Version française
Home     About     Download     Resources     Contact us    

This site is updated infrequently. For up-to-date information, please visit the new OCaml website at

Browse thread
[Caml-list] Bp_val and Data_custom_val
[ Home ] [ Index: by date | by threads ]
[ Search: ]

[ Message by date: previous | next ] [ Message in thread: previous | next ] [ Thread: previous | next ]
Date: -- (:)
From: Yaron M. Minsky <yminsky@C...>
Subject: Re: [Caml-list] Bp_val and Data_custom_val
Interesting.  So I guess that my code has been overwriting the pointer
to the custom operations block.

I wonder, though, what the significance of the following paragraph from
the documentation is, and in particular what is meant by "the format of
custom blocks is compatible with that of finalized blocks".

Custom blocks generalize the finalized blocks that were present in
Objective Caml prior to version 3.00. For backward compatibility, the
format of custom blocks is compatible with that of finalized blocks, and
the alloc_final function is still available to allocate a custom block
with a given finalization function, but default comparison, hashing and
serialization functions. alloc_final(n, f, used, max) returns a fresh
custom block of size n words, with finalization function f. The first
word is reserved for storing the custom operations; the other n-1 words
are available for your data. The two parameters used and max are used to
control the speed of garbage collection, as described for alloc_custom.

Christopher Quinn wrote:
 > hi,
 > Bp_val() returns a pointer to the first word(byte really since Bp ==
 > byte pointer) after the header, while Data_val_custom() returns the
 > address of the second word (value), which is because with finalisation
 > the first word is occupied by a pointer to the specific finalisation
 > struct of handlers (finalise, serialise, de-serialise et al).
 > so it is not the same layout!
 > - chris
 > Yaron M. Minsky wrote:
 >> I'm wondering if someone can help me out in understanding the
 >> difference between Bp_val and Data_custom_val.  These are both macros
 >> for accessing caml values from C.  I lifted and adapted some code from
 >> the distribution for building a berkeley DB interface.  In that code,
 >> Bp_val is used to access the contents of finalized blocks.  I adapted
 >> that code but shifted to custom blocks, which according to the docs
 >> have the same layout.  The docs also suggest using Data_custom_val for
 >> custom blocks.  But when I try to do that, I end up with segfaults.
 >> I'm wondering what precisely the difference is.  Here's a snippet of
 >> my data type definitions and the macros I use for accessing the data
 >> contained in the custom blocks holding that data:
 >> struct camldb {
 >>   DB *db;
 >>   int closed;
 >> };
 >> struct camltxn {
 >>   DB_TXN *txn;
 >>   int closed;
 >> };
 >> #define UW_db(v) (((struct camldb *)Bp_val(v))->db)
 >> #define UW_db_closed(v) (((struct camldb *)Bp_val(v))->closed)
 >> #define UW_txn(v) (((struct camltxn *)Bp_val(v))->txn)
 >> #define UW_txn_closed(v) (((struct camltxn *)Bp_val(v))->closed)
 >> Anything fishy about that?  And any problem with replacing Bp_val with
 >> Data_custom_val here?
 >> By the way, in a separate post, I put a URL for where you can download
 >> the entire wrapper in question.  Here it is again:
 >> y
 >> -------------------
 >> To unsubscribe, mail Archives:
 >> Bug reports: FAQ:
 >> Beginner's list:

To unsubscribe, mail Archives:
Bug reports: FAQ:
Beginner's list: