This passes "mandoc -Tlint".

This passes "mandoc -Tlint".

Fixes.

In D15687#459500, @antoine wrote:

Did you verify that the problem on i386 described in r251668 no longer occurs?

I think the problem in libssp_nonshared.a is brought in bogusly by the
libc.ldscript.

Its an ext4 feature, linux calls it crtime (creation time), just to be different ;).

In general I prefer we support a lower number of the protocol but that we support it well.

In D20598#450509, @gerald wrote:

Is there anything to be done for modern versions of GCC (GCC 8 or later)?

I think andreast@ and gerald@ may want to be kept in the loop for the GCC update then :).

Very exciting ... will upstream GCC require changes as well?

Adding fsu@ in case he wants to mention something else :).

Yes, we do fully support ext4 read and write in FreeBSD 12+. We also support Extended Attributes, ACLs, and most features with the exception of journalling and encryption. As mentioned in BSDNow, 12.1 will have a DTrace provider.

Thanks for working on this. For the record,

The changes seem correct.
I don't think a simple phabricator acceptance from a committer is suficient approval for a non src committer though, so I hereby explicitly approve it.

This has rusted since the last revision: fts had some changes (mostly just updated prototypes) and glob has been rewritten.
It needs some fresh eyes into it (cc'ing brooks in case he wants to take over).

Most of these don't make sense as they cant overflow, and those that do were committed.

This was committed as r337456.

Is this still a valid review? I recall something related was committed.

I still hope to hear from the DTrace experts, but I like this very much. Thanks!

Very cool (but I am biased since I suggested the fusefs-dtrace changes as a starting point)!

In D19325#415619, @mckusick wrote:

I recently added the new error EINTEGRITY which is intended for use when a cylinder group or other filesystem structure has an integrity error.

Upon further thought: the situation is very different for ext2fs than to a filesystem in which the all the system depends as in UFS.

In D19327#414960, @fsu wrote:

In D19327#413791, @pfg wrote:

Note why we specifically set ip->i_flags = 0 since r294652:

The ip->i_flags will be zero in all cases inside this function because we does not read new inode from disk at all for now and inode is allocated thru malloc with M_ZERO flag.

OK, you are rught, we were not allocating (with M_ZERO) this before.

Hi Kirk;
We have this as a panic in UFS, as well.

We should probably keep this panic for consistency with UFS, which also panics on this case.

Note why we specifically set ip->i_flags = 0 since r294652:

LGTM.
FWIW, the GPLv3 version does an ugly calloc() dance with pointers which looks a lot more confusing.

I am unsure if it is a style rule but we generally place such pragmas before and after all declarations so that they don't get in the way of easier reading.

In D13706#411805, @rgrimes wrote:

pfg I think you miss understood, and upon looking I was not clear, it is not you that I am unhappy with about "playing with copyrights and licenses", it is NetBSD and possibly Charles Hannum who as Robert points out modified a license without ALL authors approval. I would be very upset if someone disturbed my disclaimer in this manner.

In D13706#411735, @rgrimes wrote:

Upon closer examination of this there are some issues, it alters a license post fact of copyrights asserted by some, and changes the clause in the license from Author to NetBSD, which effects the since of the copyright. I really wish the playing with Copyrights and Licenses would stop, this is a good example, of a well meant change, causing an ugly problem. For examle, the disclaimer is now in the name NetBSD, so what happened to the disclaim asserted by nCircle Network, and Robert Watson?

(trasz is the expert here)

If it's not needed then just lets not spend time on it.

Note that I have no idea of MIPS asm, just noticed the commit on OpenBSD MIPS64 .. feel free to commit if it works for you.

Unsign uh_hashmask
(Don't expect any failure but will test anyways)

In D19148#409628, @mav wrote:

In D19148#409623, @pfg wrote:

I thought of that but I saw a -1 when calculating it:

hash->uh_hashmask = hash->uh_hashsize - 1;

.A hashsize of zero makes little sense but it is still a valid value (?).

hashsize of zero makes no sense to me. I haven't checked how exactly it is initialized and whether it is used for some implementation reasons, but smallest possible size of hash is 1 with mask of 0. For size of 0 there is just no meaningful mask. Plus few places where uh_hashmask is used, do not care whether it is negative or overflown unsigned.

Missing hval unsigning (pointed out by mav).

In D19148#409622, @mav wrote:

Generally I agree with it. I spotted it too when worked on my change. It just needs 8 times more RAM then what we have now to overflow, so I preferred my patch to be minimal. Just going that way, why not make uh_hashmask also unsigned?

FWIW, this is currently my favorite IDE on FreeBSD: netbeans has problems refreshing.
VSCode is not yet ported :(.

FWIW, I have a patch from the late ache@ to finish the removal of __collate_range_cmp() started in r302512.
It is practically impossible to remove a symbol as that would in theory break the ABI, so I understand we may be wary of adding new ones.