Front page | perl.perl5.porters |
Postings from March 2012
On Tue Mar 06 13:20:44 2012, zefram@fysh.org wrote:
> Father Chrysostomos via RT wrote:
> >Does anyone see anything wrong with applying that now?
>
> Yes, it looks like you've completely misunderstood
No, I just wrote the patch too quickly. :-)
> how the size
> is managed. Your change would copy a fixed number of bytes into the
> warning structure, ignoring the actual size of the input.
>
> As I noted earlier in this thread, the core's internal warning bit
> checks don't check that there are enough bits in the warning vector.
> But that's fine as long as it's only checking the statically-allocated
> warning categories and the user never writes an abnormally short bitset
> into ${^WARNING_BITS}. This does provide yet another, particularly
> easy, way to make perl break memory discipline, but it's through quite
> an explicit port into the internals. It won't arise if the user only
> writes bitsets previously read from the same place.
What Marc Lehmann says about an old common::sense being used with a
newer perl is a valid concern.
Here is a better version:
diff --git a/util.c b/util.c
index 1ff5913..a147ae0 100644
--- a/util.c
+++ b/util.c
@@ -2002,7 +2002,7 @@ S_ckwarn_common(pTHX_ U32 w)
STRLEN *
Perl_new_warnings_bitfield(pTHX_ STRLEN *buffer, const char *const bits,
STRLEN size) {
- const MEM_SIZE len_wanted = sizeof(STRLEN) + size;
+ const MEM_SIZE len_wanted = sizeof(STRLEN) + WARNsize;
PERL_UNUSED_CONTEXT;
PERL_ARGS_ASSERT_NEW_WARNINGS_BITFIELD;
@@ -2012,6 +2012,8 @@ Perl_new_warnings_bitfield(pTHX_ STRLEN *buffer,
const char *const bits,
PerlMemShared_realloc(buffer, len_wanted));
buffer[0] = size;
Copy(bits, (buffer + 1), size, char);
+ if (size < WARNsize)
+ Zero(bits, (char *)(buffer + 1) + size, WARNsize - size, char);
return buffer;
}
--
Father Chrysostomos
---
via perlbug: queue: perl5 status: open
https://rt.perl.org:443/rt3/Ticket/Display.html?id=111500
Thread Previous
|
Thread Next