summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* usb: phy: msm: Remove HSUSB prefix from regulator namesIvan T. Ivanov2014-04-301-3/+3
| | | | | | | | | Prefix did not bring any useful information. Currently none of the MSM platforms define these regulators, so it is safe to rename them. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Remove unused pclk_src_nameIvan T. Ivanov2014-04-302-30/+1
| | | | | | | | There are no references to 'pclk_src_name' in plaform code, so it is unused. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Replace custom enum usb_mode_type with enum usb_dr_modeIvan T. Ivanov2014-04-304-45/+20
| | | | | | | | | | | | Use enum usb_dr_mode and drop default usb_dr_mode from platform data. USB DT bindings states: dr_mode: "...In case this attribute isn't passed via DT, USB DRD controllers should default to OTG...", so remove redundand field. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Acked-by: David Brown <davidb@codeaurora.org> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Fix checkpatch.pl warningsIvan T. Ivanov2014-04-301-25/+14
| | | | | | | | | | This fixes following: WARNING: quoted string split across lines WARNING: Prefer seq_puts to seq_printf Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Remove unnecessarily check for valid regulators.Ivan T. Ivanov2014-04-301-10/+0
| | | | | | | | | Whether regulators are available or not is checked at driver probe. If they are not available driver will refuse to load, so no need to check them again. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Migrate to Managed Device Resource allocationIvan T. Ivanov2014-04-301-119/+68
| | | | | | | | Move memory, regulators, clocks and irq allocation to devm_* variants. Properly check for valid clk handles. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Enable deferred driver probingIvan T. Ivanov2014-04-301-2/+3
| | | | | | | | | | | Using platform_driver_probe() prevent driver from requesting probe deferral. Fix this. While at that, also switch to module_platform_driver() and remove __init annotation from probe(). Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Move global regulators variables to driver stateIvan T. Ivanov2014-04-302-43/+42
| | | | | | | | Eliminating global variables allows driver to handle multiple device instances. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: msm: Make driver selectable on ARCH_QCOMIvan T. Ivanov2014-04-301-3/+3
| | | | | | | | Controller could be found on APQ and MSM platforms, make configuration description more generic. Signed-off-by: Ivan T. Ivanov <iivanov@mm-sol.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: gadget: pretty print Generic CMDsFelipe Balbi2014-04-251-0/+29
| | | | | | | | this makes it slightly easier to read generic CMD logs. It also helps make sure we're sending proper parameters for each command. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: gadget: pretty print link statesFelipe Balbi2014-04-251-3/+41
| | | | | | | this makes it slightly easier to read link state change interrupt logs. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: gadget: print both cmd string and numberFelipe Balbi2014-04-251-2/+2
| | | | | | | That way it's easy for humans looking at dmesg and humans(?) looking at Databooks. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: exynos: Make provision for vdd regulatorsVivek Gautam2014-04-211-2/+49
| | | | | | | | | | | | | | | | | Facilitate getting required 3.3V and 1.0V VDD supply for DWC3 controller on Exynos. With patches for regulators' nodes merged in 3.15: c8c253f ARM: dts: Add regulator entries to smdk5420 275dcd2 ARM: dts: add max77686 pmic node for smdk5250, certain perripherals will now need to ensure that, they request VDD regulators in their drivers, and enable them so as to make them working. Signed-off-by: Vivek Gautam <gautam.vivek@samsung.com> Cc: Anton Tikhomirov <av.tikhomirov@samsung.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: musb: dsps: compile suspend/resume only with PM_SLEEPWolfram Sang2014-04-211-1/+1
| | | | | | | | | | | Depending on PM is not enough, because only PM_RUNTIME could be selected. Fixes: drivers/usb/musb/musb_dsps.c:703:12: warning: 'dsps_suspend' defined but not used [-Wunused-function] drivers/usb/musb/musb_dsps.c:721:12: warning: 'dsps_resume' defined but not used [-Wunused-function] Signed-off-by: Wolfram Sang <wsa@sang-engineering.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: f_subset: replace PTR_RET with PTR_ERR_OR_ZERODuan Jiong2014-04-211-1/+1
| | | | | | | | PTR_RET is deprecated. Do not recommend its usage anymore. Use PTR_ERR_OR_ZERO instead. Signed-off-by: Duan Jiong <duanj.fnst@cn.fujitsu.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: add glue layer dependenciesJean Delvare2014-04-211-1/+3
| | | | | | | | | | Glue layers for the DWC3 driver only make sense on specific platforms. Add dependencies so that they are not built where they aren't needed. Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: WingMan Kwok <w-kwok2@ti.com> Signed-off-by: Jean Delvare <jdelvare@suse.de> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: musb: dsps: handle babble interruptsDaniel Mack2014-04-211-1/+24
| | | | | | | | | | | | When the dsps isr sees a babble error, pass it down to the core for fixup. Also, provide a .reset hook so the core can call us back. A babble interrupt error occured when a USB mass storage device ("CHIPSBNK v3.3.9.1", 1e3d:2093) was disconnected from a AM33xx host. Signed-off-by: Daniel Mack <zonque@gmail.com> Reported-by: Thomas Mellenthin <mellenthin@teufel.de> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: musb: add a work_struct to recover from babble errorsDaniel Mack2014-04-212-0/+36
| | | | | | | | | | | | Handle BABBLE interrupt error conditions from a work struct handler. This indirection is necessary as we can't be certain that the phy functions don't sleep. Platform layer implementation may pass a babble error down to the core in order to handle it. Signed-off-by: Daniel Mack <zonque@gmail.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: musb: add reset hook to platform opsDaniel Mack2014-04-211-0/+7
| | | | | | | | | Babble interrupts require us to reset the DSPS glue layer. In order to handle all other recovery tasks independently, add a new hook for platform-specific implementations of the actual reset. Signed-off-by: Daniel Mack <zonque@gmail.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: Use GFP_ATOMIC when allocating under held spinlockAndreas Larsson2014-04-211-2/+2
| | | | | | | | | As gr_ep_init must be called with dev->lock held, GFP_KERNEL must not be used. Reported-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: Return error code when trying to set ep.maxpacket > ↵Andreas Larsson2014-04-211-0/+4
| | | | | | | | | | ep.maxpacket_limit Make gr_ep_enable fail properly when a call requests a larger ep.maxpacket than ep.maxpacket_limit. Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: Add ep.maxpacket_limit to debugfs informationAndreas Larsson2014-04-211-0/+1
| | | | | | | | Add information on ep.maxpacket_limit for each endpoint in the debugfs information. Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: Use of_property_read_u32_index to access arraysAndreas Larsson2014-04-211-8/+4
| | | | | | | | | Use an appropriate accessor function for property arrays to make the code nicer and make the code correct if it would ever run on little endian architectures. Suggested by Mark Rutland. Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: Use platform_get_irq instead of irq_of_parse_and_mapAndreas Larsson2014-04-211-6/+8
| | | | | | | | | Use platform_get_irq as no mapping needs to be done. No functional difference for SPARC which is the typical environment for the driver though. Suggested by Mark Rutland. Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: Expand devicetree documentationAndreas Larsson2014-04-211-8/+14
| | | | | | | | Provide more information on the two different interrupt cases and more information of endpoint buffer sizes. Suggested by Mark Rutland. Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: gr_udc: improve platform_device variable nameAndreas Larsson2014-04-211-9/+9
| | | | | | | | | Rename struct platform_device pointers from ofdev to pdev for clarity, while at that, also use platform_set/get_drvdata() helpers. Suggested-by: Mark Rutland <mark.rutland@arm.com> Signed-off-by: Andreas Larsson <andreas@gaisler.com> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: generic: switch over to IS_ENABLED()Felipe Balbi2014-04-211-1/+1
| | | | | | | | | | when checking if our generic PHY is enabled, it's a lot easier to use IS_ENABLED() instead of manually checking for it. While at that, also remove the bogus defined(MODULE) at the end of the line. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: generic: allow multiples calls to usb_phy_generic_register()Felipe Balbi2014-04-216-27/+40
| | | | | | | | it's now very easy to return a platform_device pointer and have the caller pass it as argument when calling usb_phy_generic_unregister(). Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: musb: move usb_phy_generic_{un,}register calls to probe()/remove()Felipe Balbi2014-04-215-12/+11
| | | | | | | This patch is in preparation to supporting calling those functions multiple times. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: rename <linux/usb/usb_phy_gen_xceiv.h> to ↵Felipe Balbi2014-04-2115-14/+13
| | | | | | | | | | <linux/usb/usb_phy_generic.h> now that all functions match the driver name, the only missing piece is to rename the header file itself. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: rename usb_nop_xceiv to usb_phy_genericFelipe Balbi2014-04-2113-57/+57
| | | | | | | | no functional changes, just renaming the function in order to make it slightly clearer what it should be used for, also matching the driver name. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: gadget: only GPL drivers in the gadget and phy frameworkFelipe Balbi2014-04-218-64/+64
| | | | | | | | We only support GPL drivers in the USB Gadget Framework, it sounds correct to make all exported symbols GPL too. Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: core: refactor mode initialization to its own functionFelipe Balbi2014-04-211-66/+67
| | | | | | | | Move mode (Host, Peripheral, OTG) initialization to its own function in order to decrease the size of our probe() routine. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: phy: mv-u3d: switch over to writel/readlFelipe Balbi2014-04-211-10/+10
| | | | | | | by removing the _relaxed suffix, we can build this driver in other architectures. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: core: refactor PHY initializationFelipe Balbi2014-04-161-54/+66
| | | | | | | | our probe() routine is too large and we can easily refactor PHY-related code out to another function to make it slightly less painful to read. Signed-off-by: Felipe Balbi <balbi@ti.com>
* usb: dwc3: gadget: clear stall when disabling endpointFelipe Balbi2014-04-161-0/+4
| | | | | | | | | | | so it seems like DWC3 IP doesn't clear stalls automatically when we disable an endpoint, because of that, we _must_ make sure stalls are cleared before clearing the proper bit in DALEPENA register. Cc: <stable@vger.kernel.org> # v3.4+ Reported-by: Johannes Stezenbach <js@sig21.net> Signed-off-by: Felipe Balbi <balbi@ti.com>
* Linux 3.15-rc1v3.15-rc1Linus Torvalds2014-04-131-2/+2
|
* mm: Initialize error in shmem_file_aio_read()Geert Uytterhoeven2014-04-131-1/+1
| | | | | | | | | | | | | | | | | Some versions of gcc even warn about it: mm/shmem.c: In function ‘shmem_file_aio_read’: mm/shmem.c:1414: warning: ‘error’ may be used uninitialized in this function If the loop is aborted during the first iteration by one of the two first break statements, error will be uninitialized. Introduced by commit 6e58e79db8a1 ("introduce copy_page_to_iter, kill loop over iovec in generic_file_aio_read()"). Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Acked-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* cifs: Use min_t() when comparing "size_t" and "unsigned long"Geert Uytterhoeven2014-04-131-1/+1
| | | | | | | | | | | | | | | | On 32 bit, size_t is "unsigned int", not "unsigned long", causing the following warning when comparing with PAGE_SIZE, which is always "unsigned long": fs/cifs/file.c: In function ‘cifs_readdata_to_iov’: fs/cifs/file.c:2757: warning: comparison of distinct pointer types lacks a cast Introduced by commit 7f25bba819a3 ("cifs_iovec_read: keep iov_iter between the calls of cifs_readdata_to_iov()"), which changed the signedness of "remaining" and the code from min_t() to min(). Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge branch 'slab/next' of ↵Linus Torvalds2014-04-135-84/+128
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux Pull slab changes from Pekka Enberg: "The biggest change is byte-sized freelist indices which reduces slab freelist memory usage: https://lkml.org/lkml/2013/12/2/64" * 'slab/next' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/linux: mm: slab/slub: use page->list consistently instead of page->lru mm/slab.c: cleanup outdated comments and unify variables naming slab: fix wrongly used macro slub: fix high order page allocation problem with __GFP_NOFAIL slab: Make allocations with GFP_ZERO slightly more efficient slab: make more slab management structure off the slab slab: introduce byte sized index for the freelist of a slab slab: restrict the number of objects in a slab slab: introduce helper functions to get/set free object slab: factor out calculate nr objects in cache_estimate
| * mm: slab/slub: use page->list consistently instead of page->lruDave Hansen2014-04-113-8/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 'struct page' has two list_head fields: 'lru' and 'list'. Conveniently, they are unioned together. This means that code can use them interchangably, which gets horribly confusing like with this nugget from slab.c: > list_del(&page->lru); > if (page->active == cachep->num) > list_add(&page->list, &n->slabs_full); This patch makes the slab and slub code use page->lru universally instead of mixing ->list and ->lru. So, the new rule is: page->lru is what the you use if you want to keep your page on a list. Don't like the fact that it's not called ->list? Too bad. Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com> Acked-by: Christoph Lameter <cl@linux.com> Acked-by: David Rientjes <rientjes@google.com> Cc: Pekka Enberg <penberg@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * mm/slab.c: cleanup outdated comments and unify variables namingJianyu Zhan2014-04-011-34/+32
| | | | | | | | | | | | | | | | | | | | | | | | As time goes, the code changes a lot, and this leads to that some old-days comments scatter around , which instead of faciliating understanding, but make more confusion. So this patch cleans up them. Also, this patch unifies some variables naming. Acked-by: Christoph Lameter <cl@linux.com> Signed-off-by: Jianyu Zhan <nasa4836@gmail.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: fix wrongly used macroJoonsoo Kim2014-04-011-11/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | commit 'slab: restrict the number of objects in a slab' uses __builtin_constant_p() on #if macro. It is wrong usage of builtin function, but it is compiled on x86 without any problem, so I can't find it before 0 day build system find it. This commit fixes the situation by using KMALLOC_MIN_SIZE, instead of KMALLOC_SHIFT_LOW. KMALLOC_SHIFT_LOW is parsed to ilog2() on some architecture and this ilog2() uses __builtin_constant_p() and results in the problem. This problem would disappear by using KMALLOC_MIN_SIZE, since it is just constant. Tested-by: David Rientjes <rientjes@google.com> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slub: fix high order page allocation problem with __GFP_NOFAILJoonsoo Kim2014-03-271-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | SLUB already try to allocate high order page with clearing __GFP_NOFAIL. But, when allocating shadow page for kmemcheck, it missed clearing the flag. This trigger WARN_ON_ONCE() reported by Christian Casteyde. https://bugzilla.kernel.org/show_bug.cgi?id=65991 https://lkml.org/lkml/2013/12/3/764 This patch fix this situation by using same allocation flag as original allocation. Reported-by: Christian Casteyde <casteyde.christian@free.fr> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: Make allocations with GFP_ZERO slightly more efficientJoe Perches2014-02-081-8/+8
| | | | | | | | | | | | | | | | | | | | Use the likely mechanism already around valid pointer tests to better choose when to memset to 0 allocations with __GFP_ZERO Acked-by: Christoph Lameter <cl@linux.com> Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: make more slab management structure off the slabJoonsoo Kim2014-02-081-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now, the size of the freelist for the slab management diminish, so that the on-slab management structure can waste large space if the object of the slab is large. Consider a 128 byte sized slab. If on-slab is used, 31 objects can be in the slab. The size of the freelist for this case would be 31 bytes so that 97 bytes, that is, more than 75% of object size, are wasted. In a 64 byte sized slab case, no space is wasted if we use on-slab. So set off-slab determining constraint to 128 bytes. Acked-by: Christoph Lameter <cl@linux.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: introduce byte sized index for the freelist of a slabJoonsoo Kim2014-02-081-9/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, the freelist of a slab consist of unsigned int sized indexes. Since most of slabs have less number of objects than 256, large sized indexes is needless. For example, consider the minimum kmalloc slab. It's object size is 32 byte and it would consist of one page, so 256 indexes through byte sized index are enough to contain all possible indexes. There can be some slabs whose object size is 8 byte. We cannot handle this case with byte sized index, so we need to restrict minimum object size. Since these slabs are not major, wasted memory from these slabs would be negligible. Some architectures' page size isn't 4096 bytes and rather larger than 4096 bytes (One example is 64KB page size on PPC or IA64) so that byte sized index doesn't fit to them. In this case, we will use two bytes sized index. Below is some number for this patch. * Before * kmalloc-512 525 640 512 8 1 : tunables 54 27 0 : slabdata 80 80 0 kmalloc-256 210 210 256 15 1 : tunables 120 60 0 : slabdata 14 14 0 kmalloc-192 1016 1040 192 20 1 : tunables 120 60 0 : slabdata 52 52 0 kmalloc-96 560 620 128 31 1 : tunables 120 60 0 : slabdata 20 20 0 kmalloc-64 2148 2280 64 60 1 : tunables 120 60 0 : slabdata 38 38 0 kmalloc-128 647 682 128 31 1 : tunables 120 60 0 : slabdata 22 22 0 kmalloc-32 11360 11413 32 113 1 : tunables 120 60 0 : slabdata 101 101 0 kmem_cache 197 200 192 20 1 : tunables 120 60 0 : slabdata 10 10 0 * After * kmalloc-512 521 648 512 8 1 : tunables 54 27 0 : slabdata 81 81 0 kmalloc-256 208 208 256 16 1 : tunables 120 60 0 : slabdata 13 13 0 kmalloc-192 1029 1029 192 21 1 : tunables 120 60 0 : slabdata 49 49 0 kmalloc-96 529 589 128 31 1 : tunables 120 60 0 : slabdata 19 19 0 kmalloc-64 2142 2142 64 63 1 : tunables 120 60 0 : slabdata 34 34 0 kmalloc-128 660 682 128 31 1 : tunables 120 60 0 : slabdata 22 22 0 kmalloc-32 11716 11780 32 124 1 : tunables 120 60 0 : slabdata 95 95 0 kmem_cache 197 210 192 21 1 : tunables 120 60 0 : slabdata 10 10 0 kmem_caches consisting of objects less than or equal to 256 byte have one or more objects than before. In the case of kmalloc-32, we have 11 more objects, so 352 bytes (11 * 32) are saved and this is roughly 9% saving of memory. Of couse, this percentage decreases as the number of objects in a slab decreases. Here are the performance results on my 4 cpus machine. * Before * Performance counter stats for 'perf bench sched messaging -g 50 -l 1000' (10 runs): 229,945,138 cache-misses ( +- 0.23% ) 11.627897174 seconds time elapsed ( +- 0.14% ) * After * Performance counter stats for 'perf bench sched messaging -g 50 -l 1000' (10 runs): 218,640,472 cache-misses ( +- 0.42% ) 11.504999837 seconds time elapsed ( +- 0.21% ) cache-misses are reduced by this patchset, roughly 5%. And elapsed times are improved by 1%. Acked-by: Christoph Lameter <cl@linux.com> Acked-by: David Rientjes <rientjes@google.com> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: restrict the number of objects in a slabJoonsoo Kim2014-02-082-0/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To prepare to implement byte sized index for managing the freelist of a slab, we should restrict the number of objects in a slab to be less or equal to 256, since byte only represent 256 different values. Setting the size of object to value equal or more than newly introduced SLAB_OBJ_MIN_SIZE ensures that the number of objects in a slab is less or equal to 256 for a slab with 1 page. If page size is rather larger than 4096, above assumption would be wrong. In this case, we would fall back on 2 bytes sized index. If minimum size of kmalloc is less than 16, we use it as minimum object size and give up this optimization. Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: introduce helper functions to get/set free objectJoonsoo Kim2014-02-081-7/+13
| | | | | | | | | | | | | | | | | | | | In the following patches, to get/set free objects from the freelist is changed so that simple casting doesn't work for it. Therefore, introduce helper functions. Acked-by: Christoph Lameter <cl@linux.com> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
| * slab: factor out calculate nr objects in cache_estimateJoonsoo Kim2014-02-081-21/+27
| | | | | | | | | | | | | | | | | | | | | | This logic is not simple to understand so that making separate function helping readability. Additionally, we can use this change in the following patch which implement for freelist to have another sized index in according to nr objects. Acked-by: Christoph Lameter <cl@linux.com> Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com> Signed-off-by: Pekka Enberg <penberg@kernel.org>
OpenPOWER on IntegriCloud