summaryrefslogtreecommitdiffstats
path: root/arch
Commit message (Collapse)AuthorAgeFilesLines
* omap: headers: Create headers necessary for compile under mach-omap1 and ↵Tony Lindgren2009-10-1930-0/+100
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | mach-omap2 Create the headers needed for compiling under mach-omap1/include/mach and mach-omap2/include/mach. This was done with the following script: #!/bin/bash mach_files="clkdev.h gpio.h hardware.h io.h irqs.h memory.h \ smp.h system.h timex.h uncompress.h vmalloc.h" omaps="mach-omap1 mach-omap2" mach_dir_old="arch/arm/plat-omap/include/mach" plat_dir_new="arch/arm/plat-omap/include/plat" mkdir -p $plat_dir_new git add $plat_dir_new for dir in $omaps; do mach_dir_new="arch/arm/$dir/include/mach" for header in $mach_files; do file="$mach_dir_new/$header" if [ ! -f $file ]; then echo -ne "/*\n * $file\n */\n\n#include <plat/$header>\n" > $file git add $file if [ ! -f $plat_dir_new/$header ]; then git mv $mach_dir_old/$header $plat_dir_new/$header fi fi done done Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: headers: Move mtd-xip.h to be mach-omap1 specificTony Lindgren2009-10-191-0/+0
| | | | | | These registers are omap1 specific. Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Split vmalloc.h for mach-omap1 and mach-omap2Tony Lindgren2009-10-192-5/+21
| | | | | | | | | | | | | | | Earlier patch "omap: Remap L3, L4 to get more kernel io address space" changed the VMALLOC_END. However, this change causes problems on mach-omap1: BUG: mapping for 0xe0000000 at 0xe0000000 overlaps vmalloc space BUG: mapping for 0xe1000000 at 0xe1000000 overlaps vmalloc space Fix this by creating separate vmalloc.h files for mach-omap1 and mach-omap2. Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: headers: Split entry-macro.S for mach-omap1 and mach-omap2Tony Lindgren2009-10-192-50/+58
| | | | | | Split entry-macro.S for mach-omap1 and mach-omap2 Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: headers: Split debug-macro.S for mach-omap1 and mach-omap2Tony Lindgren2009-10-192-13/+47
| | | | | | | This also creates the include/mach subdirectories under mach-omap1 and mach-omap2. Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: headers: Add mach path to include filesTony Lindgren2009-10-191-6/+6
| | | | | | | This is to prepare for moving hardware.h to live under plat instead of mach. Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Add OMAP4 L3 and L4 peripherals.Santosh Shilimkar2009-10-194-3/+55
| | | | | | | This patch adds few necessary peripherals for OMAP4. Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Fix DEBUG_LL UART io addressSantosh Shilimkar2009-10-1915-18/+18
| | | | | | | | This patch fixes the low level debug UART io address as per this series. The change is essential to have CONFIG_DEBUG_LL working. Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Move SRAM map to claim more io spaceSantosh Shilimkar2009-10-191-3/+3
| | | | | | | This patch moves SRAM map to free up more kernel address io space. Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Remap L3, L4 to get more kernel io address spaceSantosh Shilimkar2009-10-192-40/+64
| | | | | | | | | | | | | | This patch remap L3 and L4 io space to get more kernel address space. With this patch, 512 MB of IO space is reclaimed. Some more combinations are possible but to make it uniform across OMAP24XX, OMAP34XX and OMAP4430, these io combinations are chosen Once this is reviewed and tested sufficiently, a documentation entry can be created to ease up reading and debugging. Like "Documentation/arm/omap/io_map.txt" Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Split OMAP2_IO_ADDRESS to L3 and L4Santosh Shilimkar2009-10-1913-66/+75
| | | | | | | | | | | | | | | This patch splits OMAP2_IO_ADDRESS to OMAP2_L3_IO_ADDRESS and OMAP2_L4_IO_ADDRESS to reclaim more IO space. The omap_read*() and omap_write*() functions will work only over L4 address space. Current omap kernel stack uses these functions only to access registers over L4 io address space Note that these macros should only be used when ioremap does not work. Please use ioremap instead in all new code. Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use ioremap for omap4 L4 codeTony Lindgren2009-10-197-16/+42
| | | | | | | Use ioremap for omap4 L4 code Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use getnstimeofday for omap_deviceTony Lindgren2009-10-191-31/+19
| | | | | | | Use getnstimeofday for omap_device Acked-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use ioremap in omap_hwmod.cTony Lindgren2009-10-191-6/+12
| | | | | | | | Use ioremap in omap_hwmod.c Acked-by: Paul Walmsley <paul@pwsan.com> Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use ioremap in gpio.cTony Lindgren2009-10-191-95/+129
| | | | | | Use ioremap in gpio.c Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use ioremap in dmtimer.cTony Lindgren2009-10-191-7/+9
| | | | | | Use ioremap in dmtimer.c Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use ioremap in dma.cTony Lindgren2009-10-191-10/+24
| | | | | | Use ioremap in dma.c Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Use ioremap in irq.cTony Lindgren2009-10-191-2/+10
| | | | | | Use ioremap in irq.c Signed-off-by: Tony Lindgren <tony@atomide.com>
* omap: Change low-level serial init to use ioremapTony Lindgren2009-10-162-7/+18
| | | | | | Change low-level serial init to use ioremap Signed-off-by: Tony Lindgren <tony@atomide.com>
* Merge branch 'omap7xx-fortony-rc3' of git://robotfuzz.com/linwizard-kernel ↵Tony Lindgren2009-10-1522-676/+476
|\ | | | | | | into omap7xx
| * OMAP850: Fix zImage bootingAlistair Buxton2009-10-071-1/+2
| | | | | | | | | | | | | | This adds the OMAP850 JTAG ID to the IDs checked by OMAP uncompress.h putc. Without this putc hangs up trying to check the uarts and zImage crashes. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com>
| * OMAP850: PM: Add an ARCH_OMAP850 checkAlistair Buxton2009-10-071-0/+1
| | | | | | | | | | | | This just makes the same warning be printed on omap850 and omap730. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com>
| * OMAP7XX: Clocks: Add ck_ref and armxorAngelo Arrifano2009-10-071-2/+2
| | | | | | | | | | | | | | | | These clocks are required for booting. Signed-off-by: Angelo Arrifano <miknix@gmail.com> Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Rename all the rest of the omap730 references in omap1 coreAlistair Buxton2009-10-0712-191/+191
| | | | | | | | | | | | | | | | | | | | | | This patch is part of a series which removes references to omap730 in code which is shared with omap850, replacing them with references to omap7xx. This updates all the remaining omap730 references in miscellaneous local variables, macros and similar. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Replace omap730 references in irqs.h and all usersAlistair Buxton2009-10-0710-102/+102
| | | | | | | | | | | | | | | | | | | | This patch is part of a series which removes references to omap730 in code which is shared with omap850, replacing them with references to omap7xx. Turns INT_730_* to INT_7XX_* for definitions in irqs.h and all users. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Update core omap1 files to use omap7xx.hAlistair Buxton2009-10-077-33/+33
| | | | | | | | | | | | | | | | This patch is part of a series which removes references to omap730 in code which is shared with omap850, replacing them with references to omap7xx. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Create omap7xx.hAlistair Buxton2009-10-071-0/+104
| | | | | | | | | | | | | | | | | | | | | | This patch is part of a series which removes references to omap730 in code which is shared with omap850, replacing them with references to omap7xx. This include file is intended to replace omap730.h and omap850.h All values defined herein are identical to those in both the old files. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: McBSP: Add omap850 supportAlistair Buxton2009-10-072-4/+4
| | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Clocks: Add omap850 supportAlistair Buxton2009-10-071-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. This file had no omap850 specific code. Initial clock support was done in the Linwizard tree by Zebediah C. McClure. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: PM: Add omap850 supportAlistair Buxton2009-10-071-10/+10
| | | | | | | | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. This file had no omap850 specific code. Original omap850 support in Linwizard was done by cloning the omap730 code. That work was done by Zebediah C. McClure. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: IRQ: Remove duplicate omap850 codeAlistair Buxton2009-10-073-104/+7
| | | | | | | | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. This fixes a bug which prevents IRQs from being enabled on omap850 due to a missing check in entry-macro.S, which was found by Cory Maccarrone. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: USB: Remove duplicate omap850 codeAlistair Buxton2009-10-071-3/+3
| | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Mux: Remove duplicate omap850 codeAlistair Buxton2009-10-072-76/+2
| | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: IO: Remove duplicate omap850 codeAlistair Buxton2009-10-072-27/+4
| | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: GPIO: Remove duplicate omap850 codeAlistair Buxton2009-10-071-119/+14
| | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
| * OMAP7XX: Serial: Remove duplicate omap850 codeAlistair Buxton2009-10-071-8/+1
| | | | | | | | | | | | | | | | | | This patch is part of a series which unifies all duplicated code between omap730 and omap850. All cpu checks are converted to cpu_is_omap7xx() and CONFIG_ARCH_OMAP850 is added to all CONFIG_ARCH_OMAP730 checks. Signed-off-by: Alistair Buxton <a.j.buxton@gmail.com> Reviewed-by: Zebediah C. McClure <zmc@lurian.net>
* | Merge branch 'x86-fixes-for-linus' of ↵Linus Torvalds2009-10-153-12/+13
|\ \ | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip * 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: Revert "x86: linker script syntax nits" x86, perf_event: Rename 'performance counter interrupt'
| * | Revert "x86: linker script syntax nits"Ingo Molnar2009-10-152-10/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit e9a63a4e559fbdc522072281d05e6b13c1022f4b. This breaks older binutils, where sink-less asserts are broken. See this commit for further details: d2ba8b2: x86: Fix assert syntax in vmlinux.lds.S Acked-by: "H. Peter Anvin" <hpa@zytor.com> Acked-by: Sam Ravnborg <sam@ravnborg.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> LKML-Reference: <4AD6523D.5030909@zytor.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | Merge branch 'linus' into x86/urgentIngo Molnar2009-10-15331-5084/+1936
| |\ \ | | | | | | | | | | | | | | | | | | | | Merge reason: pull in latest, to be able to revert a patch there. Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * | | x86, perf_event: Rename 'performance counter interrupt'Li Hong2009-10-141-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In 'cdd6c482c9ff9c55475ee7392ec8f672eddb7be6', we renamed Performance Counters -> Performance Events. The name showed up in /proc/interrupts also needs a change. I use PMI (Performance monitoring interrupt) here, since it is the official name used in Intel's documents. Signed-off-by: Li Hong <lihong.hi@gmail.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Paul Mackerras <paulus@samba.org> LKML-Reference: <20091014105039.GA22670@uhli> Signed-off-by: Ingo Molnar <mingo@elte.hu>
* | | | Merge branch 'merge' of ↵Linus Torvalds2009-10-1511-28/+28
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 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/pci: Fix MODPOST warning powerpc/oprofile: Add ppc750 CL as supported by oprofile powerpc: warning: allocated section `.data_nosave' not in segment powerpc/kgdb: Fix build failure caused by "kgdb.c: unused variable 'acc'" powerpc: Fix hypervisor TLB batching powerpc/mm: Fix hang accessing top of vmalloc space powerpc: Fix memory leak in axon_msi.c powerpc/pmac: Fix issues with sleep on some powerbooks powerpc64/ftrace: use PACA to retrieve TOC in mod_return_to_handler powerpc/ftrace: show real return addresses in modules
| * \ \ \ Merge commit 'ftrace/ppc' into mergeBenjamin Herrenschmidt2009-10-152-5/+8
| |\ \ \ \
| | * | | | powerpc64/ftrace: use PACA to retrieve TOC in mod_return_to_handlerSteven Rostedt2009-10-131-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The mod_return_to_handler needs to switch to the kernel TOC before jumping to a the kernel code. It currently does this by looking at the kernel function data and retrieves the TOC that way. Not only is this inefficient, it also breaks with a relocatable kernel. The PACA contains the kernel TOC and we can easily retrieve it that way. Reported-by: Sachin Sant <sachinp@in.ibm.com> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
| | * | | | powerpc/ftrace: show real return addresses in modulesSteven Rostedt2009-10-131-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the function graph tracer is enabled, it replaces the return address with a hook back to the tracer. This makes back traces see the hook instead of the actual return address. The current code also shows the real address by checking if the return address jumps to the return_to_handler. If it is, is also prints out the saved real return address. On powerpc64, some modules may return to mod_return_to_handler, which is not checked. This patch will also show the real address if a return is to mod_return_to_handler as well. Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
| * | | | | powerpc/pci: Fix MODPOST warningHeiko Schocher2009-10-141-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | making a powerpc target with PCI support, shows the following warning: MODPOST vmlinux.o WARNING: vmlinux.o(.text+0x10430): Section mismatch in reference from the function pcibios_allocate_bus_resources() to the function .init.text:reparent_resources() The function pcibios_allocate_bus_resources() references the function __init reparent_resources(). This is often because pcibios_allocate_bus_resources lacks a __init annotation or the annotation of reparent_resources is wrong. This patch fix this warning by removing the __init annotation before reparent_resources. Signed-off-by: Heiko Schocher <hs@denx.de> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
| * | | | | powerpc/oprofile: Add ppc750 CL as supported by oprofileDragos Tatulea2009-10-141-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Here's a patch that adds the ppc750 CL cpu as supported by oprofile. Signed-off-by: Dragos Tatulea <dtatulea@ixiacom.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
| * | | | | powerpc: warning: allocated section `.data_nosave' not in segmentSean MacLennan2009-10-141-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We need to align before the output section. Having the align inside the output section causes the linker to put some filler in there, which makes it a non-empty section, but this section isn't assigned to a segment so you get a warning from the linker. Signed-off-by: Sean MacLennan <smaclennan@pikatech.com> Acked-by: Segher Boessenkool <segher@kernel.crashing.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
| * | | | | powerpc/kgdb: Fix build failure caused by "kgdb.c: unused variable 'acc'"Anton Vorontsov2009-10-141-6/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 'acc' isn't used anywhere and thus triggers gcc warning, which causes build error with CONFIG_PPC_DISABLE_WERROR=n (default): cc1: warnings being treated as errors arch/powerpc/kernel/kgdb.c: In function 'gdb_regs_to_pt_regs': arch/powerpc/kernel/kgdb.c:289: warning: unused variable 'acc' make[1]: *** [arch/powerpc/kernel/kgdb.o] Error 1 Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
| * | | | | powerpc: Fix hypervisor TLB batchingAnton Blanchard2009-10-142-7/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Profiling of a page fault scalability microbenchmark shows flush_hash_range is not calling the batch hpte invalidate hcall (H_BULK_REMOVE). It turns out we have a duplicate firmware feature for hcall-bulk and the current setup code stops after finding the first match. This meant we never batch and always do individual invalidates. The patch below removes the duplicate and shifts FW_FEATURE_CMO to close the gap. With the patch applied the single threaded page fault rate improves from 217169 to 238755 per second on a POWER5 test box, a 10% improvement. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
| * | | | | powerpc/mm: Fix hang accessing top of vmalloc spaceBenjamin Herrenschmidt2009-10-141-6/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On pSeries, we always force the IO space to be mapped using 4K pages even with a 64K base page size to cope with some limitations in the HV interface to some devices. However, the SLB miss handler code to discriminate between vmalloc and ioremap space uses a CPU feature section such that the code is nop'ed out when the processor support large pages non-cachable mappings. Thus, we end up always using the ioremap page size for vmalloc segments on such processors, causing a discrepency between the segment and the hash table, and thus a hang continously hashing the page. It works for the first segment of the vmalloc space since that segment is "bolted" in by C code correctly, and thankfully we almost never use the vmalloc space beyond the first segment, but the new percpu code made the bug happen. This fixes it by removing the feature section from the assembly, we now always do the comparison between vmalloc and ioremap. Signed-off-by; Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
OpenPOWER on IntegriCloud