Re: [Patch,AVR]: Fix PR27663

2011-05-16 Thread Denis Chertykov
2011/5/16 Georg-Johann Lay : > Denis Chertykov schrieb: >> 2011/5/11 Georg-Johann Lay : >>> Denis Chertykov schrieb: 2011/5/2 Georg-Johann Lay : > This is a fix for an optimization flaw when a long value is composed > from byte values. > > For -fsplit-wide-types (which is

Re: [Patch,AVR]: Fix PR27663

2011-05-16 Thread Georg-Johann Lay
Denis Chertykov schrieb: > 2011/5/11 Georg-Johann Lay : >> Denis Chertykov schrieb: >>> 2011/5/2 Georg-Johann Lay : >>> This is a fix for an optimization flaw when a long value is composed from byte values. For -fsplit-wide-types (which is still default for avr) the code is

Re: [Patch,AVR]: Fix PR27663

2011-05-13 Thread Denis Chertykov
2011/5/11 Georg-Johann Lay : > Denis Chertykov schrieb: >> >> 2011/5/2 Georg-Johann Lay : >> >>> This is a fix for an optimization flaw when a long value is composed >>> from byte values. >>> >>> For -fsplit-wide-types (which is still default for avr) the code is >>> worse than with -fno-split-wide

Re: [Patch,AVR]: Fix PR27663

2011-05-11 Thread Georg-Johann Lay
Denis Chertykov schrieb: 2011/5/2 Georg-Johann Lay : This is a fix for an optimization flaw when a long value is composed from byte values. For -fsplit-wide-types (which is still default for avr) the code is worse than with -fno-split-wide-types. The code for the test case is better in either

Re: [Patch,AVR]: Fix PR27663

2011-05-06 Thread Georg-Johann Lay
Denis Chertykov schrieb: > 2011/5/2 Georg-Johann Lay : >> This is a fix for an optimization flaw when a long value is composed >> from byte values. >> >> For -fsplit-wide-types (which is still default for avr) the code is >> worse than with -fno-split-wide-types. The code for the test case is >> be

Re: [Patch,AVR]: Fix PR27663

2011-05-02 Thread Denis Chertykov
2011/5/2 Georg-Johann Lay : > This is a fix for an optimization flaw when a long value is composed > from byte values. > > For -fsplit-wide-types (which is still default for avr) the code is > worse than with -fno-split-wide-types. The code for the test case is > better in either situations, i.e. c

[Patch,AVR]: Fix PR27663

2011-05-02 Thread Georg-Johann Lay
This is a fix for an optimization flaw when a long value is composed from byte values. For -fsplit-wide-types (which is still default for avr) the code is worse than with -fno-split-wide-types. The code for the test case is better in either situations, i.e. compared to code without the patch, but