Further issues from the safename() change at D56615 popped up at the same PR here https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=295875

There is some discussion at PR295875 around some characters that can be seen in filenames (~ and ? come to mind) being unexpectedly escaped in our output after this change, beyond the missing locale handling that the PR was originally opened for (and patched in D57562).

The other thing that SMR does in this respect is it puts free buckets at the tail of uzd_buckets in zone_put_bucket. Did you think about whether you want that for KASAN also, to extend the length of time before reuse further?

Please see the wireless mailing list where I sent a patch for test and did not get feedback. Changing the driver is the wrong thing to do (not that it works well currently).

Account for EINPROGRESS

Remove ARGUSED comment.

Looks good.

Looks good.

In D58281#1337131, @glebius wrote:

In D58281#1337102, @bnovkov wrote:

Thank you for the offer, but I was trying to avoid unnecessary churn if the same thing can be achieved with ifa_ifwithnet. If you have future uses for a fib-aware version of ifa_ifwithaddr() I can implement it as well.

There will be less churn if you use standard KPI in the right place instead of current version. Pouria also needs this function. Even if he didn't, still creating this KPI on the first need is the right thing to do.

In D58302#1337168, @aokblast wrote:

In D58302#1337163, @kib wrote:

Could f6ee1d20d36813e990691 fixed your issue?

Anyway, the useful tool is gdb' pinfo /o type.

It should be the correct reason. Thanks! I don't have ia32 machine so I planned to check the dwarf info.

You made some interesting points that probably deserve to be explored in more detail and in a different context.

Properly nest calls.

• Don't forget to clear the bit on inpcb disconnect/free. Thanks, Vova!

In D58302#1337163, @kib wrote:

Could f6ee1d20d36813e990691 fixed your issue?

Anyway, the useful tool is gdb' pinfo /o type.

Reflect comments

Could f6ee1d20d36813e990691 fixed your issue?

In D58302#1337157, @kib wrote:

Why? p_asig and p_tree_refcount are at the end of the struct proc.

Why? p_asig and p_tree_refcount are at the end of the struct proc.

Address review feedback

Address review feedback

worth to release the bit on closeing?

In D58262#1336336, @ngie wrote:

Side thought: it would be nice if these were actually broken down into subpackages, e.g., indent-tests, etc.

In D58281#1337102, @bnovkov wrote:

Thank you for the offer, but I was trying to avoid unnecessary churn if the same thing can be achieved with ifa_ifwithnet. If you have future uses for a fib-aware version of ifa_ifwithaddr() I can implement it as well.

I'm pretty sure this header needs to be in /usr/include due, well, its seniority status among headers (ie, what software expects.)

In D58281#1337090, @pouria wrote:

In D58281#1336902, @bnovkov wrote:

In D58281#1336697, @glebius wrote:

Instead of calling fib4_lookup() that has a different purpose, we need ifa_ifwithaddr() version that has fibnum argument.

Looking through net/if.c I noticed that ifa_ifwithnet seems to match what you're suggesting here, do you think we can use this function instead of adding a fibnum-aware version of `ifa_ifwithaddr()`?

I'll create a fib-aware version of ifa_ifwithaddr() and will let you know.

Remove useless patch

In D58281#1336902, @bnovkov wrote:

In D58281#1336697, @glebius wrote:

Instead of calling fib4_lookup() that has a different purpose, we need ifa_ifwithaddr() version that has fibnum argument.

Looking through net/if.c I noticed that ifa_ifwithnet seems to match what you're suggesting here, do you think we can use this function instead of adding a fibnum-aware version of `ifa_ifwithaddr()`?

In D53978#1336329, @cracauer wrote:

A benchmark that is easy to reproduce for everybody:

cd /usr/src/sys && find . -name \*.[ch] | ctime xargs grep -i -n preempt_thresh

no -i old code:
0:00.61 0.61 real 0.48 user 0.10 sys 95% CPU 0/1114 faults
old code -i:
0:24.59 24.59 real 24.44 user 0.11 sys 99% CPU 0/1047 faults
patch -i:
0:00.65 0.65 real 0.53 user 0.09 sys 95% CPU 0/1091 faults
patch no -i:
0:00.64 0.64 real 0.55 user 0.05 sys 95% CPU 0/1081 faults

In D56474#1336929, @ivy wrote:

In D56474#1317500, @markj wrote:

I don't really see how this is going to stay up-to-date: this seems to depend on developers remembering to update the file each time they add a dependency.

i agree and i object to landing this as-is; this simply isn't a workable solution.

when the original SBOM proposal was announced, i was disappointed when FF declared that pkgbase was out-of-scope for that work, but it seems like this and pkgbase and trying to solve similar problems when it comes to dependencies. can we not find a solution that works for both without requiring manual upkeep?

/usr/include is the standard location for profile.h. See below on a Red Hat server.

Upload the right patch

The current commit from the topic branch.

In D58264#1337050, @markj wrote:

In D58264#1336261, @kib wrote:

Note that the patch breaks existing tests, but IMO the current behavior is not logical. For instance, when I truss -f /bin/sh using process descriptors, shell is missing the zombie exit status. In other words, pdopenpid() breaks correct code.

Maybe pdwait() shouldn't let you reap a zombie that isn't your child?

Update the man page with the review comments.

In D58298#1337045, @kib wrote:

For the discussion, I am still somewhat surprised that so much of the FreeBSD symbols are leaked into linuxkpi headers, Most amusing and dangerous is PAGE_MASK (AFAIR).

In D58264#1336261, @kib wrote:

Note that the patch breaks existing tests, but IMO the current behavior is not logical. For instance, when I truss -f /bin/sh using process descriptors, shell is missing the zombie exit status. In other words, pdopenpid() breaks correct code.