The Chief wrote (in personal mail):
I have just intercepted a thread on the Mingw mailing lists - see below. It seems that the problem that I reported with "-fomit-frame-pointer" is a gcc bug.
Fine (as far as I'm concerned, so I won't have to fix it ;-).
I hope you don't mind me replying to the list. I'm doing so just to support my point that switching completely to gcc-3 (i.e., dropping gcc-2 support or doing all new development in the gcc-3) might still be a little risky ...
But David Bryan has experienced similar problems without using that switch. So, there may well be a problem with the code generation under gcc-3.2.1.
Do they also seem backend-related? Otherwise, of course, I'd need a test case to tell more ...
[ Mingw list mail and C example of the bug skipped ]
Frank
On 19 Dec 2002 at 5:28, Frank Heckenbach wrote:
The Chief wrote (in personal mail):
I have just intercepted a thread on the Mingw mailing lists - see below. It seems that the problem that I reported with "-fomit-frame-pointer" is a gcc bug.
Fine (as far as I'm concerned, so I won't have to fix it ;-).
I hope you don't mind me replying to the list. I'm doing so just to support my point that switching completely to gcc-3 (i.e., dropping gcc-2 support or doing all new development in the gcc-3) might still be a little risky ...
True.
But David Bryan has experienced similar problems without using that switch. So, there may well be a problem with the code generation under gcc-3.2.1.
Do they also seem backend-related? Otherwise, of course, I'd need a test case to tell more ...
I am not sure how to ascertain whether the problem is backend-related ...
Best regards, The Chief -------- Prof. Abimbola A. Olowofoyeku (The African Chief) web: http://www.bigfoot.com/~african_chief/
Prof A Olowofoyeku (The African Chief) wrote:
But David Bryan has experienced similar problems without using that switch. So, there may well be a problem with the code generation under gcc-3.2.1.
Do they also seem backend-related? Otherwise, of course, I'd need a test case to tell more ...
I am not sure how to ascertain whether the problem is backend-related ...
It might not be apparent indeed, though if they're similar, I suppose they're backend-related ...
Frank