From Newsgroup: comp.protocols.time.ntp

--0000000000008a490b064e64e411
Content-Type: text/plain; charset="UTF-8"

After spending an entirely unreasonable amount of time adding SO_TIMESTAMP kernel-mode UDP receive timestamps to Windows ntpd and testing it, and a
much shorter time working around the failure of essentially all FreeBSD releases prior to 15.0 to support those timestamps for IPv6 with phk's SO_BINTIME, I've fmerged a few years of fixes into a private test release
of ntpd.

In addition to those issues, symmetric auth bugs in 4.2.8p18 and build
issues with newer OpenSSL 3.x have been corrected. KoDs from Chrony are
once again respected. IPv6-only clients using the pool now work. A bug removing dynamic restrict entries generated from "restrict source" is
fixed, and structure alignment improved to help speed restrictions checks
on the input path. Build breaks triggered by newer gcc, clang, and glibc
are corrected. Upstream appliances which have questionable packet poll claiming 1s can be used as sources again. A longstanding bug with the
daemon loop going haywire at poll intervals longer than 1024s (maxpoll 11
and up) is fixed.

https://bk.davehart.net/ntp-stable-dlh-260328/ChangeLog?PAGE=anno&REV=69c85abbX0AuaX-DsPHhmzo7DD_BHg
has
more details.

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Windows binaries that can be overlaid on an existing Meinberg ntpd install
are in:

https://davehart.net/ntp/win/ntp-stable-dlh-260328.zip

To use those you'll probably need to install a recent-enough OpenSSL such
as Shining Light Productions' https://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi

Sadly, the Windows SO_TIMESTAMP support depends on setting registry keys on each network adapter that are not easy to find, and require Administrator
group membership to modify. "ntpd --enable-udp-timestamps" from an
elevated (Administrator) shell will do that, then the network interfaces
need to be cycled or the machine restarted. You can see the improvement on both FreeBSD IPv6 and Windows UDP receive timestamps by adding "logconfig =allall" or "logconfig +sysinfo" to ntp.conf, which will report after 3
hours the average difference between kernel receive timestamps and the one
ntpd grabs. Then after another 6 hours, then another 12, and every 24
hours after that.

Cheers,
Dave Hart
After spending an entirely unreasonable amount of time adding SO_TIMESTAMP kernel-mode UDP receive timestamps to Windows ntpd and testing it, and a
much shorter time working around the failure of essentially all FreeBSD releases prior to 15.0 to support those timestamps for IPv6 with phk's SO_BINTIME, I've fmerged a few years of fixes into a private test release
of ntpd.

In addition to those issues, symmetric auth bugs in 4.2.8p18 and build
issues with newer OpenSSL 3.x have been corrected. KoDs from Chrony are
once again respected. IPv6-only clients using the pool now work. A bug removing dynamic restrict entries generated from "restrict source" is
fixed, and structure alignment improved to help speed restrictions checks
on the input path. Build breaks triggered by newer gcc, clang, and glibc
are corrected. Upstream appliances which have questionable packet poll claiming 1s can be used as sources again. A longstanding bug with the
daemon loop going haywire at poll intervals longer than 1024s (maxpoll 11
and up) is fixed.

https://bk.davehart.net/ntp-stable-dlh-260328/ChangeLog?PAGE=anno&REV=69c85abbX0AuaX-DsPHhmzo7DD_BHg
has
more details.

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Windows binaries that can be overlaid on an existing Meinberg ntpd install
are in:

https://davehart.net/ntp/win/ntp-stable-dlh-260328.zip

To use those you'll probably need to install a recent-enough OpenSSL such
as Shining Light Productions' https://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi

Sadly, the Windows SO_TIMESTAMP support depends on setting registry keys on each network adapter that are not easy to find, and require Administrator
group membership to modify. "ntpd --enable-udp-timestamps" from an
elevated (Administrator) shell will do that, then the network interfaces
need to be cycled or the machine restarted. You can see the improvement on both FreeBSD IPv6 and Windows UDP receive timestamps by adding "logconfig =allall" or "logconfig +sysinfo" to ntp.conf, which will report after 3
hours the average difference between kernel receive timestamps and the one
ntpd grabs. Then after another 6 hours, then another 12, and every 24
hours after that.

Cheers,
Dave Hart
After spending an entirely unreasonable amount of time adding SO_TIMESTAMP kernel-mode UDP receive timestamps to Windows ntpd and testing it, and a
much shorter time working around the failure of essentially all FreeBSD releases prior to 15.0 to support those timestamps for IPv6 with phk's SO_BINTIME, I've fmerged a few years of fixes into a private test release
of ntpd.

In addition to those issues, symmetric auth bugs in 4.2.8p18 and build
issues with newer OpenSSL 3.x have been corrected. KoDs from Chrony are
once again respected. IPv6-only clients using the pool now work. A bug removing dynamic restrict entries generated from "restrict source" is
fixed, and structure alignment improved to help speed restrictions checks
on the input path. Build breaks triggered by newer gcc, clang, and glibc
are corrected. Upstream appliances which have questionable packet poll claiming 1s can be used as sources again. A longstanding bug with the
daemon loop going haywire at poll intervals longer than 1024s (maxpoll 11
and up) is fixed.

https://bk.davehart.net/ntp-stable-dlh-260328/ChangeLog?PAGE=anno&REV=69c85abbX0AuaX-DsPHhmzo7DD_BHg
has
more details.

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Windows binaries that can be overlaid on an existing Meinberg ntpd install
are in:

https://davehart.net/ntp/win/ntp-stable-dlh-260328.zip

To use those you'll probably need to install a recent-enough OpenSSL such
as Shining Light Productions' https://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi

Sadly, the Windows SO_TIMESTAMP support depends on setting registry keys on each network adapter that are not easy to find, and require Administrator
group membership to modify. "ntpd --enable-udp-timestamps" from an
elevated (Administrator) shell will do that, then the network interfaces
need to be cycled or the machine restarted. You can see the improvement on both FreeBSD IPv6 and Windows UDP receive timestamps by adding "logconfig =allall" or "logconfig +sysinfo" to ntp.conf, which will report after 3
hours the average difference between kernel receive timestamps and the one
ntpd grabs. Then after another 6 hours, then another 12, and every 24
hours after that.

Cheers,
Dave Hart

--0000000000008a490b064e64e411
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><div class=3D"gmail_default" style=3D""><= div style=3D"min-height:100%;color:rgb(32,33,36);font-size:medium"><div sty= le=3D"width:1920px"><div style=3D""><div style=3D"display:flex;background-c= olor:transparent"><div style=3D"overflow:hidden"><div style=3D"border-radiu= s:16px;margin-bottom:16px;overflow-y:hidden"><div style=3D""><div style=3D"= "><div style=3D""><div id=3D"m_2344152423409506380gmail-:3" style=3D"backgr= ound-image:initial;background-position:initial;background-size:initial;back= ground-repeat:initial;background-origin:initial;background-clip:initial;ove= rflow-y:scroll;padding-right:0px;height:817px"><div id=3D"m_234415242340950= 6380gmail-:1" style=3D"padding:0px;vertical-align:bottom;min-height:627px">= <div style=3D""><div role=3D"main" style=3D""><div style=3D""><div style=3D= "background-image:initial;background-position:initial;background-size:initi= al;background-repeat:initial;background-origin:initial;background-clip:init= ial;min-height:64ex;min-width:502px;margin:0px;padding-right:16px"><div sty= le=3D""><div style=3D"background-color:transparent;color:rgb(34,34,34);min-= width:502px;padding:0px"><div role=3D"list" style=3D""><div role=3D"listite=
m" aria-expanded=3D"true" style=3D"padding-bottom:0px;max-width:100000px;cl= ear:both;outline:none"><div style=3D"margin-bottom:0px;border-width:0px;bor= der-top-style:solid;border-right-style:initial;border-bottom-style:initial;= border-left-style:initial;border-top-color:rgb(239,239,239);border-right-co= lor:initial;border-left-color:initial;border-bottom-color:initial;border-ra= dius:0px;float:left;width:1508px"><div style=3D"border-top:none;padding-top= :0px;background-color:transparent;border-right:0px;border-bottom:0px rgba(1= 00,121,143,0.12);border-left:0px;border-radius:0px;margin-bottom:0px;margin= -left:0px;margin-right:0px"><div style=3D""><div id=3D"m_234415242340950638= 0gmail-avWBGd-5024" style=3D""><div id=3D"m_2344152423409506380gmail-avWBGd= -5025" style=3D""><div style=3D"border-left:none;padding:0px;display:flex">= <div style=3D"margin:0px;min-width:0px;padding:0px 0px 20px;width:auto"><di=
v style=3D""><div id=3D"m_2344152423409506380gmail-:62i" style=3D"direction= :ltr;margin:8px 0px 0px;padding:0px;font-size:0.875rem;overflow-x:hidden"><= div id=3D"m_2344152423409506380gmail-:62h" style=3D"direction:ltr;font-vari= ant-numeric:normal;font-variant-east-asian:normal;font-variant-alternates:n= ormal;font-size-adjust:none;font-kerning:auto;font-feature-settings:normal;= font-stretch:normal;font-size:small;line-height:1.5;overflow:auto hidden"><= div id=3D"m_2344152423409506380gmail-avWBGd-5041" style=3D""><div dir=3D"lt=
r" style=3D""><div style=3D"font-family:Arial,Helvetica,sans-serif"><div cl= ass=3D"gmail_default"><font face=3D"verdana, sans-serif">After spending an = entirely unreasonable amount of time adding SO_TIMESTAMP kernel-mode UDP re= ceive timestamps to Windows ntpd and testing it, and a much shorter time wo= rking around the failure of=C2=A0essentially all FreeBSD releases prior to = 15.0 to support those timestamps for IPv6 with phk&#39;s SO_BINTIME, I&#39;=
ve fmerged a few years of fixes into a private test release of ntpd.</font>= </div><div class=3D"gmail_default"><font face=3D"verdana, sans-serif"><br><= /font></div><font face=3D"verdana, sans-serif">In addition to those issues,=
symmetric auth bugs in 4.2.8p18 and build issues with newer OpenSSL 3.x ha=
ve been corrected.<span class=3D"gmail_default">=C2=A0=C2=A0</span>KoDs fro=
m Chrony are once again respected.=C2=A0 IPv6-only clients using the pool n=
ow work.=C2=A0 A bug removing dynamic restrict entries generated from &quot= ;restrict source&quot; is fixed, and=C2=A0<span class=3D"gmail_default">str= ucture=C2=A0</span>alignment improved to<span class=3D"gmail_default">=C2= =A0</span>help speed restrictions checks on the input path.<span class=3D"g= mail_default">=C2=A0=C2=A0</span>Build breaks triggered by newer gcc, clang=
, and glibc are corrected.=C2=A0 Upstream appliances which have questionabl=
e packet poll claiming 1s can be used as sources again.=C2=A0 A longstandin=
g bug with the daemon loop going ha<span class=3D"gmail_default">y</span>wi=
re at poll intervals longer than 1024s (maxpoll 11 and up) is fixed.</font>= </div><div style=3D"font-family:Arial,Helvetica,sans-serif"><font face=3D"v= erdana, sans-serif"><br></font></div><div style=3D"font-family:Arial,Helvet= ica,sans-serif"><font face=3D"verdana, sans-serif"><a href=3D"https://bk.da= vehart.net/ntp-stable-dlh-260328/ChangeLog?PAGE=3Danno&amp;REV=3D69c85abbX0= AuaX-DsPHhmzo7DD_BHg" target=3D"_blank">https://bk.davehart.net/ntp-stable-= dlh-260328/ChangeLog?PAGE=3Danno&amp;REV=3D69c85abbX0AuaX-DsPHhmzo7DD_BHg</= a>=C2=A0has more details.<br><br></font></div><div style=3D"font-family:Ari= al,Helvetica,sans-serif"><font face=3D"verdana, sans-serif"><a href=3D"http= s://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz" target=3D"_blank">h= ttps://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz</a><br><br></font= ></div><div style=3D"font-family:Arial,Helvetica,sans-serif"><font face=3D"= verdana, sans-serif">Windows binaries that can be overlaid on an existing M= einberg ntpd install are in:</font></div><div style=3D"font-family:Arial,He= lvetica,sans-serif"><font face=3D"verdana, sans-serif"><br></font></div><di=
v style=3D"font-family:Arial,Helvetica,sans-serif"><font face=3D"verdana, s= ans-serif"><a href=3D"https://davehart.net/ntp/win/ntp-stable-dlh-260328.zi=
p" target=3D"_blank">https://davehart.net/ntp/win/ntp-stable-dlh-260328.zip= </a></font></div><div style=3D"font-family:Arial,Helvetica,sans-serif"><fon=
t face=3D"verdana, sans-serif"><br></font></div><div style=3D"font-family:A= rial,Helvetica,sans-serif"><font face=3D"verdana, sans-serif"><span class= =3D"gmail_default">To use those y</span>ou&#39;ll probably need to install =
a recent-enough OpenSSL=C2=A0such as=C2=A0<span class=3D"gmail_default">Shi= ning Light Productions&#39;=C2=A0</span><span class=3D"gmail_default"></spa= n><a href=3D"https://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi" ta= rget=3D"_blank">https://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi<= /a></font></div><div style=3D""><font style=3D""><br><font face=3D"verdana,=
sans-serif">Sadly, the Windows SO_TIMESTAMP support depends on setting reg= istry keys on each network adapter that are not easy to </font>find,<font f= ace=3D"verdana, sans-serif"> and require Administrator group membership to = modify. =C2=A0</font><span class=3D"gmail_default" style=3D"font-family:ver= dana,sans-serif">&quot;</span><font face=3D"monospace">ntpd=C2=A0--enable-u= dp-timestamps</font><span class=3D"gmail_default" style=3D"font-family:verd= ana,sans-serif">&quot;</span><font face=3D"verdana, sans-serif">=C2=A0from =
an elevated (Administrator) shell will do that, then the network interfaces=
need to be cycled or the machine restarted.=C2=A0 You can see the improvem= ent on both FreeBSD</font><span class=3D"gmail_default" style=3D"font-famil= y:verdana,sans-serif">=C2=A0</span><font face=3D"verdana, sans-serif">IPv6 = and Windows UDP receive</font><span class=3D"gmail_default" style=3D"font-f= amily:verdana,sans-serif">=C2=A0</span><font face=3D"verdana, sans-serif">t= imestamps by adding &quot;</font><font face=3D"monospace">logconfig =3Dalla= ll</font><font face=3D"verdana, sans-serif">&quot; or &quot;</font><font fa= ce=3D"monospace">logconfig +sysinfo</font><font face=3D"verdana, sans-serif= ">&quot; to ntp.conf, which will report after 3 hours the average differenc=
e between kernel receive timestamps and the one ntpd grabs.=C2=A0 Then</fon= t><span class=3D"gmail_default" style=3D"font-family:verdana,sans-serif">= =C2=A0</span><font face=3D"verdana, sans-serif">after another 6 hours, then=
another 12, and every 24 hours after that.</font><br><br></font></div><div=
style=3D"font-family:Arial,Helvetica,sans-serif"><div dir=3D"ltr" class=3D= "gmail_signature"><div dir=3D"ltr"><font color=3D"#666666" face=3D"trebuche=
t ms, sans-serif">Cheers,<br>Dave Hart</font></div></div></div></div><div s= tyle=3D"font-family:Arial,Helvetica,sans-serif"></div><div style=3D"font-fa= mily:Arial,Helvetica,sans-serif"></div></div></div></div><div id=3D"m_23441= 52423409506380gmail-avWBGd-5042" style=3D"font-family:&quot;Google Sans&quo= t;,Roboto,RobotoDraft,Helvetica,Arial,sans-serif;clear:both"></div></div></= div><div style=3D"font-family:&quot;Google Sans&quot;,Roboto,RobotoDraft,He= lvetica,Arial,sans-serif;clear:both"></div></div><div style=3D"font-family:= &quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-serif;font-= size:0.875rem;padding:0px;width:auto;border-bottom-left-radius:0px;border-b= ottom-right-radius:0px;border-top:none;margin:0px;background:transparent"><= div style=3D"border-top:0px;padding:0px"><div><div style=3D"margin-left:72p= x"></div></div><div style=3D"clear:both;margin:0px;padding:0px;border-top:n= one"><div></div><div id=3D"m_2344152423409506380gmail-avWBGd-5043"></div></= div></div></div></div></div></div></div></div></div></div><div style=3D"fon= t-family:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-se= rif"></div></div></div><div style=3D"font-family:&quot;Google Sans&quot;,Ro= boto,RobotoDraft,Helvetica,Arial,sans-serif;padding:16px 0px 32px 72px"><di= v><div><div style=3D"padding:0px;width:1436px"><div style=3D"color:rgb(34,3= 4,34);padding:0px;box-sizing:border-box;border-radius:1px;border:none;margi= n:0px"><div style=3D"height:auto;padding:0px;display:flex;line-height:20px"= ><div style=3D"margin-right:8px"><button aria-label=3D"Add reaction" style= =3D"margin-top:0px;background-image:initial;background-position:initial;bac= kground-size:initial;background-repeat:initial;background-origin:initial;ba= ckground-clip:initial;border-width:initial;border-style:none;border-color:i= nitial;display:inline-flex;line-height:inherit;outline:none;color:rgb(68,71= ,70)"></button></div></div></div></div></div></div></div><div style=3D"font= -family:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-ser= if"><div style=3D"margin:0px 0px 16px;padding:0px 16px;text-align:center;ba= ckground-image:url(&quot;&quot;);background-position:center center;backgrou= nd-repeat:no-repeat;background-size:20px;min-height:56px"></div></div></div= ></div></div></div></div></div></div></div></div></div></div><div style=3D"= font-family:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans= -serif"><div style=3D"display:flex;height:865px"><div style=3D"height:865px= ;min-width:56px;width:56px"><div role=3D"complementary" aria-label=3D"Side = panel" style=3D"background:rgb(248,250,253);border:none;box-sizing:border-b= ox;display:flex;height:865px;width:56px"><div style=3D"display:flex;margin-= bottom:56px"><div role=3D"tablist" style=3D"height:100px;outline:none;overf= low:hidden"><div id=3D"m_2344152423409506380gmail-gsc-gab-6" role=3D"tab" a= ria-label=3D"Calendar" aria-disabled=3D"false" aria-selected=3D"false" styl= e=3D"height:56px;outline:none;width:56px"></div><div id=3D"m_23441524234095= 06380gmail-gsc-gab-2" role=3D"tab" aria-label=3D"Keep" aria-disabled=3D"fal= se" aria-selected=3D"false" style=3D"height:56px;outline:none;width:56px"><= /div><div id=3D"m_2344152423409506380gmail-gsc-gab-4" role=3D"tab" aria-lab= el=3D"Tasks" aria-disabled=3D"false" aria-selected=3D"false" style=3D"heigh= t:56px;outline:none;width:56px"></div><div id=3D"m_2344152423409506380gmail= -gsc-gab-9" role=3D"tab" aria-label=3D"Contacts" aria-disabled=3D"false" ar= ia-selected=3D"false" style=3D"height:56px;outline:none;width:56px"></div><= div role=3D"tab" aria-label=3D"Get Add-ons" aria-selected=3D"false" id=3D"m= _2344152423409506380gmail-qJTzr" style=3D"height:56px;outline:none;width:56= px"></div></div></div></div><div style=3D"display:flex;height:56px;overflow= :hidden;width:56px"></div></div></div></div></div></div></div></div><div st= yle=3D"font-family:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Ari= al,sans-serif;width:1920px;height:945px;color:rgb(32,33,36);font-size:mediu= m"></div><div style=3D"font-family:&quot;Google Sans&quot;,Roboto,RobotoDra= ft,Helvetica,Arial,sans-serif;color:rgb(32,33,36);font-size:medium"><div></= div></div><div style=3D"font-family:&quot;Google Sans&quot;,Roboto,RobotoDr= aft,Helvetica,Arial,sans-serif;display:flex;width:1920px;height:945px;overf= low:hidden;color:rgb(32,33,36);font-size:medium"><div style=3D"min-height:1= px;height:945px;width:56px"></div><div><div style=3D"display:flex;width:179= 8px;height:945px"><div style=3D"height:945px"><div style=3D"float:right"><d=
iv style=3D"min-height:1px;float:left;width:333px;height:945px"><div style= =3D"float:left"><div style=3D"min-height:1px;float:left;width:328px;height:= 945px"><div style=3D"width:328px"><div style=3D"border-radius:8px 8px 0px 0= px"><div><div><div style=3D"float:none;width:328px;background:rgb(242,246,2= 52);border-radius:8px 8px 0px 0px"><div style=3D"background:none;border-rad= ius:8px 8px 0px 0px;margin:0px 0px 0px 3px"><div style=3D"background:none;p= adding:0px 0px 4px;border-radius:8px 0px 0px;margin:0px 3px 0px 0px"><div><= /div></div></div></div></div></div></div></div></div></div></div></div></di= v></div></div></div><div style=3D"font-family:&quot;Google Sans&quot;,Robot= o,RobotoDraft,Helvetica,Arial,sans-serif;min-height:100%;color:rgb(32,33,36= );font-size:medium"><div style=3D"width:1920px"><div><div style=3D"display:= flex;background-color:transparent"><div style=3D"overflow:hidden"><div styl= e=3D"border-radius:16px;margin-bottom:16px;overflow-y:hidden"><div><div><di= v><div id=3D"m_2344152423409506380gmail-:3" style=3D"background-image:initi= al;background-position:initial;background-size:initial;background-repeat:in= itial;background-origin:initial;background-clip:initial;overflow-y:scroll;p= adding-right:0px;height:817px"><div id=3D"m_2344152423409506380gmail-:1" st= yle=3D"padding:0px;vertical-align:bottom;min-height:627px"><div><div role= =3D"main"><div><div style=3D"background-image:initial;background-position:i= nitial;background-size:initial;background-repeat:initial;background-origin:= initial;background-clip:initial;min-height:64ex;min-width:502px;margin:0px;= padding-right:16px"><div><div style=3D"background-color:transparent;color:r= gb(34,34,34);min-width:502px;padding:0px"><div role=3D"list"><div role=3D"l= istitem" aria-expanded=3D"true" style=3D"padding-bottom:0px;max-width:10000= 0px;clear:both;outline:none"><div style=3D"margin-bottom:0px;border-width:0= px;border-top-style:solid;border-right-style:initial;border-bottom-style:in= itial;border-left-style:initial;border-top-color:rgb(239,239,239);border-ri= ght-color:initial;border-left-color:initial;border-bottom-color:initial;bor= der-radius:0px;float:left;width:1508px"><div style=3D"border-top:none;paddi= ng-top:0px;background-color:transparent;border-right:0px;border-bottom:0px = rgba(100,121,143,0.12);border-left:0px;border-radius:0px;margin-bottom:0px;= margin-left:0px;margin-right:0px"><div><div id=3D"m_2344152423409506380gmai= l-avWBGd-5024"><div id=3D"m_2344152423409506380gmail-avWBGd-5025"><div styl= e=3D"border-left:none;padding:0px;display:flex"><div style=3D"margin:0px;mi= n-width:0px;padding:0px 0px 20px;width:auto"><div><div id=3D"m_234415242340= 9506380gmail-:62i" style=3D"direction:ltr;margin:8px 0px 0px;padding:0px;fo= nt-size:0.875rem;overflow-x:hidden"><div id=3D"m_2344152423409506380gmail-:= 62h" style=3D"direction:ltr;font-variant-numeric:normal;font-variant-east-a= sian:normal;font-variant-alternates:normal;font-size-adjust:none;font-kerni= ng:auto;font-feature-settings:normal;font-stretch:normal;font-size:small;li= ne-height:1.5;font-family:Arial,Helvetica,sans-serif;overflow:auto hidden">= <div id=3D"m_2344152423409506380gmail-avWBGd-5041"><div dir=3D"ltr"><div><d=
iv class=3D"gmail_default"><font face=3D"verdana, sans-serif">After spendin=
g an entirely unreasonable amount of time adding SO_TIMESTAMP kernel-mode U=
DP receive timestamps to Windows ntpd and testing it, and a much shorter ti=
me working around the failure of=C2=A0essentially all FreeBSD releases prio=
r to 15.0 to support those timestamps for IPv6 with phk&#39;s SO_BINTIME, I= &#39;ve fmerged a few years of fixes into a private test release of ntpd.</= font></div><div class=3D"gmail_default"><font face=3D"verdana, sans-serif">= <br></font></div><font face=3D"verdana, sans-serif">In addition to those is= sues, symmetric auth bugs in 4.2.8p18 and build issues with newer OpenSSL 3=
.x have been corrected.<span class=3D"gmail_default">=C2=A0=C2=A0</span>KoD=
s from Chrony are once again respected.=C2=A0 IPv6-only clients using the p= ool now work.=C2=A0 A bug removing dynamic restrict entries generated from = &quot;restrict source&quot; is fixed, and=C2=A0<span class=3D"gmail_default= ">structure=C2=A0</span>alignment improved to<span class=3D"gmail_default">= =C2=A0</span>help speed restrictions checks on the input path.<span class= =3D"gmail_default">=C2=A0=C2=A0</span>Build breaks triggered by newer gcc, = clang, and glibc are corrected.=C2=A0 Upstream appliances which have questi= onable packet poll claiming 1s can be used as sources again.=C2=A0 A longst= anding bug with the daemon loop going ha<span class=3D"gmail_default">y</sp= an>wire at poll intervals longer than 1024s (maxpoll 11 and up) is fixed.</= font></div><div><font face=3D"verdana, sans-serif"><br></font></div><div><f= ont face=3D"verdana, sans-serif"><a href=3D"https://bk.davehart.net/ntp-sta= ble-dlh-260328/ChangeLog?PAGE=3Danno&amp;REV=3D69c85abbX0AuaX-DsPHhmzo7DD_B= Hg" target=3D"_blank">https://bk.davehart.net/ntp-stable-dlh-260328/ChangeL= og?PAGE=3Danno&amp;REV=3D69c85abbX0AuaX-DsPHhmzo7DD_BHg</a>=C2=A0has more d= etails.<br><br></font></div><div><font face=3D"verdana, sans-serif"><a href= =3D"https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz" target=3D"_= blank">https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz</a><br><b= r></font></div><div><font face=3D"verdana, sans-serif">Windows binaries tha=
t can be overlaid on an existing Meinberg ntpd install are in:</font></div>= <div><font face=3D"verdana, sans-serif"><br></font></div><div><font face=3D= "verdana, sans-serif"><a href=3D"https://davehart.net/ntp/win/ntp-stable-dl= h-260328.zip" target=3D"_blank">https://davehart.net/ntp/win/ntp-stable-dlh= -260328.zip</a></font></div><div><font face=3D"verdana, sans-serif"><br></f= ont></div><div><font face=3D"verdana, sans-serif"><span class=3D"gmail_defa= ult">To use those y</span>ou&#39;ll probably need to install a recent-enoug=
h OpenSSL=C2=A0such as=C2=A0<span class=3D"gmail_default">Shining Light Pro= ductions&#39;=C2=A0</span><span class=3D"gmail_default"></span><a href=3D"h= ttps://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi" target=3D"_blank= ">https://slproweb.com/download/Win64OpenSSL_Light-3_6_1.msi</a></font></di= v><div><font face=3D"verdana, sans-serif"><br>Sadly, the Windows SO_TIMESTA=
MP support depends on setting registry keys on each network adapter that ar=
e not easy to find, and require Administrator group membership to modify. = =C2=A0<span class=3D"gmail_default">&quot;</span>ntpd=C2=A0--enable-udp-tim= estamps<span class=3D"gmail_default">&quot;</span>=C2=A0from an elevated (A= dministrator) shell will do that, then the network interfaces need to be cy= cled or the machine restarted.=C2=A0 You can see the improvement on both Fr= eeBSD<span class=3D"gmail_default">=C2=A0</span>IPv6 and Windows UDP receiv= e<span class=3D"gmail_default">=C2=A0</span>timestamps by adding &quot;logc= onfig =3Dallall&quot; or &quot;logconfig +sysinfo&quot; to ntp.conf, which = will report after 3 hours the average difference between kernel receive tim= estamps and the one ntpd grabs.=C2=A0 Then<span class=3D"gmail_default">=C2= =A0</span>after another 6 hours, then another 12, and every 24 hours after = that.<br><br></font></div><div><div dir=3D"ltr" class=3D"gmail_signature"><= div dir=3D"ltr"><font color=3D"#666666" face=3D"trebuchet ms, sans-serif">C= heers,<br>Dave Hart</font></div></div></div></div><div></div><div></div></d= iv></div></div><div id=3D"m_2344152423409506380gmail-avWBGd-5042" style=3D"= clear:both"></div></div></div><div style=3D"clear:both"></div></div><div st= yle=3D"font-size:0.875rem;padding:0px;width:auto;border-bottom-left-radius:= 0px;border-bottom-right-radius:0px;border-top:none;margin:0px;background:tr= ansparent"><div style=3D"border-top:0px;padding:0px"><div><div style=3D"mar= gin-left:72px"></div></div><div style=3D"clear:both;margin:0px;padding:0px;= border-top:none"><div></div><div id=3D"m_2344152423409506380gmail-avWBGd-50= 43"></div></div></div></div></div></div></div></div></div></div></div><div>= </div></div></div><div style=3D"padding:16px 0px 32px 72px"><div><div><div = style=3D"padding:0px;width:1436px"><div style=3D"color:rgb(34,34,34);paddin= g:0px;box-sizing:border-box;border-radius:1px;border:none;margin:0px"><div = style=3D"height:auto;padding:0px;display:flex;line-height:20px"><div style= =3D"margin-right:8px"><button aria-label=3D"Add reaction" style=3D"margin-t= op:0px;background-image:initial;background-position:initial;background-size= :initial;background-repeat:initial;background-origin:initial;background-cli= p:initial;border-width:initial;border-style:none;border-color:initial;displ= ay:inline-flex;line-height:inherit;outline:none;color:rgb(68,71,70)"></butt= on></div></div></div></div></div></div></div><div><div style=3D"margin:0px = 0px 16px;padding:0px 16px;text-align:center;background-image:url(&quot;&quo= t;);background-position:center center;background-repeat:no-repeat;backgroun= d-size:20px;min-height:56px"></div></div></div></div></div></div></div></di= v></div></div></div></div></div><div><div style=3D"display:flex;height:865p= x"><div style=3D"height:865px;min-width:56px;width:56px"><div role=3D"compl= ementary" aria-label=3D"Side panel" style=3D"background:rgb(248,250,253);bo= rder:none;box-sizing:border-box;display:flex;height:865px;width:56px"><div = style=3D"display:flex;margin-bottom:56px"><div role=3D"tablist" style=3D"he= ight:100px;outline:none;overflow:hidden"><div id=3D"m_2344152423409506380gm= ail-gsc-gab-6" role=3D"tab" aria-label=3D"Calendar" aria-disabled=3D"false"=
aria-selected=3D"false" style=3D"height:56px;outline:none;width:56px"></di= v><div id=3D"m_2344152423409506380gmail-gsc-gab-2" role=3D"tab" aria-label= =3D"Keep" aria-disabled=3D"false" aria-selected=3D"false" style=3D"height:5= 6px;outline:none;width:56px"></div><div id=3D"m_2344152423409506380gmail-gs= c-gab-4" role=3D"tab" aria-label=3D"Tasks" aria-disabled=3D"false" aria-sel= ected=3D"false" style=3D"height:56px;outline:none;width:56px"></div><div id= =3D"m_2344152423409506380gmail-gsc-gab-9" role=3D"tab" aria-label=3D"Contac= ts" aria-disabled=3D"false" aria-selected=3D"false" style=3D"height:56px;ou= tline:none;width:56px"></div><div role=3D"tab" aria-label=3D"Get Add-ons" a= ria-selected=3D"false" id=3D"m_2344152423409506380gmail-qJTzr" style=3D"hei= ght:56px;outline:none;width:56px"></div></div></div></div><div style=3D"dis= play:flex;height:56px;overflow:hidden;width:56px"></div></div></div></div><= /div></div></div></div><div style=3D"font-family:&quot;Google Sans&quot;,Ro= boto,RobotoDraft,Helvetica,Arial,sans-serif;width:1920px;height:945px;color= :rgb(32,33,36);font-size:medium"></div><div style=3D"font-family:&quot;Goog=
le Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-serif;color:rgb(32,33= ,36);font-size:medium"><div></div></div><div style=3D"font-family:&quot;Goo= gle Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-serif;display:flex;w= idth:1920px;height:945px;overflow:hidden;color:rgb(32,33,36);font-size:medi= um"><div style=3D"min-height:1px;height:945px;width:56px"></div><div><div s= tyle=3D"display:flex;width:1798px;height:945px"><div style=3D"height:945px"= ><div style=3D"float:right"><div style=3D"min-height:1px;float:left;width:3= 33px;height:945px"><div style=3D"float:left"><div style=3D"min-height:1px;f= loat:left;width:328px;height:945px"><div style=3D"width:328px"><div style= =3D"border-radius:8px 8px 0px 0px"><div><div><div style=3D"float:none;width= :328px;background:rgb(242,246,252);border-radius:8px 8px 0px 0px"><div styl= e=3D"background:none;border-radius:8px 8px 0px 0px;margin:0px 0px 0px 3px">= <div style=3D"background:none;padding:0px 0px 4px;border-radius:8px 0px 0px= ;margin:0px 3px 0px 0px"><div></div></div></div></div></div></div></div></d= iv></div></div></div></div></div></div></div></div><div style=3D"font-famil= y:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-serif;min= -height:100%;color:rgb(32,33,36);font-size:medium"><div style=3D"width:1920= px"><div><div style=3D"display:flex;background-color:transparent"><div styl= e=3D"overflow:hidden"><div style=3D"border-radius:16px;margin-bottom:16px;o= verflow-y:hidden"><div><div><div><div id=3D"m_2344152423409506380gmail-:3" = style=3D"background-image:initial;background-position:initial;background-si= ze:initial;background-repeat:initial;background-origin:initial;background-c= lip:initial;overflow-y:scroll;padding-right:0px;height:817px"><div id=3D"m_= 2344152423409506380gmail-:1" style=3D"padding:0px;vertical-align:bottom;min= -height:627px"><div><div role=3D"main"><div><div style=3D"background-image:= initial;background-position:initial;background-size:initial;background-repe= at:initial;background-origin:initial;background-clip:initial;min-height:64e= x;min-width:502px;margin:0px;padding-right:16px"><div><div style=3D"backgro= und-color:transparent;color:rgb(34,34,34);min-width:502px;padding:0px"><div=
role=3D"list"><div role=3D"listitem" aria-expanded=3D"true" style=3D"paddi= ng-bottom:0px;max-width:100000px;clear:both;outline:none"><div style=3D"mar= gin-bottom:0px;border-width:0px;border-top-style:solid;border-right-style:i= nitial;border-bottom-style:initial;border-left-style:initial;border-top-col= or:rgb(239,239,239);border-right-color:initial;border-left-color:initial;bo= rder-bottom-color:initial;border-radius:0px;float:left;width:1508px"><div s= tyle=3D"border-top:none;padding-top:0px;background-color:transparent;border= -right:0px;border-bottom:0px rgba(100,121,143,0.12);border-left:0px;border-= radius:0px;margin-bottom:0px;margin-left:0px;margin-right:0px"><div><div id= =3D"m_2344152423409506380gmail-avWBGd-5024"><div id=3D"m_234415242340950638= 0gmail-avWBGd-5025"><div style=3D"border-left:none;padding:0px;display:flex= "><div style=3D"margin:0px;min-width:0px;padding:0px 0px 20px;width:auto"><= div><div id=3D"m_2344152423409506380gmail-:62i" style=3D"direction:ltr;marg= in:8px 0px 0px;padding:0px;font-size:0.875rem;overflow-x:hidden"><div id=3D= "m_2344152423409506380gmail-:62h" style=3D"direction:ltr;font-variant-numer= ic:normal;font-variant-east-asian:normal;font-variant-alternates:normal;fon= t-size-adjust:none;font-kerning:auto;font-feature-settings:normal;font-stre= tch:normal;font-size:small;line-height:1.5;font-family:Arial,Helvetica,sans= -serif;overflow:auto hidden"><div id=3D"m_2344152423409506380gmail-avWBGd-5= 041"><div dir=3D"ltr"><div><div class=3D"gmail_default"><font face=3D"verda= na, sans-serif">After spending an entirely unreasonable amount of time addi=
ng SO_TIMESTAMP kernel-mode UDP receive timestamps to Windows ntpd and test= ing it, and a much shorter time working around the failure of=C2=A0essentia= lly all FreeBSD releases prior to 15.0 to support those timestamps for IPv6=
with phk&#39;s SO_BINTIME, I&#39;ve fmerged a few years of fixes into a pr= ivate test release of ntpd.</font></div><div class=3D"gmail_default"><font = face=3D"verdana, sans-serif"><br></font></div><font face=3D"verdana, sans-s= erif">In addition to those issues, symmetric auth bugs in 4.2.8p18 and buil=
d issues with newer OpenSSL 3.x have been corrected.<span class=3D"gmail_de= fault">=C2=A0=C2=A0</span>KoDs from Chrony are once again respected.=C2=A0 = IPv6-only clients using the pool now work.=C2=A0 A bug removing dynamic res= trict entries generated from &quot;restrict source&quot; is fixed, and=C2= =A0<span class=3D"gmail_default">structure=C2=A0</span>alignment improved t= o<span class=3D"gmail_default">=C2=A0</span>help speed restrictions checks =
on the input path.<span class=3D"gmail_default">=C2=A0=C2=A0</span>Build br= eaks triggered by newer gcc, clang, and glibc are corrected.=C2=A0 Upstream=
appliances which have questionable packet poll claiming 1s can be used as = sources again.=C2=A0 A longstanding bug with the daemon loop going ha<span = class=3D"gmail_default">y</span>wire at poll intervals longer than 1024s (m= axpoll 11 and up) is fixed.</font></div><div><font face=3D"verdana, sans-se= rif"><br></font></div><div><font face=3D"verdana, sans-serif"><a href=3D"ht= tps://bk.davehart.net/ntp-stable-dlh-260328/ChangeLog?PAGE=3Danno&amp;REV= =3D69c85abbX0AuaX-DsPHhmzo7DD_BHg" target=3D"_blank">https://bk.davehart.ne= t/ntp-stable-dlh-260328/ChangeLog?PAGE=3Danno&amp;REV=3D69c85abbX0AuaX-DsPH= hmzo7DD_BHg</a>=C2=A0has more details.<br><br></font></div><div><font face= =3D"verdana, sans-serif"><a href=3D"https://davehart.net/ntp/test/ntp-stabl= e-dlh-260328.tar.gz" target=3D"_blank">https://davehart.net/ntp/test/ntp-st= able-dlh-260328.tar.gz</a><br><br></font></div><div><font face=3D"verdana, = sans-serif">Windows binaries that can be overlaid on an existing Meinberg n= tpd install are in:</font></div><div><font face=3D"verdana, sans-serif"><br= ></font></div><div><font face=3D"verdana, sans-serif"><a href=3D"https://da= vehart.net/ntp/win/ntp-stable-dlh-260328.zip" target=3D"_blank">https://dav= ehart.net/ntp/win/ntp-stable-dlh-260328.zip</a></font></div><div><font face= =3D"verdana, sans-serif"><br></font></div><div><font face=3D"verdana, sans-= serif"><span class=3D"gmail_default">To use those y</span>ou&#39;ll probabl=
y need to install a recent-enough OpenSSL=C2=A0such as=C2=A0<span class=3D"= gmail_default">Shining Light Productions&#39;=C2=A0</span><span class=3D"gm= ail_default"></span><a href=3D"https://slproweb.com/download/Win64OpenSSL_L= ight-3_6_1.msi" target=3D"_blank">https://slproweb.com/download/Win64OpenSS= L_Light-3_6_1.msi</a></font></div><div><font face=3D"verdana, sans-serif"><= br>Sadly, the Windows SO_TIMESTAMP support depends on setting registry keys=
on each network adapter that are not easy to find, and require Administrat=
or group membership to modify. =C2=A0<span class=3D"gmail_default">&quot;</= span>ntpd=C2=A0--enable-udp-timestamps<span class=3D"gmail_default">&quot;<= /span>=C2=A0from an elevated (Administrator) shell will do that, then the n= etwork interfaces need to be cycled or the machine restarted.=C2=A0 You can=
see the improvement on both FreeBSD<span class=3D"gmail_default">=C2=A0</s= pan>IPv6 and Windows UDP receive<span class=3D"gmail_default">=C2=A0</span>= timestamps by adding &quot;logconfig =3Dallall&quot; or &quot;logconfig +sy= sinfo&quot; to ntp.conf, which will report after 3 hours the average differ= ence between kernel receive timestamps and the one ntpd grabs.=C2=A0 Then<s= pan class=3D"gmail_default">=C2=A0</span>after another 6 hours, then anothe=
r 12, and every 24 hours after that.<br><br></font></div><div><div dir=3D"l= tr" class=3D"gmail_signature"><div dir=3D"ltr"><font color=3D"#666666" face= =3D"trebuchet ms, sans-serif">Cheers,<br>Dave Hart</font></div></div></div>= </div><div></div><div></div></div></div></div><div id=3D"m_2344152423409506= 380gmail-avWBGd-5042" style=3D"clear:both"></div></div></div><div style=3D"= clear:both"></div></div><div style=3D"font-size:0.875rem;padding:0px;width:= auto;border-bottom-left-radius:0px;border-bottom-right-radius:0px;border-to= p:none;margin:0px;background:transparent"><div style=3D"border-top:0px;padd= ing:0px"><div><div style=3D"margin-left:72px"></div></div><div style=3D"cle= ar:both;margin:0px;padding:0px;border-top:none"><div></div><div id=3D"m_234= 4152423409506380gmail-avWBGd-5043"></div></div></div></div></div></div></di= v></div></div></div></div><div></div></div></div><div style=3D"padding:16px=
0px 32px 72px"><div><div><div style=3D"padding:0px;width:1436px"><div styl= e=3D"color:rgb(34,34,34);padding:0px;box-sizing:border-box;border-radius:1p= x;border:none;margin:0px"><div style=3D"height:auto;padding:0px;display:fle= x;line-height:20px"><div style=3D"margin-right:8px"><button aria-label=3D"A=
dd reaction" style=3D"margin-top:0px;background-image:initial;background-po= sition:initial;background-size:initial;background-repeat:initial;background= -origin:initial;background-clip:initial;border-width:initial;border-style:n= one;border-color:initial;display:inline-flex;line-height:inherit;outline:no= ne;color:rgb(68,71,70)"></button></div></div></div></div></div></div></div>= <div><div style=3D"margin:0px 0px 16px;padding:0px 16px;text-align:center;b= ackground-image:url(&quot;&quot;);background-position:center center;backgro= und-repeat:no-repeat;background-size:20px;min-height:56px"></div></div></di= v></div></div></div></div></div></div></div></div></div></div><div><div sty= le=3D"display:flex;height:865px"><div style=3D"height:865px;min-width:56px;= width:56px"><div role=3D"complementary" aria-label=3D"Side panel" style=3D"= background:rgb(248,250,253);border:none;box-sizing:border-box;display:flex;= height:865px;width:56px"><div style=3D"display:flex;margin-bottom:56px"><di=
v role=3D"tablist" style=3D"height:100px;outline:none;overflow:hidden"><div=
id=3D"m_2344152423409506380gmail-gsc-gab-6" role=3D"tab" aria-label=3D"Cal= endar" aria-disabled=3D"false" aria-selected=3D"false" style=3D"height:56px= ;outline:none;width:56px"></div><div id=3D"m_2344152423409506380gmail-gsc-g= ab-2" role=3D"tab" aria-label=3D"Keep" aria-disabled=3D"false" aria-selecte= d=3D"false" style=3D"height:56px;outline:none;width:56px"></div><div id=3D"= m_2344152423409506380gmail-gsc-gab-4" role=3D"tab" aria-label=3D"Tasks" ari= a-disabled=3D"false" aria-selected=3D"false" style=3D"height:56px;outline:n= one;width:56px"></div><div id=3D"m_2344152423409506380gmail-gsc-gab-9" role= =3D"tab" aria-label=3D"Contacts" aria-disabled=3D"false" aria-selected=3D"f= alse" style=3D"height:56px;outline:none;width:56px"></div><div role=3D"tab"=
aria-label=3D"Get Add-ons" aria-selected=3D"false" id=3D"m_234415242340950= 6380gmail-qJTzr" style=3D"height:56px;outline:none;width:56px"></div></div>= </div></div><div style=3D"display:flex;height:56px;overflow:hidden;width:56= px"></div></div></div></div></div></div></div></div><div style=3D"font-fami= ly:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Arial,sans-serif;wi= dth:1920px;height:945px;color:rgb(32,33,36);font-size:medium"></div><div st= yle=3D"font-family:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Ari= al,sans-serif;color:rgb(32,33,36);font-size:medium"><div></div></div><div s= tyle=3D"font-family:&quot;Google Sans&quot;,Roboto,RobotoDraft,Helvetica,Ar= ial,sans-serif;display:flex;width:1920px;height:945px;overflow:hidden;color= :rgb(32,33,36);font-size:medium"><div style=3D"min-height:1px;height:945px;= width:56px"></div><div><div style=3D"display:flex;width:1798px;height:945px= "><div style=3D"height:945px"><div style=3D"float:right"><div style=3D"min-= height:1px;float:left;width:333px;height:945px"><div style=3D"float:left"><= div style=3D"min-height:1px;float:left;width:328px;height:945px"><div style= =3D"width:328px"><div style=3D"border-radius:8px 8px 0px 0px"><div><div><di=
v style=3D"float:none;width:328px;background:rgb(242,246,252);border-radius= :8px 8px 0px 0px"><div style=3D"background:none;border-radius:8px 8px 0px 0= px;margin:0px 0px 0px 3px"><div style=3D"background:none;padding:0px 0px 4p= x;border-radius:8px 0px 0px;margin:0px 3px 0px 0px"><div></div></div></div>= </div></div></div></div></div></div></div></div></div></div></div></div></d= iv><br></div></div>
</div>

--0000000000008a490b064e64e411--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--00000000000049f3d6064e69fe45
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart <[email protected]> wrote=
:

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Hello Dave,

As a smoke test, I substituted this tarball into the Slackware64 15.0 NTP package, built, and installed on my server. So far it passes the smoke test
-- syncs and keeps time as well as before.

The only change necessary was to delete the file fix_ipv6_locallink_regression.diff from the Slackware build. This was a workaround for bug 3928. The patch failed for the good reason that you
fixed bug 3928 in your private release.

Cheers!
Edward

--00000000000049f3d6064e69fe45
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart &lt;<a hr= ef=3D"mailto:[email protected]">[email protected]</a>&gt; wrote:<br>&gt; =
<a href=3D"https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz">http= s://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz</a><br><br>Hello Dav= e,<br><br>As a smoke test, I substituted this tarball into the Slackware64 = 15.0 NTP package, built, and installed on my server. So far it passes the s= moke test -- syncs and keeps time as well as before.<br><br>The only change=
necessary was to delete the file fix_ipv6_locallink_regression.diff from t=
he Slackware build. This was a workaround for bug 3928. The patch failed fo=
r the good reason that you fixed bug 3928 in your private release.<br><br>C= heers!<br>Edward</div>

--00000000000049f3d6064e69fe45--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--000000000000c9c6d5064f7f7fd4
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Edward,

Chip here. I tested the stable release on my raspberry and the configure script failed.
The configure script failed to set the HAVE_PPSAPI and the HAVE_THREADS in
the config.h

I tested the config process on ntp-4.2.8p15 and the config.h is set correctly.

The result of these failures, I could not use the pps to lock on the gps devices PPS signal

Be very careful, because I dont think you are compiling what you think you
are compiling.

Chip


On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire <[email protected]> wr= ote:

On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart <[email protected]> wro=

te:

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Hello Dave,

As a smoke test, I substituted this tarball into the Slackware64 15.0 NTP package, built, and installed on my server. So far it passes the smoke te=

st

-- syncs and keeps time as well as before.

The only change necessary was to delete the file fix_ipv6_locallink_regression.diff from the Slackware build. This was a workaround for bug 3928. The patch failed for the good reason that you
fixed bug 3928 in your private release.

Cheers!
Edward

--=20
Ralph "Chip" Blach
(919) 260-0097

--000000000000c9c6d5064f7f7fd4
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Edward,</div><div><br></div><div>Chip here.=C2=A0 I t= ested the stable release on my raspberry and the configure script failed.</= div><div>The configure script failed to set the HAVE_PPSAPI and the HAVE_TH= READS in the config.h</div><div><br></div><div>I tested the config process = on=C2=A0=C2=A0ntp-4.2.8p15=C2=A0 and the config.h is set correctly.</div><d= iv><br></div><div>The result of these failures, I could not use the pps to = lock on the gps devices PPS signal</div><div><br></div><div>Be very careful=
, because I dont think you are compiling what you think you are compiling.<= /div><div><br></div><div>Chip</div><div><br></div></div><br><div class=3D"g= mail_quote gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On = Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire &lt;<a href=3D"mailto:me= [email protected]">[email protected]</a>&gt; wrote:<br></div><blockquote class= =3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rg= b(204,204,204);padding-left:1ex"><div dir=3D"ltr">On Wed, Apr 1, 2026 at 7:= 05=E2=80=AFAM Dave Hart &lt;<a href=3D"mailto:[email protected]" target=3D= "_blank">[email protected]</a>&gt; wrote:<br>&gt; <a href=3D"https://daveh= art.net/ntp/test/ntp-stable-dlh-260328.tar.gz" target=3D"_blank">https://da= vehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz</a><br><br>Hello Dave,<br>= <br>As a smoke test, I substituted this tarball into the Slackware64 15.0 N=
TP package, built, and installed on my server. So far it passes the smoke t= est -- syncs and keeps time as well as before.<br><br>The only change neces= sary was to delete the file fix_ipv6_locallink_regression.diff from the Sla= ckware build. This was a workaround for bug 3928. The patch failed for the = good reason that you fixed bug 3928 in your private release.<br><br>Cheers!= <br>Edward</div>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div>

--000000000000c9c6d5064f7f7fd4--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --------------HBHiz5e85OPPamRO4qXSIvr0
Content-Type: multipart/mixed; boundary="------------vlJQUA7XzcQ75kGE4EE6zfyb";
protected-headers="v1"
From: Martin Burnicki <[email protected]>
To: Ralph Blach <[email protected]>, Edward McGuire <[email protected]>
Cc: Dave Hart <[email protected]>,
NTPD Questions List <[email protected]>
Message-ID: <[email protected]>
Subject: Re: Test release of ntpd fixing a boatload of bugs
References: <CAMbSiYDeyX6q2j-DjWoogund+RDgQAZW3OF9u--xk+mCE=[email protected]>
<CAGdkUnVkXFkEMncGBRAjn3PAV6CgnSq13sdX6RRVhg0+iZwDuA@mail.gmail.com>
<CADbvWFPdhnov3GQBgG=[email protected]> In-Reply-To: <CADbvWFPdhnov3GQBgG=[email protected]>

--------------vlJQUA7XzcQ75kGE4EE6zfyb
Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: base64

UmFscGgsIEVkd2FyZCwNCg0KUmFscGggQmxhY2ggd3JvdGU6DQo+IEVkd2FyZCwNCj4gDQo+ IENoaXAgaGVyZS7CoCBJIHRlc3RlZCB0aGUgc3RhYmxlIHJlbGVhc2Ugb24gbXkgcmFzcGJl cnJ5IGFuZCB0aGUgDQo+IGNvbmZpZ3VyZSBzY3JpcHQgZmFpbGVkLg0KPiBUaGUgY29uZmln dXJlIHNjcmlwdCBmYWlsZWQgdG8gc2V0IHRoZSBIQVZFX1BQU0FQSSBhbmQgdGhlIEhBVkVf VEhSRUFEUyANCj4gaW4gdGhlIGNvbmZpZy5oDQo+IA0KPiBJIHRlc3RlZCB0aGUgY29uZmln IHByb2Nlc3Mgb27CoMKgbnRwLTQuMi44cDE1wqAgYW5kIHRoZSBjb25maWcuaCBpcyBzZXQg DQo+IGNvcnJlY3RseS4NCj4gDQo+IFRoZSByZXN1bHQgb2YgdGhlc2UgZmFpbHVyZXMsIEkg Y291bGQgbm90IHVzZSB0aGUgcHBzIHRvIGxvY2sgb24gdGhlIGdwcyANCj4gZGV2aWNlcyBQ UFMgc2lnbmFsDQo+IA0KPiBCZSB2ZXJ5IGNhcmVmdWwsIGJlY2F1c2UgSSBkb250IHRoaW5r IHlvdSBhcmUgY29tcGlsaW5nIHdoYXQgeW91IHRoaW5rIA0KPiB5b3UgYXJlIGNvbXBpbGlu Zy4NCj4gDQo+IENoaXANCg0KSSBjYW4gY29uZmlybSB0aGF0IFBQUyBzdXBwb3J0IGlzIG5v dCBkZXRlY3RlZCBieSBEYXZlJ3MgcGFja2FnZSwgYW5kIA0KdGhlcmVmb3JlLCBIQVZFX1BQ U0FQSSBpcyBub3QgZGVmaW5lZCwgZXZlbiB0aG91Z2ggdGhlIHBwcy10b29scy1kZXZlbCAN CnBhY2thZ2UgaXMgaW5zdGFsbGVkLCBhbmQgdGltZXBwcy5oIGlzIGF2YWlsYWJsZSBpbiAv dXNyL2luY2x1ZGUvc3lzLg0KDQoNCk1hcnRpbg0KLS0gDQpNYXJ0aW4gQnVybmlja2kNCg0K U2VuaW9yIFNvZnR3YXJlIEVuZ2luZWVyDQoNCkVtYWlsOiBtYXJ0aW4uYnVybmlja2lAbWVp bmJlcmcuZGUNClBob25lOiArNDkgNTI4MSA5MzA5LTQxNA0KTGlua2VkaW46IGh0dHBzOi8v d3d3LmxpbmtlZGluLmNvbS9pbi9tYXJ0aW5idXJuaWNraS8NCg0KTUVJTkJFUkcgRnVua3Vo cmVuIEdtYkggJiBDby4gS0cNCkxhbmdlIFdhbmQgOQ0KMzE4MTIgQmFkIFB5cm1vbnQsIEdl cm1hbnkNCg0KUmVnaXN0cnkgQ291cnQ6IEFtdHNnZXJpY2h0IEhhbm5vdmVyIDE3IEhSQSAx MDAzMjINCk1hbmFnaW5nIERpcmVjdG9yczogTmF0YWxpZSBNZWluYmVyZywgRGFuaWVsIEJv bGR0LCBBbmRyZSBIYXJ0bWFubiwgDQpIZWlrbyBHZXJzdHVuZw0KDQpXZWJzaXRlczogaHR0 cHM6Ly93d3cubWVpbmJlcmcuZGUgIGh0dHBzOi8vd3d3Lm1laW5iZXJnZ2xvYmFsLmNvbQ0K DQpNZWluYmVyZyAtIFRoZSBTeW5jaHJvbml6YXRpb24gRXhwZXJ0cy4NCg0K

--------------vlJQUA7XzcQ75kGE4EE6zfyb--

--------------HBHiz5e85OPPamRO4qXSIvr0
Content-Type: application/pgp-signature; name="OpenPGP_signature.asc" Content-Description: OpenPGP digital signature
Content-Disposition: attachment; filename="OpenPGP_signature.asc"

-----BEGIN PGP SIGNATURE-----

wsF5BAABCAAjFiEEB4vtcjCE6s9kni+wQXIcDHUJ+cgFAmnflKkFAwAAAAAACgkQQXIcDHUJ+cgm jw/+KCPXSMjl1u08nBxZAl60OjwT9i+Gbv25OJPPuJK1CjOX+u0sndHLEb3YKzYghvL3tM0LfRC3 mho7u1lCD6o8C8l4wMfAaVDnjB/mTUkN7Pqfv4p0bbQJmFaU8Bv5UEBuI0fKor8GjGvFvkCFt7pe l/3nQUBxJ5KYlVVAE276B+fYczYUWTRb092RYrVYX7+RueSlZgzDguRWq5BjfcZB7KvRPUTeIVuv RPpSQE+s9nGGvT7bwgK2z4yzPFscIbJuh2A60NtMUOfdrJdnhAYueTndjYYcK0CKUGPp0sxFjegd xqPr396EBtCGxwn95cEbYAjjUvUjlECGOo7Ftog+1udAJIq3sVNRLT9ZVr8tmDZx6peljYbFcwlV FQFf05ntnxnsmlol1Jf5Sko2sLasNj4t++m68q1Wqf2PEqONw//aBMEeqI7KXo9qGI82QtHBV9ox JbGPPpu7YQ/thsGL1ZR9nePoK3FNupMwD1l5MvnzvASHVzTwlJiJLdhcKck9ly+jx+VUFW0iqeIG 0bpQKuWXU4940pCt/PyFYmMoBMqlgpYJFBKppA4pConGkgOpMX/zLaYaMVG1uWUdubDSVNHUrnkI /U1X7t6M2XIqwrQ7PNjd0ttdlHBoA7byKz1104kl52f2UHvl4zeMq47QquypsSsURYTPKQ5VvUH8 KGY=
=CbwQ
-----END PGP SIGNATURE-----

--------------HBHiz5e85OPPamRO4qXSIvr0--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--0000000000008f3bb7064f85727b
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Edward and Dave,

I found part of the problem

From the configure script ac_cv_c_inline must be set to inline and it is

not. The configure process is very
opaque on how this variable get set. I hope you have some insight


# Check for header timepps.h, if found then we have PPS API (Draft RFC)
stuff.

# The PPSAPI headers need "inline" ($ac_cv_c_inline=3D'inline')
# The PPSAPI needs struct timespec.
# The PPSAPI also needs a timepps header.

Chip



On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire <[email protected]> wr= ote:

On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart <[email protected]> wro=

te:

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Hello Dave,

As a smoke test, I substituted this tarball into the Slackware64 15.0 NTP package, built, and installed on my server. So far it passes the smoke te=

st

-- syncs and keeps time as well as before.

The only change necessary was to delete the file fix_ipv6_locallink_regression.diff from the Slackware build. This was a workaround for bug 3928. The patch failed for the good reason that you
fixed bug 3928 in your private release.

Cheers!
Edward

--=20
Ralph "Chip" Blach
(919) 260-0097

--0000000000008f3bb7064f85727b
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Edward and Dave,</div><div><br></div><div>I found par=
t of the problem</div><div><br></div><div>From the configure script ac_cv_c= _inline must be set to inline and it is not.=C2=A0 The configure process is=
very=C2=A0</div><div>opaque on how this variable get set.=C2=A0 I hope you=
have some insight</div><div><br></div><div><br></div><div># Check for head=
er timepps.h, if found then we have PPS API (Draft RFC) stuff.<br><br># The=
PPSAPI headers need &quot;inline&quot; ($ac_cv_c_inline=3D&#39;inline&#39;= )<br># The PPSAPI needs struct timespec.<br># The PPSAPI also needs a timep=
ps header.</div><div><br></div><div>Chip</div><div><br><br></div></div><br>= <div class=3D"gmail_quote gmail_quote_container"><div dir=3D"ltr" class=3D"= gmail_attr">On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire &lt;<a hr= ef=3D"mailto:[email protected]">[email protected]</a>&gt; wrote:<br></div><bl= ockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-lef= t:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr">On Wed, Apr=
1, 2026 at 7:05=E2=80=AFAM Dave Hart &lt;<a href=3D"mailto:davehart@gmail.= com" target=3D"_blank">[email protected]</a>&gt; wrote:<br>&gt; <a href=3D= "https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz" target=3D"_bla= nk">https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz</a><br><br>H= ello Dave,<br><br>As a smoke test, I substituted this tarball into the Slac= kware64 15.0 NTP package, built, and installed on my server. So far it pass=
es the smoke test -- syncs and keeps time as well as before.<br><br>The onl=
y change necessary was to delete the file fix_ipv6_locallink_regression.dif=
f from the Slackware build. This was a workaround for bug 3928. The patch f= ailed for the good reason that you fixed bug 3928 in your private release.<= br><br>Cheers!<br>Edward</div>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div>

--0000000000008f3bb7064f85727b--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--000000000000c144f6064fab84fb
Content-Type: text/plain; charset="UTF-8"

I am not familiar with `autoconf` so I don't follow the logic in ` configure.ac` very well. But I see there were changes related to
`syscall.h`, and I see the resulting `configure` no longer sets `ac_cv_c_inline`.

Dave, are you able to compare the output of

$ grep -r ac_cv_c_inline ntp-4.2.8p18
$ grep -r ac_cv_c_inline ntp-stable-dlh-260328

and do you have any insight into why `ac_cv_c_inline` is set in ntp-4.2.8p18/configure but is not set in ntp-stable-dlh-260328/configure?

Cheers!
Edward

P.S. Chip I'm sorry for the double post. Some of the names were lost on the message I first replied to.

--000000000000c144f6064fab84fb
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">I am not familiar with `autoconf` so I don&#39;t follow th=
e logic in `<a href=3D"http://configure.ac">configure.ac</a>` very well. Bu=
t I see there were changes related to `syscall.h`, and I see the resulting = `configure` no longer sets `ac_cv_c_inline`.<br><br>Dave, are you able to c= ompare the output of<br><br>$ grep -r ac_cv_c_inline ntp-4.2.8p18<br>$ grep=
-r ac_cv_c_inline ntp-stable-dlh-260328<br><br>and do you have any insight=
into why `ac_cv_c_inline` is set in ntp-4.2.8p18/configure but is not set =
in ntp-stable-dlh-260328/configure?<br><br>Cheers!<br>Edward<div><br></div>= <div>P.S. Chip I&#39;m sorry for the double post. Some of the names were lo=
st on the message I first replied to.</div></div>

--000000000000c144f6064fab84fb--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--00000000000093424b064fabaeb6
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

I would not use the slackware ntpd package. Can you please send me a patch file and I can make those changes to ntp-4.2.8p18.
There is a big hint on this when slackware removed the PPS api

Chip

On Fri, Apr 17, 2026 at 1:45=E2=80=AFPM Edward McGuire <[email protected]> w= rote:

I am not familiar with `autoconf` so I don't follow the logic in ` configure.ac` very well. But I see there were changes related to
`syscall.h`, and I see the resulting `configure` no longer sets `ac_cv_c_inline`.

Dave, are you able to compare the output of

$ grep -r ac_cv_c_inline ntp-4.2.8p18
$ grep -r ac_cv_c_inline ntp-stable-dlh-260328

and do you have any insight into why `ac_cv_c_inline` is set in ntp-4.2.8p18/configure but is not set in ntp-stable-dlh-260328/configure?

Cheers!
Edward

P.S. Chip I'm sorry for the double post. Some of the names were lost on
the message I first replied to.

--=20
Ralph "Chip" Blach
(919) 260-0097

--00000000000093424b064fabaeb6
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>I would not use the slackware ntpd package.=C2=A0 Can=
you please send me a patch file and I can make those changes to=C2=A0ntp-4= .2.8p18.</div><div>There is a big hint on this when slackware removed the P=
PS api</div><div><br></div><div>Chip</div></div><br><div class=3D"gmail_quo=
te gmail_quote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Apr=
17, 2026 at 1:45=E2=80=AFPM Edward McGuire &lt;<a href=3D"mailto:metaed@gm= ail.com">[email protected]</a>&gt; wrote:<br></div><blockquote class=3D"gmai= l_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,20= 4,204);padding-left:1ex"><div dir=3D"ltr">I am not familiar with `autoconf`=
so I don&#39;t follow the logic in `<a href=3D"http://configure.ac" target= =3D"_blank">configure.ac</a>` very well. But I see there were changes relat=
ed to `syscall.h`, and I see the resulting `configure` no longer sets `ac_c= v_c_inline`.<br><br>Dave, are you able to compare the output of<br><br>$ gr=
ep -r ac_cv_c_inline ntp-4.2.8p18<br>$ grep -r ac_cv_c_inline ntp-stable-dl= h-260328<br><br>and do you have any insight into why `ac_cv_c_inline` is se=
t in ntp-4.2.8p18/configure but is not set in ntp-stable-dlh-260328/configu= re?<br><br>Cheers!<br>Edward<div><br></div><div>P.S. Chip I&#39;m sorry for=
the double post. Some of the names were lost on the message I first replie=
d to.</div></div>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div>

--00000000000093424b064fabaeb6--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--000000000000a67fd1064fac0b5a
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Thanks, Ralph, for reporting the problem with HAVE_PPSAPI and HAVE_THREADS
in 4.2.8p18. I believe the HAVE_THREADS issue is resolved in my 260328
test release. The pthreads detection issue rings a bell for me as
something already handled by a pending fix that the test release includes. Thanks to Edward and Martin for further testing and spelunking of the
configure process.

I'm very familiar with the Autoconf code in configure.ac and .m4 files, so
I expect little difficulty fixing this when I can find time, perhaps this weekend or sometime next week.

It would be spiffy if one of you would file a bug report at https://bugs.ntp.org about the missing HAVE_PPSAPI so we can track it
formally.

I'm happy to report Gmail users can once again create bugs.ntp.org accounts self-serve, without the speed bump of emailing a support address. Many
thanks to Dru and Harlan for handling that change.


Cheers,
Dave Hart


On Wed, Apr 15, 2026 at 1:14=E2=80=AFPM Ralph Blach <[email protected]=

wrote:

Edward,

Chip here. I tested the stable release on my raspberry and the configure script failed.
The configure script failed to set the HAVE_PPSAPI and the HAVE_THREADS i=

n

the config.h

I tested the config process on ntp-4.2.8p15 and the config.h is set correctly.

The result of these failures, I could not use the pps to lock on the gps devices PPS signal

Be very careful, because I dont think you are compiling what you think yo=

u

are compiling.

Chip

On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire <[email protected]> =

wrote:

On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart <[email protected]> wr= ote:

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Hello Dave,

As a smoke test, I substituted this tarball into the Slackware64 15.0 NT=

P

package, built, and installed on my server. So far it passes the smoke t= est
-- syncs and keeps time as well as before.

The only change necessary was to delete the file
fix_ipv6_locallink_regression.diff from the Slackware build. This was a
workaround for bug 3928. The patch failed for the good reason that you
fixed bug 3928 in your private release.

Cheers!
Edward

--
Ralph "Chip" Blach
(919) 260-0097

--000000000000a67fd1064fac0b5a
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div><div class=3D"gmail_default" style=3D"font-family:&qu= ot;trebuchet ms&quot;,sans-serif">Thanks, Ralph, for reporting the problem = with HAVE_PPSAPI and HAVE_THREADS in 4.2.8p18.=C2=A0 I believe the HAVE_THR= EADS issue is resolved in my 260328 test release.=C2=A0 The pthreads detect= ion issue rings a bell for me as something already handled by a pending fix=
that the test release includes.=C2=A0 Thanks to Edward and Martin for furt= her testing and spelunking of the configure process.</div><div class=3D"gma= il_default" style=3D"font-family:&quot;trebuchet ms&quot;,sans-serif"><br><= /div><div class=3D"gmail_default" style=3D"font-family:&quot;trebuchet ms&q= uot;,sans-serif">I&#39;m very familiar with the Autoconf code in <a href=3D= "http://configure.ac">configure.ac</a> and .m4 files, so I expect little di= fficulty fixing this when I can find time, perhaps this weekend or sometime=
next week.</div><div class=3D"gmail_default" style=3D"font-family:&quot;tr= ebuchet ms&quot;,sans-serif"><br></div><div class=3D"gmail_default" style= =3D"font-family:&quot;trebuchet ms&quot;,sans-serif">It would be spiffy if = one of you would file a bug report at <a href=3D"https://bugs.ntp.org">http= s://bugs.ntp.org</a> about the missing HAVE_PPSAPI so we can track it forma= lly.</div><div class=3D"gmail_default" style=3D"font-family:&quot;trebuchet=
ms&quot;,sans-serif"><br></div><div class=3D"gmail_default" style=3D"font-= family:&quot;trebuchet ms&quot;,sans-serif">I&#39;m happy to report Gmail u= sers can once again create <a href=3D"http://bugs.ntp.org">bugs.ntp.org</a>=
accounts self-serve, without the speed bump of emailing a support address.= =C2=A0 Many thanks to Dru and Harlan for handling that change.</div><br cle= ar=3D"all"></div><div><div dir=3D"ltr" class=3D"gmail_signature" data-smart= mail=3D"gmail_signature"><div dir=3D"ltr"><div><font face=3D"tahoma, sans-s= erif" color=3D"#666666"><br></font></div><font face=3D"tahoma, sans-serif" = color=3D"#666666">Cheers,<br>Dave Hart</font></div></div></div><br></div><b= r><div class=3D"gmail_quote gmail_quote_container"><div dir=3D"ltr" class= =3D"gmail_attr">On Wed, Apr 15, 2026 at 1:14=E2=80=AFPM Ralph Blach &lt;<a = href=3D"mailto:[email protected]">[email protected]</a>&gt; wrote= :<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.= 8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D"lt= r"><div>Edward,</div><div><br></div><div>Chip here.=C2=A0 I tested the stab=
le release on my raspberry and the configure script failed.</div><div>The c= onfigure script failed to set the HAVE_PPSAPI and the HAVE_THREADS in the c= onfig.h</div><div><br></div><div>I tested the config process on=C2=A0=C2=A0= ntp-4.2.8p15=C2=A0 and the config.h is set correctly.</div><div><br></div><= div>The result of these failures, I could not use the pps to lock on the gp=
s devices PPS signal</div><div><br></div><div>Be very careful, because I do=
nt think you are compiling what you think you are compiling.</div><div><br>= </div><div>Chip</div><div><br></div></div><br><div class=3D"gmail_quote"><d=
iv dir=3D"ltr" class=3D"gmail_attr">On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM = Edward McGuire &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">me= [email protected]</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" st= yle=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padd= ing-left:1ex"><div dir=3D"ltr">On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave = Hart &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">davehart@g= mail.com</a>&gt; wrote:<br>&gt; <a href=3D"https://davehart.net/ntp/test/nt= p-stable-dlh-260328.tar.gz" target=3D"_blank">https://davehart.net/ntp/test= /ntp-stable-dlh-260328.tar.gz</a><br><br>Hello Dave,<br><br>As a smoke test=
, I substituted this tarball into the Slackware64 15.0 NTP package, built, = and installed on my server. So far it passes the smoke test -- syncs and ke= eps time as well as before.<br><br>The only change necessary was to delete = the file fix_ipv6_locallink_regression.diff from the Slackware build. This = was a workaround for bug 3928. The patch failed for the good reason that yo=
u fixed bug 3928 in your private release.<br><br>Cheers!<br>Edward</div> </blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div> </blockquote></div>

--000000000000a67fd1064fac0b5a--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--000000000000778f9b064facdd95
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

All, the bug is created and reported.

Just let me know when the code is updated and I will test it.

Chip

On Fri, Apr 17, 2026 at 2:23=E2=80=AFPM Dave Hart <[email protected]> wrot=
e:

Thanks, Ralph, for reporting the problem with HAVE_PPSAPI and HAVE_THREAD=

S

in 4.2.8p18. I believe the HAVE_THREADS issue is resolved in my 260328
test release. The pthreads detection issue rings a bell for me as
something already handled by a pending fix that the test release includes=

.

Thanks to Edward and Martin for further testing and spelunking of the configure process.

I'm very familiar with the Autoconf code in configure.ac and .m4 files,
so I expect little difficulty fixing this when I can find time, perhaps
this weekend or sometime next week.

It would be spiffy if one of you would file a bug report at https://bugs.ntp.org about the missing HAVE_PPSAPI so we can track it formally.

I'm happy to report Gmail users can once again create bugs.ntp.org
accounts self-serve, without the speed bump of emailing a support address=

.

Many thanks to Dru and Harlan for handling that change.

Cheers,
Dave Hart

On Wed, Apr 15, 2026 at 1:14=E2=80=AFPM Ralph Blach <[email protected]=

wrote:

Edward,

Chip here. I tested the stable release on my raspberry and the configur=

e

script failed.
The configure script failed to set the HAVE_PPSAPI and the HAVE_THREADS
in the config.h

I tested the config process on ntp-4.2.8p15 and the config.h is set
correctly.

The result of these failures, I could not use the pps to lock on the gps
devices PPS signal

Be very careful, because I dont think you are compiling what you think
you are compiling.

Chip


On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire <[email protected]>=

wrote:

On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart <[email protected]> w= rote:

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Hello Dave,

As a smoke test, I substituted this tarball into the Slackware64 15.0
NTP package, built, and installed on my server. So far it passes the sm= oke
test -- syncs and keeps time as well as before.

The only change necessary was to delete the file
fix_ipv6_locallink_regression.diff from the Slackware build. This was a
workaround for bug 3928. The patch failed for the good reason that you
fixed bug 3928 in your private release.

Cheers!
Edward

--
Ralph "Chip" Blach
(919) 260-0097

--=20
Ralph "Chip" Blach
(919) 260-0097

--000000000000778f9b064facdd95
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>All, the bug is created and reported.</div><div><br><= /div><div>Just let me know when the code is updated and I will test it.</di= v><div><br></div><div>Chip</div></div><br><div class=3D"gmail_quote gmail_q= uote_container"><div dir=3D"ltr" class=3D"gmail_attr">On Fri, Apr 17, 2026 =
at 2:23=E2=80=AFPM Dave Hart &lt;<a href=3D"mailto:[email protected]">dave= [email protected]</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" st= yle=3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padd= ing-left:1ex"><div dir=3D"ltr"><div><div class=3D"gmail_default" style=3D"f= ont-family:&quot;trebuchet ms&quot;,sans-serif">Thanks, Ralph, for reportin=
g the problem with HAVE_PPSAPI and HAVE_THREADS in 4.2.8p18.=C2=A0 I believ=
e the HAVE_THREADS issue is resolved in my 260328 test release.=C2=A0 The p= threads detection issue rings a bell for me as something already handled by=
a pending fix that the test release includes.=C2=A0 Thanks to Edward and M= artin for further testing and spelunking of the configure process.</div><di=
v class=3D"gmail_default" style=3D"font-family:&quot;trebuchet ms&quot;,san= s-serif"><br></div><div class=3D"gmail_default" style=3D"font-family:&quot;= trebuchet ms&quot;,sans-serif">I&#39;m very familiar with the Autoconf code=
in <a href=3D"http://configure.ac" target=3D"_blank">configure.ac</a> and = .m4 files, so I expect little difficulty fixing this when I can find time, = perhaps this weekend or sometime next week.</div><div class=3D"gmail_defaul=
t" style=3D"font-family:&quot;trebuchet ms&quot;,sans-serif"><br></div><div=
class=3D"gmail_default" style=3D"font-family:&quot;trebuchet ms&quot;,sans= -serif">It would be spiffy if one of you would file a bug report at <a href= =3D"https://bugs.ntp.org" target=3D"_blank">https://bugs.ntp.org</a> about = the missing HAVE_PPSAPI so we can track it formally.</div><div class=3D"gma= il_default" style=3D"font-family:&quot;trebuchet ms&quot;,sans-serif"><br><= /div><div class=3D"gmail_default" style=3D"font-family:&quot;trebuchet ms&q= uot;,sans-serif">I&#39;m happy to report Gmail users can once again create =
<a href=3D"http://bugs.ntp.org" target=3D"_blank">bugs.ntp.org</a> accounts=
self-serve, without the speed bump of emailing a support address.=C2=A0 Ma=
ny thanks to Dru and Harlan for handling that change.</div><br clear=3D"all= "></div><div><div dir=3D"ltr" class=3D"gmail_signature"><div dir=3D"ltr"><d= iv><font face=3D"tahoma, sans-serif" color=3D"#666666"><br></font></div><fo=
nt face=3D"tahoma, sans-serif" color=3D"#666666">Cheers,<br>Dave Hart</font= ></div></div></div><br></div><br><div class=3D"gmail_quote"><div dir=3D"ltr=
" class=3D"gmail_attr">On Wed, Apr 15, 2026 at 1:14=E2=80=AFPM Ralph Blach = &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">chip.from.n= [email protected]</a>&gt; wrote:<br></div><blockquote class=3D"gmail_quote" style= =3D"margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding= -left:1ex"><div dir=3D"ltr"><div>Edward,</div><div><br></div><div>Chip here= .=C2=A0 I tested the stable release on my raspberry and the configure scrip=
t failed.</div><div>The configure script failed to set the HAVE_PPSAPI and = the HAVE_THREADS in the config.h</div><div><br></div><div>I tested the conf=
ig process on=C2=A0=C2=A0ntp-4.2.8p15=C2=A0 and the config.h is set correct= ly.</div><div><br></div><div>The result of these failures, I could not use = the pps to lock on the gps devices PPS signal</div><div><br></div><div>Be v= ery careful, because I dont think you are compiling what you think you are = compiling.</div><div><br></div><div>Chip</div><div><br></div></div><br><div=
class=3D"gmail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Apr 1,=
2026 at 2:10=E2=80=AFPM Edward McGuire &lt;<a href=3D"mailto:metaed@gmail.= com" target=3D"_blank">[email protected]</a>&gt; wrote:<br></div><blockquote=
class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px so= lid rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr">On Wed, Apr 1, 2026=
at 7:05=E2=80=AFAM Dave Hart &lt;<a href=3D"mailto:[email protected]" tar= get=3D"_blank">[email protected]</a>&gt; wrote:<br>&gt; <a href=3D"https:/= /davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz" target=3D"_blank">http= s://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz</a><br><br>Hello Dav= e,<br><br>As a smoke test, I substituted this tarball into the Slackware64 = 15.0 NTP package, built, and installed on my server. So far it passes the s= moke test -- syncs and keeps time as well as before.<br><br>The only change=
necessary was to delete the file fix_ipv6_locallink_regression.diff from t=
he Slackware build. This was a workaround for bug 3928. The patch failed fo=
r the good reason that you fixed bug 3928 in your private release.<br><br>C= heers!<br>Edward</div>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div> </blockquote></div>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div>

--000000000000778f9b064facdd95--

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

I found a problem with the C "inline" check and I have an idea of how to
fix it.

H

On 4/17/2026 12:21 PM, Ralph Blach wrote:

All, the bug is created and reported.

Just let me know when the code is updated and I will test it.

Chip

On Fri, Apr 17, 2026 at 2:23 PM Dave Hart <[email protected] <mailto:[email protected]>> wrote:

Thanks, Ralph, for reporting the problem with HAVE_PPSAPI and
HAVE_THREADS in 4.2.8p18.  I believe the HAVE_THREADS issue is
resolved in my 260328 test release.  The pthreads detection issue
rings a bell for me as something already handled by a pending fix
that the test release includes.  Thanks to Edward and Martin for
further testing and spelunking of the configure process.

I'm very familiar with the Autoconf code in configure.ac <http://
configure.ac> and .m4 files, so I expect little difficulty fixing
this when I can find time, perhaps this weekend or sometime next week.

It would be spiffy if one of you would file a bug report at https://
bugs.ntp.org <https://bugs.ntp.org> about the missing HAVE_PPSAPI so
we can track it formally.

I'm happy to report Gmail users can once again create bugs.ntp.org
<http://bugs.ntp.org> accounts self-serve, without the speed bump of
emailing a support address.  Many thanks to Dru and Harlan for
handling that change.

Cheers,
Dave Hart

On Wed, Apr 15, 2026 at 1:14 PM Ralph Blach <[email protected]
<mailto:[email protected]>> wrote:

Edward,

Chip here.  I tested the stable release on my raspberry and the
configure script failed.
The configure script failed to set the HAVE_PPSAPI and the
HAVE_THREADS in the config.h

I tested the config process on  ntp-4.2.8p15  and the config.h
is set correctly.

The result of these failures, I could not use the pps to lock on
the gps devices PPS signal

Be very careful, because I dont think you are compiling what you
think you are compiling.

Chip

On Wed, Apr 1, 2026 at 2:10 PM Edward McGuire <[email protected]
<mailto:[email protected]>> wrote:

On Wed, Apr 1, 2026 at 7:05 AM Dave Hart <[email protected]
<mailto:[email protected]>> wrote:
> https://davehart.net/ntp/test/ntp-stable-
dlh-260328.tar.gz <https://davehart.net/ntp/test/ntp-stable-
dlh-260328.tar.gz>

Hello Dave,

As a smoke test, I substituted this tarball into the
Slackware64 15.0 NTP package, built, and installed on my
server. So far it passes the smoke test -- syncs and keeps
time as well as before.

The only change necessary was to delete the file
fix_ipv6_locallink_regression.diff from the Slackware build.
This was a workaround for bug 3928. The patch failed for the
good reason that you fixed bug 3928 in your private release.

Cheers!
Edward

--
Ralph "Chip" Blach
(919) 260-0097

--
Ralph "Chip" Blach
(919) 260-0097

--
Harlan Stenn <[email protected]>
https://networktimefoundation.org - be a member!

--- Synchronet 3.21f-Linux NewsLink 1.2

From Newsgroup: comp.protocols.time.ntp

--0000000000005fb9ff064ff37e88
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

I think I understand the regression detecting PPSAPI support. Edward and
Chip were well along the path to the root cause, thanks for the
assistance. It was my flub in removing the code that set $ac_cv_c_inline.
I'm going to reuse a Bugzilla comment for the benefit of the list archives
for the remainder of this message. Feel free to check out https://bugs.ntp.org/4022 (thanks Chip for filing) and https://bugs.ntp.org/3982, the patch for which introduced the regression.

chip.from.nc, thanks for testing my private release and thanks especially
for taking the time to file the bug after reporting the issue on [email protected].

Your diagnosis is spot-on. In trying to clean up our humongous configure scripts, I got a bit over my skis while patching [Bug 3982]. I noticed
none of our code used the HAVE_INLINE #define and removed cod=C3=A8 from ntp_libntp.m4 that provided it.

So far so good.

I mistakenly thought the C89 standard (AKA ANSI C) mandated support for "inline", reasoned that we require ANSI C, and also removed the configure.a=
c
macro invocation that tests for that support, AC_C_INLINE. That broke our PPSAPI detection logic, which relies on the undocumented side-effect
variable $ac_cv_c_inline. That's understandable as there's no documented
way to check for its support in configure.ac/*.m4 (m4sh Autoconf script).

As you probably already know, but for the benefit of others, a quick ugly
hack would be to edit config.h after configure has already run, adding a "#define HAVE_PPSAPI 1" line.

For a minimal fix, plop AC_C_INLINE either back where it was in
ntp_libntp.m4, or at the bottom of sntp/m4/ntp_compiler.m4 alongside the
other AC_C_* invocations. Make sure you run "autoreconf -i" afterwards in
the top level of the source tree (not the build tree, if you build
somewhere other than the source directory). That should regenerate a
top-level configure script which sets ac_cv_c_inline to "inline" after
testing the compiler, so that HAVE_PPSAPI ends up in your config.h after configure is run.

While digging into this I found some other code to break (I'm kidding). I found some other configure.ac/m4 stuff that could be better so I'm not attaching a patch yet. I want to first spend more time testing those
changes alongside this one across several different OSes.

Cheers,
Dave Hart


On Fri, Apr 17, 2026 at 6:23=E2=80=AFPM Dave Hart <[email protected]> wrot=
e:

Thanks, Ralph, for reporting the problem with HAVE_PPSAPI and HAVE_THREAD=

S

in 4.2.8p18. I believe the HAVE_THREADS issue is resolved in my 260328
test release. The pthreads detection issue rings a bell for me as
something already handled by a pending fix that the test release includes=

.

Thanks to Edward and Martin for further testing and spelunking of the configure process.

I'm very familiar with the Autoconf code in configure.ac and .m4 files,
so I expect little difficulty fixing this when I can find time, perhaps
this weekend or sometime next week.

It would be spiffy if one of you would file a bug report at https://bugs.ntp.org about the missing HAVE_PPSAPI so we can track it formally.

I'm happy to report Gmail users can once again create bugs.ntp.org
accounts self-serve, without the speed bump of emailing a support address=

.

Many thanks to Dru and Harlan for handling that change.

Cheers,
Dave Hart

On Wed, Apr 15, 2026 at 1:14=E2=80=AFPM Ralph Blach <[email protected]=

wrote:

Edward,

Chip here. I tested the stable release on my raspberry and the configur=

e

script failed.
The configure script failed to set the HAVE_PPSAPI and the HAVE_THREADS
in the config.h

I tested the config process on ntp-4.2.8p15 and the config.h is set
correctly.

The result of these failures, I could not use the pps to lock on the gps
devices PPS signal

Be very careful, because I dont think you are compiling what you think
you are compiling.

Chip


On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire <[email protected]>=

wrote:

On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart <[email protected]> w= rote:

https://davehart.net/ntp/test/ntp-stable-dlh-260328.tar.gz

Hello Dave,

As a smoke test, I substituted this tarball into the Slackware64 15.0
NTP package, built, and installed on my server. So far it passes the sm= oke
test -- syncs and keeps time as well as before.

The only change necessary was to delete the file
fix_ipv6_locallink_regression.diff from the Slackware build. This was a
workaround for bug 3928. The patch failed for the good reason that you
fixed bug 3928 in your private release.

Cheers!
Edward

--
Ralph "Chip" Blach
(919) 260-0097

--0000000000005fb9ff064ff37e88
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><div><div class=3D"gmail_default" style= =3D"font-family:&quot;trebuchet ms&quot;,sans-serif">I think I understand t=
he regression detecting PPSAPI support.=C2=A0 Edward and Chip were well alo=
ng the path to the root cause, thanks for the assistance.=C2=A0 It was my f= lub in removing the code that set $ac_cv_c_inline.=C2=A0 I&#39;m going to r= euse a Bugzilla comment for the benefit of the list archives for the remain= der of this message.=C2=A0 Feel free to check out <a href=3D"https://bugs.n= tp.org/4022">https://bugs.ntp.org/4022</a> (thanks Chip for filing) and <a = href=3D"https://bugs.ntp.org/3982">https://bugs.ntp.org/3982</a>, the patch=
for which introduced the regression.</div><font face=3D"trebuchet ms, sans= -serif" color=3D"#073763"><br clear=3D"all"></font></div><div><font face=3D= "trebuchet ms, sans-serif" color=3D"#073763"><a href=3D"http://chip.from.nc= ">chip.from.nc</a>, thanks for testing my private release and thanks especi= ally for taking the time to file the bug after reporting the issue on <a hr= ef=3D"mailto:[email protected]">[email protected]</a>.<br><br>Y= our diagnosis is spot-on.=C2=A0 In trying to clean up our humongous configu=
re scripts, I got a bit over my skis while patching [Bug 3982].=C2=A0 I not= iced none of our code used the HAVE_INLINE #define and removed cod=C3=A8 fr=
om ntp_libntp.m4 that provided it. =C2=A0<br><br>So far so good.<br><br>I m= istakenly thought the C89 standard (AKA ANSI C) mandated support for &quot;= inline&quot;, reasoned that we require ANSI C, and also removed the <a href= =3D"http://configure.ac">configure.ac</a> macro invocation that tests for t= hat support, AC_C_INLINE.=C2=A0 That broke our PPSAPI detection logic, whic=
h relies on the undocumented side-effect variable $ac_cv_c_inline.=C2=A0 Th= at&#39;s understandable as there&#39;s no documented way to check for its s= upport in <a href=3D"http://configure.ac/*.m4">configure.ac/*.m4</a> (m4sh = Autoconf script).<br><br>As you probably already know, but for the benefit =
of others, a quick ugly hack would be to edit config.h after configure has = already run, adding a &quot;#define HAVE_PPSAPI 1&quot; line.<br><br>For a = minimal fix, plop AC_C_INLINE either back where it was in ntp_libntp.m4, or=
at the bottom of sntp/m4/ntp_compiler.m4 alongside the other AC_C_* invoca= tions.=C2=A0 Make sure you run &quot;autoreconf -i&quot; afterwards in the = top level of the source tree (not the build tree, if you build somewhere ot= her than the source directory).=C2=A0 That should regenerate a top-level co= nfigure script which sets ac_cv_c_inline to &quot;inline&quot; after testin=
g the compiler, so that HAVE_PPSAPI ends up in your config.h after configur=
e is run.<br><br>While digging into this I found some other code to break (= I&#39;m kidding).=C2=A0 I found some other <a href=3D"http://configure.ac/m= 4">configure.ac/m4</a> stuff that could be better so I&#39;m not attaching =
a patch yet.=C2=A0 I want to first spend more time testing those changes al= ongside this one across several different OSes.</font></div><div><div dir= =3D"ltr" class=3D"gmail_signature"><div dir=3D"ltr"><div><font face=3D"taho= ma, sans-serif" color=3D"#666666"><br></font></div><font face=3D"tahoma, sa= ns-serif" color=3D"#666666">Cheers,<br>Dave Hart</font></div></div></div><b= r></div><br><div class=3D"gmail_quote gmail_quote_container"><div dir=3D"lt=
r" class=3D"gmail_attr">On Fri, Apr 17, 2026 at 6:23=E2=80=AFPM Dave Hart &= lt;<a href=3D"mailto:[email protected]">[email protected]</a>&gt; wrote:<= br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8e= x;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr"= ><div><div style=3D"font-family:&quot;trebuchet ms&quot;,sans-serif">Thanks=
, Ralph, for reporting the problem with HAVE_PPSAPI and HAVE_THREADS in 4.2= .8p18.=C2=A0 I believe the HAVE_THREADS issue is resolved in my 260328 test=
release.=C2=A0 The pthreads detection issue rings a bell for me as somethi=
ng already handled by a pending fix that the test release includes.=C2=A0 T= hanks to Edward and Martin for further testing and spelunking of the config= ure process.</div><div style=3D"font-family:&quot;trebuchet ms&quot;,sans-s= erif"><br></div><div style=3D"font-family:&quot;trebuchet ms&quot;,sans-ser= if">I&#39;m very familiar with the Autoconf code in <a href=3D"http://confi= gure.ac" target=3D"_blank">configure.ac</a> and .m4 files, so I expect litt=
le difficulty fixing this when I can find time, perhaps this weekend or som= etime next week.</div><div style=3D"font-family:&quot;trebuchet ms&quot;,sa= ns-serif"><br></div><div style=3D"font-family:&quot;trebuchet ms&quot;,sans= -serif">It would be spiffy if one of you would file a bug report at <a href= =3D"https://bugs.ntp.org" target=3D"_blank">https://bugs.ntp.org</a> about = the missing HAVE_PPSAPI so we can track it formally.</div><div style=3D"fon= t-family:&quot;trebuchet ms&quot;,sans-serif"><br></div><div style=3D"font-= family:&quot;trebuchet ms&quot;,sans-serif">I&#39;m happy to report Gmail u= sers can once again create <a href=3D"http://bugs.ntp.org" target=3D"_blank= ">bugs.ntp.org</a> accounts self-serve, without the speed bump of emailing =
a support address.=C2=A0 Many thanks to Dru and Harlan for handling that ch= ange.</div><br clear=3D"all"></div><div><div dir=3D"ltr" class=3D"gmail_sig= nature"><div dir=3D"ltr"><div><font face=3D"tahoma, sans-serif" color=3D"#6= 66666"><br></font></div><font face=3D"tahoma, sans-serif" color=3D"#666666"= >Cheers,<br>Dave Hart</font></div></div></div><br></div><br><div class=3D"g= mail_quote"><div dir=3D"ltr" class=3D"gmail_attr">On Wed, Apr 15, 2026 at 1= :14=E2=80=AFPM Ralph Blach &lt;<a href=3D"mailto:[email protected]" ta= rget=3D"_blank">[email protected]</a>&gt; wrote:<br></div><blockquote = class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1px sol=
id rgb(204,204,204);padding-left:1ex"><div dir=3D"ltr"><div>Edward,</div><d= iv><br></div><div>Chip here.=C2=A0 I tested the stable release on my raspbe= rry and the configure script failed.</div><div>The configure script failed =
to set the HAVE_PPSAPI and the HAVE_THREADS in the config.h</div><div><br><= /div><div>I tested the config process on=C2=A0=C2=A0ntp-4.2.8p15=C2=A0 and = the config.h is set correctly.</div><div><br></div><div>The result of these=
failures, I could not use the pps to lock on the gps devices PPS signal</d= iv><div><br></div><div>Be very careful, because I dont think you are compil= ing what you think you are compiling.</div><div><br></div><div>Chip</div><d= iv><br></div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr" class=3D= "gmail_attr">On Wed, Apr 1, 2026 at 2:10=E2=80=AFPM Edward McGuire &lt;<a h= ref=3D"mailto:[email protected]" target=3D"_blank">[email protected]</a>&gt; = wrote:<br></div><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0=
px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir= =3D"ltr">On Wed, Apr 1, 2026 at 7:05=E2=80=AFAM Dave Hart &lt;<a href=3D"ma= ilto:[email protected]" target=3D"_blank">[email protected]</a>&gt; wrote= :<br>&gt; <a href=3D"https://davehart.net/ntp/test/ntp-stable-dlh-260328.ta= r.gz" target=3D"_blank">https://davehart.net/ntp/test/ntp-stable-dlh-260328= .tar.gz</a><br><br>Hello Dave,<br><br>As a smoke test, I substituted this t= arball into the Slackware64 15.0 NTP package, built, and installed on my se= rver. So far it passes the smoke test -- syncs and keeps time as well as be= fore.<br><br>The only change necessary was to delete the file fix_ipv6_loca= llink_regression.diff from the Slackware build. This was a workaround for b=
ug 3928. The patch failed for the good reason that you fixed bug 3928 in yo=
ur private release.<br><br>Cheers!<br>Edward</div>
</blockquote></div><div><br clear=3D"all"></div><br><span class=3D"gmail_si= gnature_prefix">-- </span><br><div dir=3D"ltr" class=3D"gmail_signature">Ra= lph &quot;Chip&quot; Blach<br>(919) 260-0097<br><br><br><br></div> </blockquote></div>
</blockquote></div></div>

--0000000000005fb9ff064ff37e88--

--- Synchronet 3.21f-Linux NewsLink 1.2