aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* x86: mark read_crX() asm code as volatileKirill Korotaev2007-10-172-5/+5
| | | | | | | | | Some gcc versions (I checked at least 4.1.1 from RHEL5 & 4.1.2 from gentoo) can generate incorrect code with read_crX()/write_crX() functions mix up, due to cached results of read_crX(). The small app for x8664 below compiled with -O2 demonstrates this (i686 does the same thing):
* x86: initialize 64bit registers for a.out executablesAndi Kleen2007-10-171-0/+2
| | | | | | | | | | | Previously the data from before the exec was kept in there. Zero them instead. [ tglx: arch/x86 adaptation ] Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* x86: return correct error code from child_rip in x86_64 entry.SAndrey Mirkin2007-10-171-1/+1
| | | | | | | | | | | | | | | Right now register edi is just cleared before calling do_exit. That is wrong because correct return value will be ignored. Value from rax should be copied to rdi instead of clearing edi. AK: changed to 32bit move because it's strictly an int [ tglx: arch/x86 adaptation ] Signed-off-by: Andrey Mirkin <major@openvz.org> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* i386: avoid temporarily inconsistent pte-sJan Beulich2007-10-171-2/+1
| | | | | | | | | | | | | | | | | | One more of these issues (which were considered fixed a few releases back): other than on x86-64, i386 allows set_fixmap() to replace already present mappings. Consequently, on PAE, care must be taken to not update the high half of a pte while the low half is still holding the old value. [ tglx: arch/x86 adaptation ] Signed-off-by: Jan Beulich <jbeulich@novell.com> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> arch/x86/mm/pgtable_32.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
* i386: fix section mismatch warning in intel.cSam Ravnborg2007-10-172-16/+15
| | | | | | | | | | | | | | | | | | | | | | Fix following section mismatch warning: WARNING: vmlinux.o(.text+0xc88c): Section mismatch: reference to .init.text:trap_init_f00f_bug (between 'init_intel' and 'cpuid4_cache_lookup') init_intel are __cpuint where trap_init_f00f_bug is __init. Fixed by declaring trap_init_f00f_bug __cpuinit. Moved the defintion of trap_init_f00f_bug to the sole user in init.c so the ugly prototype in intel.c could get killed. Frank van Maarseveen <frankvm@frankvm.com> supplied the .config used to reproduce the warning. [ tglx: arch/x86 adaptation ] Cc: Frank van Maarseveen <frankvm@frankvm.com> Signed-off-by: Sam Ravnborg <sam@ravnborg.org> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* i386: Fix section mismatchSatyam Sharma2007-10-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Fix bugzilla #8679 WARNING: arch/i386/kernel/built-in.o(.data+0x2148): Section mismatch: reference to .init.text: (between 'thermal_throttle_cpu_notifier' and 'mtrr_mutex') comes because struct notifier_block thermal_throttle_cpu_notifier in arch/i386/kernel/cpu/mcheck/therm_throt.c goes in .data section but the notifier callback function itself has been marked __cpuinit which becomes __init == .init.text when HOTPLUG_CPU=n. The warning is bogus because the callback will never be called out if HOTPLUG_CPU=n in the first place (as one can see from kernel/cpu.c, the cpu_chain itself is __cpuinitdata :-) So, let's mark thermal_throttle_cpu_notifier as __cpuinitdata to fix the section mismatch warning. [ tglx: arch/x86 adaptation ] Signed-off-by: Satyam Sharma <satyam@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* i386: fix 4 bit apicid assumption of mach-defaultSiddha, Suresh B2007-10-171-1/+7
| | | | | | | | | | | | | | | | | | | Fix get_apic_id() in mach-default, so that it uses 8 bits incase of xAPIC case and 4 bits for legacy APIC case. This fixes the i386 kernel assumption that apic id is less than 16 for xAPIC platforms with 8 cpus or less and makes the kernel boot on such platforms. [ tglx: arch/x86 adaptation ] Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Signed-off-by: Andi Kleen <ak@suse.de> Cc: Andi Kleen <ak@suse.de> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* x86: fix off-by-one in find_next_zero_stringAndrew Hastings2007-10-171-1/+1
| | | | | | | | | | | | Fix an off-by-one error in find_next_zero_string which prevents allocating the last bit. [ tglx: arch/x86 adaptation ] Signed-off-by: Andrew Hastings <abh@cray.com> on behalf of Cray Inc. Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* i386: export i386 smp_call_function_mask() to modulesLaurent Vivier2007-10-172-6/+10
| | | | | | | | | | | | | This patch export i386 smp_call_function_mask() with EXPORT_SYMBOL(). This function is needed by KVM to call a function on a set of CPUs. [ tglx: arch/x86 adaptation ] Signed-off-by: Laurent Vivier <Laurent.Vivier@bull.net> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* x86: Install unstripped copy of 64bit vdso to diskRoland McGrath2007-10-172-1/+17
| | | | | | | | | | | | | | | This keeps an unstripped copy of the 64bit vDSO images built before they are stripped and embedded in the kernel. The unstripped copies get installed in $(MODLIB)/vdso/ by "make install" (or you can explicitly use the subtarget "make vdso_install"). These files can be useful when they contain source-level debugging information. [ tglx: arch/x86 adaptation ] Signed-off-by: Roland McGrath <roland@redhat.com> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* x86_64: install unstripped copies of compat vdso on diskRoland McGrath2007-10-172-5/+27
| | | | | | | | | | | | | | | This keeps an unstripped copy of the vDSO images built before they are stripped and embedded in the kernel. The unstripped copies get installed in $(MODLIB)/vdso/ by "make install" (or you can explicitly use the subtarget "make vdso_install"). These files can be useful when they contain source-level debugging information. [ tglx: arch/x86 adaptation ] Signed-off-by: Roland McGrath <roland@redhat.com> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* i386: setup_trampoline() must be __cpuinitAdrian Bunk2007-10-171-1/+1
| | | | | | | | | | | | | | | | | WARNING: arch/i386/kernel/built-in.o(.text+0xf201): Section mismatch: reference to .init.data:trampoline_end (between 'setup_trampoline' and 'cpu_coregroup_map') WARNING: arch/i386/kernel/built-in.o(.text+0xf207): Section mismatch: reference to .init.data:trampoline_data (between 'setup_trampoline' and 'cpu_coregroup_map') WARNING: arch/i386/kernel/built-in.o(.text+0xf21a): Section mismatch: reference to .init.data:trampoline_data (between 'setup_trampoline' and 'cpu_coregroup_map') Harmless but annoying warnings present when building an i386 SMP kernel with CONFIG_HOTPLUG_CPU=n and gcc < 4.0 . [ tglx: arch/x86 adaptation ] Signed-off-by: Adrian Bunk <bunk@stusta.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* x86: clean up apicid_to_node declarationAndrew Morton2007-10-171-1/+2
| | | | | | | | | | | | | | Use the correct #define in the declaration of apicid_to_node[], to match the definition. [ tglx: arch/x86 adaptation ] Cc: Andi Kleen <ak@suse.de> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* i386: make Oprofile call shutdown() only once per sessionStephane Eranian2007-10-171-1/+1
| | | | | | | | | | | | Oprofile: call model->shutdown() only once to avoid calling release_ev*() multiple times [ tglx: arch/x86 adaptation ] Signed-off-by: Stephane Eranian <eranian@hpl.hp.com> Signed-off-by: Andi Kleen <ak@suse.de> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
* x86: C1E late detection fix. Really switch off lapic timerThomas Gleixner2007-10-172-17/+4
| | | | | | | | | | | Doh, I completely missed that devices marked DUMMY are not running the set_mode function. So we force broadcasting, but we keep the local APIC timer running. Let the clock event layer mark the device _after_ switching it off. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@elte.hu>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-schedLinus Torvalds2007-10-174-20/+49
|\ | | | | | | | | | | | | | | * git://git.kernel.org/pub/scm/linux/kernel/git/mingo/linux-2.6-sched: sched: fix new task startup crash sched: fix !SYSFS build breakage sched: fix improper load balance across sched domain sched: more robust sd-sysctl entry freeing
| * sched: fix new task startup crashSrivatsa Vaddagiri2007-10-172-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Child task may be added on a different cpu that the one on which parent is running. In which case, task_new_fair() should check whether the new born task's parent entity should be added as well on the cfs_rq. Patch below fixes the problem in task_new_fair. This could fix the put_prev_task_fair() crashes reported. Reported-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com> Reported-by: Andy Whitcroft <apw@shadowen.org> Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * sched: fix !SYSFS build breakageDhaval Giani2007-10-172-8/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When CONFIG_SYSFS is not set, CONFIG_FAIR_USER_SCHED fails to build with kernel/built-in.o: In function `uids_kobject_init': (.init.text+0x1488): undefined reference to `kernel_subsys' kernel/built-in.o: In function `uids_kobject_init': (.init.text+0x1490): undefined reference to `kernel_subsys' kernel/built-in.o: In function `uids_kobject_init': (.init.text+0x1480): undefined reference to `kernel_subsys' kernel/built-in.o: In function `uids_kobject_init': (.init.text+0x1494): undefined reference to `kernel_subsys' This patch fixes this build error. Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> Signed-off-by: Dhaval Giani <dhaval@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * sched: fix improper load balance across sched domainKen Chen2007-10-171-4/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We recently discovered a nasty performance bug in the kernel CPU load balancer where we were hit by 50% performance regression. When tasks are assigned to a subset of CPUs that span across sched_domains (either ccNUMA node or the new multi-core domain) via cpu affinity, kernel fails to perform proper load balance at these domains, due to several logic in find_busiest_group() miss identified busiest sched group within a given domain. This leads to inadequate load balance and causes 50% performance hit. To give you a concrete example, on a dual-core, 2 socket numa system, there are 4 logical cpu, organized as: CPU0 attaching sched-domain: domain 0: span 0003 groups: 0001 0002 domain 1: span 000f groups: 0003 000c CPU1 attaching sched-domain: domain 0: span 0003 groups: 0002 0001 domain 1: span 000f groups: 0003 000c CPU2 attaching sched-domain: domain 0: span 000c groups: 0004 0008 domain 1: span 000f groups: 000c 0003 CPU3 attaching sched-domain: domain 0: span 000c groups: 0008 0004 domain 1: span 000f groups: 000c 0003 If I run 2 tasks with CPU affinity set to 0x5. There are situation where cpu0 has run queue length of 2, and cpu2 will be idle. The kernel load balancer is unable to balance out these two tasks over cpu0 and cpu2 due to at least three logics in find_busiest_group() that heavily bias load balance towards power saving mode. e.g. while determining "busiest" variable, kernel only set it when "sum_nr_running > group_capacity". This test is flawed that "sum_nr_running" is not necessary same as sum-tasks-allowed-to-run-within-the sched-group. The end result is that kernel "think" everything is balanced, but in reality we have an imbalance and thus causing one CPU to be over-subscribed and leaving other idle. There are two other logic in the same function will also causing similar effect. The nastiness of this bug is that kernel not be able to get unstuck in this unfortunate broken state. From what we've seen in our environment, kernel will stuck in imbalanced state for extended period of time and it is also very easy for the kernel to stuck into that state (it's pretty much 100% reproducible for us). So proposing the following fix: add addition logic in find_busiest_group to detect intrinsic imbalance within the busiest group. When such condition is detected, load balance goes into spread mode instead of default grouping mode. Signed-off-by: Ken Chen <kenchen@google.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
| * sched: more robust sd-sysctl entry freeingMilton Miller2007-10-171-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It occurred to me this morning that the procname field was dynamically allocated and needed to be freed. I started to put in break statements when allocation failed but it was approaching 50% error handling code. I came up with this alternative of looping while entry->mode is set and checking proc_handler instead of ->table. Alternatively, the string version of the domain name and cpu number could be stored the structs. I verified by compiling CONFIG_DEBUG_SLAB and checking the allocation counts after taking a cpuset exclusive and back. Signed-off-by: Ingo Molnar <mingo@elte.hu>
* | Merge branch 'for-linus' of git://git.kernel.dk/data/git/linux-2.6-blockLinus Torvalds2007-10-1710-47/+62
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'for-linus' of git://git.kernel.dk/data/git/linux-2.6-block: [SCSI] Remove full sg table memset() [SCSI] ide-scsi: remove usage of sg_last() Fix loop terminating conditions in fill_sg(). [BLOCK] Clear sg entry before filling in blk_rq_map_sg() IA64: iommu uses sg_next with an invalid sg element cciss: disable DMA refetch on Smart Array P600 swiotlb: fix map_sg failure handling SPARC64: fix iommu sg chaining [SCSI] ide-scsi: use scsi_sg_count() instead of ->use_sg
| * | [SCSI] Remove full sg table memset()FUJITA Tomonori2007-10-171-2/+0
| | | | | | | | | | | | | | | | | | | | | We don't need to do that anymore, since blk_rq_map_sg() clears individual entries. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | [SCSI] ide-scsi: remove usage of sg_last()Jens Axboe2007-10-171-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | We want to remove sg_last(), it's a very expensive interface. So keep track of number of sg entries in the sg list, instead of comparing with the last entry. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | Fix loop terminating conditions in fill_sg().David S. Miller2007-10-172-12/+15
| | | | | | | | | | | | | | | Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | [BLOCK] Clear sg entry before filling in blk_rq_map_sg()Jens Axboe2007-10-171-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | The memset() of the sg entry was originally removed, because it could overwrite a chain pointer. But it's quite OK to memset() it when we know it's a valid entry, since it can't contain a chain pointer. Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | IA64: iommu uses sg_next with an invalid sg elementFUJITA Tomonori2007-10-171-2/+2
| | | | | | | | | | | | | | | | | | | | | sg list elements might not be continuous. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | cciss: disable DMA refetch on Smart Array P600Mike Miller (OS Dev)2007-10-171-3/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch disables DMA refetch in the PCI bridge. We have disabled DMA prefetch for quite some time. Testing with XEN revealed another ASIC bug. If dom0 resides on a P600 the board can can an MCA bi accessing invalid memory addresses. Apparently, we need to disable both prefetch and refetch. My understanding is a refetch operation should not occur but it is a valid thing to do if prefetched data is no longer available for whatever reason. Please consider this patch for inclusion. Signed-off-by: Mike Miller <mike.miller@hp.com> Signed-off-by: Alex Chiang <achiang@hp.com> -------------------------------------------------------------------------------- Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | swiotlb: fix map_sg failure handlingFUJITA Tomonori2007-10-171-1/+1
| | | | | | | | | | | | | | | | | | | | | sg list elements might not be continuous. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | SPARC64: fix iommu sg chainingFUJITA Tomonori2007-10-174-23/+31
| | | | | | | | | | | | | | | | | | | | | | | | Commit 2c941a204070ab32d92d40318a3196a7fb994c00 looks incomplete. The helper functions like prepare_sg() need to support sg chaining too. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
| * | [SCSI] ide-scsi: use scsi_sg_count() instead of ->use_sgBoaz Harrosh2007-10-171-1/+1
| |/ | | | | | | Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
* | Merge branch 'merge' of ↵Linus Torvalds2007-10-1788-483/+3010
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc * 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc: (24 commits) [POWERPC] Fix vmemmap warning in init_64.c [POWERPC] Fix 64 bits vDSO DWARF info for CR register [POWERPC] Add 1TB workaround for PA6T [POWERPC] Enable NO_HZ and high res timers for pseries and ppc64 configs [POWERPC] Quieten cache information at boot [POWERPC] Quieten clockevent printk [POWERPC] Enable SLUB in *_defconfig [POWERPC] Fix 1TB segment detection [POWERPC] Fix iSeries_hpte_insert prototype [POWERPC] Fix copyright symbol [POWERPC] ibmebus: Move to of_device and of_platform_driver, match eHCA and eHEA drivers [POWERPC] ibmebus: Add device creation and bus probing based on of_device [POWERPC] ibmebus: Remove bus match/probe/remove functions [POWERPC] Move of_device allocation into of_device.[ch] [POWERPC] mpc52xx: device tree changes for FEC and MDIO [POWERPC] bestcomm: GenBD task support [POWERPC] bestcomm: FEC task support [POWERPC] bestcomm: ATA task support [POWERPC] bestcomm: core bestcomm support for Freescale MPC5200 [POWERPC] mpc52xx: Update mpc52xx_psc structure with B revision changes ...
| * \ Merge branch 'for-2.6.24' of git://git.secretlab.ca/git/linux-2.6-mpc52xx ↵Paul Mackerras2007-10-1727-7/+2591
| |\ \ | | | | | | | | | | | | into merge
| | * | [POWERPC] mpc52xx: device tree changes for FEC and MDIODomen Puncer2007-10-161-1/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Add device tree entries for lite5200b's FEC's PHY. Signed-off-by: Domen Puncer <domen.puncer@telargo.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] bestcomm: GenBD task supportSylvain Munaut2007-10-166-0/+449
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is the microcode for the GenBD task and the associated support code. This is a generic task that copy data to/from a hardware FIFO. This is currently locked to 32bits wide access but could be extended as needed. The microcode itself comes directly from the offical API (v2.2) Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] bestcomm: FEC task supportSylvain Munaut2007-10-166-0/+509
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is the microcode for the FEC task and the associated support code. The microcode itself comes directly from the offical API (v2.2) Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] bestcomm: ATA task supportSylvain Munaut2007-10-165-0/+267
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is the microcode for the ATA task and the associated support code. The microcode itself comes directly from the offical API (v2.2) Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] bestcomm: core bestcomm support for Freescale MPC5200Sylvain Munaut2007-10-169-0/+1312
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds support for the core of the BestComm API for the Freescale MPC5200(b). The BestComm engine is a microcode-controlled / tasks-based DMA used by several of the onchip devices. Setting up the tasks / memory allocation and all common low level functions are handled by this patch. The specifics details of each tasks and their microcode are split-out in separate patches. This is not the official API, but a much cleaner one. (hopefully) Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] mpc52xx: Update mpc52xx_psc structure with B revision changesSylvain Munaut2007-10-161-2/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On the mpc5200b the ccr register is 32 bits wide while on the mpc5200 it's only 16 bits. It's up to the driver to use the correct format depending on the chip it's running on. The 5200b also offers some more registers & status in AC97 mode. Again, if not running on a 5200b the driver should not use those. Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] rheap: Changes config mechanismSylvain Munaut2007-10-165-4/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of having in the makefile all the option that requires rheap, we define a configuration symbol and when needed we make sure it's selected. Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| | * | [POWERPC] exports rheap symbol to modulesSylvain Munaut2007-10-161-0/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Theses can be useful in modules too. So we export them. Signed-off-by: Sylvain Munaut <tnt@246tNt.com> Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
| * | | Merge branch 'fixes-2.6.24' of ↵Paul Mackerras2007-10-172-5/+10
| |\ \ \ | | | | | | | | | | | | | | | master.kernel.org:/pub/scm/linux/kernel/git/galak/powerpc into merge
| | * | | [POWERPC] Add missing semicolon for fsl_pci.cTony Li2007-10-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Tony Li <tony.li@freescale.com> Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
| | * | | [POWERPC] Fix handling of stfiwx math emulationKumar Gala2007-10-161-4/+9
| | |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Its legal for the stfiwx instruction to have RA = 0 as part of its effective address calculation. This is illegal for all other XE form instructions. Add code to compute the proper effective address for stfiwx if RA = 0 rather than treating it as illegal. Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
| * | | [POWERPC] Fix vmemmap warning in init_64.cTony Breeds2007-10-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Use the right printk format to silence the following warning. CC arch/powerpc/mm/init_64.o arch/powerpc/mm/init_64.c: In function 'vmemmap_populate': arch/powerpc/mm/init_64.c:243: warning: format '%p' expects type 'void *', but argument 4 has type 'long unsigned int' Signed-off-by: Tony Breeds <tony@bakeyournoodle.com> Signed-off-by: Paul Mackerras <paulus@samba.org>
| * | | [POWERPC] Fix 64 bits vDSO DWARF info for CR registerBenjamin Herrenschmidt2007-10-171-4/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current DWARF info for CR are incorrect, causing the gcc unwinder to go to lunch if we take a segfault in the vdso. This fixes it. Problem identified by Andrew Haley, and fix provided by Jakub Jelinek (thanks !). Unfortunately, a bug in gcc cause it to not quite work either, but that is being fixed separately with something around the lines of: linux-unwind.h: fs->regs.reg[R_CR2].loc.offset = (long) &regs->ccr - new_cfa; + /* CR? regs are just 32-bit and PPC is big-endian. */ + fs->regs.reg[R_CR2].loc.offset += sizeof (long) - 4; (According to Jakub) Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
| * | | [POWERPC] Add 1TB workaround for PA6TOlof Johansson2007-10-174-2/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | PA6T has a bug where the slbie instruction does not honor the large segment bit. As a result, we have to always use slbia when switching context. We don't have to worry about changing the slbie's during fault processing, since they should never be replacing one VSID with another using the same ESID. I.e. there's no risk for inserting duplicate entries due to a failed slbie of the old entry. So as long as we clear it out on context switch we should be fine. Signed-off-by: Olof Johansson <olof@lixom.net> Signed-off-by: Paul Mackerras <paulus@samba.org>
| * | | [POWERPC] Enable NO_HZ and high res timers for pseries and ppc64 configsAnton Blanchard2007-10-172-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Enable NO_HZ and high res timers for the ppc64 and pseries defconfigs. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
| * | | [POWERPC] Quieten cache information at bootAnton Blanchard2007-10-171-5/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | After 6 years the ppc64 kernel still thinks its important to tell me my cache line size is 0x80 bytes. I think most people who care know that by now. The rest probably cant even understand the hex output. Since we might have misconfigured firmware or cpus that have a linesize that isnt 128 bytes, I still print it out for those cases. If people would prefer to remove it completely, lets do it. Also for lpar remove the htab_address printout since its not used. Anton ppc64 boot log usability expert Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
| * | | [POWERPC] Quieten clockevent printkAnton Blanchard2007-10-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | The clockevent bootup message only needs to be KERN_INFO. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
| * | | [POWERPC] Enable SLUB in *_defconfigAnton Blanchard2007-10-1738-106/+143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When checking out the new NO_HZ support in powerpc, I noticed we never slept for more than 2 seconds. It turns out SLAB has a 2 second per cpu timer that causes this. After switching to SLUB I see some nice 4 second sleeps which is the limit on this POWER6 box (the decrementer ticks at 512MHz): slept 4.19 sec slept 4.19 sec slept 4.19 sec slept 4.19 sec slept 3.96 sec slept 3.80 sec slept 2.99 sec Since SLUB is now the default and some powerpc defconfigs already enable it, lets enable SLUB across the board for consistency. While doing this I also noticed that the maple defconfig has SLAB debugging enabled which is sure to make your box nice and slow. Fix that too. Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Paul Mackerras <paulus@samba.org>
generated by cgit v1.1 at 2025-02-19 13:03:29 +0000