58 lines · plain
1===============================================2Release notes for Linux Kernel VFP support code3===============================================4 5Date: 20 May 20046 7Author: Russell King8 9This is the first release of the Linux Kernel VFP support code. It10provides support for the exceptions bounced from VFP hardware found11on ARM926EJ-S.12 13This release has been validated against the SoftFloat-2b library by14John R. Hauser using the TestFloat-2a test suite. Details of this15library and test suite can be found at:16 17 http://www.jhauser.us/arithmetic/SoftFloat.html18 19The operations which have been tested with this package are:20 21 - fdiv22 - fsub23 - fadd24 - fmul25 - fcmp26 - fcmpe27 - fcvtd28 - fcvts29 - fsito30 - ftosi31 - fsqrt32 33All the above pass softfloat tests with the following exceptions:34 35- fadd/fsub shows some differences in the handling of +0 / -0 results36 when input operands differ in signs.37- the handling of underflow exceptions is slightly different. If a38 result underflows before rounding, but becomes a normalised number39 after rounding, we do not signal an underflow exception.40 41Other operations which have been tested by basic assembly-only tests42are:43 44 - fcpy45 - fabs46 - fneg47 - ftoui48 - ftosiz49 - ftouiz50 51The combination operations have not been tested:52 53 - fmac54 - fnmac55 - fmsc56 - fnmsc57 - fnmul58