Re: [3/4] RFC: add DWARF index support

[Jan Kratochvil <jan.kratochvil@redhat.com>]

On Tue, 06 Jul 2010 18:50:32 +0200, Tom Tromey wrote:
> My thinking is that we only need 64 bit for CU offsets.  It seems very
> unlikely to me that an index file could be bigger than 4G (not 2G, since
> the offsets are all unsigned).
> 
> Also I have been thinking a little about the 64 bit obstack patch.  I
> think we can avoid it by going a different route:
[...]
> Second, if we have a section we cannot map (say it is compressed or
> something), we could just malloc memory for it instead of using the
> obstack.  The last "read DWARF in the background" patch has the needed
> infrastructure to make this work ok, basically having a destructor
> function for sections in dwarf2read.c.
> 
> What do you think?

Frankly, I do not want to reinvent the optimizations for tiny, small, medium,
compact, large and huge memory models. :-)  Any size should be 64bit and it is
a bug if it is not.  I was reluctant to make review notes on more places using
`int' memory sizes in your patch.  I find needlessly labor-expensive to
evaluate each memory size if it really can always fit in 32bit or not.


> Tom> +   [1] The mtime of the objfile, as a 64-bit little-endian value.
> 
> Jan> Do you have some specific needs for storing timestamp into the file?
> Jan> When such index files will be packaged by distros it may create needless
> Jan> differences across various verifications.  Isn't make(1)-like timestamp
> Jan> dependency enough?
> 
> If the distro processes modify timestamps, then it seems like we also
> could not rely on them to remain ordered.

With various filesystems and network filesystems having interesting timestamp
rules it would be best to avoid depending on it at all.


> One idea would be to store the build-id and only verify it if the
> objfile has a .note.gnu.build-id section.

Good point, I find .gdb-index file should have exactly the same verification
rules as the .debug file has.  (That is crc32 if build-id is not available.)


Thanks,
Jan