[GIT PULL v4] Another round of build script tweaks

[Akira Yokosawa <akiyks@gmail.com>]

On 2016/05/11 02:17:27 -0700, Paul E. McKenney wrote:
> On Sun, May 08, 2016 at 12:23:49AM +0900, Akira Yokosawa wrote:
>> On 2016/05/07 06:20:39 -0700, Paul E. McKenney wrote:
>>> On Sat, May 07, 2016 at 09:06:18AM +0900, Akira Yokosawa wrote:
>>>> Hi, Paul.
>>>>
>>>> On 2016/05/06 04:09:35 -0700, Paul E. McKenney wrote:
>>>>> On Fri, May 06, 2016 at 08:47:13AM +0900, Akira Yokosawa wrote:
>>>>>> Is it OK to commit the fixes on the makefile-tweaks-v2 branch?
>>>>>
>>>>> Make the above changes, and I will commit it.  "git rebase -i" is
>>>>> of course your friend for this task.
>>>>>
>>>>
>>>> It seems there were no advance in master since then, so rebase was
>>>> not necessary.
>>>
>>> The issue is not master moving.  If I do move it in the next little
>>> bit, it will be for changes under CodeSamples, which won't conflict
>>> with your changes anyway.  So for me it will be a simple merge, not
>>> a problem at all.
>>>
>>> I am instead asking you to rebase in order to put the changes I
>>> requested into a few of the earlier patches.  For example:
>>>
>>> 	git branch makefile-tweaks-v3
>>> 	git rebase -i origin/master origin/master makefile-tweaks-v3
>>>
>>> Creating the branch allows you to keep the history.  This is important,
>>> as it allows you to do this as a check after you are done:
>>>
>>> 	git diff makefile-tweaks-v2
>>>
>>> Any diffs should of course correspond to intentional changes.
>>>
>>> Alternatively, you can do the following:
>>>
>>> 	git checkout -b makefile-tweaks-v3 origin/master
>>>
>>> Then for each commit ID "X":
>>>
>>> 	git cherry-pick -n X
>>> 	# Do any needed editing
>>> 	git commit -a -s
>>>
>>> At the end, you can again use "git diff" to check your work.
>>>
>>> You don't have a huge number of commits, so either way will work well.
>>> If you had (say) 20 commits, "git rebase -i" would work much better.
>>>
>>>> Pull request follows. Note that this is relative to the last one.
>>>> Do you prefer a request relative to the current master?
>>>
>>> Please!  That allows you to create FAQ-BUILD.txt to start with and
>>> have your changes relative to that.
>>
>> I see. Attached is the retry of the PULL request. I hope it is what you are
>> expecting. In the future, I'll do things this way. There are a lot of things
>> I need to learn regarding the workflow using git.
> 
> Understood, "git rebase" was a bit scary for me the first time I got a
> conflict.  But could you please at least give it a try?  It is the only
> way to become comfortable with it, and that skill is quite valuable.
> 
> 							Thanx, Paul

Hi, Paul.

This time, I tried "git rebase -i".

Made a branch makefile-tweaks-v4 on makefile-tweaks-v2, then did "git rebase -i".
There actually were commits you pushed after makefile-tweaks-v2 was branched, so
the rebase does have some meaning. (no conflict -- went smoothly)

Thank you for giving me a chance to try rebasing. Well, this is an interesting, or
rather, amazing feature of git! 

Appended is the pull request of v4. Does this seem right?

                                                        Thanks, Akira
---
Change from v3:
    Rebased to the current master 08259a39f7dc ("Make spinlock primitives abort
    on error for easier debugging") using "git rebase -i"

Changes from v2:
    changes from "Fix regression in Makefile" to "Makefile: Add comment on why
    'extraction' is an empty target".  In short, 
        1) regression in v2 ("make distclean") is fixed
        2) questions on build issues in FAQ.txt are moved to a new file FAQ-BUILD.txt,
        3) use of empty targets "perfbook_aux", "perfbook-1c_aux" and
           "perfbook-hb_aux" are abandoned and actual "perfbook.aux", etc. are
           used instead,
        4) rule for qqz.tex is added and qqz.tex is removed from the repository, 
        5) rule for (.dot -> pdf) is removed and rule of (.dot -> .eps) is used
           instead. store15tred.esp is added to the repository,
        6) rules related to perfbook_html is removed,

Changes from v1:
    Reorganized the entire series so that minor changes would come
    first.

---
The following changes since commit 08259a39f7dccdd3aead9add883291678a017209:

  Make spinlock primitives abort on error for easier debugging (2016-05-06 10:46:51 -0700)

are available in the git repository at:

  https://github.com/akiyks/perfbook.git makefile-tweaks-v4

for you to fetch changes up to 7f46bce80811f24a1d623da37920964294c623be:

  Makefile: Add comment on why 'extraction' is an empty target (2016-05-11 22:54:49 +0900)

----------------------------------------------------------------
Akira Yokosawa (13):
      Add font installation check
      Add short name targets in Makefile
      sig-theft: Fix .dot source for dot - graphviz version 2.36.0
      Make default target of "make" overridable
      Makefile: Reorder rules
      Makefile: Use wildcards
      Further improvement of build scripts
      Fix regression in Makefile
      Place trouble-shooting questions to FAQ-BUILD.txt
      Further tweaks of Makefile
      rm qqz.tex and add store15tred.esp to repository
      Remove rules related to perfbook_html
      Makefile: Add comment on why 'extraction' is an empty target

 .gitignore              |   11 +-
 FAQ-BUILD.txt           |   80 +
 FAQ.txt                 |   68 +-
 Makefile                |  353 +--
 advsync/store15tred.eps |  653 ++++
 count/sig-theft.dot     |   17 +-
 count/sig-theft.eps     |  338 +--
 qqz.tex                 | 7707 -----------------------------------------------
 utilities/eps2pdf.sh    |   12 +
 utilities/runlatex.sh   |   13 +-
 10 files changed, 1062 insertions(+), 8190 deletions(-)
 create mode 100644 FAQ-BUILD.txt
 create mode 100644 advsync/store15tred.eps
 delete mode 100644 qqz.tex