diff mbox series

[Ada] Fix glitch in comment of System.Powten_Table

Message ID 20201216131535.GA69879@adacore.com
State New
Headers show
Series [Ada] Fix glitch in comment of System.Powten_Table | expand

Commit Message

Pierre-Marie de Rodat Dec. 16, 2020, 1:15 p.m. UTC 
There is a typo in the formula to compute Maxpow exposed in the comment.

Tested on x86_64-pc-linux-gnu, committed on trunk

gcc/ada/

	* libgnat/s-powtab.ads (Maxpow): Use explicit formula in comment.
diff mbox series

Patch

diff --git a/gcc/ada/libgnat/s-powtab.ads b/gcc/ada/libgnat/s-powtab.ads
--- a/gcc/ada/libgnat/s-powtab.ads
+++ b/gcc/ada/libgnat/s-powtab.ads
@@ -36,9 +36,9 @@  package System.Powten_Table is
 
    Maxpow : constant := 22;
    --  The number of entries in this table is chosen to include powers of ten
-   --  that are exactly representable with long_long_float. Assuming that on
-   --  all targets we have 53 bits of mantissa for the type, the upper bound is
-   --  given by 53/(log 5). If the scaling factor for a string is greater than
+   --  that are exactly representable with Long_Long_Float. Assuming that on
+   --  all targets we have 53 bits of mantissa for the type, the upper bound
+   --  is given by 53 * log 2 / log 5. If the scaling factor is greater than
    --  Maxpow, it can be obtained by several multiplications, which is less
    --  efficient than with a bigger table, but avoids anomalies at end points.