_MEM_WRAP_NEEDS_RUNTIME_CHECK (malloc wrapping) in handy.h broken #16231
Comments
From @maukeCreated by @maukeIn handy.h there is this piece of code: # define _MEM_WRAP_NEEDS_RUNTIME_CHECK(n,t) \ A comment claims: * It's mathematically equivalent to This is false because _MEM_WRAP_NEEDS_RUNTIME_CHECK currently always returns It was lasted touched in commit 004073b The old version was: # define _MEM_WRAP_NEEDS_RUNTIME_CHECK(n,t) \ I don't understand the new version, so I wrote a little C program to compare ( uchar, char ) max(1) * 1 > UINT4_MAX? | old:no | new:yes | expected:no (!) Perl Info |
From @maukeOn Tue, 07 Nov 2017 16:31:51 -0800, mauke- wrote:
Code attached. |
From @mauke#include <stdio.h>
#include <limits.h>
#include <stdint.h>
#ifndef MEM_SIZE
#define MEM_SIZE size_t
#endif
#define MEM_SIZE_MAX ((MEM_SIZE)-1)
#define MEM_WRAP_NEEDS_RUNTIME_CHECK_1(n, t) \
( sizeof(MEM_SIZE) < sizeof(n) \
|| sizeof(t) > ((MEM_SIZE)1 << 8*(sizeof(MEM_SIZE) - sizeof(n))))
#define MEM_WRAP_NEEDS_RUNTIME_CHECK_2(n, t) \
(8 * sizeof(n) + sizeof(t) > sizeof(MEM_SIZE))
#define ACTUALLY_OVERFLOWS(n, t) \
((n) > MEM_SIZE_MAX / sizeof(t))
#define TRY_WITH2(n, t) \
printf("(%6s, %-15s) max(%zu) * %5zu > UINT%zu_MAX? | old:%-3s | new:%-3s | expected:%-3s %s\n", \
#n, #t, \
sizeof (n), sizeof (t), \
sizeof (MEM_SIZE), \
MEM_WRAP_NEEDS_RUNTIME_CHECK_1(n, t) ? "yes" : "no", \
MEM_WRAP_NEEDS_RUNTIME_CHECK_2(n, t) ? "yes" : "no", \
ACTUALLY_OVERFLOWS(n, t) ? "yes" : "no", \
MEM_WRAP_NEEDS_RUNTIME_CHECK_1(n, t) != ACTUALLY_OVERFLOWS(n, t) || \
MEM_WRAP_NEEDS_RUNTIME_CHECK_2(n, t) != ACTUALLY_OVERFLOWS(n, t) ? "(!)" : "")
#define TRY_WITH(t) \
do { \
const unsigned char uchar = UCHAR_MAX; \
const unsigned short ushort = USHRT_MAX; \
const unsigned int uint = UINT_MAX; \
const unsigned long ulong = ULONG_MAX; \
const unsigned long long ullong = ULLONG_MAX; \
TRY_WITH2(uchar, t); \
TRY_WITH2(ushort, t); \
TRY_WITH2(uint, t); \
TRY_WITH2(ulong, t); \
TRY_WITH2(ulong, t); \
TRY_WITH2(ullong, t); \
} while (0)
int main(void) {
TRY_WITH(char);
TRY_WITH(char [2]);
TRY_WITH(char [0xffff]);
TRY_WITH(char [0xffff+1]);
}
|
From @maukeOn Tue, 07 Nov 2017 16:31:51 -0800, mauke- wrote:
I have now reverted this commit in 445198b. I believe the new version could have been fixed by changing (8 * sizeof(n) + sizeof(t) > sizeof(MEM_SIZE)) to (8 * sizeof(n) + ILOG2(sizeof(t)) > 8 * sizeof(MEM_SIZE)) but then I'd have to define a macro to compute the integer log base 2 (rounded up) of sizeof(t) at compile time. This is possible but annoying (64 separate comparisons). It's also the opposite of a simplification. So unless we really need to do that, I'd rather not. |
From @iabynOn Sat, Nov 11, 2017 at 04:57:27AM -0800, l.mai@web.de via RT wrote:
I belatedly agree with your analysis. I clearly had a brain fart that -- |
|
The RT System itself - Status changed from 'new' to 'open' |
From @khwilliamsonOfflist On 03/06/2018 06:08 AM, Dave Mitchell wrote:
I'm glad to find out that happens to younger people than me, too. |
From @demerphqOn 6 March 2018 at 16:21, Karl Williamson <public@khwilliamson.com> wrote:
Happens to the younger folks more probably, shorter attention sp Ooh! Look! A butterfly! Yves -- |
|
@iabyn - Status changed from 'open' to 'resolved' |
Migrated from rt.perl.org#132415 (status was 'resolved')
Searchable as RT132415$
The text was updated successfully, but these errors were encountered: