summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* s390/lowcore: replace lowcore irb array with a per-cpu variableMartin Schwidefsky2014-05-288-19/+16
| | | | | | | | | Remove the 96-byte irb array from the lowcore and create a per-cpu variable instead. That way we will pick up any change in the definition of the struct irb automatically. Acked-By: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/lowcore: reserve 96 bytes for IRB in lowcoreChristian Borntraeger2014-05-281-5/+6
| | | | | | | | | | | | | | | | | | | | | The IRB might be 96 bytes if the extended-I/O-measurement facility is used. This feature is currently not used by Linux, but struct irb already has the emw defined. So let's make the irb in lowcore match the size of the internal data structure to be future proof. We also have to add a pad, to correctly align the paste. The bigger irb field also circumvents a bug in some QEMU versions that always write the emw field on test subchannel and therefore destroy the paste definitions of this CPU. Running under these QEMU version broke some timing functions in the VDSO and all users of these functions, e.g. some JREs. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Cornelia Huck <cornelia.huck@de.ibm.com> Cc: stable@vger.kernel.org
* s390/facilities: remove extract-cpu-time facility checkHeiko Carstens2014-05-281-4/+4
| | | | | | | | | | | | | Remove the check for the extract-cpu-time facility within early startup code. Both kernel and user space work if the facility is not installed. The vdso code has a run time check if the ectg is available. Besides that there is no known user. Reported-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390: require mvcos facility for z10 and newer machinesHeiko Carstens2014-05-271-3/+3
| | | | | | | | | With inlined uaccess functions we always need the mvcos facility. Checking at each inline place if mvcos is available would make the inlining of get_user/put_user pointless. Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/boot: fix boot of compressed kernel built with gcc 4.9Martin Schwidefsky2014-05-271-1/+1
| | | | | | | | Add -fno-delete-null-pointer-checks to CFLAGS for the code in arch/s390/boot. Without the option a compressed kernel built with gcc 4.9 won't boot. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/cio: remove weird assignment during argument evaluationSebastian Ott2014-05-271-1/+1
| | | | | | | | | Get rid of a useless assignment during argument evaluation. No functional change. Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/time: cast tv_nsec to u64 prior to shift in update_vsyscallMartin Schwidefsky2014-05-271-1/+1
| | | | | | | | | | Analog to git commit 28b92e09e25bdc0ae864b22eacf195a74f861389 first cast tk->wall_to_monotonic.tv_nsec to u64 before doing the shift with tk->shift to avoid loosing relevant bits on a 32-bit kernel. Cc: stable@vger.kernel.org # 3.13+ Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/oprofile: make return of 0 explicitJulia Lawall2014-05-271-10/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Delete unnecessary local variable whose value is always 0 and that hides the fact that the result is always 0. A simplified version of the semantic patch that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @r exists@ local idexpression ret; expression e; position p; @@ -ret = 0; ... when != ret = e return - ret + 0 ; // </smpl> [heiko.carstens: turn prepare_cpu_buffers into a void returning function] Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/spinlock: refactor arch_spin_lock_wait[_flags]Martin Schwidefsky2014-05-201-34/+47
| | | | | | Reorder the spinlock wait code to make it more readable. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/rwlock: add missing local_irq_restore callsMartin Schwidefsky2014-05-201-0/+2
| | | | | | | | | | The out of line _raw_read_lock_wait_flags/_raw_write_lock_wait_flags functions for the arch_read_lock_flags/arch_write_lock_flags calls fail to re-enable the interrupts after another unsuccessful try to get the lock with compare-and-swap. The following wait would be done with interrupts disabled which is suboptimal. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/spinlock,rwlock: always to a load-and-test firstMartin Schwidefsky2014-05-202-33/+46
| | | | | | | | | | | | | | | | In case a lock is contended it is better to do a load-and-test first before trying to get the lock with compare-and-swap. This helps to avoid unnecessary cache invalidations of the cacheline for the lock if the CPU has to wait for the lock. For an uncontended lock doing the compare-and-swap directly is a bit better, if the CPU does not have the cacheline in its cache yet the compare-and-swap will get it read-write immediately while a load-and-test would get it read-only first. Always to the load-and-test first to avoid the cacheline invalidations for the contended case outweight the potential read-only to read-write cacheline upgrade for the uncontended case. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/cio: fix multiple structure definitionsSebastian Ott2014-05-207-26/+27
| | | | | | | | | | Fix multiple definitions of struct channel_path_desc by moving it to asm/chpid.h . Also change ccw_device_get_chp_desc to use proper types. Reviewed-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/spinlock: fix system hang with spin_retry <= 0Gerald Schaefer2014-05-201-6/+8
| | | | | | | | | | On LPAR, when spin_retry is set to <= 0, arch_spin_lock_wait() and arch_spin_lock_wait_flags() may end up in a while(1) loop w/o doing any compare and swap operation. To fix this, use do/while instead of for loop. Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/appldata: add slab.h for kzalloc/kfreeJeff Mahoney2014-05-201-0/+1
| | | | | | | | | | This fixes: arch/s390/appldata/appldata_mem.c:135:2: error: implicit declaration of function 'kzalloc' [-Werror=implicit-function-declaration] arch/s390/appldata/appldata_mem.c:141:3: error: implicit declaration of function 'kfree' [-Werror=implicit-function-declaration] Signed-off-by: Jeff Mahoney <jeffm@suse.com> Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/uaccess: provide inline variants of get_user/put_userHeiko Carstens2014-05-201-0/+30
| | | | | | | | | | | This shortens the code by ~17k (performace_defconfig, march=z196). The number of exception table entries however increases from 164 entries to 2500 entries (+~18k). However the executed code is shorter and also faster since we save the branches to the out-of-line copy_to/from_user implementations. Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/pci: add some new arch specific pci attributesSebastian Ott2014-05-204-3/+65
| | | | | | | | | Add a bunch of s390 specific pci attributes to help identifying pci functions. Reviewed-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/pci: use pdev->dev.groups for attribute creationSebastian Ott2014-05-203-40/+16
| | | | | | | | | | | Let the driver core handle attribute creation by putting all s390 specific pci attributes in an attribute group which is referenced by pdev->dev.groups in pcibios_add_device. Link: https://lkml.kernel.org/r/alpine.LFD.2.11.1404141101500.1529@denkbrett Reviewed-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/pci: use macro for attribute creationSebastian Ott2014-05-201-38/+17
| | | | | | | | | Introduce the zpci_attr macro to create read only sysfs attributes to avoid duplicate code. Reviewed-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/pci: improve state check when processing hotplug eventsSebastian Ott2014-05-201-2/+3
| | | | | | | | | | | | Processing pci hotplug events can fail when a pci function is in an unexpected state. This can happen when we already processed the change associated with the hotplug event (especially when receiving hotplug events during early boot). Just ignore the event in this case. Reviewed-by: Gerald Schaefer <gerald.schaefer@de.ibm.com> Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390: split TIF bits into CIF, PIF and TIF bitsMartin Schwidefsky2014-05-2016-126/+173
| | | | | | | | | | The oi and ni instructions used in entry[64].S to set and clear bits in the thread-flags are not guaranteed to be atomic in regard to other CPUs. Split the TIF bits into CPU, pt_regs and thread-info specific bits. Updates on the TIF bits are done with atomic instructions, updates on CPU and pt_regs bits are done with non-atomic instructions. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/uaccess: simplify control register updatesMartin Schwidefsky2014-05-208-43/+33
| | | | | | | | | | | | | Always switch to the kernel ASCE in switch_mm. Load the secondary space ASCE in finish_arch_post_lock_switch after checking that any pending page table operations have completed. The primary ASCE is loaded in entry[64].S. With this the update_primary_asce call can be removed from the switch_to macro and from the start of switch_mm function. Remove the load_primary argument from update_user_asce/clear_user_asce, rename update_user_asce to set_user_asce and rename update_primary_asce to load_kernel_asce. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/smp: Avoid busy loop after halt and "begin" on z/VMMichael Holzheu2014-05-202-12/+2
| | | | | | | | | | | Currently the smp_stop_cpu() function for SMP kernels enters a busy loop when "begin" is entered on the z/VM console after Linux is halted. To avoid this behavior, use the non-SMP variant of smp_stop_cpu() which stops the CPU again after "begin" is entered. As a side effect we now have consistent behavior for SMP and non-SMP Linux. Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390: fix new ccwgroup.h kernel-doc warningRandy Dunlap2014-05-201-0/+2
| | | | | | | | | | | | | | | | Fix new s390 kernel-doc warning: Warning(arch/s390/include/asm/ccwgroup.h:27): No description found for parameter 'ungroup_work' Signed-off-by: Randy Dunlap <rdunlap@infradead.org> Cc: Sebastian Ott <sebott@linux.vnet.ibm.com> Cc: Peter Oberparleiter <oberpar@linux.vnet.ibm.com> Cc: linux-s390@vger.kernel.org Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: linux390@de.ibm.com Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/zfcpdump: Update documentation in zfcpdump.txtMichael Holzheu2014-05-201-53/+18
| | | | | | | | | | | | | | Do the following changes: - Document new /proc/vmcore interface - Document partition dump external initramfs with s390-tools-1.24.0 - Remove initramfs config file because initramfs is now built automatically in s390-tools - Replace description of kernel config options with "make zfcpdump_defconfig" - Some editorial changes Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/dump: Remove CONFIG_ZFCPDUMPMichael Holzheu2014-05-205-20/+11
| | | | | | | | | | | Currently there are two s390 kernel dump config options "CONFIG_ZFCPDUMP" and "CONFIG_CRASH_DUMP". In order to keep things simple and because the "CONFIG_ZFCPDUMP" option already has a dependency to "CONFIG_CRASH_DUMP" remove the CONFIG_ZFCPDUMP option. Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com> Reviewed-by: Eric Farman <farman@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/cmma: remove "cmma disable" code in case of dump againHeiko Carstens2014-05-201-10/+0
| | | | | | | | This is not necessary anymore, since the offending code is gone with the conversion to the memblock code. Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>- Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/spinlock: optimize spinlock code sequencePhilipp Hachtmann2014-05-205-10/+21
| | | | | | | | | Use lowcore constant to improve the code generated for spinlocks. [ Martin Schwidefsky: patch breakdown and code beautification ] Signed-off-by: Philipp Hachtmann <phacht@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/spinlock: cleanup spinlock codePhilipp Hachtmann2014-05-203-77/+82
| | | | | | | | | | | | | | Improve the spinlock code in several aspects: - Have _raw_compare_and_swap return true if the operation has been successful instead of returning the old value. - Remove the "volatile" from arch_spinlock_t and arch_rwlock_t - Rename 'owner_cpu' to 'lock' - Add helper functions arch_spin_trylock_once / arch_spin_tryrelease_once [ Martin Schwidefsky: patch breakdown and code beautification ] Signed-off-by: Philipp Hachtmann <phacht@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* s390/mm: Convert bootmem to memblockPhilipp Hachtmann2014-05-2010-498/+270
| | | | | | | | | | | The original bootmem allocator is getting replaced by memblock. To cover the needs of the s390 kdump implementation the physical memory list is used. With this patch the bootmem allocator and its bitmaps are completely removed from s390. Signed-off-by: Philipp Hachtmann <phacht@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* mm/memblock: add physical memory listPhilipp Hachtmann2014-05-203-0/+19
| | | | | | | | | | | Add the physmem list to the memblock structure. This list only exists if HAVE_MEMBLOCK_PHYS_MAP is selected and contains the unmodified list of physically available memory. It differs from the memblock memory list as it always contains all memory ranges even if the memory has been restricted, e.g. by use of the mem= kernel parameter. Signed-off-by: Philipp Hachtmann <phacht@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* mm/memblock: Do some refactoring, enhance APIPhilipp Hachtmann2014-05-202-85/+183
| | | | | | | | | | | | | | | | | | | | | Refactor the memblock code and extend the memblock API to make it more flexible. With the extended API it is simple to define and work with additional memory lists. The static functions memblock_add_region and __memblock_remove are renamed to memblock_add_range and meblock_remove_range and added to the memblock API. The __next_free_mem_range and __next_free_mem_range_rev functions are replaced with calls to the more generic list walkers __next_mem_range and __next_mem_range_rev. To walk an arbitrary memory list two new macros for_each_mem_range and for_each_mem_range_rev are added. These new macros are used to define for_each_free_mem_range and for_each_free_mem_range_reverse. Signed-off-by: Philipp Hachtmann <phacht@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* Merge branch 'parisc-3.15-4' of ↵Linus Torvalds2014-05-204-46/+65
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux Pull parisc fixes from Helge Deller: "There are two patches in here: The first patch greatly improves latency and corrects the memory ordering in our light-weight atomic locking syscall. The second patch ratelimits printing of userspace segfaults in the same way as it's done on other platforms. This fixes a possible DOS on parisc since it prevents the syslog to grow too fast. For example, when the debian acl2 package was built on our debian buildd servers, this package produced lots of gigabytes in syslog in very short time and thus filled our harddisks, which then turned the server nearly completely unaccessible and unresponsive" * 'parisc-3.15-4' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux: parisc: Improve LWS-CAS performance parisc: ratelimit userspace segfault printing
| * parisc: Improve LWS-CAS performanceJohn David Anglin2014-05-151-3/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The attached change significantly improves the performance of the LWS-CAS code in syscall.S. This allows a number of packages to build (e.g., zeromq3, gtest and libxs) that previously failed because slow LWS-CAS performance under contention. In particular, interrupts taken while the lock was taken degraded performance significantly. The change does the following: 1) Disables interrupts around the CAS operation, and 2) Changes the loads and stores to use the ordered completer, "o", on PA 2.0. "o" and "ma" with a zero offset are equivalent. The latter is accepted on both PA 1.X and 2.0. The use of ordered loads and stores probably makes no difference on all existing hardware, but it seemed pedantically correct. In particular, the CAS operation must complete before LDCW lock is released. As written before, a processor could reorder the operations. I don't believe the period interrupts are disabled is long enough to significantly increase interrupt latency. For example, the TLB insert code is longer. Worst case is a memory fault in the CAS operation. Signed-off-by: John David Anglin <dave.anglin@bell.net> Cc: stable@vger.kernel.org # 3.13+ Signed-off-by: Helge Deller <deller@gmx.de>
| * parisc: ratelimit userspace segfault printingHelge Deller2014-05-153-43/+56
| | | | | | | | | | | | | | | | | | | | Ratelimit printing of userspace segfaults and make it runtime configurable via the /proc/sys/debug/exception-trace variable. This should resolve syslog from growing way too fast and thus prevents possible system service attacks. Signed-off-by: Helge Deller <deller@gmx.de> Cc: stable@vger.kernel.org # 3.13+
* | Merge tag 'arm64-fixes' of ↵Linus Torvalds2014-05-202-3/+11
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux Pull two arm64 fixes from Catalin Marinas: - arm64 migrate_irqs() fix following commit ffde1de64012 (irqchip: Gic: Support forced affinity setting) - fix arm64 pud_huge() to return 0 when only 2 levels page tables are used (__PAGETABLE_PMD_FOLDED defined and pmd_huge already covers block entries at the first level), otherwise KVM gets confused * tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: arm64: fix pud_huge() for 2-level pagetables arm64: use cpu_online_mask when using forced irq_set_affinity
| * | arm64: fix pud_huge() for 2-level pagetablesMark Salter2014-05-161-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The following happens when trying to run a kvm guest on a kernel configured for 64k pages. This doesn't happen with 4k pages: BUG: failure at include/linux/mm.h:297/put_page_testzero()! Kernel panic - not syncing: BUG! CPU: 2 PID: 4228 Comm: qemu-system-aar Tainted: GF 3.13.0-0.rc7.31.sa2.k32v1.aarch64.debug #1 Call trace: [<fffffe0000096034>] dump_backtrace+0x0/0x16c [<fffffe00000961b4>] show_stack+0x14/0x1c [<fffffe000066e648>] dump_stack+0x84/0xb0 [<fffffe0000668678>] panic+0xf4/0x220 [<fffffe000018ec78>] free_reserved_area+0x0/0x110 [<fffffe000018edd8>] free_pages+0x50/0x88 [<fffffe00000a759c>] kvm_free_stage2_pgd+0x30/0x40 [<fffffe00000a5354>] kvm_arch_destroy_vm+0x18/0x44 [<fffffe00000a1854>] kvm_put_kvm+0xf0/0x184 [<fffffe00000a1938>] kvm_vm_release+0x10/0x1c [<fffffe00001edc1c>] __fput+0xb0/0x288 [<fffffe00001ede4c>] ____fput+0xc/0x14 [<fffffe00000d5a2c>] task_work_run+0xa8/0x11c [<fffffe0000095c14>] do_notify_resume+0x54/0x58 In arch/arm/kvm/mmu.c:unmap_range(), we end up doing an extra put_page() on the stage2 pgd which leads to the BUG in put_page_testzero(). This happens because a pud_huge() test in unmap_range() returns true when it should always be false with 2-level pages tables used by 64k pages. This patch removes support for huge puds if 2-level pagetables are being used. Signed-off-by: Mark Salter <msalter@redhat.com> [catalin.marinas@arm.com: removed #ifndef around PUD_SIZE check] Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Cc: <stable@vger.kernel.org> # v3.11+
| * | arm64: use cpu_online_mask when using forced irq_set_affinitySudeep Holla2014-05-121-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 01f8fa4f01d8("genirq: Allow forcing cpu affinity of interrupts") enabled the forced irq_set_affinity which previously refused to route an interrupt to an offline cpu. Commit ffde1de64012("irqchip: Gic: Support forced affinity setting") implements this force logic and disables the cpu online check for GIC interrupt controller. When __cpu_disable calls migrate_irqs, it disables the current cpu in cpu_online_mask and uses forced irq_set_affinity to migrate the IRQs away from the cpu but passes affinity mask with the cpu being offlined also included in it. When calling irq_set_affinity with force == true in a cpu hotplug path, the caller must ensure that the cpu being offlined is not present in the affinity mask or it may be selected as the target CPU, leading to the interrupt not being migrated. This patch uses cpu_online_mask when using forced irq_set_affinity so that the IRQs are properly migrated away. Signed-off-by: Sudeep Holla <sudeep.holla@arm.com> Acked-by: Mark Rutland <mark.rutland@arm.com> Cc: Will Deacon <will.deacon@arm.com> Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
* | | Merge tag 'metag-for-v3.15-2' of ↵Linus Torvalds2014-05-2010-22/+33
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag Pull Metag architecture and related fixes from James Hogan: "Mostly fixes for metag and parisc relating to upgrowing stacks. - Fix missing compiler barriers in metag memory barriers. - Fix BUG_ON on metag when RLIMIT_STACK hard limit is increased beyond safe value. - Make maximum stack size configurable. This reduces the default user stack size back to 80MB (especially on parisc after their removal of _STK_LIM_MAX override). This only affects metag and parisc. - Remove metag _STK_LIM_MAX override to match other arches and follow parisc, now that it is safe to do so (due to the BUG_ON fix mentioned above). - Finally now that both metag and parisc _STK_LIM_MAX overrides have been removed, it makes sense to remove _STK_LIM_MAX altogether" * tag 'metag-for-v3.15-2' of git://git.kernel.org/pub/scm/linux/kernel/git/jhogan/metag: asm-generic: remove _STK_LIM_MAX metag: Remove _STK_LIM_MAX override parisc,metag: Do not hardcode maximum userspace stack size metag: Reduce maximum stack size to 256MB metag: fix memory barriers
| * | | asm-generic: remove _STK_LIM_MAXJames Hogan2014-05-152-8/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | _STK_LIM_MAX could be used to override the RLIMIT_STACK hard limit from an arch's include/uapi/asm-generic/resource.h file, but is no longer used since both parisc and metag removed the override. Therefore remove it entirely, setting the hard RLIMIT_STACK limit to RLIM_INFINITY directly in include/asm-generic/resource.h. Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: linux-arch@vger.kernel.org Cc: Helge Deller <deller@gmx.de> Cc: John David Anglin <dave.anglin@bell.net>
| * | | metag: Remove _STK_LIM_MAX overrideJames Hogan2014-05-152-8/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Meta overrode _STK_LIM_MAX (the default RLIMIT_STACK hard limit) to 256MB, apparently in an attempt to prevent setup_arg_pages's STACK_GROWSUP code from choosing the maximum stack size of 1GB, which is far too large for Meta's limited virtual address space and hits a BUG_ON (stack_top is usually 0x3ffff000). However the commit "metag: Reduce maximum stack size to 256MB" reduces the absolute stack size limit to a safe value for metag. This allows the default _STK_LIM_MAX override to be removed, bringing the default behaviour in line with all other architectures. Parisc in particular recently removed their override of _STK_LIMT_MAX in commit e0d8898d76a7 (parisc: remove _STK_LIM_MAX override) since it subtly affects stack allocation semantics in userland. Meta's uapi/asm/resource.h can now be removed and switch to using generic-y. Suggested-by: Helge Deller <deller@gmx.de> Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: linux-metag@vger.kernel.org Cc: John David Anglin <dave.anglin@bell.net>
| * | | parisc,metag: Do not hardcode maximum userspace stack sizeHelge Deller2014-05-154-5/+23
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch affects only architectures where the stack grows upwards (currently parisc and metag only). On those do not hardcode the maximum initial stack size to 1GB for 32-bit processes, but make it configurable via a config option. The main problem with the hardcoded stack size is, that we have two memory regions which grow upwards: stack and heap. To keep most of the memory available for heap in a flexmap memory layout, it makes no sense to hard allocate up to 1GB of the memory for stack which can't be used as heap then. This patch makes the stack size for 32-bit processes configurable and uses 80MB as default value which has been in use during the last few years on parisc and which hasn't showed any problems yet. Signed-off-by: Helge Deller <deller@gmx.de> Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: linux-parisc@vger.kernel.org Cc: linux-metag@vger.kernel.org Cc: John David Anglin <dave.anglin@bell.net>
| * | | metag: Reduce maximum stack size to 256MBJames Hogan2014-05-153-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Specify the maximum stack size for arches where the stack grows upward (parisc and metag) in asm/processor.h rather than hard coding in fs/exec.c so that metag can specify a smaller value of 256MB rather than 1GB. This fixes a BUG on metag if the RLIMIT_STACK hard limit is increased beyond a safe value by root. E.g. when starting a process after running "ulimit -H -s unlimited" it will then attempt to use a stack size of the maximum 1GB which is far too big for metag's limited user virtual address space (stack_top is usually 0x3ffff000): BUG: failure at fs/exec.c:589/shift_arg_pages()! Signed-off-by: James Hogan <james.hogan@imgtec.com> Cc: Helge Deller <deller@gmx.de> Cc: "James E.J. Bottomley" <jejb@parisc-linux.org> Cc: linux-parisc@vger.kernel.org Cc: linux-metag@vger.kernel.org Cc: John David Anglin <dave.anglin@bell.net> Cc: stable@vger.kernel.org # only needed for >= v3.9 (arch/metag)
| * | | metag: fix memory barriersMikulas Patocka2014-05-151-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Volatile access doesn't really imply the compiler barrier. Volatile access is only ordered with respect to other volatile accesses, it isn't ordered with respect to general memory accesses. Gcc may reorder memory accesses around volatile access, as we can see in this simple example (if we compile it with optimization, both increments of *b will be collapsed to just one): void fn(volatile int *a, long *b) { (*b)++; *a = 10; (*b)++; } Consequently, we need the compiler barrier after a write to the volatile variable, to make sure that the compiler doesn't reorder the volatile write with something else. Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Cc: stable@vger.kernel.org Acked-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: James Hogan <james.hogan@imgtec.com>
* | | | Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linuxLinus Torvalds2014-05-207-29/+141
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull drm/intel fixes from Dave Airlie: "Just some intel fixes. I have some radeon ones but I need to get some patches dropped from the pull req" * 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: drm/i915: Increase WM memory latency values on SNB drm/i915: restore backlight precision when converting from ACPI drm/i915: Use the first mode if there is no preferred mode in the EDID drm/i915/dp: force eDP lane count to max available lanes on BDW drm/i915/vlv: reset VLV media force wake request register drm/i915/SDVO: For sysfs link put directory and target in correct order drm/i915: use lane count and link rate from VBT as minimums for eDP drm/i915: clean up VBT eDP link param decoding drm/i915: consider the source max DP lane count too
| * \ \ \ Merge tag 'drm-intel-fixes-2014-05-16' of ↵Dave Airlie2014-05-207-29/+141
| |\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://anongit.freedesktop.org/drm-intel into drm-fixes Intel fixes for regressions, black screens and hangs, for 3.15. * tag 'drm-intel-fixes-2014-05-16' of git://anongit.freedesktop.org/drm-intel: drm/i915: Increase WM memory latency values on SNB drm/i915: restore backlight precision when converting from ACPI drm/i915: Use the first mode if there is no preferred mode in the EDID drm/i915/dp: force eDP lane count to max available lanes on BDW drm/i915/vlv: reset VLV media force wake request register drm/i915/SDVO: For sysfs link put directory and target in correct order drm/i915: use lane count and link rate from VBT as minimums for eDP drm/i915: clean up VBT eDP link param decoding drm/i915: consider the source max DP lane count too
| | * | | | drm/i915: Increase WM memory latency values on SNBVille Syrjälä2014-05-151-0/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On SNB the BIOS provided WM memory latency values seem insufficient to handle high resolution displays. In this particular case the display mode was a 2560x1440@60Hz, which makes the pixel clock 241.5 MHz. It was empirically found that a memory latency value if 1.2 usec is enough to avoid underruns, whereas the BIOS provided value of 0.7 usec was clearly too low. Incidentally 1.2 usec is what the typical BIOS provided values are on IVB systems. Increase the WM memory latency values to at least 1.2 usec on SNB. Hopefully this won't have a significant effect on power consumption. v2: Increase the latency values regardless of the pixel clock Cc: Robert N <crshman@gmail.com> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70254 Tested-by: Robert Navarro <crshman@gmail.com> Tested-by: Vitaly Minko <vitaly.minko@gmail.com> Cc: stable@vger.kernel.org Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
| | * | | | drm/i915: restore backlight precision when converting from ACPIAaron Lu2014-05-151-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we set backlight on behalf of ACPI opregion, we will convert the backlight value in the 0-255 range defined in opregion to the actual hardware level. Commit 22505b82a2 (drm/i915: avoid brightness overflow when doing scale) is meant to fix the overflow problem when doing the conversion, but it also caused a problem that the converted hardware level doesn't quite represent the intended value: say user wants maximum backlight level(255 in opregion's range), then we will calculate the actual hardware level to be: level = freq / max * level, where freq is the hardware's max backlight level(937 on an user's box), and max and level are all 255. The converted value should be 937 but the above calculation will yield 765. To fix this issue, just use 64 bits to do the calculation to keep the precision and avoid overflow at the same time. Buglink: https://bugzilla.kernel.org/show_bug.cgi?id=72491 Reported-by: Nico Schottelius <nico-bugzilla.kernel.org@schottelius.org> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: stable@vger.kernel.org Signed-off-by: Aaron Lu <aaron.lu@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
| | * | | | drm/i915: Use the first mode if there is no preferred mode in the EDIDChris Wilson2014-05-151-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This matches the algorithm used by earlier kernels when selecting the mode for the fbcon. And only if there is no modes at all, do we fall back to using the BIOS configuration. Seamless transition is still preserved (from the BIOS configuration to ours) so long as the BIOS has also chosen what we hope is the native configuration. Reported-by: Knut Petersen <Knut_Petersen@t-online.de> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=78655 Reviewed-by: Jesse Barnes <jbarnes@virtuousgeek.org> Tested-by: Knut Petersen <Knut_Petersen@t-online.de> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> [Jani: applied Chris' "Please imagine that I wrote this correctly."] Signed-off-by: Jani Nikula <jani.nikula@intel.com>
| | * | | | drm/i915/dp: force eDP lane count to max available lanes on BDWJani Nikula2014-05-151-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are certain BDW high res eDP machines that regressed due to commit 38aecea0ccbb909d635619cba22f1891e589b434 Author: Daniel Vetter <daniel.vetter@ffwll.ch> Date: Mon Mar 3 11:18:10 2014 +0100 drm/i915: reverse dp link param selection, prefer fast over wide again The commit lead to 2 lanes at 5.4 Gbps being used instead of 4 lanes at 2.7 Gbps on the affected machines. Link training succeeded for both, but the screen remained blank with the former config. Further investigation showed that 4 lanes at 5.4 Gbps worked also. The root cause for the blank screen using 2 lanes remains unknown, but apparently the driver for a certain other operating system by default uses the max available lanes. Follow suit on Broadwell eDP, for at least until we figure out what is going on. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=76711 Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@gmail.com> Tested-by: Rodrigo Vivi <rodrigo.vivi@gmail.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com>
| | * | | | drm/i915/vlv: reset VLV media force wake request registerJani Nikula2014-05-091-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Media force wake get hangs the machine when the system is booted without displays attached. The assumption is that (at least some versions of) the firmware has skipped some initialization in that case. Empirical evidence suggests we need to reset the media force wake request register in addition to the render one to avoid hangs. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=75895 Reported-by: Imre Deak <imre.deak@intel.com> Reported-by: Darren Hart <dvhart@linux.intel.com> Tested-by: Darren Hart <dvhart@linux.intel.com> Reviewed-by: Mika Kuoppala <mika.kuoppala@intel.com> Cc: stable@vger.kernel.org Signed-off-by: Jani Nikula <jani.nikula@intel.com>
OpenPOWER on IntegriCloud