Format casts so that there is a non-breaking space after the

type, and remove unnecessary parentheses around the casted value.
Thus, for example, the following:

     .IR "(size_t) (\-1)" .

becomes:
    .IR "(size_t)\ \-1" .

man2/getsid.2

@@ -47,7 +47,7 @@ returns the session ID of the process with process ID
 session leader.)
 .SH RETURN VALUE
 On success, a session ID is returned.
-On error, \fI(pid_t) \-1\fP will be returned, and
+On error, \fI(pid_t)\ \-1\fP will be returned, and
 .I errno
 is set appropriately.
 .SH ERRORS

man2/mmap.2

@@ -327,7 +327,7 @@ returns a pointer to the mapped area.
 On error, the value
 .B MAP_FAILED
 (that is,
-.IR "(void *) \-1" )
+.IR "(void\ *)\ \-1" )
 is returned, and
 .I errno
 is set appropriately.

man2/mremap.2

@@ -125,7 +125,7 @@ On success
 returns a pointer to the new virtual memory area.
 On error, the value
 .B MAP_FAILED
-(that is, \fI(void *) \-1\fP) is returned,
+(that is, \fI(void\ *)\ \-1\fP) is returned,
 and \fIerrno\fP is set appropriately.
 .SH ERRORS
 .TP

man2/shmop.2

@@ -170,7 +170,7 @@ all attached shared memory segments are detached from the process.
 On success
 .BR shmat ()
 returns the address of the attached shared memory segment; on error
-.I (void *) \-1
+.I (void\ *)\ \-1
 is returned, and
 .I errno
 is set to indicate the cause of the error.

man3/clock.3

@@ -44,7 +44,7 @@ to get the number of seconds used, divide by
 .BR CLOCKS_PER_SEC .
 If the processor time used is not available or its value cannot
 be represented, the function returns the value
-.IR (clock_t)\-1 .
+.IR (clock_t)\ \-1 .
 .SH "CONFORMING TO"
 C89, C99, POSIX.1-2001.
 POSIX requires that

man3/ctime.3

@@ -232,7 +232,7 @@ If the specified broken-down
 time cannot be represented as calendar time (seconds since the Epoch),
 .BR mktime ()
 returns a value of
-.I (time_t)(\-1)
+.I (time_t)\ \-1
 and does not alter the
 \fItm_wday\fP and \fItm_yday\fP members of the broken-down time structure.
 .SH "RETURN VALUE"

man3/iconv.3

@@ -49,7 +49,7 @@ The conversion can stop for four reasons:
 1. An invalid multibyte sequence is encountered in the input.
 In this case
 it sets \fIerrno\fP to \fBEILSEQ\fP and returns
-.IR (size_t)(\-1) .
+.IR (size_t)\ \-1 .
 \fI*inbuf\fP
 is left pointing to the beginning of the invalid multibyte sequence.
 .PP
@@ -64,13 +64,13 @@ non-reversible conversions performed during this call.
 input byte sequence terminates after it.
 In this case it sets \fIerrno\fP to
 \fBEINVAL\fP and returns
-.IR (size_t)(\-1) .
+.IR (size_t)\ \-1 .
 \fI*inbuf\fP is left pointing to the
 beginning of the incomplete multibyte sequence.
 .PP
 4. The output buffer has no more room for the next converted character.
 In this case it sets \fIerrno\fP to \fBE2BIG\fP and returns
-.IR (size_t)(\-1) .
+.IR (size_t)\ \-1 .
 .PP
 A different case is when \fIinbuf\fP is NULL or \fI*inbuf\fP is NULL, but
 \fIoutbuf\fP is not NULL and \fI*outbuf\fP is not NULL.
@@ -81,7 +81,7 @@ initial state and store a corresponding shift sequence at \fI*outbuf\fP.
 At most \fI*outbytesleft\fP bytes, starting at \fI*outbuf\fP, will be written.
 If the output buffer has no more room for this reset sequence, it sets
 \fIerrno\fP to \fBE2BIG\fP and returns
-.IR (size_t)(\-1) .
+.IR (size_t)\ \-1 .
 Otherwise it increments
 \fI*outbuf\fP and decrements \fI*outbytesleft\fP by the number of bytes
 written.
@@ -97,7 +97,7 @@ The
 function returns the number of characters converted in a
 non-reversible way during this call; reversible conversions are not counted.
 In case of error, it sets \fIerrno\fP and returns
-.IR (size_t)(\-1) .
+.IR (size_t)\ \-1 .
 .SH ERRORS
 The following errors can occur, among others:
 .TP

man3/iconv_open.3

@@ -72,7 +72,7 @@ The
 function returns a freshly allocated conversion
 descriptor.
 In case of error, it sets \fIerrno\fP and returns
-.IR (iconv_t)(\-1) .
+.IR (iconv_t)\ \-1 .
 .SH ERRORS
 The following error can occur, among others:
 .TP

man3/mbrlen.3

@@ -35,7 +35,7 @@ shift state \fI*ps\fP to the initial state and returns 0.
 If the \fIn\fP bytes starting at \fIs\fP do not contain a complete multibyte
 character,
 .BR mbrlen ()
-returns \fI(size_t)(\-2)\fP.
+returns \fI(size_t)\ \-2\fP.
 This can happen even if
 \fIn\fP >= \fIMB_CUR_MAX\fP, if the multibyte string contains redundant shift
 sequences.
@@ -44,7 +44,7 @@ If the multibyte string starting at \fIs\fP contains an invalid multibyte
 sequence before the next complete character,
 .BR mbrlen ()
 returns
-\fI(size_t) \-1\fP and sets \fIerrno\fP to \fBEILSEQ\fP.
+\fI(size_t)\ \-1\fP and sets \fIerrno\fP to \fBEILSEQ\fP.
 In this case,
 the effects on \fI*ps\fP are undefined.
 .PP
@@ -59,10 +59,10 @@ parsed from the multibyte
 sequence starting at \fIs\fP, if a non-null wide character was recognized.
 It returns 0, if a null wide character was recognized.
 It returns
-.I "(size_t) \-1"
+.I "(size_t)\ \-1"
 and sets \fIerrno\fP to \fBEILSEQ\fP, if an invalid multibyte sequence was
 encountered.
-It returns \fI(size_t)(\-2)\fP if it couldn't parse a complete multibyte
+It returns \fI(size_t)\ \-2\fP if it couldn't parse a complete multibyte
 character, meaning that \fIn\fP should be increased.
 .SH "CONFORMING TO"
 C99

man3/mbrtowc.3

@@ -40,7 +40,7 @@ state \fI*ps\fP to the initial state and returns 0.
 If the \fIn\fP bytes starting at \fIs\fP do not contain a complete multibyte
 character,
 .BR mbrtowc ()
-returns \fI(size_t)(\-2)\fP.
+returns \fI(size_t)\ \-2\fP.
 This can happen even if
 \fIn\fP >= \fIMB_CUR_MAX\fP, if the multibyte string contains redundant shift
 sequences.
@@ -91,10 +91,10 @@ multibyte sequence starting at \fIs\fP, if a non-L'\\0' wide character
 was recognized.
 It returns 0, if a L'\\0' wide character was recognized.
 It returns
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 and sets \fIerrno\fP to \fBEILSEQ\fP, if an invalid multibyte sequence was
 encountered.
-It returns \fI(size_t)(\-2)\fP if it couldn't parse a complete multibyte
+It returns \fI(size_t)\ \-2\fP if it couldn't parse a complete multibyte
 character, meaning that \fIn\fP should be increased.
 .SH "CONFORMING TO"
 C99

man3/mbsnrtowcs.3

@@ -50,7 +50,7 @@ conversion can stop for three reasons:
 1. An invalid multibyte sequence has been encountered.
 In this case \fI*src\fP
 is left pointing to the invalid multibyte sequence,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned,
 and \fIerrno\fP is set to \fBEILSEQ\fP.
 .PP
@@ -85,7 +85,7 @@ that make up the converted part of the wide-character string,
 not including the terminating null wide character.
 If an invalid multibyte sequence was
 encountered,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned, and \fIerrno\fP set to \fBEILSEQ\fP.
 .SH "CONFORMING TO"
 This function is a GNU extension.

man3/mbsrtowcs.3

@@ -40,7 +40,7 @@ The conversion can stop for three reasons:
 1. An invalid multibyte sequence has been encountered.
 In this case \fI*src\fP
 is left pointing to the invalid multibyte sequence,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned,
 and \fIerrno\fP is set to \fBEILSEQ\fP.
 .PP
@@ -79,7 +79,7 @@ up the converted part of the wide-character string, not including the
 terminating null wide character.
 If an invalid multibyte sequence was
 encountered,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned, and \fIerrno\fP set to \fBEILSEQ\fP.
 .SH "CONFORMING TO"
 C99

man3/mbstowcs.3

@@ -33,7 +33,7 @@ The conversion can stop for three reasons:
 .PP
 1. An invalid multibyte sequence has been encountered.
 In this case
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned.
 .PP
 2. \fIn\fP non-L'\\0' wide characters have been stored at \fIdest\fP.
@@ -63,7 +63,7 @@ up the converted part of the wide-character string, not including the
 terminating null wide character.
 If an invalid multibyte sequence was
 encountered,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned.
 .SH "CONFORMING TO"
 C99

man3/wcrtomb.3

@@ -67,7 +67,7 @@ bytes that have been or would
 have been written to the byte array at \fIs\fP.
 If \fIwc\fP can not be
 represented as a multibyte sequence (according to the current locale),
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned, and \fIerrno\fP set to \fBEILSEQ\fP.
 .SH "CONFORMING TO"
 C99.

man3/wcsnrtombs.3

@@ -53,7 +53,7 @@ The conversion can stop for three reasons:
 multibyte sequence (according to the current locale).
 In this case \fI*src\fP
 is left pointing to the invalid wide character,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned,
 and \fIerrno\fP is set to \fBEILSEQ\fP.
 .PP
@@ -91,7 +91,7 @@ converted part of multibyte sequence,
 not including the terminating null byte.
 If a wide character was encountered which
 could not be converted,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned, and \fIerrno\fP set to \fBEILSEQ\fP.
 .SH "CONFORMING TO"
 This function is a GNU extension.

man3/wcsrtombs.3

@@ -44,7 +44,7 @@ The conversion can stop for three reasons:
 multibyte sequence (according to the current locale).
 In this case \fI*src\fP
 is left pointing to the invalid wide character,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned,
 and
 .I errno
@@ -83,7 +83,7 @@ converted part of multibyte sequence,
 not including the terminating null byte.
 If a wide character was encountered
 which could not be converted,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned, and
 .I errno
 set to \fBEILSEQ\fP.

man3/wcstombs.3

@@ -34,7 +34,7 @@ The conversion can stop for three reasons:
 1. A wide character has been encountered that can not be represented as a
 multibyte sequence (according to the current locale).
 In this case
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned.
 .PP
 2. The length limit forces a stop.
@@ -63,7 +63,7 @@ function returns the number of bytes that make up the
 converted part of multibyte sequence, not including the terminating null byte.
 If a wide character was encountered which could not be
 converted,
-.I (size_t)(\-1)
+.I (size_t)\ \-1
 is returned.
 .SH "CONFORMING TO"
 C99.