kernel-fxtec-pro1x/include
Eric Dumazet 6a2d7a955d [PATCH] SLAB: use a multiply instead of a divide in obj_to_index()
When some objects are allocated by one CPU but freed by another CPU we can
consume lot of cycles doing divides in obj_to_index().

(Typical load on a dual processor machine where network interrupts are
handled by one particular CPU (allocating skbufs), and the other CPU is
running the application (consuming and freeing skbufs))

Here on one production server (dual-core AMD Opteron 285), I noticed this
divide took 1.20 % of CPU_CLK_UNHALTED events in kernel.  But Opteron are
quite modern cpus and the divide is much more expensive on oldest
architectures :

On a 200 MHz sparcv9 machine, the division takes 64 cycles instead of 1
cycle for a multiply.

Doing some math, we can use a reciprocal multiplication instead of a divide.

If we want to compute V = (A / B)  (A and B being u32 quantities)
we can instead use :

V = ((u64)A * RECIPROCAL(B)) >> 32 ;

where RECIPROCAL(B) is precalculated to ((1LL << 32) + (B - 1)) / B

Note :

I wrote pure C code for clarity. gcc output for i386 is not optimal but
acceptable :

mull   0x14(%ebx)
mov    %edx,%eax // part of the >> 32
xor     %edx,%edx // useless
mov    %eax,(%esp) // could be avoided
mov    %edx,0x4(%esp) // useless
mov    (%esp),%ebx

[akpm@osdl.org: small cleanups]
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Cc: Christoph Lameter <clameter@sgi.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-13 09:05:49 -08:00
..
acpi [PATCH] slab: remove kmem_cache_t 2006-12-07 08:39:25 -08:00
asm-alpha [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-arm Merge branch 'i2c-for-linus' of git://jdelvare.pck.nerim.net/jdelvare-2.6 2006-12-12 09:57:55 -08:00
asm-arm26 [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-avr32 Merge branch 'for-linus' of git://www.atmel.no/~hskinnemoen/linux/kernel/avr32 2006-12-11 18:28:59 -08:00
asm-cris [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-frv [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-generic Make sure we populate the initroot filesystem late enough 2006-12-11 12:12:04 -08:00
asm-h8300 [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-i386 Merge ../linus 2006-12-12 17:41:41 -05:00
asm-ia64 [IA64] enable trap code on slot 1 2006-12-12 12:00:55 -08:00
asm-m32r [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-m68k [PATCH] Sun3: General updates 2006-12-09 09:41:18 -08:00
asm-m68knommu [PATCH] cleanup asm/setup.h userspace visibility 2006-12-07 08:39:46 -08:00
asm-mips [MIPS] IP27: Don't drag <asm/sn/arch.h> into topology.h. 2006-12-12 01:46:24 +00:00
asm-parisc [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-powerpc Merge git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc 2006-12-11 18:24:58 -08:00
asm-ppc Merge branch 'for_paulus' of master.kernel.org:/pub/scm/linux/kernel/git/galak/powerpc 2006-12-11 16:31:42 +11:00
asm-s390 Merge branch 'for-linus' of git://git390.osdl.marist.edu/pub/scm/linux-2.6 2006-12-08 11:21:55 -08:00
asm-sh sh: SH-MobileR SH7722 CPU support. 2006-12-12 08:42:09 +09:00
asm-sh64 [PATCH] cleanup asm/setup.h userspace visibility 2006-12-07 08:39:46 -08:00
asm-sparc [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-sparc64 [SPARC64]: Fix several kprobes bugs. 2006-12-10 02:42:03 -08:00
asm-um [PATCH] UML: add generic BUG support 2006-12-08 08:28:39 -08:00
asm-v850 [PATCH] tty: preparatory structures for termios revamp 2006-12-08 08:28:56 -08:00
asm-x86_64 Merge ../linus 2006-12-12 17:41:41 -05:00
asm-xtensa [PATCH] xtensa: fix system call interface 2006-12-10 09:55:39 -08:00
crypto [CRYPTO] lib: table driven multiplications in GF(2^128) 2006-12-06 18:38:55 -08:00
keys
linux [PATCH] SLAB: use a multiply instead of a divide in obj_to_index() 2006-12-13 09:05:49 -08:00
math-emu
media V4L/DVB (4935): Added the capability of selecting fm gain by tuner 2006-12-10 09:05:52 -02:00
mtd Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6 2006-10-01 17:55:53 +01:00
net [HAMRADIO]: Fix baycom_epp.c compile failure. 2006-12-11 14:35:01 -08:00
pcmcia [PATCH] pcmcia: multifunction card handling fixes 2006-12-04 20:09:15 -05:00
rdma IB: Add DMA mapping functions to allow device drivers to interpose 2006-12-12 14:27:41 -08:00
rxrpc
scsi [PATCH] slab: remove kmem_cache_t 2006-12-07 08:39:25 -08:00
sound Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6 2006-12-05 17:01:28 +00:00
video [PATCH] video: pm3fb macros fix 2006-12-08 08:29:08 -08:00
Kbuild