Page MenuHomeFreeBSD
Differential D11231

Correct link metadata created when installing with -DNO_ROOT.
ClosedPublic
Actions

Authored by brooks on Jun 15 2017, 11:02 PM.
Tags
None
Referenced Files
F115282182: D11231.id29690.diff
Tue, Apr 22, 5:17 AM
Unknown Object (File)
Tue, Apr 15, 7:08 PM
Unknown Object (File)
Mon, Apr 14, 1:12 PM
Unknown Object (File)
Fri, Apr 11, 7:55 PM
Unknown Object (File)
Fri, Apr 4, 10:16 PM
Unknown Object (File)
Wed, Mar 26, 3:35 PM
Unknown Object (File)
Mar 10 2025, 2:39 AM
Unknown Object (File)
Jan 26 2025, 10:22 PM
View All 81 Files
Subscribers
emaste
imp

Details

Reviewers
bdrewery
Commits
rS335797: Correct link metadata created when installing with -DNO_ROOT.
Summary

Explicitly specify owner/group/mode metadata when creating links.

More consistently use INSTALL_SYMLINK to install symlinks

Diff Detail

Repository
rS FreeBSD src repository - subversion
Lint
Lint Not Applicable
Unit
Tests Not Applicable

Event Timeline

brooks created this revision.Jun 15 2017, 11:02 PM
Herald added a subscriber: emaste. · View Herald TranscriptJun 15 2017, 11:02 PM
brooks updated this revision to Diff 29736.Jun 17 2017, 12:10 AM
  • Add an INSTALL_MLINK to install manpage links.
Harbormaster completed remote builds in B9942: Diff 29736.Jun 17 2017, 12:10 AM
brooks added a comment.Jun 17 2017, 12:11 AM
In D11231#232526, @brooks wrote:
  • Add an INSTALL_MLINK to install manpage links.

With this change, I've verified the result builds working images. Some more testing required on releases.

bdrewery requested changes to this revision.Jun 19 2017, 5:15 PM

At least update bsd.README too please. Then +1 from me.

share/mk/bsd.own.mk
205–210 ↗(On Diff #29736)

Can you please add these to share/mk/bsd.README?

217 ↗(On Diff #29736)

Can you maybe add a comment above MLINK for it being MANPAGE-specific? Doesn't feel obvious like the MHRDLINK one is.

share/mk/bsd.prog.mk
285 ↗(On Diff #29736)

I feel like you should have LINK{OWN|GRP}= here too. It's a coincidence that they default to ${BIN*} today and may not tomorrow. I'm not strong on this opinion though.

This revision now requires changes to proceed.Jun 19 2017, 5:15 PM
brooks added a comment.Jun 20 2017, 10:39 PM

I'll submit a revised version shortly. I think I've got a test plan for release tar balls so I'll be trying that as well.

share/mk/bsd.own.mk
205–210 ↗(On Diff #29736)

I've added them at the top of this file. I'll try to scatter these in the right places. It does feel like bsd.README's coverage of bsd.own.mk is lacking most of the contents.

217 ↗(On Diff #29736)

In the changes I'm about to push I've changed the M prefix to MAN which should reduce confusion. I'm also adding a similar INSTALL_LIBSYMLINK to deal with an issue in /sbin/geom

share/mk/bsd.prog.mk
285 ↗(On Diff #29736)

I debated this myself and I'll go with your suggestion here.

brooks updated this revision to Diff 29878.Jun 20 2017, 10:44 PM
brooks edited edge metadata.
  • More cleanups of link permissions.
  • The geom classes build libraries, but install links to programs so
Harbormaster completed remote builds in B10022: Diff 29878.Jun 20 2017, 10:44 PM
brooks marked 4 inline comments as done.Jun 20 2017, 11:03 PM
bdrewery accepted this revision.Jun 19 2018, 6:24 PM
This revision is now accepted and ready to land.Jun 19 2018, 6:24 PM
brooks updated this revision to Diff 44605.Jun 28 2018, 10:43 PM
  • Rebase.
This revision now requires review to proceed.Jun 28 2018, 10:43 PM
Harbormaster completed remote builds in B17706: Diff 44605.Jun 28 2018, 10:43 PM
bdrewery accepted this revision.Jun 28 2018, 11:12 PM
This revision is now accepted and ready to land.Jun 28 2018, 11:12 PM
Closed by commit rS335797: Correct link metadata created when installing with -DNO_ROOT. (authored by brooks). · Explain WhyJun 29 2018, 4:08 PM
This revision was automatically updated to reflect the committed changes.
Herald added a subscriber: imp. · View Herald TranscriptJun 29 2018, 4:08 PM

Revision Contents
Changeset List

PathSize
head/
share/
mk/
40 lines
11 lines
2 lines
28 lines
4 lines

Diff 44642

View Options

head/share/mk/bsd.README

.\" .\"
.Dd September 28, 2023 .Dd December 14, 2023
.Dt IPFW 8 .Dt IPFW 8
.Os .Os
.Sh NAME .Sh NAME
.Nm ipfw , dnctl .Nm ipfw , dnctl
.Nd User interface for firewall, traffic shaper, packet scheduler, .Nd User interface for firewall, traffic shaper, packet scheduler,
in-kernel NAT. in-kernel NAT.
.Sh SYNOPSIS .Sh SYNOPSIS
.Ss FIREWALL CONFIGURATION .Ss FIREWALL CONFIGURATION
▲ Show 20 Lines • Show All 3,023 Lines • ▼ Show 20 Lines
is normally reported to the caller routine in the kernel, in the is normally reported to the caller routine in the kernel, in the
same way as it happens when a device queue fills up. same way as it happens when a device queue fills up.
Setting this Setting this
option reports the packet as successfully delivered, which can be option reports the packet as successfully delivered, which can be
needed for some experimental setups where you want to simulate needed for some experimental setups where you want to simulate
loss or congestion at a remote router. loss or congestion at a remote router.
.Pp .Pp
.It Cm plr Ar packet-loss-rate .It Cm plr Ar packet-loss-rate
.It Cm plr Ar K,p,H,r
Packet loss rate. Packet loss rate.
Argument Argument
.Ar packet-loss-rate .Ar packet-loss-rate
is a floating-point number between 0 and 1, with 0 meaning no is a floating-point number between 0 and 1, with 0 meaning no
loss, 1 meaning 100% loss. loss, 1 meaning 100% loss.
The loss rate is internally represented on 31 bits. .Pp
When invoked with four arguments, the simple Gilbert-Elliott
channel model with two states (Good and Bad) is used.
.Bd -literal -offset indent
r
.----------------.
v |
.------------. .------------.
| G | | B |
| drop (K) | | drop (H) |
'------------' '------------'
| ^
'----------------'
p
.Ed
This has the associated probabilities
.Po Ar K
and
.Ar H Pc
for the loss probability. This is different from the literature,
where this model is described with probabilities of successful
transmission k and h. However, converting from literature is
easy:
.Pp
K = 1 - k ; H = 1 - h
.Pp
This is to retain consistency within the interface and allow the
quick re-use of loss probability when giving only a single argument.
In addition the state change probabilities
.Po Ar p
and
.Ar r Pc
are given.
All of the above probabilities are internally represented on 31 bits.
.Pp .Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes .It Cm queue Brq Ar slots | size Ns Cm Kbytes
pauamma_gundo.comUnsubmitted
Done
Inline Actions

Worth being explicit that these arguments are comma-separated and mention their internal representation if relevant?

pauamma_gundo.com: Worth being explicit that these arguments are comma-separated and mention their internal…
rscheffAuthorUnsubmitted
Done
Inline Actions

Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert-elliott model give these (k, h) as transmission probabilities - the inverse of what the simple PLR loss probability is;
Adjusted the code accordingly.
Alternatively, I could call these parameters K and H, with K = (1-k) and H = (1-h), keep the code more streamlined and the probabilities would align nicely with the simple PLR loss model...

Any opinions?

rscheff: Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert…
pauamma_gundo.comUnsubmitted
Not Done
Inline Actions

I have no basis for an informed opinion on that point myself.

pauamma_gundo.com: I have no basis for an informed opinion on that point myself.
Queue size, in Queue size, in
.Ar slots .Ar slots
or or
.Cm KBytes . .Cm KBytes .
Default value is 50 slots, which Default value is 50 slots, which
is the typical queue size for Ethernet devices. is the typical queue size for Ethernet devices.
Note that for slow speed links you should keep the queue Note that for slow speed links you should keep the queue
size short or your traffic might be affected by a significant size short or your traffic might be affected by a significant
queueing delay. queueing delay.
pauamma_gundo.comUnsubmitted
Done
Inline Actions
.Po Ar change-to-B
and
- .Ar change-to-A Pc
+ .Ar change-to-G Pc
.
.Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes

Or explain what A is.

pauamma_gundo.com: Or explain what A is.
E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit
or 20s of queue on a 30Kbit/s pipe. or 20s of queue on a 30Kbit/s pipe.
Even worse effects can result if you get packets from an Even worse effects can result if you get packets from an
interface with a much larger MTU, e.g.\& the loopback interface interface with a much larger MTU, e.g.\& the loopback interface
with its 16KB packets. with its 16KB packets.
The The
.Xr sysctl 8 .Xr sysctl 8
variables variables
▲ Show 20 Lines • Show All 1,890 Lines • Show Last 20 Lines
View Options

head/share/mk/bsd.lib.mk

.\" .\"
.Dd September 28, 2023 .Dd December 14, 2023
.Dt IPFW 8 .Dt IPFW 8
.Os .Os
.Sh NAME .Sh NAME
.Nm ipfw , dnctl .Nm ipfw , dnctl
.Nd User interface for firewall, traffic shaper, packet scheduler, .Nd User interface for firewall, traffic shaper, packet scheduler,
in-kernel NAT. in-kernel NAT.
.Sh SYNOPSIS .Sh SYNOPSIS
.Ss FIREWALL CONFIGURATION .Ss FIREWALL CONFIGURATION
▲ Show 20 Lines • Show All 3,023 Lines • ▼ Show 20 Lines
is normally reported to the caller routine in the kernel, in the is normally reported to the caller routine in the kernel, in the
same way as it happens when a device queue fills up. same way as it happens when a device queue fills up.
Setting this Setting this
option reports the packet as successfully delivered, which can be option reports the packet as successfully delivered, which can be
needed for some experimental setups where you want to simulate needed for some experimental setups where you want to simulate
loss or congestion at a remote router. loss or congestion at a remote router.
.Pp .Pp
.It Cm plr Ar packet-loss-rate .It Cm plr Ar packet-loss-rate
.It Cm plr Ar K,p,H,r
Packet loss rate. Packet loss rate.
Argument Argument
.Ar packet-loss-rate .Ar packet-loss-rate
is a floating-point number between 0 and 1, with 0 meaning no is a floating-point number between 0 and 1, with 0 meaning no
loss, 1 meaning 100% loss. loss, 1 meaning 100% loss.
The loss rate is internally represented on 31 bits. .Pp
When invoked with four arguments, the simple Gilbert-Elliott
channel model with two states (Good and Bad) is used.
.Bd -literal -offset indent
r
.----------------.
v |
.------------. .------------.
| G | | B |
| drop (K) | | drop (H) |
'------------' '------------'
| ^
'----------------'
p
.Ed
This has the associated probabilities
.Po Ar K
and
.Ar H Pc
for the loss probability. This is different from the literature,
where this model is described with probabilities of successful
transmission k and h. However, converting from literature is
easy:
.Pp
K = 1 - k ; H = 1 - h
.Pp
This is to retain consistency within the interface and allow the
quick re-use of loss probability when giving only a single argument.
In addition the state change probabilities
.Po Ar p
and
.Ar r Pc
are given.
All of the above probabilities are internally represented on 31 bits.
.Pp .Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes .It Cm queue Brq Ar slots | size Ns Cm Kbytes
pauamma_gundo.comUnsubmitted
Done
Inline Actions

Worth being explicit that these arguments are comma-separated and mention their internal representation if relevant?

pauamma_gundo.com: Worth being explicit that these arguments are comma-separated and mention their internal…
rscheffAuthorUnsubmitted
Done
Inline Actions

Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert-elliott model give these (k, h) as transmission probabilities - the inverse of what the simple PLR loss probability is;
Adjusted the code accordingly.
Alternatively, I could call these parameters K and H, with K = (1-k) and H = (1-h), keep the code more streamlined and the probabilities would align nicely with the simple PLR loss model...

Any opinions?

rscheff: Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert…
pauamma_gundo.comUnsubmitted
Not Done
Inline Actions

I have no basis for an informed opinion on that point myself.

pauamma_gundo.com: I have no basis for an informed opinion on that point myself.
Queue size, in Queue size, in
.Ar slots .Ar slots
or or
.Cm KBytes . .Cm KBytes .
Default value is 50 slots, which Default value is 50 slots, which
is the typical queue size for Ethernet devices. is the typical queue size for Ethernet devices.
Note that for slow speed links you should keep the queue Note that for slow speed links you should keep the queue
size short or your traffic might be affected by a significant size short or your traffic might be affected by a significant
queueing delay. queueing delay.
pauamma_gundo.comUnsubmitted
Done
Inline Actions
.Po Ar change-to-B
and
- .Ar change-to-A Pc
+ .Ar change-to-G Pc
.
.Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes

Or explain what A is.

pauamma_gundo.com: Or explain what A is.
E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit
or 20s of queue on a 30Kbit/s pipe. or 20s of queue on a 30Kbit/s pipe.
Even worse effects can result if you get packets from an Even worse effects can result if you get packets from an
interface with a much larger MTU, e.g.\& the loopback interface interface with a much larger MTU, e.g.\& the loopback interface
with its 16KB packets. with its 16KB packets.
The The
.Xr sysctl 8 .Xr sysctl 8
variables variables
▲ Show 20 Lines • Show All 1,890 Lines • Show Last 20 Lines
View Options

head/share/mk/bsd.man.mk

.\" .\"
.Dd September 28, 2023 .Dd December 14, 2023
.Dt IPFW 8 .Dt IPFW 8
.Os .Os
.Sh NAME .Sh NAME
.Nm ipfw , dnctl .Nm ipfw , dnctl
.Nd User interface for firewall, traffic shaper, packet scheduler, .Nd User interface for firewall, traffic shaper, packet scheduler,
in-kernel NAT. in-kernel NAT.
.Sh SYNOPSIS .Sh SYNOPSIS
.Ss FIREWALL CONFIGURATION .Ss FIREWALL CONFIGURATION
▲ Show 20 Lines • Show All 3,023 Lines • ▼ Show 20 Lines
is normally reported to the caller routine in the kernel, in the is normally reported to the caller routine in the kernel, in the
same way as it happens when a device queue fills up. same way as it happens when a device queue fills up.
Setting this Setting this
option reports the packet as successfully delivered, which can be option reports the packet as successfully delivered, which can be
needed for some experimental setups where you want to simulate needed for some experimental setups where you want to simulate
loss or congestion at a remote router. loss or congestion at a remote router.
.Pp .Pp
.It Cm plr Ar packet-loss-rate .It Cm plr Ar packet-loss-rate
.It Cm plr Ar K,p,H,r
Packet loss rate. Packet loss rate.
Argument Argument
.Ar packet-loss-rate .Ar packet-loss-rate
is a floating-point number between 0 and 1, with 0 meaning no is a floating-point number between 0 and 1, with 0 meaning no
loss, 1 meaning 100% loss. loss, 1 meaning 100% loss.
The loss rate is internally represented on 31 bits. .Pp
When invoked with four arguments, the simple Gilbert-Elliott
channel model with two states (Good and Bad) is used.
.Bd -literal -offset indent
r
.----------------.
v |
.------------. .------------.
| G | | B |
| drop (K) | | drop (H) |
'------------' '------------'
| ^
'----------------'
p
.Ed
This has the associated probabilities
.Po Ar K
and
.Ar H Pc
for the loss probability. This is different from the literature,
where this model is described with probabilities of successful
transmission k and h. However, converting from literature is
easy:
.Pp
K = 1 - k ; H = 1 - h
.Pp
This is to retain consistency within the interface and allow the
quick re-use of loss probability when giving only a single argument.
In addition the state change probabilities
.Po Ar p
and
.Ar r Pc
are given.
All of the above probabilities are internally represented on 31 bits.
.Pp .Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes .It Cm queue Brq Ar slots | size Ns Cm Kbytes
pauamma_gundo.comUnsubmitted
Done
Inline Actions

Worth being explicit that these arguments are comma-separated and mention their internal representation if relevant?

pauamma_gundo.com: Worth being explicit that these arguments are comma-separated and mention their internal…
rscheffAuthorUnsubmitted
Done
Inline Actions

Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert-elliott model give these (k, h) as transmission probabilities - the inverse of what the simple PLR loss probability is;
Adjusted the code accordingly.
Alternatively, I could call these parameters K and H, with K = (1-k) and H = (1-h), keep the code more streamlined and the probabilities would align nicely with the simple PLR loss model...

Any opinions?

rscheff: Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert…
pauamma_gundo.comUnsubmitted
Not Done
Inline Actions

I have no basis for an informed opinion on that point myself.

pauamma_gundo.com: I have no basis for an informed opinion on that point myself.
Queue size, in Queue size, in
.Ar slots .Ar slots
or or
.Cm KBytes . .Cm KBytes .
Default value is 50 slots, which Default value is 50 slots, which
is the typical queue size for Ethernet devices. is the typical queue size for Ethernet devices.
Note that for slow speed links you should keep the queue Note that for slow speed links you should keep the queue
size short or your traffic might be affected by a significant size short or your traffic might be affected by a significant
queueing delay. queueing delay.
pauamma_gundo.comUnsubmitted
Done
Inline Actions
.Po Ar change-to-B
and
- .Ar change-to-A Pc
+ .Ar change-to-G Pc
.
.Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes

Or explain what A is.

pauamma_gundo.com: Or explain what A is.
E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit
or 20s of queue on a 30Kbit/s pipe. or 20s of queue on a 30Kbit/s pipe.
Even worse effects can result if you get packets from an Even worse effects can result if you get packets from an
interface with a much larger MTU, e.g.\& the loopback interface interface with a much larger MTU, e.g.\& the loopback interface
with its 16KB packets. with its 16KB packets.
The The
.Xr sysctl 8 .Xr sysctl 8
variables variables
▲ Show 20 Lines • Show All 1,890 Lines • Show Last 20 Lines
View Options

head/share/mk/bsd.own.mk

.\" .\"
.Dd September 28, 2023 .Dd December 14, 2023
.Dt IPFW 8 .Dt IPFW 8
.Os .Os
.Sh NAME .Sh NAME
.Nm ipfw , dnctl .Nm ipfw , dnctl
.Nd User interface for firewall, traffic shaper, packet scheduler, .Nd User interface for firewall, traffic shaper, packet scheduler,
in-kernel NAT. in-kernel NAT.
.Sh SYNOPSIS .Sh SYNOPSIS
.Ss FIREWALL CONFIGURATION .Ss FIREWALL CONFIGURATION
▲ Show 20 Lines • Show All 3,023 Lines • ▼ Show 20 Lines
is normally reported to the caller routine in the kernel, in the is normally reported to the caller routine in the kernel, in the
same way as it happens when a device queue fills up. same way as it happens when a device queue fills up.
Setting this Setting this
option reports the packet as successfully delivered, which can be option reports the packet as successfully delivered, which can be
needed for some experimental setups where you want to simulate needed for some experimental setups where you want to simulate
loss or congestion at a remote router. loss or congestion at a remote router.
.Pp .Pp
.It Cm plr Ar packet-loss-rate .It Cm plr Ar packet-loss-rate
.It Cm plr Ar K,p,H,r
Packet loss rate. Packet loss rate.
Argument Argument
.Ar packet-loss-rate .Ar packet-loss-rate
is a floating-point number between 0 and 1, with 0 meaning no is a floating-point number between 0 and 1, with 0 meaning no
loss, 1 meaning 100% loss. loss, 1 meaning 100% loss.
The loss rate is internally represented on 31 bits. .Pp
When invoked with four arguments, the simple Gilbert-Elliott
channel model with two states (Good and Bad) is used.
.Bd -literal -offset indent
r
.----------------.
v |
.------------. .------------.
| G | | B |
| drop (K) | | drop (H) |
'------------' '------------'
| ^
'----------------'
p
.Ed
This has the associated probabilities
.Po Ar K
and
.Ar H Pc
for the loss probability. This is different from the literature,
where this model is described with probabilities of successful
transmission k and h. However, converting from literature is
easy:
.Pp
K = 1 - k ; H = 1 - h
.Pp
This is to retain consistency within the interface and allow the
quick re-use of loss probability when giving only a single argument.
In addition the state change probabilities
.Po Ar p
and
.Ar r Pc
are given.
All of the above probabilities are internally represented on 31 bits.
.Pp .Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes .It Cm queue Brq Ar slots | size Ns Cm Kbytes
pauamma_gundo.comUnsubmitted
Done
Inline Actions

Worth being explicit that these arguments are comma-separated and mention their internal representation if relevant?

pauamma_gundo.com: Worth being explicit that these arguments are comma-separated and mention their internal…
rscheffAuthorUnsubmitted
Done
Inline Actions

Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert-elliott model give these (k, h) as transmission probabilities - the inverse of what the simple PLR loss probability is;
Adjusted the code accordingly.
Alternatively, I could call these parameters K and H, with K = (1-k) and H = (1-h), keep the code more streamlined and the probabilities would align nicely with the simple PLR loss model...

Any opinions?

rscheff: Ideally, I wanted to have the parameters in loss probabilities; the literature of the gilbert…
pauamma_gundo.comUnsubmitted
Not Done
Inline Actions

I have no basis for an informed opinion on that point myself.

pauamma_gundo.com: I have no basis for an informed opinion on that point myself.
Queue size, in Queue size, in
.Ar slots .Ar slots
or or
.Cm KBytes . .Cm KBytes .
Default value is 50 slots, which Default value is 50 slots, which
is the typical queue size for Ethernet devices. is the typical queue size for Ethernet devices.
Note that for slow speed links you should keep the queue Note that for slow speed links you should keep the queue
size short or your traffic might be affected by a significant size short or your traffic might be affected by a significant
queueing delay. queueing delay.
pauamma_gundo.comUnsubmitted
Done
Inline Actions
.Po Ar change-to-B
and
- .Ar change-to-A Pc
+ .Ar change-to-G Pc
.
.Pp
.It Cm queue Brq Ar slots | size Ns Cm Kbytes

Or explain what A is.

pauamma_gundo.com: Or explain what A is.
E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit E.g., 50 max-sized Ethernet packets (1500 bytes) mean 600Kbit
or 20s of queue on a 30Kbit/s pipe. or 20s of queue on a 30Kbit/s pipe.
Even worse effects can result if you get packets from an Even worse effects can result if you get packets from an
interface with a much larger MTU, e.g.\& the loopback interface interface with a much larger MTU, e.g.\& the loopback interface
with its 16KB packets. with its 16KB packets.
The The
.Xr sysctl 8 .Xr sysctl 8
variables variables
▲ Show 20 Lines • Show All 1,890 Lines • Show Last 20 Lines
View Options

head/share/mk/bsd.prog.mk

Loading...