.email.ne.jp retitled D26542: [ggated/ggatec] Simplify Gate Handshake. from Simplify Gate Handshake. to [ggated/ggatec] Simplify Gate Handshake..
Sep 24 2020, 3:48 AM
ota_j.email.ne.jp added a reviewer for D26503: [systat] Improve number reading by rounding up to a next unit earlier for memory display.: mr.
Sep 24 2020, 3:43 AM
ota_j.email.ne.jp requested review of D26542: [ggated/ggatec] Simplify Gate Handshake..
Sep 24 2020, 2:07 AM

Sep 20 2020

ota_j.email.ne.jp requested review of D26503: [systat] Improve number reading by rounding up to a next unit earlier for memory display..
Sep 20 2020, 11:57 PM

Sep 9 2020

ota_j.email.ne.jp updated the diff for D21388: Fix ggated/ggatec debug print of offsets..

Thank you for checking this again.
It was also my bad that the problem wasn't described well enough.

Sep 9 2020, 1:55 AM

Sep 7 2020

ota_j.email.ne.jp updated the summary of D21388: Fix ggated/ggatec debug print of offsets..
Sep 7 2020, 2:09 AM
ota_j.email.ne.jp updated the diff for D21388: Fix ggated/ggatec debug print of offsets..

Stop using PR* formatter.
Tested on i386 and amd64.

Sep 7 2020, 2:06 AM

Sep 6 2020

ota_j.email.ne.jp added a comment to D21388: Fix ggated/ggatec debug print of offsets..

ggate defines its offset in unsigned int 64 as below.
So, printing negative numbers from printf indicates using a wrong formatter.

Sep 6 2020, 8:38 PM

Aug 29 2020

ota_j.email.ne.jp added a reviewer for D21388: Fix ggated/ggatec debug print of offsets.: markj.
Aug 29 2020, 2:14 AM
ota_j.email.ne.jp added a comment to D26168: Bug 132845: ggated: Fix double file open and file descriptor leak..

Thank you for your support, Mark.
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=132845 is the corresponding bugzilla entry.

Aug 29 2020, 2:12 AM

Aug 28 2020

ota_j.email.ne.jp updated the diff for D26168: Bug 132845: ggated: Fix double file open and file descriptor leak..

Verify paths are the same between handshakes for a single session.

Aug 28 2020, 3:24 AM
ota_j.email.ne.jp added inline comments to D26168: Bug 132845: ggated: Fix double file open and file descriptor leak..
Aug 28 2020, 3:01 AM

Aug 27 2020

ota_j.email.ne.jp updated the diff for D26168: Bug 132845: ggated: Fix double file open and file descriptor leak..

Moved close() to connection_remove() and aligned with send/recv fds.

Aug 27 2020, 3:50 AM

Aug 26 2020

ota_j.email.ne.jp added a comment to D26168: Bug 132845: ggated: Fix double file open and file descriptor leak..

First of all, thank you for your quick response.

Aug 26 2020, 4:39 AM

Aug 25 2020

ota_j.email.ne.jp added a reviewer for D26168: Bug 132845: ggated: Fix double file open and file descriptor leak.: cem.
Aug 25 2020, 3:14 AM

Aug 24 2020

ota_j.email.ne.jp added a reviewer for D26168: Bug 132845: ggated: Fix double file open and file descriptor leak.: pjd.
Aug 24 2020, 2:18 AM
ota_j.email.ne.jp added a reviewer for D26168: Bug 132845: ggated: Fix double file open and file descriptor leak.: delphij.
Aug 24 2020, 2:16 AM
ota_j.email.ne.jp requested review of D26168: Bug 132845: ggated: Fix double file open and file descriptor leak..
Aug 24 2020, 2:12 AM

Aug 2 2020

ota_j.email.ne.jp updated the test plan for D25920: Fault in the parent of zombie/killed processes before killing a process in OOM..
Aug 2 2020, 4:04 PM
ota_j.email.ne.jp requested review of D25920: Fault in the parent of zombie/killed processes before killing a process in OOM..
Aug 2 2020, 3:59 PM

Jun 12 2020

ota_j.email.ne.jp updated the diff for D25226: PR 219829 Handle device removal and removal+add cases to fix infinity rate..

Added static to implementatoin; follow style of other existing static functions.

Jun 12 2020, 3:44 AM
ota_j.email.ne.jp updated the diff for D25226: PR 219829 Handle device removal and removal+add cases to fix infinity rate..

Drop unnecessary NULL check. This is a field on a struct and won't be NULL.

Jun 12 2020, 3:42 AM

Jun 11 2020

ota_j.email.ne.jp retitled D25226: PR 219829 Handle device removal and removal+add cases to fix infinity rate. from PR 219829 -Handle device removal and removal+add cases to fix infinity rate. to PR 219829 Handle device removal and removal+add cases to fix infinity rate..
Jun 11 2020, 3:41 AM
ota_j.email.ne.jp added a reviewer for D25226: PR 219829 Handle device removal and removal+add cases to fix infinity rate.: mr.
Jun 11 2020, 3:39 AM
ota_j.email.ne.jp requested review of D25226: PR 219829 Handle device removal and removal+add cases to fix infinity rate..
Jun 11 2020, 3:05 AM

May 29 2020

ota_j.email.ne.jp added a comment to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

I think I addressed all of feed backs so far.
I'm wondering if someone can take a look.

May 29 2020, 4:18 AM

May 14 2020

ota_j.email.ne.jp added a reviewer for D24834: systat: Introduced sysputpage() to display large page size with human readable format.: f20190109_eeeit.de.
May 14 2020, 2:33 AM
ota_j.email.ne.jp retitled D24834: systat: Introduced sysputpage() to display large page size with human readable format. from Introduced sysputpage() to display large page size with human readable format. to systat: Introduced sysputpage() to display large page size with human readable format..
May 14 2020, 2:27 AM
ota_j.email.ne.jp requested review of D24834: systat: Introduced sysputpage() to display large page size with human readable format..
May 14 2020, 2:25 AM
ota_j.email.ne.jp added inline comments to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
May 14 2020, 2:11 AM
ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Address other review comments.

May 14 2020, 2:10 AM

Mar 20 2020

ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Fixed style, moved 0 assignment to *before and *after when SWWAPBLK_NONE,
and adjusted if/else statement for *after case.

Mar 20 2020, 2:04 AM

Mar 15 2020

ota_j.email.ne.jp added inline comments to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Mar 15 2020, 1:05 PM
ota_j.email.ne.jp added inline comments to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Mar 15 2020, 12:28 AM
ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Split backward and forward search into separate functions for ease of reading.

Mar 15 2020, 12:24 AM

Mar 5 2020

ota_j.email.ne.jp added a comment to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Thank you for quick response, Mark.

Mar 5 2020, 4:37 AM

Feb 25 2020

ota_j.email.ne.jp updated the test plan for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Feb 25 2020, 6:13 AM

Jan 20 2020

ota_j.email.ne.jp updated the diff for D23274: Use swp_swdevt as struct swdevt and swp_swblk as struct swblk function prefix, respectively..

Updated function name in comment.

Jan 20 2020, 2:49 AM
ota_j.email.ne.jp created D23274: Use swp_swdevt as struct swdevt and swp_swblk as struct swblk function prefix, respectively..
Jan 20 2020, 2:46 AM
ota_j.email.ne.jp updated the diff for D23222: Check swblk size if a swap device has enoush space before calling blist_alloc.

Changed its prefix to swdevt.

Jan 20 2020, 2:44 AM

Jan 18 2020

ota_j.email.ne.jp abandoned D20675: Eliminate incorrect 2 pagesized block skipping while preparing swap device..
Jan 18 2020, 4:12 AM

Jan 17 2020

ota_j.email.ne.jp created D23222: Check swblk size if a swap device has enoush space before calling blist_alloc.
Jan 17 2020, 3:07 AM

Nov 22 2019

ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Check NULL pointers explicitly.

Nov 22 2019, 12:36 AM
ota_j.email.ne.jp added inline comments to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Nov 22 2019, 12:34 AM

Nov 21 2019

ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Restore KASSERT.

Nov 21 2019, 3:27 AM

Nov 20 2019

ota_j.email.ne.jp added inline comments to D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Nov 20 2019, 5:31 AM
ota_j.email.ne.jp retitled D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks. from swp_pager_meta_lookup() to find swap address and additionally search for the number of continuous blocks. to Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Nov 20 2019, 1:45 AM
ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Update and merge D22437.

Nov 20 2019, 1:41 AM

Nov 18 2019

ota_j.email.ne.jp added a comment to D22324: try to free blocks by upgrading object locks.

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 18 2019, 4:35 AM

Nov 16 2019

ota_j.email.ne.jp updated the diff for D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..

Dropped self-test.

Nov 16 2019, 6:22 PM
ota_j.email.ne.jp created D22409: Optimize swp_pager_meta_lookup() to find additionally the numbers of continuous blocks..
Nov 16 2019, 2:39 PM
ota_j.email.ne.jp created D22402: Create swp_pager_free_empty_swblk function to free if a swblk is free..
Nov 16 2019, 1:16 AM

Nov 15 2019

ota_j.email.ne.jp abandoned D22256: Skip SWAPBLK_NONEs with swap_pager_find_least() in swap_pager_copy()..

D22280 has been committed.

Nov 15 2019, 8:42 AM

Nov 11 2019

ota_j.email.ne.jp added inline comments to D22280: exploit sparsity in swap_pager_copy.
Nov 11 2019, 2:12 AM

Nov 9 2019

ota_j.email.ne.jp updated the diff for D22256: Skip SWAPBLK_NONEs with swap_pager_find_least() in swap_pager_copy()..

Correct swblk index and exclude swp_pager_swblk_free_empty() changes.

Nov 9 2019, 6:38 PM