Status of removing SIGSEGV from GC

"J. Gareth Williams" <[email protected]> Fri, 3 Apr 2026 16:51:38 -0400
Newsgroups gmane.lisp.steel-bank.general
Message-ID <CALfqwAqS92T21GYT6t6fKSW0Bjv_FFDd-Tk-ZOE-oCW7Gu3mYQ@mail.gmail.com>
--===============8307728870212901483==
Content-Type: multipart/alternative; boundary="00000000000071964b064e947c4f"

--00000000000071964b064e947c4f
Content-Type: text/plain; charset="UTF-8"

hi SBCL devs,

About a year ago I kicked off a thread
https://sourceforge.net/p/sbcl/mailman/sbcl-help/thread/CALfqwAo0OFGU0TYntX1Bzqm0WQe_4SUG2rUUFQGOGhFUW4UO2Q%40mail.gmail.com/#msg59163647)
in which I learned that the GC relies on SIGSEGV, but that this mechanism
has largely but not completely been removed.

It looks like SEGV is still used at least as of 2.6.0 or so.

I found a work-around in my application to accommodate this scheme, but
it's a bit awkward.

So my questions are:
- is it still in the cards to complete this work?
- is it a significant task?
- any other users out there interested in seeing this?
- would this be something that would be reasonable for someone
completely-unfamiliar with the codebase to tackle?

Any insight appreciated!

Gareth

--00000000000071964b064e947c4f
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">hi SBCL devs,<div><br></div><div>About a year ago I kicked=
 off a thread <a href=3D"https://sourceforge.net/p/sbcl/mailman/sbcl-help/t=
hread/CALfqwAo0OFGU0TYntX1Bzqm0WQe_4SUG2rUUFQGOGhFUW4UO2Q%40mail.gmail.com/=
#msg59163647">https://sourceforge.net/p/sbcl/mailman/sbcl-help/thread/CALfq=
wAo0OFGU0TYntX1Bzqm0WQe_4SUG2rUUFQGOGhFUW4UO2Q%40mail.gmail.com/#msg5916364=
7</a>) in which I learned that the GC relies on SIGSEGV, but that this mech=
anism has largely but not completely been removed.</div><div><div><br></div=
><div>It looks like SEGV is still used at least as of 2.6.0 or so.</div><br=
 class=3D"gmail-Apple-interchange-newline"></div><div>I found a work-around=
 in my application to accommodate=C2=A0this scheme, but it&#39;s a bit awkw=
ard.</div><div><br></div><div>So my questions are:</div><div>- is it still =
in the cards to complete this work?</div><div>- is it a significant task?</=
div><div>- any other users out there interested in seeing this?=C2=A0=C2=A0=
</div><div>- would this be something=C2=A0that would be reasonable for some=
one completely-unfamiliar with the codebase to tackle?</div><div><br></div>=
<div>Any insight=C2=A0appreciated!</div><div><br></div><div>Gareth</div><di=
v><br></div><div>=C2=A0</div></div>

--00000000000071964b064e947c4f--


--===============8307728870212901483==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline


--===============8307728870212901483==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Sbcl-help mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/sbcl-help

--===============8307728870212901483==--