net: increase optmem_max default value

For many years, /proc/sys/net/core/optmem_max default value
on a 64bit kernel has been 20 KB.

Regular usage of TCP tx zerocopy needs a bit more.

Google has used 128KB as the default value for 7 years without
any problem.

Signed-off-by: Eric Dumazet <[email protected]>
Reviewed-by: Willem de Bruijn <[email protected]>
Reviewed-by: Willem de Bruijn <[email protected]>
Acked-by: Neal Cardwell <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

Author: edumazet

Documentation/admin-guide/sysctl/net.rst

@@ -345,7 +345,10 @@ optmem_max
 ----------
 
 Maximum ancillary buffer size allowed per socket. Ancillary data is a sequence
-of struct cmsghdr structures with appended data.
+of struct cmsghdr structures with appended data. TCP tx zerocopy also uses
+optmem_max as a limit for its internal structures.
+
+Default : 128 KB
 
 fb_tunnels_only_for_init_net
 ----------------------------

net/core/sock.c

@@ -283,8 +283,10 @@ EXPORT_SYMBOL(sysctl_rmem_max);
 __u32 sysctl_wmem_default __read_mostly = SK_WMEM_MAX;
 __u32 sysctl_rmem_default __read_mostly = SK_RMEM_MAX;
 
-/* Maximal space eaten by iovec or ancillary data plus some space */
-int sysctl_optmem_max __read_mostly = sizeof(unsigned long)*(2*UIO_MAXIOV+512);
+/* Limits per socket sk_omem_alloc usage.
+ * TCP zerocopy regular usage needs 128 KB.
+ */
+int sysctl_optmem_max __read_mostly = 128 * 1024;
 EXPORT_SYMBOL(sysctl_optmem_max);
 
 int sysctl_tstamp_allow_data __read_mostly = 1;