aboutsummaryrefslogtreecommitdiff
path: root/arch
AgeCommit message (Collapse)Author
2009-11-10omap: Fix race condition in omap dma driverTao Hu
The bug could cause irq enable bit of one DMA channel is cleared/set unexpectedly when 2 (or more) drivers are calling omap_request_dma()/omap_free_dma() simultaneously Signed-off-by: Fei Yang <AFY095@motorola.com> Signed-off-by: Tao Hu <taohu@motorola.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
2009-11-10Merge branch 'for-rmk' of git://linux-arm.org/linux-2.6Russell King
2009-11-10x86: Add iommu_init to x86_init_ops, fix buildIngo Molnar
Most of the time x86_init.h is included in pci-dma.c - but not always, leading to this rare build failure: arch/x86/kernel/pci-dma.c:296: error: 'x86_init' undeclared (first use in this function) So include asm/x86_init.h explicitly. Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com LKML-Reference: <1257849980-22640-2-git-send-email-fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86, 32-bit: Fix swiotlb boot crashFUJITA Tomonori
Ingo Molnar reported this boot crash: [ 8.655620] pata_amd 0000:00:06.0: version 0.4.1 [ 8.660286] BUG: unable to handle kernel NULL pointer dereference at 00000034 [ 8.663572] IP: [<c100617b>] dma_supported+0x3b/0xa4 [ 8.663572] *pde = 00000000 Initialize dma_ops properly in the 32-bit case. Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86: Handle HW IOMMU initialization failure gracefullyFUJITA Tomonori
If HW IOMMU initialization fails (Intel VT-d often does this, typically due to BIOS bugs), we fall back to nommu. It doesn't work for the majority since nowadays we have more than 4GB memory so we must use swiotlb instead of nommu. The problem is that it's too late to initialize swiotlb when HW IOMMU initialization fails. We need to allocate swiotlb memory earlier from bootmem allocator. Chris explained the issue in detail: http://marc.info/?l=linux-kernel&m=125657444317079&w=2 The current x86 IOMMU initialization sequence is too complicated and handling the above issue makes it more hacky. This patch changes x86 IOMMU initialization sequence to handle the above issue cleanly. The new x86 IOMMU initialization sequence are: 1. we initialize the swiotlb (and setting swiotlb to 1) in the case of (max_pfn > MAX_DMA32_PFN && !no_iommu). dma_ops is set to swiotlb_dma_ops or nommu_dma_ops. if swiotlb usage is forced by the boot option, we finish here. 2. we call the detection functions of all the IOMMUs 3. the detection function sets x86_init.iommu.iommu_init to the IOMMU initialization function (so we can avoid calling the initialization functions of all the IOMMUs needlessly). 4. if the IOMMU initialization function doesn't need to swiotlb then sets swiotlb to zero (e.g. the initialization is sucessful). 5. if we find that swiotlb is set to zero, we free swiotlb resource. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com LKML-Reference: <1257849980-22640-10-git-send-email-fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10swiotlb: Defer swiotlb init printing, export swiotlb_print_info()FUJITA Tomonori
This enables us to avoid printing swiotlb memory info when we initialize swiotlb. After swiotlb initialization, we could find that we don't need swiotlb. This patch removes the code to print swiotlb memory info in swiotlb_init() and exports the function to do that. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com Cc: tony.luck@intel.com Cc: benh@kernel.crashing.org LKML-Reference: <1257849980-22640-9-git-send-email-fujita.tomonori@lab.ntt.co.jp> [ -v2: merge up conflict ] Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86: intel-iommu: Convert detect_intel_iommu to use iommu_init hookFUJITA Tomonori
This changes detect_intel_iommu() to set intel_iommu_init() to iommu_init hook if detect_intel_iommu() finds the IOMMU. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com LKML-Reference: <1257849980-22640-6-git-send-email-fujita.tomonori@lab.ntt.co.jp> [ -v2: build fix for the !CONFIG_DMAR case ] Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86: amd_iommu: Convert amd_iommu_detect() to use iommu_init hookFUJITA Tomonori
This changes amd_iommu_detect() to set amd_iommu_init to iommu_init hook if amd_iommu_detect() finds the AMD IOMMU. We can kill the code to check if we found the IOMMU in amd_iommu_init() since amd_iommu_detect() sets amd_iommu_init() only when it found the IOMMU. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com LKML-Reference: <1257849980-22640-5-git-send-email-fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86: GART: Convert gart_iommu_hole_init() to use iommu_init hookFUJITA Tomonori
This changes gart_iommu_hole_init() to set gart_iommu_init() to iommu_init hook if gart_iommu_hole_init() finds the GART IOMMU. We can kill the code to check if we found the IOMMU in gart_iommu_init() since gart_iommu_hole_init() sets gart_iommu_init() only when it found the IOMMU. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com LKML-Reference: <1257849980-22640-4-git-send-email-fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86: Calgary: Convert detect_calgary() to use iommu_init hookFUJITA Tomonori
This changes detect_calgary() to set init_calgary() to iommu_init hook if detect_calgary() finds the Calgary IOMMU. We can kill the code to check if we found the IOMMU in init_calgary() since detect_calgary() sets init_calgary() only when it found the IOMMU. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Muli Ben-Yehuda <muli@il.ibm.com> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com LKML-Reference: <1257849980-22640-3-git-send-email-fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86: Add iommu_init to x86_init_opsFUJITA Tomonori
We call the detections functions of all the IOMMUs then all their initialization functions. The latter is pointless since we don't detect multiple different IOMMUs. What we need to do is calling the initialization function of the detected IOMMU. This adds iommu_init hook to x86_init_ops so if an IOMMU detection function can set its initialization function to the hook. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: chrisw@sous-sol.org Cc: dwmw2@infradead.org Cc: joerg.roedel@amd.com Cc: muli@il.ibm.com LKML-Reference: <1257849980-22640-2-git-send-email-fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10ARM: Use a definition for the userspace cmpxchg emulation syscallRussell King
Use a definition for the cmpxchg SWI instead of hard-coding the number. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk> Acked-by: Nicolas Pitre <nico@fluxnic.net>
2009-11-10x86: GART: pci-gart_64.c: Use correct length in strncmpJoe Perches
Signed-off-by: Joe Perches <joe@perches.com> Cc: <stable@kernel.org> # .3x.x LKML-Reference: <1257818330.12852.72.camel@Joe-Laptop.home> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-10x86, amd-ucode: Check UCODE_MAGIC before loading the container fileBorislav Petkov
Signed-off-by: Borislav Petkov <borislav.petkov@amd.com> Signed-off-by: Andreas Herrmann <andreas.herrmann3@amd.com> LKML-Reference: <20091029134552.GC30802@alberich.amd.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-09ARM: S3C: Merge fixes-s3c64xx-dmaBen Dooks
Merge branch 'fixes-s3c64xx-dma' into fixes-s3c-2632-rc6
2009-11-09ARM: S3C64XX: DMA: Free node for non-circular queuesJassi Brar
We need to free the buff and lli nodes if the buffer queue is not CIRCULAR. Signed-off-by: Jassi Brar <jassi.brar@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: S3C64XX: DMA: Callback with correct buffer pointerJassi Brar
buffdone callback should be called per buffer request with pointer to the latest serviced request. 'next' should point to the one next to currently active. Signed-off-by: Jassi Brar <jassi.brar@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: S3C64XX: DMA: Make src and dst transfer size sameJassi Brar
Some devices don't seem to work if the source and desitnation transfer widths are not same. For example, SPI dma xfers, with 8bits/word, don't work without this patch. Signed-off-by: Jassi Brar <jassi.brar@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: S3C64XX: DMA: Unify callback functions for success/failureJassi Brar
Replace s3c64xx_dma_tcirq and s3c64xx_dma_errirq with the common s3c64xx_dma_buffdone. Signed-off-by: Jassi Brar <jassi.brar@samsung.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: S3C64XX: DMA: Protect buffer pointers while manipulationJassi Brar
Ensure the DMA buffer points are not updated from another source during the process of enquing a buffer. Signed-off-by: Jassi Brar <jassi.brar@samsung.com> [ben-linux@fluff.org: Updated patch comment] Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: S3C64XX: Tidy definition and comments in s3c_dma_has_circular()Ben Dooks
The recent changes to arch/arm/mach-s3c6400/include/mach/dma.h have left an out of date comment in there as well as accidentally changing the type of the function. Fix the commit 54489cd46a3a268ed981c681726c6d690883f076 Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: S3C64XX: Remove duplicate s3c_dma_has_circular() definition for S3C64xx.Maurus Cuelenaere
This patch removes the duplicated s3c_dma_has_circular() definition and so fixes compilation for S3C64xx. Signed-off-by: Maurus Cuelenaere <mcuelenaere@gmail.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: SMDK6410: Allocate more GPIO space for WM1190-EV1Mark Brown
The WM835x has some GPIOs on it, allocate some space so we can use them with gpiolib. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09ARM: SMDK6410: Configure GPIO pull up for WM835x IRQ lineMark Brown
When used with the WM1190-EV1 board we can use the internal pull up resistor of the CPU to provide the required pull for the IRQ line. Without this interrupts from the WM835x don't work in the default WM1190-EV1 hardwaer configuration. Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Signed-off-by: Ben Dooks <ben-linux@fluff.org>
2009-11-09Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6Linus Torvalds
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6: sparc: Move of_set_property_mutex acquisition outside of devtree_lock grab. sparc64: replace parentheses in pmul() sparc64: Add a comment about why we only use certain memory barriers these days.
2009-11-09Merge branch 'sh/for-2.6.32' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6 * 'sh/for-2.6.32' of git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6: sh: Replace old style lock initializer sh: Account for cache aliases in flush_icache_range() sh: unwinder: Fix up invalid PC refetch in dwarf unwinder. serial: sh-sci: disable callback typo fix
2009-11-09sh: Replace old style lock initializerThomas Gleixner
SPIN_LOCK_UNLOCKED is deprecated. Use __SPIN_LOCK_UNLOCKED instead. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: linux-sh@vger.kernel.org Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-11-09sh: Account for cache aliases in flush_icache_range()Matt Fleming
The icache may also contain aliases so we must account for them just like we do when manipulating the dcache. We usually get away with aliases in the icache because the instructions that are read from memory are read-only, i.e. they never change. However, the place where this bites us is when the code has been modified. Signed-off-by: Matt Fleming <matt@console-pimps.org> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-11-08sparc: Move of_set_property_mutex acquisition outside of devtree_lock grab.David S. Miller
Otherwise we try to sleep with preemption disabled, etc. Noticed by Thomas Gleixner. Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-09m32r: fix arch/m32r/boot/compressed/MakefileHirokazu Takata
- Fix a comment string - Fix a typo of $(suffix-y) Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
2009-11-08[ARM] Fix test for unimplemented ARM syscallsRussell King
The existing test always failed since 'no' was always greater than 0x7ff. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2009-11-08Merge branch 'for-rmk' of git://git.marvell.com/orionRussell King
2009-11-08x86/PCI: Adjust GFP mask handling for coherent allocationsJan Beulich
Rather than forcing GFP flags and DMA mask to be inconsistent, GFP flags should be determined even for the fallback device through dma_alloc_coherent_mask()/dma_alloc_coherent_gfp_flags(). This restores 64-bit behavior as it was prior to commits 8965eb19386fdf5ccd0ef8b02593eb8560aa3416 and 4a367f3a9dbf2e7ffcee4702203479809236ee6e (not sure why there are two of them), where GFP_DMA was forced on for 32-bit, but not for 64-bit, with the slight adjustment that afaict even 32-bit doesn't need this without CONFIG_ISA. Signed-off-by: Jan Beulich <jbeulich@novell.com> Acked-by: Takashi Iwai <tiwai@suse.de> LKML-Reference: <4AF18187020000780001D8AA@vpn.id2.novell.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
2009-11-08x86: Fix iommu=nodac parameter handlingTejun Heo
iommu=nodac should forbid dac instead of enabling it. Fix it. Signed-off-by: Tejun Heo <tj@kernel.org> Acked-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: Matteo Frigo <athena@fftw.org> Cc: <stable@kernel.org> # .32.x and older LKML-Reference: <4AE5B52A.4050408@kernel.org> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-08x86: Use x86_platform for iommu_shutdownFUJITA Tomonori
This patch cleans up pci_iommu_shutdown() a bit to use x86_platform (similar to how IA64 initializes an IOMMU driver). This adds iommu_shutdown() to x86_platform to avoid calling every IOMMUs' shutdown functions in pci_iommu_shutdown() in order. The IOMMU shutdown functions are platform specific (we don't have multiple different IOMMU hardware) so the current way is pointless. An IOMMU driver sets x86_platform.iommu_shutdown to the shutdown function if necessary. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Cc: joerg.roedel@amd.com LKML-Reference: <20091027163358F.fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-08ARM: 5784/1: fix early boot machine ID mismatch error displayNicolas Pitre
That code was refactored a long time ago, but one particular label didn't get adjusted properly which broke the listing of supported machines. Signed-off-by: Nicolas Pitre <nico@marvell.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2009-11-08x86: Fix error return sequence in __ioremap_caller()Xiaotian Feng
kernel missed to free memtype if get_vm_area_caller failed in __ioremap_caller. This patch introduces error path to fix this and cleans up the repetitive error return sequences that contributed to the creation of the bug. Signed-off-by: Xiaotian Feng <dfeng@redhat.com> Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Cc: H. Peter Anvin <hpa@zytor.com> LKML-Reference: <1257389031-20429-1-git-send-email-dfeng@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-11-08sparc64: replace parentheses in pmul()Roel Kluin
`>>' has a higher precedence than `?' so src2 evaluated to either 16 or 0 dependent on the bits set in rs2. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-07[ARM] orion5x: update defconfigNicolas Pitre
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2009-11-07[ARM] Kirkwood: update defconfigNicolas Pitre
Signed-off-by: Nicolas Pitre <nico@marvell.com>
2009-11-07[ARM] Kirkwood: clarify PCIe MEM bus/physical address distinctionLennert Buytenhek
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: Nicolas Pitre <nico@marvell.com>
2009-11-07[ARM] kirkwood: fix PCI I/O port assignmentLennert Buytenhek
Instead of allocating PCI devices I/O port bus addresses from the 000xxxxx I/O port range as intended, due to a bus versus physical address mixup, the Kirkwood PCIe handling code inadvertently allocated I/O port bus addresses from the f20xxxxx address range (which is the physical address range of the PCIe I/O mapping window), but then direct all I/O port accesses to bus addresses 000xxxxx, which would then not be decoded at all. Fix this by setting the base address of the PCIe I/O space struct resource to KIRKWOOD_PCIE_IO_BUS_BASE instead of the incorrect KIRKWOOD_PCIE_IO_PHYS_BASE, and fix up __io() to expect addresses offsetted by the former instead of the latter. (The suggested fix of directing I/O port accesses from the host to bus addresses f20xxxxx instead has the problem that assigning full 32bit I/O port bus addresses (f20xxxxx) doesn't work on all PCI devices, as not all PCI devices implement full 32 bit BAR registers for I/O ports. We should really try to allocate I/O port bus addresses that fit in 16 bits.) Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: Nicolas Pitre <nico@marvell.com>
2009-11-06x86: Add Phoenix/MSC BIOSes to lowmem corruption listSimon Kagstrom
We have a board with a Phoenix/MSC BIOS which also corrupts the low 64KB of RAM, so add an entry to the table. Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net> LKML-Reference: <20091106154404.002648d9@marrow.netinsight.se> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-11-06sh: unwinder: Fix up invalid PC refetch in dwarf unwinder.Paul Mundt
The dwarf unwinder presently attempts to provide a sane PC value if none is provided, however the logic is broken and cases where a previous valid dwarf frame exists along with a bogus PC value can still proceed. This fixes up the test and prevents the unwinder from blowing up. Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-11-05sparc64: Add a comment about why we only use certain memory barriers these days.David S. Miller
Based upon feedback from Mathieu Desnoyers. Signed-off-by: David S. Miller <davem@davemloft.net>
2009-11-05[ARM] kirkwood: fix section mismatchLi Jie
kirkwood_timer_init() and kirkwood_pcie_setup() lack of __init which causes following warnings: WARNING: vmlinux.o(.text+0x9568): Section mismatch in reference from the function kirkwood_timer_init() to the function .init.text:kirkwood_find_tclk() The function kirkwood_timer_init() references the function __init kirkwood_find_tclk(). This is often because kirkwood_timer_init lacks a __init annotation or the annotation of kirkwood_find_tclk is wrong. WARNING: vmlinux.o(.text+0x979c): Section mismatch in reference from the function kirkwood_pcie_setup() to the function .init.text:orion_pcie_setup() The function kirkwood_pcie_setup() references the function __init orion_pcie_setup(). This is often because kirkwood_pcie_setup lacks a __init annotation or the annotation of orion_pcie_setup is wrong. Signed-off-by: lijie <eltshanli@gmail.com> Signed-off-by: Nicolas Pitre <nico@fluxnic.net>
2009-11-05[ARM] OpenRD base: Initialize PCI express and i2cSimon Kagstrom
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net> Acked-by: Dieter Kiermaier <dk-arm-linux@gmx.de> Signed-off-by: Nicolas Pitre <nico@marvell.com>
2009-11-05[ARM] properly report mv78100 stepping A1Lennert Buytenhek
Signed-off-by: Lennert Buytenhek <buytenh@marvell.com> Signed-off-by: Nicolas Pitre <nico@marvell.com>
2009-11-05Merge branch 'kvm-updates/2.6.32' of git://git.kernel.org/pub/scm/virt/kvm/kvmLinus Torvalds
* 'kvm-updates/2.6.32' of git://git.kernel.org/pub/scm/virt/kvm/kvm: KVM: get_tss_base_addr() should return a gpa_t KVM: x86: Catch potential overrun in MCE setup
2009-11-05Merge branch 'merge' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc * 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: powerpc/kvm: Remove problematic BUILD_BUG_ON statement powerpc/pci: Fix regression in powerpc MSI-X powerpc: Avoid giving out RTC dates below EPOCH powerpc/mm: Remove debug context clamping from nohash code powerpc: Cleanup Kconfig selection of hugetlbfs support