comparison PROBLEMS @ 263:727739f917cb r20-5b30

Import from CVS: tag r20-5b30
author cvs
date Mon, 13 Aug 2007 10:24:41 +0200
parents 11cf20601dec
children ca9a9ec9c1c1
comparison
equal deleted inserted replaced
262:9d8607af9e13 263:727739f917cb
31 ** Don't use -O2 with gcc 2.7.2 under Intel/XXX without also using 31 ** Don't use -O2 with gcc 2.7.2 under Intel/XXX without also using
32 `-fno-strength-reduce'. 32 `-fno-strength-reduce'.
33 33
34 gcc will generate incorrect code otherwise. This bug is present in at 34 gcc will generate incorrect code otherwise. This bug is present in at
35 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and 35 least 2.6.x and 2.7.[0-2]. This bug has been fixed in GCC 2.7.2.1 and
36 later. This bug is also O/S independent. 36 later. This bug is O/S independent, but is limited to x86 architectures.
37
38 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
37 39
38 ** Don't use -O2 with gcc 2.7.2 under Intel architectures without also 40 ** Don't use -O2 with gcc 2.7.2 under Intel architectures without also
39 using `-fno-caller-saves'. 41 using `-fno-caller-saves'.
40 42
41 gcc will generate incorrect code otherwise. This bug is still present 43 gcc will generate incorrect code otherwise. This bug is still
42 in gcc 2.7.2.3. There have been no reports to indicate the bug is 44 present in gcc 2.7.2.3. There have been no reports to indicate the
43 present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S independent. 45 bug is present in egcs 1.0 (or pgcc 1.0) or later. This bug is O/S
46 independent, but limited to x86 architectures.
47
48 This problem is known to be fixed in egcs (or pgcc) 1.0 or later.
44 49
45 ** Excessive optimization with pgcc can break XEmacs 50 ** Excessive optimization with pgcc can break XEmacs
46 51
47 It has been reported on some systems that compiling with -O6 can lead 52 It has been reported on some systems that compiling with -O6 can lead
48 to XEmacs failures. The workaround is to use a lower optimization 53 to XEmacs failures. The workaround is to use a lower optimization