[illumos-gate] 18156 Sun::Solaris::Pg::Pg.pm produces warnings under perl 5.42 -- Andy Fiddaman <illumos⊙fn>

[illumos-gate] 6043 mdb: pipe shell output back into mdb -- Andy Fiddaman <illumos⊙fn>

[illumos-gate] 18263 Want uio_copyin(9F) and uio_copyout(9F) -- Andy Fiddaman <illumos⊙fn>

I have an mdb question that I put in a gist: gist.github.com/rzezeski/99fd47546ab88a571c259fd64b825897

I suspect.... or it smells like the formatting is applied on first address line count times. thats why you get only first address.

as the output is mostly 0 and once 80, manual inspection should be able to confirm.

the 80/0 are values I expected, it's printing the value based on the 20 different addresses, the question is why only the first line shows its address. I can see if I just do `addr/56+Bn` that the address is printed, which I guest is just part of how formatting dcmds work, but I found it odd that I only see the first one when using it on a pipeline

I added a comment to that guest that uses `/a56+Bn` to clarify what I'm asking and prove that this isn't the case of a repeat.

just FYI, CVE-2026-16461 appears to be fixed by illumos/illumos-gate 791dfaa

→ GitHub commit 791dfaa: 6908961 rpcinfo -s can not deal with large number of versions (committed)

not that anyone still uses code older than snv_132

you never know...

I mean, yes, more people than you'd expect still send us money to continue running S10, but that's not your code base

same change also appears to fix CVE-2026-16277

rzezeski does /na56+Bn make difference?

I did an experiment, took fb address from fb_info, and printed: 0xfffffd684c825000,8 with: /a3+Bn /3+Bn /n3+Bn and /na3+Bn -- that is, 8 times from given offset.

Nice! Thank you alanc!

also mentioned in openwall.com/lists/oss-security/2026/07/23/8 - OpenSolaris beat NetBSD to the fix by 1 year, Linux by 16 years 8-)