Page MenuHomeFreeBSD

bhyveUmbrella
ActivePublic

Details

Description

bhyve reviewers group

Recent Activity

Wed, Jan 16

richard_meric.id.au added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Not sure about the seg_max value (even in the current code).
Looking at the qemu implementation, I suspect it should be queue length -2, which I guess is what the patched code happens to achieve (based on the definition of VTBLK_RINGSZ), but it could be more explicit.

Wed, Jan 16, 6:01 AM · bhyve

Tue, Jan 15

araujo added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.
NOTE: I blocked this patch, till I finish my analysis, nothing wrong with that...
Tue, Jan 15, 12:19 AM · bhyve
araujo requested changes to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.
Tue, Jan 15, 12:17 AM · bhyve

Mon, Jan 14

araujo added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Hi Richard,

Do you mind share the tests you made??

Best,

I've currently got these changes running in a FreeNAS-11.2-RELEASE-U1 instance (FreeBSD 11.2 kernel) running a couple of Windows 2016 VMs plus one Centos 7 - all using virtio-blk disk devices without issue. I also have pfSense (FreeBSD 11.2) and OpenSuSE (Leap 42.3) VMs which I haven't tested yet - but can do if required.

Mon, Jan 14, 9:43 AM · bhyve
richard_meric.id.au added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Hi Richard,

Do you mind share the tests you made??

Best,

Mon, Jan 14, 6:11 AM · bhyve
araujo added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Hi Richard,

Mon, Jan 14, 3:31 AM · bhyve
richard_meric.id.au added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

This is effectively what we did in SmartOS/illumos to address the issue:
https://github.com/joyent/illumos-joyent/commit/192e1e6405f98e4b0a12f9488793c5dd000f3f7e

I would suggest including a comment where vbc_seg_max is set to note why that value is clamped as it is.

Mon, Jan 14, 12:51 AM · bhyve
richard_meric.id.au updated the diff for D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Added comment where vbc_seg_max is set to note why that value is clamped as it is (as recommended by pmooney_pfmooney.com).

Mon, Jan 14, 12:48 AM · bhyve

Sun, Jan 13

pmooney_pfmooney.com added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

This is effectively what we did in SmartOS/illumos to address the issue:
https://github.com/joyent/illumos-joyent/commit/192e1e6405f98e4b0a12f9488793c5dd000f3f7e

Sun, Jan 13, 11:19 PM · bhyve
araujo added a comment to D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Thanks for the patch, I will take a look on within the next two weeks.

Sun, Jan 13, 7:09 AM · bhyve
richard_meric.id.au updated the diff for D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.

Fix cut-and-paste error in patch

Sun, Jan 13, 7:08 AM · bhyve
araujo added a reviewer for D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests: araujo.
Sun, Jan 13, 7:05 AM · bhyve
richard_meric.id.au created D18831: Increase bhyve BLOCKIF_IOV_MAX and VTBLK_RINGSZ to 128 to support modern Windows guests.
Sun, Jan 13, 6:49 AM · bhyve

Fri, Jan 4

chuck closed D18703: Fix bhyve's NVMe Completion Queue entry values.
Fri, Jan 4, 3:03 PM · bhyve
chuck closed D18702: Fix bhyve's NVMe queue bookkeeping.
Fri, Jan 4, 3:03 PM · bhyve

Wed, Jan 2

imp added a comment to D18703: Fix bhyve's NVMe Completion Queue entry values.

LGTM! Could you please set a MFC?

Absolutely. I have one in the real commit message, but Differential didn't seem to like it in the summary.

Wed, Jan 2, 6:21 PM · bhyve
chuck added a comment to D18703: Fix bhyve's NVMe Completion Queue entry values.

LGTM! Could you please set a MFC?

Wed, Jan 2, 5:44 PM · bhyve
imp accepted D18703: Fix bhyve's NVMe Completion Queue entry values.

lgtm

Wed, Jan 2, 7:10 AM · bhyve
imp accepted D18702: Fix bhyve's NVMe queue bookkeeping.

This looks good to me as well...

Wed, Jan 2, 7:09 AM · bhyve
araujo accepted D18703: Fix bhyve's NVMe Completion Queue entry values.

LGTM! Could you please set a MFC?

Wed, Jan 2, 4:21 AM · bhyve
araujo added a reviewer for D18703: Fix bhyve's NVMe Completion Queue entry values: bhyve.
Wed, Jan 2, 4:21 AM · bhyve
araujo accepted D18702: Fix bhyve's NVMe queue bookkeeping.
Wed, Jan 2, 4:20 AM · bhyve
araujo added a reviewer for D18702: Fix bhyve's NVMe queue bookkeeping: bhyve.
Wed, Jan 2, 4:20 AM · bhyve
araujo accepted D18702: Fix bhyve's NVMe queue bookkeeping.

LGTM! Do you mind set a MFC?

Wed, Jan 2, 4:20 AM · bhyve

Mon, Dec 31

chuck created D18703: Fix bhyve's NVMe Completion Queue entry values.
Mon, Dec 31, 6:58 PM · bhyve
chuck updated the diff for D18702: Fix bhyve's NVMe queue bookkeeping.

Use ONE_BASED() macro for CQ creation as well as SQ

Mon, Dec 31, 6:54 PM · bhyve
chuck created D18702: Fix bhyve's NVMe queue bookkeeping.
Mon, Dec 31, 6:53 PM · bhyve

Dec 11 2018

mav closed D18504: Allow CTL device specification in bhyve virtio-scsi.
Dec 11 2018, 8:47 PM · bhyve
araujo accepted D18504: Allow CTL device specification in bhyve virtio-scsi.

LGTM! Thanks for the patch!

Dec 11 2018, 5:05 AM · bhyve

Dec 10 2018

mav added a reviewer for D18504: Allow CTL device specification in bhyve virtio-scsi: araujo.
Dec 10 2018, 9:52 PM · bhyve
mav created D18504: Allow CTL device specification in bhyve virtio-scsi.
Dec 10 2018, 9:51 PM · bhyve

Dec 3 2018

avg removed a member for bhyve: avg.
Dec 3 2018, 1:20 PM

Nov 5 2018

araujo removed a watcher for bhyve: araujo.
Nov 5 2018, 5:57 PM
araujo added a watcher for bhyve: araujo.
Nov 5 2018, 5:56 PM

Sep 6 2018

jhb closed D16822: bhyve: Use MAP_GUARD guard pages.
Sep 6 2018, 8:30 PM · bhyve

Sep 4 2018

araujo accepted D16822: bhyve: Use MAP_GUARD guard pages.

I have tested it with FreeBSD HEAD as a guest running for couple days.

Sep 4 2018, 1:51 AM · bhyve
lattera-gmail.com added a comment to D16822: bhyve: Use MAP_GUARD guard pages.
In D16822#361197, @jhb wrote:

This looks right to me now. I'll try to test it locally in the next day or so.

Sep 4 2018, 12:22 AM · bhyve

Aug 28 2018

jhb added a comment to D16822: bhyve: Use MAP_GUARD guard pages.

This looks right to me now. I'll try to test it locally in the next day or so.

Aug 28 2018, 11:04 PM · bhyve
lattera-gmail.com updated the diff for D16822: bhyve: Use MAP_GUARD guard pages.

Minimize diff with suggestions by jhb.

Aug 28 2018, 9:54 PM · bhyve
lattera-gmail.com added inline comments to D16822: bhyve: Use MAP_GUARD guard pages.
Aug 28 2018, 9:49 PM · bhyve
jhb added inline comments to D16822: bhyve: Use MAP_GUARD guard pages.
Aug 28 2018, 9:45 PM · bhyve
lattera-gmail.com added a comment to D16822: bhyve: Use MAP_GUARD guard pages.
In D16822#361129, @kib wrote:

The only useful feature of the phabricator is to easily see context around the patch, which you successfully botched.

Aug 28 2018, 9:23 PM · bhyve
lattera-gmail.com updated the diff for D16822: bhyve: Use MAP_GUARD guard pages.

Reflect changes requested by both kib and jhb.

Aug 28 2018, 9:23 PM · bhyve
jhb added a comment to D16822: bhyve: Use MAP_GUARD guard pages.

You should only be replacing the existing MAP_ANON / PROT_NONE mmap() calls with MAP_GUARD instead. You shouldn't be adding new mmap() calls. The "real" thing is already being mmap()'d later. As @kib notes, it's good to use 'svn diff -x -U 99999' or the like to generate context when uploading to phabricator.

Aug 28 2018, 9:06 PM · bhyve
kib added a comment to D16822: bhyve: Use MAP_GUARD guard pages.

The only useful feature of the phabricator is to easily see context around the patch, which you successfully botched.

Aug 28 2018, 8:55 PM · bhyve
lattera-gmail.com updated the diff for D16822: bhyve: Use MAP_GUARD guard pages.

Missed a spot. Cover another mapping with MAP_GUARD.

Aug 28 2018, 8:38 PM · bhyve
lattera-gmail.com updated the diff for D16822: bhyve: Use MAP_GUARD guard pages.

Update the patch to use John Baldwin's suggestion on mapping the entire range first with MAP_GUARD.

Aug 28 2018, 8:30 PM · bhyve
lattera-gmail.com added a comment to D16822: bhyve: Use MAP_GUARD guard pages.
In D16822#361106, @jhb wrote:

Adding @kib since he added MAP_GUARD. I think you should instead MAP_GUARD the entire range first and then remap the middle. The middle is already remapped on line 518, so instead of adding new mmap()'s just for the guards, you should replace the mmap() on line 496 with this:

base = mmap(NULL, len2, PROT_NONE, MAP_GUARD | MAP_ALIGNED_SUPER, -1, 0);
Aug 28 2018, 8:14 PM · bhyve
jhb added a comment to D16822: bhyve: Use MAP_GUARD guard pages.

Adding @kib since he added MAP_GUARD. I think you should instead MAP_GUARD the entire range first and then remap the middle. The middle is already remapped on line 518, so instead of adding new mmap()'s just for the guards, you should replace the mmap() on line 496 with this:

Aug 28 2018, 8:03 PM · bhyve
jhb added a reviewer for D16822: bhyve: Use MAP_GUARD guard pages: kib.
Aug 28 2018, 7:52 PM · bhyve