- Aug 25, 2006
-
-
Michael Neuling authored
The rtas console doesn't have to be Cell specific. If we get both RTAS tokens, we should just enabled the console then and there. Signed-off-by:
Michael Neuling <mikey@neuling.org> Signed-off-by:
Paul Mackerras <paulus@samba.org>
-
Jon Loeliger authored
Signed-off-by:
Jon Loeliger <jdl@freescale.com> Signed-off-by:
-
Olof Johansson authored
Cleanup CPU inits a bit more, Geoff Levand already did some earlier. * Move CPU state save to cpu_setup, since cpu_setup is only ever done on cpu 0 on 64-bit and save is never done more than once. * Rename __restore_cpu_setup to __restore_cpu_ppc970 and add function pointers to the cputable to use instead. Powermac always has 970 so no need to check there. * Rename __970_cpu_preinit to __cpu_preinit_ppc970 and check PVR before calling it instead of in it, it's too early to use cputable. * Rename pSeries_secondary_smp_init to generic_secondary_smp_init since everyone but powermac and iSeries use it. Signed-off-by:
Olof Johansson <olof@lixom.net> Signed-off-by:
-
Olof Johansson authored
Rename cpu_setup_power4.S to cpu_setup_ppc970.S, since that's really what it is. No functional or other changes. Signed-off-by:
-
Geoff Levand authored
Whitespace clean up for cell/interrupt.c. Signed-off-by:
Geoff Levand <geoffrey.levand@am.sony.com> Signed-off-by:
-
Michael Neuling authored
Cleanup some of the #define magic as suggested by Milton. Signed-off-by:
-
- Aug 08, 2006
-
-
Michael Neuling authored
This adds a shadow buffer for the SLBs and regsiters it with PHYP. Only the bolted SLB entries (top 3) are shadowed. The SLB shadow buffer tells the hypervisor what the kernel needs to have in the SLB for the kernel to be able to function. The hypervisor can use this information to speed up partition context switches. Signed-off-by:
-
Matt Porter authored
The PIN_SIZE definition name changed, update 44x_mmu.c accordingly. Signed-off-by:
Matt Porter <mporter@embeddedalley.com> Signed-off-by:
-
Matt Porter authored
Removes the flush_dcache_all export for non coherent platforms. We removed the last in-kernel user of this years ago in arch/ppc so it no longer serves a purpose. Plus, it breaks the build at the moment. Signed-off-by:
-
Jon Loeliger authored
As per list discussion, let's add device tree source files under powerpc/boot/dts. If nothing else, it is a starting point. Signed-off-by:
-
Michael Ellerman authored
We don't have much in the way of doc comments, but some of those we do have don't work because they start with "/***" or "/*", not "/**" which is what kernel-doc requires. Signed-off-by:
Michael Ellerman <michael@ellerman.id.au> Signed-off-by:
-
Amos Waterland authored
I think that most people who use maple_defconfig are doing so for a JS21, so it might make sense to turn Tigon3 support on by default. Built and booted on a JS21. Signed-off-by:
Amos Waterland <apw@us.ibm.com> Signed-off-by:
-
Jake Moilanen authored
Forgot to export symbols for MSI. Signed-off-by:
Jake Moilanen <moilanen@austin.ibm.com> Acked-by:
Segher Boessenkool <segher@kernel.crashing.org> Signed-off-by:
-
Jon Loeliger authored
IRQ setup now comes from the Flat Device Tree and use the new generic IRQ code. Fixed the fsl_soc.c IRQ OF interrupt node parsing. Removed some unused MPC86xx macro definition. Signed-off-by:
Zhang Wei <wei.zhang@freescale.com> Signed-off-by:
-
Haren Myneni authored
Noticing the following might_sleep warning (dump_stack()) during kdump testing when CONFIG_DEBUG_SPINLOCK_SLEEP is enabled. All secondary CPUs will be calling rtas_set_indicator with interrupts disabled to remove them from global interrupt queue. BUG: sleeping function called from invalid context at arch/powerpc/kernel/rtas.c:463 in_atomic():1, irqs_disabled():1 Call Trace: [C00000000FFFB970] [C000000000010234] .show_stack+0x68/0x1b0 (unreliable) [C00000000FFFBA10] [C000000000059354] .__might_sleep+0xd8/0xf4 [C00000000FFFBA90] [C00000000001D1BC] .rtas_busy_delay+0x20/0x5c [C00000000FFFBB20] [C00000000001D8A8] .rtas_set_indicator+0x6c/0xcc [C00000000FFFBBC0] [C000000000048BF4] .xics_teardown_cpu+0x118/0x134 [C00000000FFFBC40] [C00000000004539C] .pseries_kexec_cpu_down_xics+0x74/0x8c [C00000000FFFBCC0] [C00000000002DF08] .crash_ipi_callback+0x15c/0x188 [C00000000FFFBD50] [C0000000000296EC] .smp_message_recv+0x84/0xdc [C00000000FFFBDC0] [C000000000048E08] .xics_ipi_dispatch+0xf0/0x130 [C00000000FFFBE50] [C00000000009EF10] .handle_IRQ_event+0x7c/0xf8 [C00000000FFFBF00] [C0000000000A0A14] .handle_percpu_irq+0x90/0x10c [C00000000FFFBF90] [C00000000002659C] .call_handle_irq+0x1c/0x2c [C00000000058B9C0] [C00000000000CA10] .do_IRQ+0xf4/0x1a4 [C00000000058BA50] [C0000000000044EC] hardware_interrupt_entry+0xc/0x10 --- Exception: 501 at .plpar_hcall_norets+0x14/0x1c LR = .pseries_dedicated_idle_sleep+0x190/0x1d4 [C00000000058BD40] [C00000000058BDE0] 0xc00000000058bde0 (unreliable) [C00000000058BDF0] [C00000000001270C] .cpu_idle+0x10c/0x1e0 [C00000000058BE70] [C000000000009274] .rest_init+0x44/0x5c To fix this issue, rtas_set_indicator_fast() is added so that will not wait for RTAS 'busy' delay and this new function is used for kdump (in xics_teardown_cpu()) and for CPU hotplug ( xics_migrate_irqs_away() and xics_setup_cpu()). Note that the platform architecture spec says that set-indicator on the indicator we're using here is not permitted to return the busy or extended busy status codes. Signed-off-by:
Haren Myneni <haren@us.ibm.com> Signed-off-by:
-
Sonny Rao authored
We should not be calling power4_enable_pmcs() in pseries_lpar_enable_pmcs(); just doing the hypercall is sufficient. Prior to 2.6.15 we did not call power4_enable_pmcs() for an lpar. power4_enable_pmcs() tries to read the hid0 register which is no longer legal for an lpar in newer Power processors. Signed-off-by:
-
- Aug 07, 2006
-
-
Michael Neuling authored
We have CPU_FTR_PURR now, so let's use it. Signed-off-by:
-
- Aug 01, 2006
-
-
Anton Blanchard authored
Our pseries hcall interfaces are out of control: plpar_hcall_norets plpar_hcall plpar_hcall_8arg_2ret plpar_hcall_4out plpar_hcall_7arg_7ret plpar_hcall_9arg_9ret Create 3 interfaces to cover all cases: plpar_hcall_norets: 7 arguments no returns plpar_hcall: 6 arguments 4 returns plpar_hcall9: 9 arguments 9 returns There are only 2 cases in the kernel that need plpar_hcall9, hopefully we can keep it that way. Pass in a buffer to stash return parameters so we avoid the &dummy1, &dummy2 madness. Signed-off-by:
Anton Blanchard <anton@samba.org> -- Signed-off-by:
-
- Jul 31, 2006
-
-
Michael Hanselmann authored
This patch fixes several problems: - The legacy backlight value might be set at interrupt time. Introduced a worker to prevent it from directly calling the backlight code. - via-pmu allows the backlight to be grabbed, in which case we need to prevent other kernel code from changing the brightness. - Don't send PMU requests in via-pmu-backlight when the machine is about to sleep or waking up. - More Kconfig fixes. Signed-off-by:
Michael Hanselmann <linux-kernel@hansmi.ch> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: "Antonino A. Daplas" <adaplas@pol.net> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Roland McGrath authored
The latest toolchains can produce a new ELF section in DSOs and dynamically-linked executables. The new section ".gnu.hash" replaces ".hash", and allows for more efficient runtime symbol lookups by the dynamic linker. The new ld option --hash-style={sysv|gnu|both} controls whether to produce the old ".hash", the new ".gnu.hash", or both. In some new systems such as Fedora Core 6, gcc by default passes --hash-style=gnu to the linker, so that a standard invocation of "gcc -shared" results in producing a DSO with only ".gnu.hash". The new ".gnu.hash" sections need to be dealt with the same way as ".hash" sections in all respects; only the dynamic linker cares about their contents. To work with older dynamic linkers (i.e. preexisting releases of glibc), a binary must have the old ".hash" section. The --hash-style=both option produces binaries that a new dynamic linker can use more efficiently, but an old dynamic linker can still handle. The new section runs afoul of the custom linker scripts used to build vDSO images for the kernel. On ia64, the failure mode for this is a boot-time panic because the vDSO's PT_IA_64_UNWIND segment winds up ill-formed. This patch addresses the problem in two ways. First, it mentions ".gnu.hash" in all the linker scripts alongside ".hash". This produces correct vDSO images with --hash-style=sysv (or old tools), with --hash-style=gnu, or with --hash-style=both. Second, it passes the --hash-style=sysv option when building the vDSO images, so that ".gnu.hash" is not actually produced. This is the most conservative choice for compatibility with any old userland. There is some concern that some ancient glibc builds (though not any known old production system) might choke on --hash-style=both binaries. The optimizations provided by the new style of hash section do not really matter for a DSO with a tiny number of symbols, as the vDSO has. If someone wants to use =gnu or =both for their vDSO builds and worry less about that compatibility, just change the option and the linker script changes will make any choice work fine. Signed-off-by:Roland McGrath <roland@redhat.com> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Paul Mackerras <paulus@samba.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Jeff Dike <jdike@addtoit.com> Cc: Andi Kleen <ak@muc.de> Cc: Sam Ravnborg <sam@ravnborg.org> Signed-off-by:
-
Chandra Seetharaman authored
Few of the callback functions and notifier blocks that are associated with cpu notifications incorrectly have __devinit and __devinitdata. They should be __cpuinit and __cpuinitdata instead. It makes no functional difference but wastes text area when CONFIG_HOTPLUG is enabled and CONFIG_HOTPLUG_CPU is not. This patch fixes all those instances. Signed-off-by:
Chandra Seetharaman <sekharan@us.ibm.com> Cc: Ashok Raj <ashok.raj@intel.com> Signed-off-by:
-
Horms authored
This patch is part of an effort to unify the panic_on_oops behaviour across all architectures that implement it. It was pointed out to me by Andi Kleen that if an oops has occured in interrupt context, then calling sleep() in the oops path will only cause a panic, and that it would be really better for it not to be in the path at all. This patch removes the ssleep() call and reworks the console message accordinly. I have a slght concern that the resulting console message is too long, feedback welcome. For powerpc it also unifies the 32bit and 64bit behaviour. Fror x86_64, this patch only updates the console message, as ssleep() is already not present. Signed-off-by:
Horms <horms@verge.net.au> Acked-by:
-
Jeremy Kerr authored
Use BUG_ON rather than BUG to simplify the dma_ops handing, and remove the now-unnecessary return cases. Booted on pseries. Signed-off-by:
Jeremy Kerr <jk@ozlabs.org> Signed-off-by:
-
Jeremy Kerr authored
Previous changes have treated the return values of get_property as const, so now we can make the actual change to get_property(). There shouldn't be a need to cast the return values anymore. We will now get compiler warnings when property values are assigned to a non-const variable. If properties need to be updated, there's still the of_find_property function. Built for cell_defconfig, chrp32_defconfig, g5_defconfig, iseries_defconfig, maple_defconfig, pmac32_defconfig, ppc64_defconfig and pseries_defconfig. Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. tsi108 driver changes. Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. powermac platform & macintosh driver changes. Built for pmac32_defconfig, g5_defconfig Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. maple platform changes. Built for maple_defconfig Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. chrp platform changes. Built for chrp32_defconfig Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. cell platform changes. Built for cell_defconfig Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. mpc* platform changes. Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. iseries platform changes. Built for iseries_defconfig Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. pseries platform changes. Built for pseries_defconfig Signed-off-by:
-
Jeremy Kerr authored
Now that get_property() returns a void *, there's no need to cast its return value. Also, treat the return value as const, so we can constify get_property later. powerpc core changes. Signed-off-by:
-
Geoff Levand authored
A minor comment fix for misc_64.S from Takao Shinohara. Signed-off-by:
-
Jake Moilanen authored
On the JS21 systems, they have the SPLPAR hypertas set, but are not SMT capable. So, they are not making the H_CEDE call. This is causing the hypervisor to have to queue up work for the hdecr, taking an excessive amount of time in maintenance code, and causing jitter on the box. Making the H_CEDE call helps alleviate that problem. Signed-off-by:
-
- Jul 28, 2006
-
-
Olaf Hering authored
The firmware of POWER4 and JS20 systems does not switch the cpu to 64bit mode when the registered system_reset and machine_check handlers get called. If a 32bit process runs on that cpu at the time of the event, the cpu remains in 32bit mode. xmon and kdump can not deal with it, the result is an error like 'Bad kernel stack pointer fff2aad0 at 3200'. xmon just loses some register info, but booting the kdump kernel usually fails. Both handlers are not hot paths. Duplicate the EXCEPTION_PROLOG_PSERIES macro and add two instructions to switch to 64bit: li r11,5; rldimi r10,r11,61,0; Signed-off-by:
Olaf Hering <olh@suse.de> Signed-off-by:
-
Segher Boessenkool authored
As the code comment already says, the Maple device-tree is incorrect here; make the Linux code detect the correct thing, too. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Segher Boessenkool authored
All U3/U4 based systems are big-endian, not all express it in their device trees. Signed-off-by:
-
Benjamin Herrenschmidt authored
The Xserve G5 are capable of frequency switching like other desktop G5s. This enables it. It also fix a Kconfig issue which prevented from building the G5 cpufreq support if CONFIG_PMAC_SMU was not set (the first version of that driver only worked with SMU based macs, but this isn't the case anymore). Signed-off-by:
-
- Jul 25, 2006
-
-
Michael Ellerman authored
There's a bug in my cleaned up mem= handling, if the memory limit is larger than the RMO size we'll erroneously enlarge the RMO size. Fix is to only change the RMO size if the memory limit is less than the current RMO value. Signed-off-by:
-
