From 75be1bffe67910de003ab45bec6fa08530e8531d Mon Sep 17 00:00:00 2001 From: Richard Stallman Date: Fri, 12 Jan 2024 17:03:09 -0500 Subject: [PATCH] * c.texi: fix typos. * fp.texi (Machine Epsilon, Rounding Control): Don't have text on the same line as @ignore. --- ChangeLog | 6 ++++++ c.texi | 8 ++++---- fp.texi | 6 ++++-- 3 files changed, 14 insertions(+), 6 deletions(-) diff --git a/ChangeLog b/ChangeLog index f709e8c..0ca844b 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +2024-01-12 Richard Stallman + + * c.texi: fix typos. + * fp.texi (Machine Epsilon, Rounding Control): + Don't have text on the same line as @ignore. + 2024-01-08 Richard Stallman * c.texi (Unicode Character Codes): Rewrite the initial explanation of diff --git a/c.texi b/c.texi index ba3acd0..43b4404 100644 --- a/c.texi +++ b/c.texi @@ -731,7 +731,7 @@ provides a value for it to return. @xref{return Statement}. Calculating @code{fib} using ordinary integers in C works only for @var{n} < 47 because the value of @code{fib (47)} is too large to fit in type @code{int}. In GNU C, type @code{int} holds 32 bits -(@pxref{Integer Typex}), so the addition operation that tries to add +(@pxref{Integer Types}), so the addition operation that tries to add @code{fib (46)} and @code{fib (45)} cannot deliver the correct result. This occurrence is called @dfn{integer overflow}. @@ -933,7 +933,7 @@ examine the value of @code{x} and ignore it. That is not useful.@footnote{Computing an expression and ignoring the result can be useful in peculiar cases. For instance, dereferencing a pointer and ignoring the value is a way to cause a fault if a pointer value is -invalid. @xref{signals}. But you may need to declare the pointer +invalid. @xref{Signals}. But you may need to declare the pointer target @code{volatile} or the dereference may be optimized away. @xref{volatile}.} @@ -5192,7 +5192,7 @@ int i = 2; int foo (void) -{ +@{ /* @r{Save global variable @code{i}'s address.} */ int *global_i = &i; @@ -5202,7 +5202,7 @@ foo (void) /* @r{Print value of global @code{i} and value of local @code{i}.} */ printf ("global i: %d\nlocal i: %d\n", *global_i, i); return i; -} +@} @end example Of course, in a real program it would be much cleaner to use different diff --git a/fp.texi b/fp.texi index c462081..e86749e 100644 --- a/fp.texi +++ b/fp.texi @@ -1151,7 +1151,8 @@ The @code{volatile} qualifier (@pxref{volatile}) is essential on x86 platforms to prevent an optimizing compiler from producing the same value for both bounds. -@ignore We no longer discuss the double rounding issue. +@ignore +We no longer discuss the double rounding issue. The code also needs to be compiled with the option @option{-ffloat-store} that prevents use of higher precision for the basic operations, because that would introduce double rounding @@ -1242,7 +1243,8 @@ macheps (NaN) = nan Notice that @code{macheps} has a special test for a NaN to prevent an infinite loop. -@ignore We no longer discuss double rounding. +@ignore +We no longer discuss double rounding. To ensure that no expressions are evaluated with an intermediate higher precision, we can compile with the @option{-fexcess-precision=standard} option, which tells the compiler that all calculation results, including