diff options
Diffstat (limited to 'HOWTO')
| -rw-r--r-- | HOWTO/INSTALL-CROSS.md | 21 | ||||
| -rw-r--r-- | HOWTO/INSTALL.md | 12 |
2 files changed, 31 insertions, 2 deletions
diff --git a/HOWTO/INSTALL-CROSS.md b/HOWTO/INSTALL-CROSS.md index b01de12e44..cb9d12e753 100644 --- a/HOWTO/INSTALL-CROSS.md +++ b/HOWTO/INSTALL-CROSS.md @@ -520,12 +520,31 @@ When a variable has been set, no warning will be issued. `posix_memalign` implementation that accepts larger than page size alignment. +* `erl_xcomp_ose_ldflags_pass1` - Linker flags for the OSE module (pass 1) + +* `erl_xcomp_ose_ldflags_pass2` - Linker flags for the OSE module (pass 2) + +* `erl_xcomp_ose_OSEROOT` - OSE installation root directory + +* `erl_xcomp_ose_STRIP` - Strip utility shipped with the OSE distribution + +* `erl_xcomp_ose_LM_POST_LINK` - OSE postlink tool + +* `erl_xcomp_ose_LM_SET_CONF` - Sets the configuration for an OSE load module + +* `erl_xcomp_ose_LM_ELF_SIZE` - Prints the section size information for an + OSE load module + +* `erl_xcomp_ose_LM_LCF` - OSE load module linker configuration file + +* `erl_xcomp_ose_LM_CONF` - OSE load module default configuration file + Copyright and License --------------------- %CopyrightBegin% -Copyright Ericsson AB 2009-2013. All Rights Reserved. +Copyright Ericsson AB 2009-2014. All Rights Reserved. The contents of this file are subject to the Erlang Public License, Version 1.1, (the "License"); you may not use this file except in diff --git a/HOWTO/INSTALL.md b/HOWTO/INSTALL.md index c9f4062b83..bbde5bc08c 100644 --- a/HOWTO/INSTALL.md +++ b/HOWTO/INSTALL.md @@ -268,12 +268,22 @@ Some of the available `configure` options are: default if possible) * `--{enable,disable}-hipe` - HiPE support (enabled by default on supported platforms) +* `--{enable,disable}-fp-exceptions` - Floating point exceptions (an + optimization for floating point operations). The default differs + depending on operating system and hardware platform. Note that by + enabling this you might get a seemingly working system that sometimes + fail on floating point operations. * `--enable-darwin-universal` - Build universal binaries on darwin i386. * `--enable-darwin-64bit` - Build 64-bit binaries on darwin * `--enable-m64-build` - Build 64-bit binaries using the `-m64` flag to `(g)cc` * `--enable-m32-build` - Build 32-bit binaries using the `-m32` flag to `(g)cc` +* `--with-assumed-cache-line-size=SIZE` - Set assumed cache-line size in + bytes. Default is 64. Valid values are powers of two between and + including 16 and 8192. The runtime system use this value in order to + try to avoid false sharing. A too large value wastes memory. A to + small value will increase the amount of false sharing. * `--{with,without}-termcap` - termcap (without implies that only the old Erlang shell can be used) * `--with-javac=JAVAC` - Specify Java compiler to use @@ -288,7 +298,7 @@ Some of the available `configure` options are: memory accesses. If `configure` should inform you about no native atomic implementation available, you typically want to try using the `libatomic_ops` library. It can be downloaded from - <http://www.hpl.hp.com/research/linux/atomic_ops/>. + <https://github.com/ivmai/libatomic_ops/>. * `--disable-smp-require-native-atomics` - By default `configure` will fail if an SMP runtime system is about to be built, and no implementation for native atomic memory accesses can be found. If this happens, you are |