From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeffrey A Law To: hjl@lucon.org (H.J. Lu) Cc: florian@suse.de, paubert@iram.es, egcs@cygnus.com Subject: Re: egcs-problem 1.0.2 Date: Sun, 05 Apr 1998 17:27:00 -0000 Message-id: <889.891803114@hurl.cygnus.com> References: X-SW-Source: 1998-04/msg00197.html In message < m0yLuSz-00058iC@ocean.lucon.org >you write: > > This is the "linux asm" problem that I haven't been able to look at > > for the last month or so due to personal and company commitments. > > > > Basically there's a couple problems: > > > > * The linux x86 asm for strstr is totally bogus. > > Does the code below make sense? Not sure what you're asking. If you want I can resend the analysis. Basically the clobbers are bogus and the asm itself uses too many registers to be safely compiled. This is the exact code we're using as an example of the asm problems that we need to address. > > > > * GCC does not issue a warning/error when it encouters the bogus > > asm and instead generates incorrect code. > > Agreed. I don't think it is too hard to issue an error. The problem is that many linux asms use a invalid construct, so the error has to be conditional so that the linux folks have the opportunity to transition away from the bogus way they've been writing asms. jeff