aboutsummaryrefslogtreecommitdiff
path: root/arch
AgeCommit message (Collapse)AuthorFilesLines
2009-10-28ARM: S3C2410: Fix sparse warnings in arch/arm/mach-s3c2410/gpio.cBen Dooks1-0/+1
Fix sparse warning in arch/arm/mach-s3c2410/gpio.c due to missing include of <mach/gpio-fns.h>. Fixes the following warning: warning: symbol 's3c2410_gpio_irqfilter' was not declared. Should it be static? Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C2440: mini2440: Fix spare warningsBen Dooks1-2/+2
Fix the following sparse warnings in arch/arm/mach-s3c2440/mach-mini2440.c due to missing 'static'. warning: symbol 'mini2440_lcd_cfg' was not declared. Should it be static? warning: symbol 'mini2440_fb_info' was not declared. Should it be static? Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C24XX: Fix warnings in arch/arm/plat-s3c24xx/gpio.cBen Dooks1-0/+1
Fix the following warnings from sparse in arch/arm/plat-s3c24xx/gpio. due to the missing include of <mach/gpio-fns.h> gpio.c:36:6: warning: symbol 's3c2410_gpio_cfgpin' was not declared. Should it be static? gpio.c:84:14: warning: symbol 's3c2410_gpio_getcfg' was not declared. Should it be static? gpio.c:103:6: warning: symbol 's3c2410_gpio_pullup' was not declared. Should it be static? gpio.c:125:5: warning: symbol 's3c2410_gpio_getpull' was not declared. Should it be static? gpio.c:138:6: warning: symbol 's3c2410_gpio_setpin' was not declared. Should it be static? gpio.c:157:14: warning: symbol 's3c2410_gpio_getpin' was not declared. Should it be static? gpio.c:184:5: warning: symbol 's3c2410_gpio_getirq' was not declared. Should it be static? Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C2440: mini2440: Fix missing CONFIG_S3C_DEV_USB_HOSTBen Dooks1-0/+1
Fix missing select of S3C_DEV_USB_HOST when building for mini2440 only. Fixes the following error: built-in.o: undefined reference to `s3c_device_usb` Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: Merge fixes-s3c64xx-dmaBen Dooks2-0/+12
Merge branch 'fixes-s3c64xx-dma' into fixes-s3c-2632-rc5
2009-10-28ARM: Merge fixes-s3c64xxBen Dooks2-2/+5
Merge branch 'fixes-s3c64xx' into fixes-s3c-2632-rc5
2009-10-28ARM: S3C24XX: arch/arm/plat-s3c24xx: Move dereference after NULL testJulia Lawall1-3/+3
If the NULL test on buf is needed, then the dereference should be after the NULL test. A simplified version of the semantic match that detects this problem is as follows (http://coccinelle.lip6.fr/): // <smpl> @match exists@ expression x, E; identifier fld; @@ * x->fld ... when != \(x = E\|&x\) * x == NULL // </smpl> Signed-off-by: Julia Lawall <[email protected]> Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C: Fix adc function exportsRyan Mallon1-1/+1
Fix the export of s3c_adc_read. Signed-off-by: Ryan Mallon <[email protected]> [[email protected]: remove unexport of s3c_adc_start, needed for ts] Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C2410: Fix link if CONFIG_S3C2410_IOTIMING is not setBen Dooks1-2/+7
Fix the link errors if cpu-frequency support is enabled on s3c2410 systems but there is no CONFIG_S3C2410_IOTIMING set. Fix this by ensuring the relevant symbols are defined NULL if the code is not being built in. Fixes the following error: arch/arm/mach-s3c2410/built-in.o: undefined reference to `s3c2410_iotiming_get' arch/arm/mach-s3c2410/built-in.o: undefined reference to `s3c2410_iotiming_set' arch/arm/mach-s3c2410/built-in.o: undefined reference to `s3c2410_iotiming_calc' Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C24XX: Introduce S3C2442B CPUHarald Welte1-0/+10
Add the S3C2442B CPU ID to aid support the Openmoko GTA02 / Freerunner. Signed-off-by: Harald Welte <[email protected]> Signed-off-by: Andy Green <[email protected]> Signed-off-by: Nelson Castillo <[email protected]> [[email protected]: edit description for clarity and S3C2442B as uppercase] Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C24XX: Define a macro to avoid compilation errorRamax Lo1-0/+1
Define a macro to avoid the following error during kernel build process for platforms other than s3c2410: arch/arm/plat-s3c24xx/cpu.c:84: error: ‘s3c2410a_init’ undeclared here (not in a function) Signed-off-by: Ramax Lo <[email protected]> Signed-off-by: Ben Dooks <[email protected]>
2009-10-28ARM: S3C: Add info for supporting circular DMA buffersBen Dooks2-0/+12
The S3C64XX DMA implementation will work a lot better with the ability to enqueue circular buffers as the hardware can do it's own linked-list management. Add a function s3c_dma_has_circular() to show that the system can do this and a flag for the channel. Update the s3c24xx/s3c64xx I2S DMA code to deal with this. Signed-off-by: Ben Dooks <[email protected]> Signed-off-by: Ben Dooks <[email protected]> Acked-by: Mark Brown <broonie@@opensource.wolfsonmicro.com>
2009-10-28x86/amd-iommu: Un__init function required on shutdownJoerg Roedel1-1/+1
The function iommu_feature_disable is required on system shutdown to disable the IOMMU but it is marked as __init. This may result in a panic if the memory is reused. This patch fixes this bug. Signed-off-by: Joerg Roedel <[email protected]>
2009-10-27xen: set up mmu_ops before trying to set any ptesJeremy Fitzhardinge1-1/+2
xen_setup_stackprotector() ends up trying to set page protections, so we need to have vm_mmu_ops set up before trying to do so. Failing to do so causes an early boot crash. [ Impact: Fix early crash under Xen. ] Signed-off-by: Jeremy Fitzhardinge <[email protected]>
2009-10-28Merge commit 'gcl/merge' into mergeBenjamin Herrenschmidt19-364/+1188
2009-10-27x86: Make EFI RTC function depend on 32bit againFeng Tang1-0/+2
The EFI RTC functions are only available on 32 bit. commit 7bd867df (x86: Move get/set_wallclock to x86_platform_ops) removed the 32bit dependency which leads to boot crashes on 64bit EFI systems. Add the dependency back. Solves: http://bugzilla.kernel.org/show_bug.cgi?id=14466 Tested-by: Matthew Garrett <[email protected]> Signed-off-by: Feng Tang <[email protected]> LKML-Reference: <20091020125402.028d66d5@feng-desktop> Signed-off-by: Thomas Gleixner <[email protected]>
2009-10-27powerpc/ppc64: Use preempt_schedule_irq instead of preempt_scheduleBenjamin Herrenschmidt1-20/+21
Based on an original patch by Valentine Barshak <[email protected]> Use preempt_schedule_irq to prevent infinite irq-entry and eventual stack overflow problems with fast-paced IRQ sources. This kind of problems has been observed on the PASemi Electra IDE controller. We have to make sure we are soft-disabled before calling preempt_schedule_irq and hard disable interrupts after that to avoid unrecoverable exceptions. This patch also moves the "clrrdi r9,r1,THREAD_SHIFT" out of the #ifdef CONFIG_PPC_BOOK3E scope, since r9 is clobbered and has to be restored in both cases. Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc: Fix compile errors found by new ppc64e_defconfigKumar Gala3-2/+3
Fix the following 3 issues: arch/powerpc/kernel/process.c: In function 'arch_randomize_brk': arch/powerpc/kernel/process.c:1183: error: 'mmu_highuser_ssize' undeclared (first use in this function) arch/powerpc/kernel/process.c:1183: error: (Each undeclared identifier is reported only once arch/powerpc/kernel/process.c:1183: error: for each function it appears in.) arch/powerpc/kernel/process.c:1183: error: 'MMU_SEGSIZE_1T' undeclared (first use in this function) In file included from arch/powerpc/kernel/setup_64.c:60: arch/powerpc/include/asm/mmu-hash64.h:132: error: redefinition of 'struct mmu_psize_def' arch/powerpc/include/asm/mmu-hash64.h:159: error: expected identifier or '(' before numeric constant arch/powerpc/include/asm/mmu-hash64.h:396: error: conflicting types for 'mm_context_t' arch/powerpc/include/asm/mmu-book3e.h:184: error: previous declaration of 'mm_context_t' was here cc1: warnings being treated as errors arch/powerpc/kernel/pci_64.c: In function 'pcibios_unmap_io_space': arch/powerpc/kernel/pci_64.c:100: error: unused variable 'res' Signed-off-by: Kumar Gala <[email protected]> Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc: Add a Book-3E 64-bit defconfigKumar Gala1-0/+2199
This defconfig's purpose at this time is to help catch compile errors between Book-3S and Book-3E support in ppc64. It is based on the ppc64_defconfig with some things disabled that we dont support on Book-3E right now (hugetlbfs, slices, etc.) Signed-off-by: Kumar Gala <[email protected]> Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc/booke: Fix xmon single step on PowerPC Book-EJosh Boyer1-1/+19
Prior to the arch/ppc -> arch/powerpc transition, xmon had support for single stepping on 4xx boards. The functionality was lost when arch/ppc was removed. This patch restores single step support for 44x boards, and Book-E in general. Signed-off-by: Josh Boyer <[email protected]> Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc: Align vDSO base addressAndreas Schwab1-1/+10
The ABI specifies a 64K alignment, we need to map the vDSO accordingly Signed-off-by: Andreas Schwab <[email protected]> Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc: Fix segment mapping in vdso32Andreas Schwab1-2/+2
Due to missing segment assignments the .text section was put in the NOTES segment (and marked as NOTE section), and the .got was put in the DYNAMIC segment. Signed-off-by: Andreas Schwab <[email protected]> Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc/iseries: Remove compiler version dependent hackStephen Rothwell2-42/+25
The creation of the flattened device tree depended on the compiler putting the constant strings for an object in a section with a particular name. This was changed with recent compilers. Do this explicitly instead. Without this patch, iseries kernels may silently not boot when built with some compilers. Signed-off-by: Stephen Rothwell <[email protected]> Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27powerpc/perf_events: Fix priority of MSR HV vs PR bitsMichael Neuling1-7/+10
The architecture defines that if MSR PR is set we are in problem state irrespective of the HV bit. This fixes perf events to reflect this. Also, on bare metal systems, samples taken in Linux will now be reported as kernel rather than hypervisor. Signed-off-by: Michael Neuling <[email protected]> CC: [email protected] Signed-off-by: Benjamin Herrenschmidt <[email protected]>
2009-10-27sh: Fix hugetlbfs dependencies for SH-3 && MMU configurations.Paul Mundt2-1/+6
The hugetlb dependencies presently depend on SUPERH && MMU while the hugetlb page size definitions depend on CPU_SH4 or CPU_SH5. This unfortunately allows SH-3 + MMU configurations to enable hugetlbfs without a corresponding HPAGE_SHIFT definition, resulting in the build blowing up. As SH-3 doesn't support variable page sizes, we tighten up the dependenies a bit to prevent hugetlbfs from being enabled. These days we also have a shiny new SYS_SUPPORTS_HUGETLBFS, so switch to using that rather than adding to the list of corner cases in fs/Kconfig. Reported-by: Kristoffer Ericson <[email protected]> Signed-off-by: Paul Mundt <[email protected]>
2009-10-27sh: Document uImage.bin target in archhelp.Paul Mundt1-0/+1
This was missing from the previous patch. Reported-by: Mike Frysinger <[email protected]> Signed-off-by: Paul Mundt <[email protected]>
2009-10-26ARM: S3C64XX: Set rate of crystal muxMark Brown1-0/+3
The current code assumes that the external clock mux will be set to the crystal. Set this up explicitly within the clock API. Signed-off-by: Mark Brown <[email protected]> Signed-off-by: Ben Dooks <[email protected]>
2009-10-26ARM: S3C64XX: Fix S3C64XX_CLKDIV0_ARM_MASK valueKukjin Kim1-2/+2
Fix the values of S3C6400_CLKDIV0_ARM_MASK and S3C6410_CLKDIV0_ARM_MASK. Signed-off-by: Kukjin Kim <[email protected]> Signed-off-by: Ben Dooks <[email protected]>
2009-10-26x86-64: Fix register leak in 32-bit syscall audtingJan Beulich1-3/+2
Restoring %ebp after the call to audit_syscall_exit() is not only unnecessary (because the register didn't get clobbered), but in the sysenter case wasn't even doing the right thing: It loaded %ebp from a location below the top of stack (RBP < ARGOFFSET), i.e. arbitrary kernel data got passed back to user mode in the register. Signed-off-by: Jan Beulich <[email protected]> Acked-by: Roland McGrath <[email protected]> Cc: <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-26x86: crash_dump: Fix non-pae kdump kernel memory accessesJiri Slaby1-0/+19
Non-PAE 32-bit dump kernels may wrap an address around 4G and poke unwanted space. ptes there are 32-bit long, and since pfn << PAGE_SIZE may exceed this limit, high pfn bits are cropped and wrong address mapped by kmap_atomic_pfn in copy_oldmem_page. Don't allow this behavior in non-PAE kdump kernels by checking pfns passed into copy_oldmem_page. In the case of failure, userspace process gets EFAULT. [v2] - fix comments - move ifdefs inside the function Signed-off-by: Jiri Slaby <[email protected]> Cc: Vivek Goyal <[email protected]> Cc: Eric W. Biederman <[email protected]> Cc: Simon Horman <[email protected]> Cc: Paul Mundt <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-26x86: Side-step lguest problem by only building cmpxchg8b_emu for pre-PentiumRusty Russell1-1/+1
Commit 79e1dd05d1a22 "x86: Provide an alternative() based cmpxchg64()" broke lguest, even on systems which have cmpxchg8b support. The emulation code gets used until alternatives get run, but it contains native instructions, not their paravirt alternatives. The simplest fix is to turn this code off except for 386 and 486 builds. Reported-by: Johannes Stezenbach <[email protected]> Signed-off-by: Rusty Russell <[email protected]> Acked-by: H. Peter Anvin <[email protected]> Cc: [email protected] Cc: Arjan van de Ven <[email protected]> Cc: Jeremy Fitzhardinge <[email protected]> Cc: Linus Torvalds <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-26sh: add uImage.bin targetMagnus Damm2-2/+6
Add an uImage.bin target to allow uncompressed uImages. Useful for boards with busted u-boot decompression like the rsk7203 on my desk. Signed-off-by: Magnus Damm <[email protected]> Signed-off-by: Paul Mundt <[email protected]>
2009-10-26sh: rsk7203 CONFIG_MTD=n fixMagnus Damm2-13/+14
Fix the rsk7203 board code to build with CONFIG_MTD=n. Signed-off-by: Magnus Damm <[email protected]> Signed-off-by: Paul Mundt <[email protected]>
2009-10-26sh: Check for return_to_handler when unwinding the stackMatt Fleming1-0/+22
When CONFIG_FUNCTION_GRAPH_TRACER is enabled the function graph tracer may patch return addresses on the stack with the address of return_to_handler(). This really confuses the DWARF unwinder because it will try find the caller of return_to_handler(), not the caller of the real return address. So teach the DWARF unwinder how to find the real return address whenever it encounters return_to_handler(). This patch does not cope very well when multiple return addresses on the stack have been patched. To make it work properly it would require state to track how many return_to_handler()'s have been seen so that we'd know where to look in current->curr_ret_stack[]. So for now, instead of trying to handle this, just moan if more than one return address on the stack has been patched. Signed-off-by: Matt Fleming <[email protected]> Signed-off-by: Paul Mundt <[email protected]>
2009-10-26sh: Build fix: define more __movmem* symbolsLubomir Rintel1-0/+14
ERROR: "__movmemSI12" [net/unix/unix.ko] undefined! ERROR: "__movmemSI52" [net/ipv6/sit.ko] undefined! ERROR: "__movmemSI24" [net/ipv6/ipv6.ko] undefined! ERROR: "__movmemSI60" [net/ipv6/ipv6.ko] undefined! ERROR: "__movmemSI16" [net/ipv6/ipv6.ko] undefined! ERROR: "__movmemSI20" [net/ipv6/ipv6.ko] undefined! ERROR: "__movmemSI32" [net/ipv6/ipv6.ko] undefined! Signed-off-by: Lubomir Rintel <[email protected]> Signed-off-by: Paul Mundt <[email protected]>
2009-10-26sh: __irq_entry annotate do_IRQ().Paul Mundt1-1/+2
This adds an __irq_entry annotation for do_IRQ() so that the IRQ annotation in the function graph tracer works as advertized. We already have the IRQENTRY section wired up, so this is just a trivial addition to actually make use of it. Signed-off-by: Paul Mundt <[email protected]>
2009-10-25ARM: Use GFP_DMA only for masks _less_ than 32-bitRussell King1-2/+2
We were using GFP_DMA for masks other than 0xffffffff, which is wrong when some masks are initialized to 0xffffffffffffffff. This caused such masks to obtain memory from the precious DMA pool. Signed-off-by: Russell King <[email protected]>
2009-10-25ARM: integrator: allow Integrator to be built with highmemRussell King1-0/+1
This fixes Integrator builds with highmem enabled; we need to translate from 'struct page' to a DMA address, and this is not possible without __pfn_to_bus(). Signed-off-by: Russell King <[email protected]>
2009-10-25ARM: Fix signal restart issues with NX and OABI compatRussell King3-26/+23
The signal restarting code was placed on the user stack when OABI compatibility is enabled. Unfortunately, with an EABI NX executable, this results in an attempt to run code from the non-executable stack, which segfaults the application. Fix this by placing the code in the vectors page, along side the signal return code, and directing the application to that code. Reported-by: saeed bishara <[email protected]> Tested-by: saeed bishara <[email protected]> Signed-off-by: Russell King <[email protected]>
2009-10-23Pull ticket-spinaphore into release branchTony Luck1-6/+18
2009-10-23Pull ticket4byte into release branchTony Luck2-22/+42
2009-10-23x86: Remove STACKPROTECTOR_ALLArjan van de Ven2-5/+0
STACKPROTECTOR_ALL has a really high overhead (runtime and stack footprint) and is not really worth it protection wise (the normal STACKPROTECTOR is in effect for all functions with buffers already), so lets just remove the option entirely. Reported-by: Dave Jones <[email protected]> Reported-by: Chuck Ebbert <[email protected]> Signed-off-by: Arjan van de Ven <[email protected]> Cc: Eric Sandeen <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-22omap4: Fix UART4 platform data on omap4Santosh Shilimkar1-10/+0
This patch removes the unnecessary UART4 platform which is under data is wrong because of this There is a separate platform structure for UART4 Signed-off-by: Santosh Shilimkar <[email protected]> Reviewed-By: Kevin Hilman <[email protected]> Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap4: Allow omap_serial_early_init() for OMAP4430 boardSantosh Shilimkar2-2/+4
This patch enables omap_serial_early_init() function for OMAP4430 SDP. Without this the bootup would throw oops in omap_serial_init(). Note that the ifndef CONFIG_ARCH_OMAP4 is split into two sections to enable omap_serial_early_init(). This ifndef cannot be removed until omap4 clock framework is implemented. Signed-off-by: Santosh Shilimkar <[email protected]> Reviewed-By: Tony Lindgren <[email protected]> Reviewed-By: Kevin Hilman <[email protected]> Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap3: PM: enable UART3 module wakeupsKevin Hilman1-3/+4
UART3 is in the PER powerdomain. If PER goes idle/inactive independently of CORE, for UART3 to wakeup it must have its wakeup enable bits setup in PM_WKEN_PER. This patch enables these bits. The reason it works when PER and CORE work together is because when CORE goes inactive/retention, the IOPAD wakeups are enabled and trigger UART3 wakeup. Without this patch, when the UART inactivity timer fires for UART3, its clocks are disabled and it's unable to wakeup so will be unusable until PER is awoken by another source. Another way of testing is by keeping CORE on during suspend but allowing PER to hit retention # echo 3 > /debug/pm_debug/core_pwrdm/suspend then enter suspend # echo mem > /sys/power/state Without this patch, UART3 will be unable to wakeup the system. Signed-off-by: Kevin Hilman <[email protected]> Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap2: Fix console serial port number for n8x0Tony Lindgren1-1/+1
With the recent changes omap serial ports match the physical numbering like they should. Fix the kernel CMDLINE accordingly so console works. Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap2: Fix detection of n8x0Tony Lindgren1-0/+12
Otherwise the machine_is_nokia_n8*() does not work. Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap1: Fix DSP public peripherals support for ams-deltaJanusz Krzysztofik1-0/+2
DSP public peripherals used to work on OMAP1510 based (or all OMAP1 class?) machines as long as old dspgateway code were present in the l-o tree. For several months it is no longer included, breaking support for McBSP1 based audio on Amstrad Delta, for example. This patch, derived from the old dspgateway code, corrects the problem for the board by simply taking the DSP out of reset state, I guess. That way, things should not break when a new dsp code is added to the tree, and the change can be reverted then. If there are any reports on McBSP1 or other DSP public peripherals not working for other OMAP1 machines (I've not heard of any for now), I can prepare a more general patch providing an extra include file with a helper function defined. Created and tested against linux-2.6.32-rc5 Signed-off-by: Janusz Krzysztofik <[email protected]> Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap1: Fix redundant UARTs pin muxing that can break other hardware supportJanusz Krzysztofik9-26/+72
Commit 15ac408ee5a509053a765b816e9179515329369f removed enabled_uart and OMAP_TAG_UART. This works for mach-omap2, but causes issues on mach-omap1 for some boards as the mach-omap1 serial.c was muxing pins based on the enabled_uart flag for 15xx. Fix this by muxing pins in board-*.c files for the 15xx boards for the uart ports that had enabled_uart flag set before the commit above. Tested on Amsdtrad Delta only. Note that in the future we should add support for powering down the uarts with a timer like mach-omap2/serial.c does. Otherwise the enabled uarts will be blocking retention-while-idle. Signed-off-by: Janusz Krzysztofik <[email protected]> Signed-off-by: Tony Lindgren <[email protected]>
2009-10-22omap: iommu: fix wrong condition check for SUPERSECTIONHiroshi DOYU1-1/+1
A bit (2 << 0) is set both on SECTION and SUPERSECTION. To identify SUPERSECTION correctly, other bits should be compared too. Reported-by: "Srinivas Pulukuru" <[email protected]> Signed-off-by: Hiroshi DOYU <[email protected]> Signed-off-by: Tony Lindgren <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.