* debian-merge: (206 commits)
  New upstream version: 0.2.5.4-alpha
  Fix hyphen handling in format_changelog
  reflow changelog.
  place the cherries on top and call it done
  promote two stanzas; shuffle around two more
  touchups on upcoming changelog
  resolve a typo: sanboxing->sandboxing.
  two changelog tweaks suggested by Andrea
  Write a release blurb
  Re-order and combine changelog sections
  Integrate another entry to the changelog; twiddle changelog more
  format_changelog.py now formats the first section, in-place.
  Bump version to 0.2.5.4-alpha.
  fix memory leak in dump_exit_policy_to_string tests
  Include the scan-build.sh script I used for 8793
  Changes file for scan-build fixes (#8793)
  Revise changelog verbiage; try to tighten it up.
  ChangeLog: spell-check, prefer "relay" to "node" or "server"
  Try to sort the changelog a little more
  Fwd-port: Add a missing changelog entry back from 0.2.4.11-alpha
  ...

Tor 0.2.5.4-alpha release

* tag 'tor-0.2.5.4-alpha': (205 commits)
  Fix hyphen handling in format_changelog
  reflow changelog.
  place the cherries on top and call it done
  promote two stanzas; shuffle around two more
  touchups on upcoming changelog
  resolve a typo: sanboxing->sandboxing.
  two changelog tweaks suggested by Andrea
  Write a release blurb
  Re-order and combine changelog sections
  Integrate another entry to the changelog; twiddle changelog more
  format_changelog.py now formats the first section, in-place.
  Bump version to 0.2.5.4-alpha.
  fix memory leak in dump_exit_policy_to_string tests
  Include the scan-build.sh script I used for 8793
  Changes file for scan-build fixes (#8793)
  Revise changelog verbiage; try to tighten it up.
  ChangeLog: spell-check, prefer "relay" to "node" or "server"
  Try to sort the changelog a little more
  Fwd-port: Add a missing changelog entry back from 0.2.4.11-alpha
  Reformat the changelog for 0.2.5.4-alpha.  No textual changes.
  ...

Work on style, add some sentences to blurb, explain that 10468 is more
general than had been described, etc etc

Probably releasing within ~22 hours, pending testing

Otherwise we'll never remember what to re-run before each release, or
whether we tried a given checker.

When I merged the fix for #7351, and implemented proposal 214 (4-byte
circuit IDs), I forgot to add a changes file.  Later, we never noticed
that it didn't have one.

Resolves ticket #11555.  Thanks to cypherpunks for noticing this was
missing.

This is a cherry-pick of 75e10f58 into
the master branch.

Also, add a script to do this, since doing it manually with fmt sucks.

Conflicts:
	src/or/main.c

That is, GETINFO limits/max-mem-in-queues

If we can't detect the physical memory, the new default is 8 GB on
64-bit architectures, and 1 GB on 32-bit architectures.

If we *can* detect the physical memory, the new default is
  CLAMP(256 MB, phys_mem * 0.75, MAX_DFLT)
where MAX_DFLT is 8 GB on 64-bit architectures and 2 GB on 32-bit
architectures.

You can still override the default by hand.  The logic here is simply
trying to choose a lower default value on systems with less than 12 GB
of physical RAM.

Conflicts:
	src/or/circuitbuild.c

Use a per-channel ratelim_t to control the rate at which we report
failures for each channel.

Explain why I picked N=32.

Never return a zero circID.

Thanks to Andrea and to cypherpunks.