Changeset 13748
- Timestamp:
- May 30, 2010, 1:57:18 PM (10 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
branches/arm/lisp-kernel/ARM-notes.txt
r13734 r13748 19 19 don't see any reason to try to optimize for older variants.) 20 20 21 More: recent ARM variants (ARMv7, at least) support vfp v3 with NEON 22 SIMD extenstions. If vfp v3, the exception-enable bits in the FPSCR 23 always read as 0; that means that FP exceptions can't cause SIGFPE. 24 25 On a jailbroken iPod Touch, an enabled FP exception seems to force 26 a reboot instead of (or as well as) SIGFPE. 27 28 Checking to see if an FP operation generated an interesting FP 29 exception involves something like: 30 31 (fp-operation fp1 fp2 fp3) 32 (mov temp1 fpscr) ; the "exception occurred" bits get set and can be read 33 (load temp2 (tcr.lisp-exception-bits)) 34 (ands temp3 temp1 temp2) ; set flags 35 (uuo-if (:? ne) ....) 36 37 That just sucks; it makes FP exception checking a time/space/safety tradeoff. 38 39 40 41 21 42 22 43 - instruction set
Note: See TracChangeset
for help on using the changeset viewer.