netsniff-ng.git - A high-performance Linux networking toolkit

summary refs log tree commit diff

path: root/Extra

Age	Commit message (Collapse)	Author	Files	Lines
2013-07-30	build: reduce Extra contents to bare minimum	Daniel Borkmann	1	-16/+0
	Improve the build system, so that configuration files are installed per tool basis. Also, introduce post_install targets, so that config files can be altered in some way, e.g. done by trafgen. Moreover, move custom targets from Extra to tool-specific Makefiles. Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
2013-07-30	build: Don't uninstall configuration files	Tobias Klauser	1	-12/+0
	Several of the tools install/uninstall files from $NCONF_FILES, however if we call the uninstall target for a specific tool (e.g. `make astraceroute_uninstall') but other tools depending on one or more of the config files are still installed, the file is still removed unconditionally. As it would require quite some logic in the Makefile in order to detect installed tools and their dependencies, we just don't uninstall the configuration files anymore. Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
2013-07-29	build: Use NACL_{INC,LIB}_DIR set by configure	Tobias Klauser	1	-1/+1
	The environment variables $NACL_INC_DIR/$NACL_LIB_DIR might not always be set, but the configure script will attempt to detect an existing NaCl installation if they aren't. Thus use the $CONFIG_NACL_INC_DIR/$CONFIG_NACL_LIB_DIR variables as emitted by the configure script for the build. Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
2013-07-24	build: split build system into smaller manageable pieces	Daniel Borkmann	1	-0/+32
	Split the build system into smaller manageable pieces, grouped by context. In that sense, we can keep the most important information within the Makefile itself, and group out misc bits and pieces that we just include into it. E.g. 'Cmds' contains all defined commands that we call from other locations, 'Template' contains the generic build template for all tools, 'Extra' has pieces that need special treatment for the tools to build and 'Misc' contains help and release code. This makes it more manageable on the long run. While at it, I also simplified some code and removed all Wflags expect "-Wall" as we want to convert into "-Wall -Wextra -W" anyway. Signed-off-by: Daniel Borkmann <dborkman@redhat.com>

nd up being cmpxchg and xadd. The atomic bit operations return the old value of the same bit we changed, not the value of an unrelated bit. On x86, we could put the lock bit in the high bit of the byte, and use "xadd" with that bit (where the overflow ends up not touching other bits), and look at the other bits of the result. However, an even simpler model is to just use a regular atomic "and" to clear the lock bit, and then the sign bit in eflags will indicate the resulting state of the unrelated bit #7. So by moving the PageWaiters bit up to bit #7, we can atomically clear the lock bit and test the waiters bit on x86 too. And architectures with LL/SC (which is all the usual RISC suspects), the particular bit doesn't matter, so they are fine with this approach too. This avoids the extra access to the same atomic word, and thus avoids the costly stall at page unlock time. The only downside is that the interface ends up being a bit odd and specialized: clear a bit in a byte, and test the sign bit. Nick doesn't love the resulting name of the new primitive, but I'd rather make the name be descriptive and very clear about the limitation imposed by trying to work across all relevant architectures than make it be some generic thing that doesn't make the odd semantics explicit. So this introduces the new architecture primitive clear_bit_unlock_is_negative_byte(); and adds the trivial implementation for x86. We have a generic non-optimized fallback (that just does a "clear_bit()"+"test_bit(7)" combination) which can be overridden by any architecture that can do better. According to Nick, Power has the same hickup x86 has, for example, but some other architectures may not even care. All these optimizations mean that my page locking stress-test (which is just executing a lot of small short-lived shell scripts: "make test" in the git source tree) no longer makes our page locking look horribly bad. Before all these optimizations, just the unlock_page() costs were just over 3% of all CPU overhead on "make test". After this, it's down to 0.66%, so just a quarter of the cost it used to be. (The difference on NUMA is bigger, but there this micro-optimization is likely less noticeable, since the big issue on NUMA was not the accesses to 'struct page', but the waitqueue accesses that were already removed by Nick's earlier commit). Acked-by: Nick Piggin <npiggin@gmail.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Bob Peterson <rpeterso@redhat.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Andrew Lutomirski <luto@kernel.org> Cc: Andreas Gruenbacher <agruenba@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Mel Gorman <mgorman@techsingularity.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Diffstat (limited to 'sound/pci/atiixp_modem.c')