this means that timeouts greater than 35.79 minutes are treated as infinity.
.SH SEE ALSO
.BR epoll_create (2),
diff --git a/man2/fcntl.2 b/man2/fcntl.2
index 3ec52dc4dc03..630fc55888bc 100644
--- a/man2/fcntl.2
+++ b/man2/fcntl.2
@@ -2004,7 +2004,7 @@ A limitation of the Linux system call conventions on some
architectures (notably i386) means that if a (negative)
process group ID to be returned by
.B F_GETOWN
-falls in the range \-1 to \-4095, then the return value is wrongly
+falls in the range \-1 to \-4Ki-1, then the return value is wrongly
interpreted by glibc as an error in the system call;
.\" glibc source: sysdeps/unix/sysv/linux/i386/sysdep.h
that is, the return value of
diff --git a/man2/getgroups.2 b/man2/getgroups.2
index 36300bf61b6a..f01af687ccbd 100644
--- a/man2/getgroups.2
+++ b/man2/getgroups.2
@@ -119,7 +119,7 @@ can additionally fail with the following errors:
.I size
is greater than
.B NGROUPS_MAX
-(32 before Linux 2.6.4; 65536 since Linux 2.6.4).
+(32 before Linux 2.6.4; 64Ki since Linux 2.6.4).
.TP
.B ENOMEM
Out of memory.
diff --git a/man2/ioctl_console.2 b/man2/ioctl_console.2
index 89f794c1956c..477e6fd1a7e1 100644
--- a/man2/ioctl_console.2
+++ b/man2/ioctl_console.2
@@ -171,7 +171,7 @@ bright cyan, and white.
.B GIO_FONT
Gets 256-character screen font in expanded form.
.I argp
-points to an 8192-byte array.
+points to an 8Ki-byte array.
Fails with error code
.B EINVAL
if the
@@ -211,7 +211,7 @@ Sets 256-character screen font.
Load font into the EGA/VGA character
generator.
.I argp
-points to an 8192-byte map, with 32 bytes per
+points to an 8Ki-byte map, with 32 bytes per
character.
Only the first
.I N
diff --git a/man2/iopl.2 b/man2/iopl.2
index abf1bef675fd..c967296157b7 100644
--- a/man2/iopl.2
+++ b/man2/iopl.2
@@ -34,7 +34,7 @@ Permissions are inherited from parents to children.
This call is deprecated, is significantly slower than
.BR ioperm (2),
and is only provided for older X servers which require
-access to all 65536 I/O ports.
+access to all 64Ki I/O ports.
It is mostly for the i386 architecture.
On many other architectures it does not exist or will always
return an error.
diff --git a/man2/madvise.2 b/man2/madvise.2
index 9b4652a635d3..e05e9c5de4a7 100644
--- a/man2/madvise.2
+++ b/man2/madvise.2
@@ -329,8 +329,8 @@ naturally aligned to the huge page size (see
This feature is primarily aimed at applications that use large mappings of
data and access large regions of that memory at a time (e.g., virtualization
systems such as QEMU).
-It can very easily waste memory (e.g., a 2\ MB mapping that only ever accesses
-1 byte will result in 2\ MB of wired memory instead of one 4\ KB page).
+It can very easily waste memory (e.g., a 2\ MiB mapping that only ever accesses
+1 byte will result in 2\ MiB of wired memory instead of one 4\ KiB page).
See the Linux kernel source file
.I Documentation/admin\-guide/mm/transhuge.rst
for more details.
diff --git a/man2/mmap2.2 b/man2/mmap2.2
index 1fd5732ad41b..f975c1388a77 100644
--- a/man2/mmap2.2
+++ b/man2/mmap2.2
@@ -32,7 +32,7 @@ The
system call provides the same interface as
.BR mmap (2),
except that the final argument specifies the offset into the
-file in 4096-byte units (instead of bytes, as is done by
+file in 4Ki-byte units (instead of bytes, as is done by
.BR mmap (2)).
This enables applications that use a 32-bit
.I off_t
@@ -50,8 +50,8 @@ is set to indicate the error.
Problem with getting the data from user space.
.TP
.B EINVAL
-(Various platforms where the page size is not 4096 bytes.)
-.I "offset\ *\ 4096"
+(Various platforms where the page size is not 4Ki bytes.)
+.I "offset\ *\ 4Ki"
is not a multiple of the system page size.
.PP
.BR mmap2 ()
@@ -74,7 +74,7 @@ This system call does not exist on x86-64.
.PP
On ia64, the unit for
.I offset
-is actually the system page size, rather than 4096 bytes.
+is actually the system page size, rather than 4Ki bytes.
.\" ia64 can have page sizes ranging from 4 kB to 64 kB.
.\" On cris, it looks like the unit might also be the page size,
.\" which is 8192 bytes. -- mtk, June 2007
diff --git a/man2/request_key.2 b/man2/request_key.2
index e78321e3c23f..dacc5282f3d8 100644
--- a/man2/request_key.2
+++ b/man2/request_key.2
@@ -399,7 +399,7 @@ The size of the string (including the terminating null byte) specified in
.I type
or
.I description
-exceeded the limit (32 bytes and 4096 bytes respectively).
+exceeded the limit (32 bytes and 4Ki bytes respectively).
.TP
.B EINVAL
The size of the string (including the terminating null byte) specified in
diff --git a/man2/sched_setaffinity.2 b/man2/sched_setaffinity.2
index 86a93539137d..9e7a26293e73 100644
--- a/man2/sched_setaffinity.2
+++ b/man2/sched_setaffinity.2
@@ -243,10 +243,10 @@ impose no restriction on the size of the CPU mask.
However, the
.I cpu_set_t
data type used by glibc has a fixed size of 128 bytes,
-meaning that the maximum CPU number that can be represented is 1023.
+meaning that the maximum CPU number that can be represented is 1\[aq]023.
.\" FIXME . See https://sourceware.org/bugzilla/show_bug.cgi?id=15630
.\" and https://sourceware.org/ml/libc-alpha/2013-07/msg00288.html
-If the kernel CPU affinity mask is larger than 1024,
+If the kernel CPU affinity mask is larger than 1Ki,
then calls of the form:
.PP
.in +4n
diff --git a/man2/seccomp.2 b/man2/seccomp.2
index 32706397f03e..0bb8caa75698 100644
--- a/man2/seccomp.2
+++ b/man2/seccomp.2
@@ -836,7 +836,7 @@ but the filter program pointed to by
.I args
was not valid or the length of the filter program was zero or exceeded
.B BPF_MAXINSNS
-(4096) instructions.
+(4Ki) instructions.
.TP
.B ENOMEM
Out of memory.
@@ -846,7 +846,7 @@ Out of memory.
The total length of all filter programs attached
to the calling thread would exceed
.B MAX_INSNS_PER_PATH
-(32768) instructions.
+(32Ki) instructions.
Note that for the purposes of calculating this limit,
each already existing filter program incurs an
overhead penalty of 4 instructions.
diff --git a/man2/semop.2 b/man2/semop.2
index 7a1416a26894..a0027e0706c5 100644
--- a/man2/semop.2
+++ b/man2/semop.2
@@ -434,7 +434,7 @@ On Linux, this limit can be read and modified via the third field of
.IR /proc/sys/kernel/sem .
.\" This /proc file is not available in Linux 2.2 and earlier -- MTK
.IR Note :
-this limit should not be raised above 1000,
+this limit should not be raised above 1\[aq]000,
.\" See comment in Linux 3.19 source file include/uapi/linux/sem.h
because of the risk of that
.BR semop ()
@@ -445,7 +445,7 @@ array.
.B SEMVMX
Maximum allowable value for
.IR semval :
-implementation dependent (32767).
+implementation dependent (32Ki-1).
.PP
The implementation has no intrinsic limits for
the adjust on exit maximum value
diff --git a/man2/sendmmsg.2 b/man2/sendmmsg.2
index 4e5475c45a09..3f355382ebf6 100644
--- a/man2/sendmmsg.2
+++ b/man2/sendmmsg.2
@@ -139,7 +139,7 @@ The value specified in
.I vlen
is capped to
.B UIO_MAXIOV
-(1024).
+(1Ki).
.\" commit 98382f419f32d2c12d021943b87dea555677144b
.\" net: Cap number of elements for sendmmsg
.\"
diff --git a/man2/shmget.2 b/man2/shmget.2
index c4d8df8ed619..5421fd4bf3e9 100644
--- a/man2/shmget.2
+++ b/man2/shmget.2
@@ -360,7 +360,7 @@ Because it is not possible to map just part of a shared memory segment,
the amount of virtual memory places another limit on the maximum size of a
usable segment:
for example, on i386 the largest segments that can be mapped have a
-size of around 2.8\ GB, and on x86-64 the limit is around 127 TB.
+size of around 2.8\ GB, and on x86-64 the limit is around 127\ TB.
.TP
.B SHMMIN
Minimum size in bytes for a shared memory segment: implementation
@@ -371,7 +371,7 @@ is the effective minimum size).
.B SHMMNI
System-wide limit on the number of shared memory segments.
In Linux 2.2, the default value for this limit was 128;
-since Linux 2.4, the default value is 4096.
+since Linux 2.4, the default value is 4Ki.
.IP
On Linux, this limit can be read and modified via
.IR /proc/sys/kernel/shmmni .
diff --git a/man2/syslog.2 b/man2/syslog.2
index 09c086f181e3..7d76e8cd9658 100644
--- a/man2/syslog.2
+++ b/man2/syslog.2
@@ -54,9 +54,9 @@ in which messages given as arguments to the kernel function
are stored (regardless of their log level).
In early kernels,
.B LOG_BUF_LEN
-had the value 4096;
-from Linux 1.3.54, it was 8192;
-from Linux 2.1.113, it was 16384;
+had the value 4Ki;
+from Linux 1.3.54, it was 8Ki;
+from Linux 2.1.113, it was 16Ki;
since Linux 2.4.23/2.6, the value is a kernel configuration option
.RB ( CONFIG_LOG_BUF_SHIFT ,
default value dependent on the architecture).
diff --git a/man2/vmsplice.2 b/man2/vmsplice.2
index 01ac37b3584f..08ede47361ae 100644
--- a/man2/vmsplice.2
+++ b/man2/vmsplice.2
@@ -149,7 +149,7 @@ as defined in
.IR <limits.h> .
Currently,
.\" UIO_MAXIOV in kernel source
-this limit is 1024.
+this limit is 1Ki.
.PP
.\" commit 6a14b90bb6bc7cd83e2a444bf457a2ea645cbfe7
.BR vmsplice ()
