public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
* [Bug c++/43645] New: 32 bit pointer in top half of VM casts to unsigned long long incorrectly @ 2010-04-05 10:17 alangcarter at gmail dot com 2010-04-05 10:18 ` [Bug c++/43645] " alangcarter at gmail dot com ` (3 more replies) 0 siblings, 4 replies; 5+ messages in thread From: alangcarter at gmail dot com @ 2010-04-05 10:17 UTC (permalink / raw) To: gcc-bugs A 32 bit pointer with highest bit set ( >= 0x80000000) is incorrectly cast to unsigned long long. The top 32 bits of the unsigned long long are all set 1, not 0. The problem does not occur if the pointer is less than 0x80000000 or if it is first cast to unsigned long and that is then cast to unsigned long long. Obviously the highest bit would indicate a negative value in a signed int of the same size. It looks like some signed logic is being invoked inappropriately Details provided from 4.3.3 on Ubuntu, also seen on 4.4.1 on Ubuntu and 4.1.2 on NetBSD. Exact same behaviour also seen with Microsoft Visual C++ 2008, but not on Apple's version of 4.0.1. Here is a small test program which demonstrates the problem, I'll attach the .ii and gcc output: #include <iostream> #include <iomanip> #include <string> using namespace std; template<typename T> void dump(string legend, T* data) { // Display the bytes of the supplied type and its cout operator << rendition unsigned char* puc = (unsigned char*)data; cout << legend << ":" << endl; for(unsigned int i = 0; i < sizeof(T); i++) cout << hex << setw(2) << setfill('0') << (unsigned int)(puc[i]) << " "; cout << endl << hex << setw(16) << setfill('0') << *data << endl << endl; } int main(int argc, char **argv) { void* pv1 = (void*)0x12345678; void* pv2 = (void*)0x87654321; unsigned long long int ulli; dump("small void*", &pv1); ulli = (unsigned long long int)pv1; dump("small void* cast to unsigned long long", &ulli); ulli = (unsigned long long int)(unsigned long int)pv1; dump("small void* cast to unsigned long cast to unsigned long long", &ulli); dump("large void*", &pv2); ulli = (unsigned long long int)pv2; dump("large void* cast to unsigned long long", &ulli); ulli = (unsigned long long int)(unsigned long int)pv2; dump("large void* cast to unsigned long cast to unsigned long long", &ulli); } -- Summary: 32 bit pointer in top half of VM casts to unsigned long long incorrectly Product: gcc Version: unknown Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ AssignedTo: unassigned at gcc dot gnu dot org ReportedBy: alangcarter at gmail dot com http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43645 ^ permalink raw reply [flat|nested] 5+ messages in thread
* [Bug c++/43645] 32 bit pointer in top half of VM casts to unsigned long long incorrectly 2010-04-05 10:17 [Bug c++/43645] New: 32 bit pointer in top half of VM casts to unsigned long long incorrectly alangcarter at gmail dot com @ 2010-04-05 10:18 ` alangcarter at gmail dot com 2010-04-05 10:20 ` alangcarter at gmail dot com ` (2 subsequent siblings) 3 siblings, 0 replies; 5+ messages in thread From: alangcarter at gmail dot com @ 2010-04-05 10:18 UTC (permalink / raw) To: gcc-bugs ------- Comment #1 from alangcarter at gmail dot com 2010-04-05 10:18 ------- Created an attachment (id=20307) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20307&action=view) Source code of demo program -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43645 ^ permalink raw reply [flat|nested] 5+ messages in thread
* [Bug c++/43645] 32 bit pointer in top half of VM casts to unsigned long long incorrectly 2010-04-05 10:17 [Bug c++/43645] New: 32 bit pointer in top half of VM casts to unsigned long long incorrectly alangcarter at gmail dot com 2010-04-05 10:18 ` [Bug c++/43645] " alangcarter at gmail dot com @ 2010-04-05 10:20 ` alangcarter at gmail dot com 2010-04-05 10:20 ` alangcarter at gmail dot com 2010-04-05 10:27 ` rguenth at gcc dot gnu dot org 3 siblings, 0 replies; 5+ messages in thread From: alangcarter at gmail dot com @ 2010-04-05 10:20 UTC (permalink / raw) To: gcc-bugs ------- Comment #3 from alangcarter at gmail dot com 2010-04-05 10:20 ------- Created an attachment (id=20309) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20309&action=view) g++ command and console output -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43645 ^ permalink raw reply [flat|nested] 5+ messages in thread
* [Bug c++/43645] 32 bit pointer in top half of VM casts to unsigned long long incorrectly 2010-04-05 10:17 [Bug c++/43645] New: 32 bit pointer in top half of VM casts to unsigned long long incorrectly alangcarter at gmail dot com 2010-04-05 10:18 ` [Bug c++/43645] " alangcarter at gmail dot com 2010-04-05 10:20 ` alangcarter at gmail dot com @ 2010-04-05 10:20 ` alangcarter at gmail dot com 2010-04-05 10:27 ` rguenth at gcc dot gnu dot org 3 siblings, 0 replies; 5+ messages in thread From: alangcarter at gmail dot com @ 2010-04-05 10:20 UTC (permalink / raw) To: gcc-bugs ------- Comment #2 from alangcarter at gmail dot com 2010-04-05 10:19 ------- Created an attachment (id=20308) --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20308&action=view) .ii file from compiling test program -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43645 ^ permalink raw reply [flat|nested] 5+ messages in thread
* [Bug c++/43645] 32 bit pointer in top half of VM casts to unsigned long long incorrectly 2010-04-05 10:17 [Bug c++/43645] New: 32 bit pointer in top half of VM casts to unsigned long long incorrectly alangcarter at gmail dot com ` (2 preceding siblings ...) 2010-04-05 10:20 ` alangcarter at gmail dot com @ 2010-04-05 10:27 ` rguenth at gcc dot gnu dot org 3 siblings, 0 replies; 5+ messages in thread From: rguenth at gcc dot gnu dot org @ 2010-04-05 10:27 UTC (permalink / raw) To: gcc-bugs ------- Comment #4 from rguenth at gcc dot gnu dot org 2010-04-05 10:27 ------- This is implementation-defined. See http://gcc.gnu.org/onlinedocs/gcc-4.4.3/gcc/Arrays-and-pointers-implementation.html#Arrays-and-pointers-implementation -- rguenth at gcc dot gnu dot org changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |INVALID Summary|32 bit pointer in top half |32 bit pointer in top half |of VM casts to unsigned long|of VM casts to unsigned long |long incorrectly |long incorrectly http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43645 ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2010-04-05 10:27 UTC | newest] Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2010-04-05 10:17 [Bug c++/43645] New: 32 bit pointer in top half of VM casts to unsigned long long incorrectly alangcarter at gmail dot com 2010-04-05 10:18 ` [Bug c++/43645] " alangcarter at gmail dot com 2010-04-05 10:20 ` alangcarter at gmail dot com 2010-04-05 10:20 ` alangcarter at gmail dot com 2010-04-05 10:27 ` rguenth at gcc dot gnu dot org
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).