- 13 Dec, 2006 40 commits
-
-
Eric Dumazet authored
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>
-
Paul Jackson authored
Elaborate the API for calling cpuset_zone_allowed(), so that users have to explicitly choose between the two variants: cpuset_zone_allowed_hardwall() cpuset_zone_allowed_softwall() Until now, whether or not you got the hardwall flavor depended solely on whether or not you or'd in the __GFP_HARDWALL gfp flag to the gfp_mask argument. If you didn't specify __GFP_HARDWALL, you implicitly got the softwall version. Unfortunately, this meant that users would end up with the softwall version without thinking about it. Since only the softwall version might sleep, this led to bugs with possible sleeping in interrupt context on more than one occassion. The hardwall version requires that the current tasks mems_allowed allows the node of the specified zone (or that you're in interrupt or that __GFP_THISNODE is set or that you're on a one cpuset system.) The softwall version, depending on the gfp_mask, might allow a node if it was allowed in the nearest enclusing cpuset marked mem_exclusive (which requires taking the cpuset lock 'callback_mutex' to evaluate.) This patch removes the cpuset_zone_allowed() call, and forces the caller to explicitly choose between the hardwall and the softwall case. If the caller wants the gfp_mask to determine this choice, they should (1) be sure they can sleep or that __GFP_HARDWALL is set, and (2) invoke the cpuset_zone_allowed_softwall() routine. This adds another 100 or 200 bytes to the kernel text space, due to the few lines of nearly duplicate code at the top of both cpuset_zone_allowed_* routines. It should save a few instructions executed for the calls that turned into calls of cpuset_zone_allowed_hardwall, thanks to not having to set (before the call) then check (within the call) the __GFP_HARDWALL flag. For the most critical call, from get_page_from_freelist(), the same instructions are executed as before -- the old cpuset_zone_allowed() routine it used to call is the same code as the cpuset_zone_allowed_softwall() routine that it calls now. Not a perfect win, but seems worth it, to reduce this chance of hitting a sleeping with irq off complaint again. Signed-off-by:
Paul Jackson <pj@sgi.com> Signed-off-by:
-
Christoph Lameter authored
More cleanups for slab.h 1. Remove tabs from weird locations as suggested by Pekka 2. Drop the check for NUMA and SLAB_DEBUG from the fallback section as suggested by Pekka. 3. Uses static inline for the fallback defs as also suggested by Pekka. 4. Make kmem_ptr_valid take a const * argument. 5. Separate the NUMA fallback definitions from the kmalloc_track fallback definitions. Signed-off-by:
Christoph Lameter <clameter@sgi.com> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by:
-
Christoph Lameter authored
This is a response to an earlier discussion on linux-mm about splitting slab.h components per allocator. Patch is against 2.6.19-git11. See http://marc.theaimsgroup.com/?l=linux-mm&m=116469577431008&w=2 This patch cleans up the slab header definitions. We define the common functions of slob and slab in slab.h and put the extra definitions needed for slab's kmalloc implementations in <linux/slab_def.h>. In order to get a greater set of common functions we add several empty functions to slob.c and also rename slob's kmalloc to __kmalloc. Slob does not need any special definitions since we introduce a fallback case. If there is no need for a slab implementation to provide its own kmalloc mess^H^H^Hacros then we simply fall back to __kmalloc functions. That is sufficient for SLOB. Sort the function in slab.h according to their functionality. First the functions operating on struct kmem_cache * then the kmalloc related functions followed by special debug and fallback definitions. Also redo a lot of comments. Signed-off-by: Christoph Lameter <clameter@sgi.com>? Signed-off-by:
-
Mike Miller authored
Remove calls to pci_disable_device except in fail_all_cmds. The pci_disable_device function does something nasty to Smart Array controllers that pci_enable_device does not undo. So if the driver is unloaded it cannot be reloaded. Also, customers can disable any pci device via the ROM Based Setup Utility (RBSU). If the customer has disabled the controller we should not try to blindly enable the card from the driver. Please consider this for inclusion. Signed-off-by:
Mike Miller <mike.miller@hp.com> Acked-by:
Alan Cox <alan@redhat.com> Signed-off-by:
-
Mike Miller authored
Map out more memory for our config table. It's required to reach offset 0x214 to disable DMA on the P600. I'm not sure how I lost this hunk. Please consider this for inclusion. Signed-off-by:
-
Jiri Slaby authored
When CONFIG_PCI is not defined (i.e. PCI bus is disabled), the sx driver fails to link, since some pci functions are not available. Fix this behaviour to be able to compile this driver on machines with no PCI bus (but with ISA bus support). Signed-off-by:
Jiri Slaby <jirislaby@gmail.com> Signed-off-by:
-
Jiri Slaby authored
When CONFIG_PCI is not defined (i.e. PCI bus is disabled), the mxser_new driver fails to link, since some pci functions are not available. Fix this behaviour to be able to compile this driver on machines with no PCI bus (but with ISA bus support). Signed-off-by:
-
Randy Dunlap authored
With CONFIG_PCI=n: drivers/char/isicom.c: In function 'isicom_probe': drivers/char/isicom.c:1793: warning: implicit declaration of function 'pci_request_region' drivers/char/isicom.c:1827: warning: implicit declaration of function 'pci_release_region' Let's CONFIG_ISI depend on CONFIG_PCI. Signed-off-by:
Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
Evgeniy Polyakov authored
Based on patch from Alexander Rigbo <alexander.rigbo@acgnystrom.se> Signed-off-by:
Evgeniy Polyakov <johnpol@2ka.mipt.ru> Signed-off-by:
-
Avi Kivity authored
It seems macbooks set bit 2 but not bit 0, which is an "enabled but vmxon will fault" setting. Signed-off-by:
Avi Kivity <avi@qumranet.com> Tested-by: Alex Larsson (sometimes testing helps) Signed-off-by:
-
Michael Riepe authored
Signed-off-by:
-
Avi Kivity authored
They're not on speaking terms. Signed-off-by:
-
NeilBrown authored
Thanks Jens for alerting me to this. Cc: Jens Axboe <jens.axboe@oracle.com> Cc: <raziebe@gmail.com> Signed-off-by:
Neil Brown <neilb@suse.de> Signed-off-by:
-
Jeff Dike authored
The previous checkstack fix for UML, which needs to use the host's tools, was wrong in the crossbuilding case. It would use the build host's, rather than the target's, toolchain. This patch removes the old fix and adds an explicit special case for UML, leaving everyone else alone. Signed-off-by:
Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Christoph Lameter authored
Fallback_alloc() does not do the check for GFP_WAIT as done in cache_grow(). Thus interrupts are disabled when we call kmem_getpages() which results in the failure. Duplicate the handling of GFP_WAIT in cache_grow(). Signed-off-by:
-
Eric Dumazet authored
Fields of struct pipe_buf_operations have not a precise layout (ie not optimized to fit cache lines nor reduce cache line ping pongs) The bufs[] array is *large* and is placed near the beginning of the structure, so all following fields have a large offset. This is unfortunate because many archs have smaller instructions when using small offsets relative to a base register. On x86 for example, 7 bits offsets have smaller instruction lengths. Moving bufs[] at the end of pipe_buf_operations permits all fields to have small offsets, and reduce text size, and icache pressure. # size vmlinux.pre vmlinux text data bss dec hex filename 3268989 664356 492196 4425541 438745 vmlinux.pre 3268765 664356 492196 4425317 438665 vmlinux So this patch reduces text size by 224 bytes on my x86_64 machine. Similar results on ia32. Signed-off-by: -
Karsten Wiese authored
Clean up a little. Signed-off-by:
Karsten Wiese <fzu@wemgehoertderstaat.de> Cc: Sam Ravnborg <sam@ravnborg.org> Cc: Roman Zippel <zippel@linux-m68k.org> Acked-by:
-
Karsten Wiese authored
Added function sets "void (*conf_changed_callback)(void)". Call it, if .config's changed state changes. Use above in qconf.cc to set gui's save-widget's sensitvity. Signed-off-by:
-
Karsten Wiese authored
Those two functions are void sym_set_change_count(int count) and void sym_add_change_count(int count) All write accesses to sym_change_count are replaced by calls to above functions. Variable and changer-functions are moved to confdata.c. IMO thats ok, as sym_change_count is an attribute of the .config's change state. Signed-off-by:
-
Karsten Wiese authored
Run "make xconfig" on a freshly untarred kernel-tree. Look at the floppy disk icon of the qt application, that has just started: Its in a normal, active state. Mouse click on it: .config is being saved. This patch series changes things so taht after the mouse click on the floppy disk icon, the icon is greyed out. If you mouse click on it now, nothing happens. If you change some CONFIG_*, the floppy disk icon returns to "active state", that is, if you mouse click it now, .config is written. This patch: Returns sym_change_count to reflect the .config's change state. All read only accesses of sym_change_count are replaced by calls to conf_get_changed() . mconfig.c is manipulated to ask for saving only when conf_get_changed() returned true. Signed-off-by:
-
Eric W. Biederman authored
The code has been fixed to use kill_pid instead of kill_proc fix the comments as well. Signed-off-by:
Eric W. Biederman <ebiederm@xmission.com> Signed-off-by:
-
Eric W. Biederman authored
This reverts commit 373beb35. No one is using this identifier yet. The purpose of this identifier is to export nsproxy to user space which is wrong. nsproxy is an internal implementation optimization, which should keep our fork times from getting slower as we increase the number of global namespaces you don't have to share. Adding a global identifier like this is inappropriate because it makes namespaces inherently non-recursive, greatly limiting what we can do with them in the future. Signed-off-by:
-
Eric Dumazet authored
- pipe/splice should use const pipe_buf_operations and file_operations - struct pipe_inode_info has an unused field "start" : get rid of it. Signed-off-by:
-
Avi Kivity authored
Signed-off-by:
-
Avi Kivity authored
The pcd, pwt, and pat bits on page table entries affect the cpu cache. Since the cache is a host resource, the guest should not be able to control it. Moreover, the meaning of these bits changes depending on whether pat is enabled or not. So, force these bits to zero on shadow page table entries at all times. Signed-off-by:
-
Avi Kivity authored
The arch splitting patchset left an extra put_cpu() in core code, where it can cause trouble for CONFIG_PREEMPT kernels. Reported-by:
Huihong Luo <huisinro@yahoo.com> Signed-off-by:
-
Avi Kivity authored
Signed-off-by:
-
Uri Lublin authored
This makes the SET_SREGS ioctl behave symmetrically to the GET_SREGS ioctl wrt the segment access rights flag. Signed-off-by:
Uri Lublin <uril@qumranet.com> Signed-off-by:
-
Randy Dunlap authored
Section .parainstructions should not warn about section mismatches. WARNING: drivers/net/hamradio/scc.o - Section mismatch: reference to .exit.text: from .parainstructions after '' (at offset 0x0) WARNING: drivers/net/hamradio/scc.o - Section mismatch: reference to .exit.text: from .parainstructions after '' (at offset 0x8) Signed-off-by:
Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
Ralf Baechle authored
i2o_exec_exit and i2o_driver_exit were marked as __exit which is a bug because both are invoked from __init and __exit functions. Signed-off-by:
Ralf Baechle <ralf@linux-mips.org> Cc: Markus Lidel <Markus.Lidel@shadowconnect.com> Signed-off-by:
-
Tilman Schmidt authored
Eliminate some possibilities for user processes writing to the Gigaset character device to be left sleeping indefinitely, by adding wakeup calls to error paths and properly disposing of pending write requests when the device is disconnected. It also removes unnecessary NULL checks before usb_free_urb() and usb_kill_urb() calls. Signed-off-by:
Tilman Schmidt <tilman@imap.cc> Acked-by:
Karsten Keil <kkeil@suse.de> Signed-off-by:
-
Peter Zijlstra authored
Fix up the work on stack and exit scope trouble by placing the work_struct in the uml_net_private data. Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Acked-by:
-
David Brownell authored
Some of the header file rearrangements broke the build for board-osk. Signed-off-by:
David Brownell <dbrownell@users.sourceforge.net> Cc: Russell King <rmk@arm.linux.org.uk> Cc: <tony@atomide.com> Signed-off-by:
-
David Brownell authored
More fixes to build breakage from the work_struct changes ... this updates the tps65010 driver. Plus, fix some dependencies related to the way it's used on the OMAP OSK: force static linking there, since the resulting kernel can't link. NOTE that until the i2c core gets fixed to work without SMBUS_QUICK, kernels needing this driver must still use "tps65010.force=0,0x48" on the command line. Signed-off-by:
-
Avi Kivity authored
As per akpm's request. Signed-off-by:
-
Avi Kivity authored
By letting gcc choose the temporary register for us, we lose arch dependency and some ugliness. Conceivably gcc will also generate marginally better code. Signed-off-by:
-
Avi Kivity authored
Instead of in the main drivers menu. Signed-off-by:
-
Anthony Liguori authored
load_TR_desc() lives in asm/desc.h, so #include that file. Signed-off-by:
Anthony Liguori <anthony@codemonkey.ws> Signed-off-by:
-
Heiko Carstens authored
24ec839c causes this: WARNING: "get_current_tty" [drivers/s390/char/fs3270.ko] undefined! Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Acked-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by:
-
