On 8/1/19 7:04 PM, Paul E Murphy wrote:
On 8/1/19 2:36 PM, Paul A. Clarke wrote:
Since fe{en,dis}ableexcept() and fesetmode() read-modify-write just the
"mode" (exception enable and rounding mode) bits of the Floating Point Status
Control Register (FPSCR), the lighter weight 'mffsl' instruction can be used
to read the FPSCR (enables and rounding mode), and 'mtfsf 0b00000011' can be
used to write just those bits back to the FPSCR. The net is better performance.
In addition, fe{en,dis}ableexcept() read the FPSCR again after writing it, or
they determine that it doesn't need to be written because it is not changing.
In either case, the local variable holds the current values of the enable
bits in the FPSCR. This local variable can be used instead of again reading
the FPSCR.
Also, that value of the FPSCR which is read the second time is validated
against the requested enables. Since the write can't fail, this validation
step is unnecessary, and can be removed.
The write to FPSCR may not fail, but it may not change all the
requested bits. e.g fedisableexcept(FE_INVALID_SQRT). Should the
existing behavior be preserved?
That's an interesting question. The current unpatched code will apparently accept all possible exception bits, enable (or disable) the ones it can, and return -1 if the set (or unset) bits don't match the requested bits.
For the INVALID bits, these are transformed into FE_INVALID iff all of the defined INVALID bits are set.
So, the case of "feenableexcept (FE_OVERFLOW|FE_INVALID_SQRT)" would result in FE_OVERFLOW being enabled, yet the function returns failure because FE_INVALID_SQRT is not. Is the side effect of enabling FE_OVERFLOW but returning failure appropriate? Should the enablement be reverted in that case (all-or-nothing enabled) to avoid side-effects on failure?
Are you just asking about the whether the requested bits _can_ be set? That could be determined without reading (or re-reading) the FPSCR. Should that be done up-front, before attempting to set the FPSCR at all, and returning failure if so?
Is there a crisp definition of the expected operation of these functions? The man page is wanting.