Re: [PATCH 1/2] commit-slab: document clear_$slabname()

2013-11-29 Thread Thomas Rast
Jonathan Nieder jrnie...@gmail.com writes:

 Thomas Rast wrote:

 + *
 + * - void clear_indegree(struct indegree *);
 + *
 + *   Free the slab's data structures.

 Tense shift (previous descriptions were in the present tense, while
 this one is in the imperative).

 More importantly, this doesn't answer the questions I'd have if I were
 in a hurry, which are what exactly is being freed (has the slab taken
 ownership of any memory from the user, e.g. when elemtype is a
 pointer?) and whether the slab needs to be init_ ed again.

 Maybe something like the following would work?
[...]

Ok, I see that while I was procrastinating, you sorted this out and
Junio merged it to next.

Thanks, both.

-- 
Thomas Rast
t...@thomasrast.ch
--
To unsubscribe from this list: send the line unsubscribe git in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


Re: [PATCH 1/2] commit-slab: document clear_$slabname()

2013-11-27 Thread Eric Sunshine
On Mon, Nov 25, 2013 at 3:39 PM, Junio C Hamano gits...@pobox.com wrote:
 Thomas Rast t...@thomasrast.ch writes:

 The clear_$slabname() function was only documented by source code so
 far.  Write something about it.

 Signed-off-by: Thomas Rast t...@thomasrast.ch
 ---
  commit-slab.h | 4 
  1 file changed, 4 insertions(+)

 diff --git a/commit-slab.h b/commit-slab.h
 index d4c8286..d77aaea 100644
 --- a/commit-slab.h
 +++ b/commit-slab.h
 @@ -24,6 +24,10 @@
   *   to each commit. 'stride' specifies how big each array is.  The slab
   *   that id initialied by the variant without _with_stride associates

 Is that id a typo for is?

And, s/initialied/initialized/

   *   each commit with an array of one integer.
 + *
 + * - void clear_indegree(struct indegree *);
 + *
 + *   Free the slab's data structures.
   */

  /* allocate ~512kB at once, allowing for malloc overhead */
 --
 To unsubscribe from this list: send the line unsubscribe git in
 the body of a message to majord...@vger.kernel.org
 More majordomo info at  http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line unsubscribe git in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


[PATCH 1/2] commit-slab: document clear_$slabname()

2013-11-25 Thread Thomas Rast
The clear_$slabname() function was only documented by source code so
far.  Write something about it.

Signed-off-by: Thomas Rast t...@thomasrast.ch
---
 commit-slab.h | 4 
 1 file changed, 4 insertions(+)

diff --git a/commit-slab.h b/commit-slab.h
index d4c8286..d77aaea 100644
--- a/commit-slab.h
+++ b/commit-slab.h
@@ -24,6 +24,10 @@
  *   to each commit. 'stride' specifies how big each array is.  The slab
  *   that id initialied by the variant without _with_stride associates
  *   each commit with an array of one integer.
+ *
+ * - void clear_indegree(struct indegree *);
+ *
+ *   Free the slab's data structures.
  */
 
 /* allocate ~512kB at once, allowing for malloc overhead */
-- 
1.8.5.rc3.397.g2a3acd5

--
To unsubscribe from this list: send the line unsubscribe git in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


Re: [PATCH 1/2] commit-slab: document clear_$slabname()

2013-11-25 Thread Jonathan Nieder
Thomas Rast wrote:

 The clear_$slabname() function was only documented by source code so
 far.  Write something about it.

Good idea.

[...]
 --- a/commit-slab.h
 +++ b/commit-slab.h
 @@ -24,6 +24,10 @@
   *   to each commit. 'stride' specifies how big each array is.  The slab
   *   that id initialied by the variant without _with_stride associates
   *   each commit with an array of one integer.
 + *
 + * - void clear_indegree(struct indegree *);
 + *
 + *   Free the slab's data structures.

Tense shift (previous descriptions were in the present tense, while
this one is in the imperative).

More importantly, this doesn't answer the questions I'd have if I were
in a hurry, which are what exactly is being freed (has the slab taken
ownership of any memory from the user, e.g. when elemtype is a
pointer?) and whether the slab needs to be init_ ed again.

Maybe something like the following would work?

- void clear_indegree(struct indegree *);

  Empties the slab.  The slab can be reused with the same
  stride without calling init_indegree again or can be
  reconfigured to a different stride by calling
  init_indegree_with_stride.

  Call this function before the slab falls out of scope to
  avoid leaking memory.

Thanks,
Jonathan
--
To unsubscribe from this list: send the line unsubscribe git in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


Re: [PATCH 1/2] commit-slab: document clear_$slabname()

2013-11-25 Thread Junio C Hamano
Thomas Rast t...@thomasrast.ch writes:

 The clear_$slabname() function was only documented by source code so
 far.  Write something about it.

 Signed-off-by: Thomas Rast t...@thomasrast.ch
 ---
  commit-slab.h | 4 
  1 file changed, 4 insertions(+)

 diff --git a/commit-slab.h b/commit-slab.h
 index d4c8286..d77aaea 100644
 --- a/commit-slab.h
 +++ b/commit-slab.h
 @@ -24,6 +24,10 @@
   *   to each commit. 'stride' specifies how big each array is.  The slab
   *   that id initialied by the variant without _with_stride associates

Is that id a typo for is?

   *   each commit with an array of one integer.
 + *
 + * - void clear_indegree(struct indegree *);
 + *
 + *   Free the slab's data structures.
   */
  
  /* allocate ~512kB at once, allowing for malloc overhead */
--
To unsubscribe from this list: send the line unsubscribe git in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html