summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* KVM: Remove internal timer abstractionAvi Kivity2012-08-017-73/+67
| | | | | | | | | | | kvm_timer_fn(), the sole inhabitant of timer.c, is only used by lapic.c. Move it there to make it easier to hack on it. struct kvm_timer is a thin wrapper around hrtimer, and only adds obfuscation. Move near its two users (with different names) to prepare for simplification. Signed-off-by: Avi Kivity <avi@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
* KVM: Don't update PPR on any APIC readAvi Kivity2012-07-301-2/+4
| | | | | | | | | | | | The current code will update the PPR on almost any APIC read; however that's only required if we read the PPR. kvm_update_ppr() shows up in some profiles, albeit with a low usage (~1%). This should reduce it further (it will still be called during interrupt processing). Signed-off-by: Avi Kivity <avi@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
* KVM: s390: Add implementation-specific trace eventsCornelia Huck2012-07-265-0/+241
| | | | | | | | | | | | | | | Introduce a new trace system, kvm-s390, for some kvm/s390 specific trace points: - injection of interrupts - delivery of interrupts to the guest - creation/destruction of kvm machines and vcpus - stop actions for vcpus - reset requests for userspace Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: s390: Add architectural trace eventsCornelia Huck2012-07-266-1/+368
| | | | | | | | | | | | Add trace events for several s390 architecture specifics: - SIE entry/exit - common intercepts - common instructions (sigp/diagnose) Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* s390/dis: Instruction decoding interfaceCornelia Huck2012-07-262-0/+28
| | | | | | | | | | Provide a new function, insn_to_mnemonic, by which e.g. kvm can obtain a human-readable decoding of an instruction's opcode. Reviewed-by: Christian Borntraeger <borntraeger@de.ibm.com> Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com> Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: Move KVM_IRQ_LINE to arch-generic codeChristoffer Dall2012-07-264-46/+44
| | | | | | | | | | Handle KVM_IRQ_LINE and KVM_IRQ_LINE_STATUS in the generic kvm_vm_ioctl() function and call into kvm_vm_ioctl_irq_line(). This is even more relevant when KVM/ARM also uses this ioctl. Signed-off-by: Christoffer Dall <c.dall@virtualopensystems.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: x86 emulator: drop unneeded call to get_segment()Gleb Natapov2012-07-261-6/+3
| | | | | | | | | | | setup_syscalls_segments() calls get_segment() and than overwrites all but one of the structure fields and this one should also be overwritten anyway, so we can drop call to get_segment() and avoid a couple of vmreads on vmx. Also drop zeroing ss/cs structures since most of the fields are set anyway. Just set those that were not set explicitly. Signed-off-by: Gleb Natapov <gleb@redhat.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: x86 emulator: simplify read_emulatedXiao Guangrong2012-07-261-17/+14
| | | | | | | | | | | | No need split mmio read region into 8-bits pieces since we do it in emulator_read_write_onepage Changelog: Add a WARN_ON to check read-cache overflow Acked-by: Marcelo Tosatti <mtosatti@redhat.com> Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: remove dummy pagesXiao Guangrong2012-07-263-75/+52
| | | | | | | | | | | Currently, kvm allocates some pages and use them as error indicators, it wastes memory and is not good for scalability Base on Avi's suggestion, we use the error codes instead of these pages to indicate the error conditions Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: use kvm_release_page_clean to release the pageXiao Guangrong2012-07-261-2/+2
| | | | | | | | | | | In kvm_async_pf_wakeup_all, it uses bad_page to generate broadcast wakeup, and uses put_page to release bad_page, the work depends on the fact that bad_page is the normal page. But we will use the error code instead of bad_page, so use kvm_release_page_clean to release the page which will release the error code properly Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* KVM: MMU: use kvm_release_pfn_clean to release pfnXiao Guangrong2012-07-261-1/+1
| | | | | | | | | | The current code depends on the fact that fault_page is the normal page, however, we will use the error code instead of these dummy pages in the later patch, so we use kvm_release_pfn_clean to release pfn which will release the error code properly Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Avi Kivity <avi@redhat.com>
* Merge branch 'queue' into nextAvi Kivity2012-07-2620-115/+277
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Merge patches queued during the run-up to the merge window. * queue: (25 commits) KVM: Choose better candidate for directed yield KVM: Note down when cpu relax intercepted or pause loop exited KVM: Add config to support ple or cpu relax optimzation KVM: switch to symbolic name for irq_states size KVM: x86: Fix typos in pmu.c KVM: x86: Fix typos in lapic.c KVM: x86: Fix typos in cpuid.c KVM: x86: Fix typos in emulate.c KVM: x86: Fix typos in x86.c KVM: SVM: Fix typos KVM: VMX: Fix typos KVM: remove the unused parameter of gfn_to_pfn_memslot KVM: remove is_error_hpa KVM: make bad_pfn static to kvm_main.c KVM: using get_fault_pfn to get the fault pfn KVM: MMU: track the refcount when unmap the page KVM: x86: remove unnecessary mark_page_dirty KVM: MMU: Avoid handling same rmap_pde in kvm_handle_hva_range() KVM: MMU: Push trace_kvm_age_page() into kvm_age_rmapp() KVM: MMU: Add memslot parameter to hva handlers ... Signed-off-by: Avi Kivity <avi@redhat.com>
| * KVM: Choose better candidate for directed yieldRaghavendra K T2012-07-232-0/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, on a large vcpu guests, there is a high probability of yielding to the same vcpu who had recently done a pause-loop exit or cpu relax intercepted. Such a yield can lead to the vcpu spinning again and hence degrade the performance. The patchset keeps track of the pause loop exit/cpu relax interception and gives chance to a vcpu which: (a) Has not done pause loop exit or cpu relax intercepted at all (probably he is preempted lock-holder) (b) Was skipped in last iteration because it did pause loop exit or cpu relax intercepted, and probably has become eligible now (next eligible lock holder) Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com> Reviewed-by: Rik van Riel <riel@redhat.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> # on s390x Signed-off-by: Avi Kivity <avi@redhat.com>
| * KVM: Note down when cpu relax intercepted or pause loop exitedRaghavendra K T2012-07-232-0/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | Noting pause loop exited vcpu or cpu relax intercepted helps in filtering right candidate to yield. Wrong selection of vcpu; i.e., a vcpu that just did a pl-exit or cpu relax intercepted may contribute to performance degradation. Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com> Reviewed-by: Rik van Riel <riel@redhat.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> # on s390x Signed-off-by: Avi Kivity <avi@redhat.com>
| * KVM: Add config to support ple or cpu relax optimzationRaghavendra K T2012-07-233-0/+5
| | | | | | | | | | | | | | | | | | Suggested-by: Avi Kivity <avi@redhat.com> Signed-off-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> Reviewed-by: Marcelo Tosatti <mtosatti@redhat.com> Reviewed-by: Rik van Riel <riel@redhat.com> Tested-by: Christian Borntraeger <borntraeger@de.ibm.com> # on s390x Signed-off-by: Avi Kivity <avi@redhat.com>
| * KVM: switch to symbolic name for irq_states sizeMichael S. Tsirkin2012-07-202-3/+3
| | | | | | | | | | | | | | Use PIC_NUM_PINS instead of hard-coded 16 for pic pins. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: x86: Fix typos in pmu.cGuo Chao2012-07-201-1/+1
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: x86: Fix typos in lapic.cGuo Chao2012-07-201-2/+2
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: x86: Fix typos in cpuid.cGuo Chao2012-07-201-1/+1
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: x86: Fix typos in emulate.cGuo Chao2012-07-201-5/+5
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: x86: Fix typos in x86.cGuo Chao2012-07-201-7/+7
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: SVM: Fix typosGuo Chao2012-07-201-2/+2
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: VMX: Fix typosGuo Chao2012-07-201-3/+3
| | | | | | | | | | Signed-off-by: Guo Chao <yan@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: remove the unused parameter of gfn_to_pfn_memslotXiao Guangrong2012-07-195-18/+16
| | | | | | | | | | | | | | | | The parameter, 'kvm', is not used in gfn_to_pfn_memslot, we can happily remove it Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: remove is_error_hpaXiao Guangrong2012-07-191-4/+0
| | | | | | | | | | | | | | Remove them since they are not used anymore Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: make bad_pfn static to kvm_main.cXiao Guangrong2012-07-192-4/+3
| | | | | | | | | | | | | | | | bad_pfn is not used out of kvm_main.c, so mark it static, also move it near hwpoison_pfn and fault_pfn Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: using get_fault_pfn to get the fault pfnXiao Guangrong2012-07-193-17/+7
| | | | | | | | | | | | | | Using get_fault_pfn to cleanup the code Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: MMU: track the refcount when unmap the pageXiao Guangrong2012-07-191-0/+8
| | | | | | | | | | | | | | | | It will trigger a WARN_ON if the page has been freed but it is still used in mmu, it can help us to detect mm bug early Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: x86: remove unnecessary mark_page_dirtyXiao Guangrong2012-07-191-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | fix: [ 132.474633] 3.5.0-rc1+ #50 Not tainted [ 132.474634] ------------------------------- [ 132.474635] include/linux/kvm_host.h:369 suspicious rcu_dereference_check() usage! [ 132.474636] [ 132.474636] other info that might help us debug this: [ 132.474636] [ 132.474638] [ 132.474638] rcu_scheduler_active = 1, debug_locks = 1 [ 132.474640] 1 lock held by qemu-kvm/2832: [ 132.474657] #0: (&vcpu->mutex){+.+.+.}, at: [<ffffffffa01e1636>] vcpu_load+0x1e/0x91 [kvm] [ 132.474658] [ 132.474658] stack backtrace: [ 132.474660] Pid: 2832, comm: qemu-kvm Not tainted 3.5.0-rc1+ #50 [ 132.474661] Call Trace: [ 132.474665] [<ffffffff81092f40>] lockdep_rcu_suspicious+0xfc/0x105 [ 132.474675] [<ffffffffa01e0c85>] kvm_memslots+0x6d/0x75 [kvm] [ 132.474683] [<ffffffffa01e0ca1>] gfn_to_memslot+0x14/0x4c [kvm] [ 132.474693] [<ffffffffa01e3575>] mark_page_dirty+0x17/0x2a [kvm] [ 132.474706] [<ffffffffa01f21ea>] kvm_arch_vcpu_ioctl+0xbcf/0xc07 [kvm] Actually, we do not write vcpu->arch.time at this time, mark_page_dirty should be removed. Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: MMU: Avoid handling same rmap_pde in kvm_handle_hva_range()Takuya Yoshikawa2012-07-181-10/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When we invalidate a THP page, we call the handler with the same rmap_pde argument 512 times in the following loop: for each guest page in the range for each level unmap using rmap This patch avoids these extra handler calls by changing the loop order like this: for each level for each rmap in the range unmap using rmap With the preceding patches in the patch series, this made THP page invalidation more than 5 times faster on our x86 host: the host became more responsive during swapping the guest's memory as a result. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: MMU: Push trace_kvm_age_page() into kvm_age_rmapp()Takuya Yoshikawa2012-07-181-13/+10
| | | | | | | | | | | | | | | | This restricts the tracing to page aging and makes it possible to optimize kvm_handle_hva_range() further in the following patch. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: MMU: Add memslot parameter to hva handlersTakuya Yoshikawa2012-07-181-7/+9
| | | | | | | | | | | | | | | | This is needed to push trace_kvm_age_page() into kvm_age_rmapp() in the following patch. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: Separate rmap_pde from kvm_lpage_info->write_countTakuya Yoshikawa2012-07-183-4/+15
| | | | | | | | | | | | | | | | | | This makes it possible to loop over rmap_pde arrays in the same way as we do over rmap so that we can optimize kvm_handle_hva_range() easily in the following patch. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: Introduce kvm_unmap_hva_range() for ↵Takuya Yoshikawa2012-07-185-2/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | kvm_mmu_notifier_invalidate_range_start() When we tested KVM under memory pressure, with THP enabled on the host, we noticed that MMU notifier took a long time to invalidate huge pages. Since the invalidation was done with mmu_lock held, it not only wasted the CPU but also made the host harder to respond. This patch mitigates this by using kvm_handle_hva_range(). Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Cc: Alexander Graf <agraf@suse.de> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: MMU: Make kvm_handle_hva() handle range of addressesTakuya Yoshikawa2012-07-182-18/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When guest's memory is backed by THP pages, MMU notifier needs to call kvm_unmap_hva(), which in turn leads to kvm_handle_hva(), in a loop to invalidate a range of pages which constitute one huge page: for each page for each memslot if page is in memslot unmap using rmap This means although every page in that range is expected to be found in the same memslot, we are forced to check unrelated memslots many times. If the guest has more memslots, the situation will become worse. Furthermore, if the range does not include any pages in the guest's memory, the loop over the pages will just consume extra time. This patch, together with the following patches, solves this problem by introducing kvm_handle_hva_range() which makes the loop look like this: for each memslot for each page in memslot unmap using rmap In this new processing, the actual work is converted to a loop over rmap which is much more cache friendly than before. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Cc: Alexander Graf <agraf@suse.de> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: Introduce hva_to_gfn_memslot() for kvm_handle_hva()Takuya Yoshikawa2012-07-183-5/+12
| | | | | | | | | | | | | | | | | | | | | | This restricts hva handling in mmu code and makes it easier to extend kvm_handle_hva() so that it can treat a range of addresses later in this patch series. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Cc: Alexander Graf <agraf@suse.de> Cc: Paul Mackerras <paulus@samba.org> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
| * KVM: MMU: Use __gfn_to_rmap() to clean up kvm_handle_hva()Takuya Yoshikawa2012-07-181-6/+6
| | | | | | | | | | | | | | We can treat every level uniformly. Signed-off-by: Takuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
* | Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsiLinus Torvalds2012-07-24114-1018/+2509
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull first round of SCSI updates from James Bottomley: "The most important feature of this patch set is the new async infrastructure that makes sure async_synchronize_full() synchronizes all domains and allows us to remove all the hacks (like having scsi_complete_async_scans() in the device base code) and means that the async infrastructure will "just work" in future. The rest is assorted driver updates (aacraid, bnx2fc, virto-scsi, megaraid, bfa, lpfc, qla2xxx, qla4xxx) plus a lot of infrastructure work in sas and FC. Signed-off-by: James Bottomley <JBottomley@Parallels.com>" * tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (97 commits) [SCSI] Revert "[SCSI] fix async probe regression" [SCSI] cleanup usages of scsi_complete_async_scans [SCSI] queue async scan work to an async_schedule domain [SCSI] async: make async_synchronize_full() flush all work regardless of domain [SCSI] async: introduce 'async_domain' type [SCSI] bfa: Fix to set correct return error codes and misc cleanup. [SCSI] aacraid: Series 7 Async. (performance) mode support [SCSI] aha152x: Allow use on 64bit systems [SCSI] virtio-scsi: Add vdrv->scan for post VIRTIO_CONFIG_S_DRIVER_OK LUN scanning [SCSI] bfa: squelch lockdep complaint with a spin_lock_init [SCSI] qla2xxx: remove unnecessary reads of PCI_CAP_ID_EXP [SCSI] qla4xxx: remove unnecessary read of PCI_CAP_ID_EXP [SCSI] ufs: fix incorrect return value about SUCCESS and FAILED [SCSI] ufs: reverse the ufshcd_is_device_present logic [SCSI] ufs: use module_pci_driver [SCSI] usb-storage: update usb devices for write cache quirk in quirk list. [SCSI] usb-storage: add support for write cache quirk [SCSI] set to WCE if usb cache quirk is present. [SCSI] virtio-scsi: hotplug support for virtio-scsi [SCSI] virtio-scsi: split scatterlist per target ...
| * | [SCSI] Revert "[SCSI] fix async probe regression"Dan Williams2012-07-201-5/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 43a8d39d0137612c336aa8bbb2cb886a79772ffb. Commit 43a8d39d fixed the fact that wait_for_device_probe() was unable to flush sd probe work. Now that sd probe work is once again flushable via wait_for_device_probe() this workaround is no longer needed. Signed-off-by: Dan Williams <dan.j.williams@intel.com> Tested-by: Eldad Zack <eldad@fogrefinery.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] cleanup usages of scsi_complete_async_scansDan Williams2012-07-203-25/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now that scsi registers its async scan work with the async subsystem, wait_for_device_probe() is sufficient for ensuring all scanning is complete. [jejb: fix merge problems with eea03c20ae38 Make wait_for_device_probe() also do scsi_complete_async_scans()] Signed-off-by: Dan Williams <dan.j.williams@intel.com> Tested-by: Eldad Zack <eldad@fogrefinery.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] queue async scan work to an async_schedule domainDan Williams2012-07-201-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is preparation to enable async_synchronize_full() to be used as a replacement for scsi_complete_async_scans(), i.e. to stop leaking scsi internal details where they are not needed. Signed-off-by: Dan Williams <dan.j.williams@intel.com> Tested-by: Eldad Zack <eldad@fogrefinery.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] async: make async_synchronize_full() flush all work regardless of domainDan Williams2012-07-203-2/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In response to an async related regression James noted: "My theory is that this is an init problem: The assumption in a lot of our code is that async_synchronize_full() waits for everything ... even the domain specific async schedules, which isn't true." ...so make this assumption true. Each domain, including the default one, registers itself on a global domain list when work is scheduled. Once all entries complete it exits that list. Waiting for the list to be empty syncs all in-flight work across all domains. Domains can opt-out of global syncing if they are declared as exclusive ASYNC_DOMAIN_EXCLUSIVE(). All stack-based domains have been declared exclusive since the domain may go out of scope as soon as the last work item completes. Statically declared domains are mostly ok, but async_unregister_domain() is there to close any theoretical races with pending async_synchronize_full waiters at module removal time. Signed-off-by: Dan Williams <dan.j.williams@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Reported-by: Meelis Roos <mroos@linux.ee> Reported-by: Eldad Zack <eldadzack@gmail.com> Tested-by: Eldad Zack <eldad@fogrefinery.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] async: introduce 'async_domain' typeDan Williams2012-07-207-27/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is in preparation for teaching async_synchronize_full() to sync all pending async work, and not just on the async_running domain. This conversion is functionally equivalent, just embedding the existing list in a new async_domain type. The .registered attribute is used in a later patch to distinguish between domains that want to be flushed by async_synchronize_full() versus those that only expect async_synchronize_{full|cookie}_domain to be used for flushing. [jejb: add async.h to scsi_priv.h for struct async_domain] Signed-off-by: Dan Williams <dan.j.williams@intel.com> Acked-by: Arjan van de Ven <arjan@linux.intel.com> Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Tested-by: Eldad Zack <eldad@fogrefinery.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] bfa: Fix to set correct return error codes and misc cleanup.Krishna Gudipati2012-07-202-4/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | - Remove unnecessary if NULL check in function bfa_fcs_vport_free(). - Set correct return error codes in case of memory allocation failure in the BSG ELS/CT passthru command handler. Signed-off-by: Krishna Gudipati <kgudipat@brocade.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] aacraid: Series 7 Async. (performance) mode supportMahesh Rajashekhara2012-07-208-125/+357
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Series 7 Async. (performance) mode support added - New scatter/gather list format for Series 7 - Driver converts s/g list to a firmware suitable list for best performance on Series 7, this can be disabled with driver parameter "aac_convert_sgl" for testing purposes - New container read/write command structure for Series 7 - Fast response support for the SCSI pass-through path added - Async. status response buffer changes Signed-off-by: Mahesh Rajashekhara <Mahesh_Rajashekhara@pmc-sierra.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] aha152x: Allow use on 64bit systemsAlan Cox2012-07-201-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | This is reported to work, known to work on PCMCIA and a code check shows no problems on the other bits of the code. Signed-off-by: Alan Cox <alan@linux.intel.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] virtio-scsi: Add vdrv->scan for post VIRTIO_CONFIG_S_DRIVER_OK LUN ↵Nicholas Bellinger2012-07-203-4/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | scanning This patch changes virtio-scsi to use a new virtio_driver->scan() callback so that scsi_scan_host() can be properly invoked once virtio_dev_probe() has set add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK) to signal active virtio-ring operation, instead of from within virtscsi_probe(). This fixes a bug where SCSI LUN scanning for both virtio-scsi-raw and virtio-scsi/tcm_vhost setups was happening before VIRTIO_CONFIG_S_DRIVER_OK had been set, causing VIRTIO_SCSI_S_BAD_TARGET to occur. This fixes a bug with virtio-scsi/tcm_vhost where LUN scan was not detecting LUNs. Tested with virtio-scsi-raw + virtio-scsi/tcm_vhost w/ IBLOCK on 3.5-rc2 code. Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org> Acked-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] bfa: squelch lockdep complaint with a spin_lock_initKyle McMartin2012-07-201-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Spotted this while looking at another issue... INFO: trying to register non-static key. the code is fine but needs lockdep annotation. turning off the locking correctness validator. Pid: 298, comm: udevd Not tainted 3.3.0 #1 Call Trace: [<ffffffff810cee63>] __lock_acquire+0x1363/0x1bb0 [<ffffffff816593d2>] ? __slab_alloc+0x4a0/0x4fd [<ffffffff81020063>] ? native_sched_clock+0x13/0x80 [<ffffffff810200d9>] ? sched_clock+0x9/0x10 [<ffffffff810cfd81>] lock_acquire+0xa1/0x1e0 [<ffffffffa0321b83>] ? bfa_fcs_lport_aen_post+0x43/0x160 [bfa] [<ffffffff816630e5>] _raw_spin_lock_irqsave+0x65/0xb0 [<ffffffffa0321b83>] ? bfa_fcs_lport_aen_post+0x43/0x160 [bfa] [<ffffffffa0321b83>] bfa_fcs_lport_aen_post+0x43/0x160 [bfa] [<ffffffffa032fa57>] bfa_fcs_lport_init+0x97/0x120 [bfa] [<ffffffffa031d4ff>] bfa_fcs_fabric_sm_uninit+0x1cf/0x250 [bfa] [<ffffffffa031c1da>] bfa_fcs_fabric_modinit+0x2a/0xb0 [bfa] [<ffffffffa031bef2>] ? bfa_fcs_fabric_attach+0xf2/0x170 [bfa] [<ffffffffa031e75a>] bfa_fcs_init+0x2a/0x40 [bfa] [<ffffffffa02facf7>] bfad_drv_init+0x107/0x1f0 [bfa] [<ffffffffa02fb057>] bfad_pci_probe+0x277/0x450 [bfa] [<ffffffff81354d4c>] local_pci_probe+0x5c/0xd0 [<ffffffff81354ed1>] pci_device_probe+0x111/0x120 [<ffffffff8141c0e6>] driver_probe_device+0x96/0x2f0 [<ffffffff8141c3eb>] __driver_attach+0xab/0xb0 [<ffffffff8141c340>] ? driver_probe_device+0x2f0/0x2f0 [<ffffffff8141a2e5>] bus_for_each_dev+0x55/0x90 [<ffffffffa037b000>] ? 0xffffffffa037afff [<ffffffff8141bbce>] driver_attach+0x1e/0x20 [<ffffffff8141b8d8>] bus_add_driver+0x1b8/0x2b0 [<ffffffffa037b000>] ? 0xffffffffa037afff [<ffffffff8141cbc7>] driver_register+0x77/0x160 [<ffffffffa037b000>] ? 0xffffffffa037afff [<ffffffff81353b23>] __pci_register_driver+0x73/0xf0 [<ffffffffa037b000>] ? 0xffffffffa037afff [<ffffffffa037b083>] bfad_init+0x83/0x1000 [bfa] [<ffffffff8100212a>] do_one_initcall+0x12a/0x180 [<ffffffff810df0d0>] sys_init_module+0xc0/0x220 [<ffffffff8166bbe9>] system_call_fastpath+0x16/0x1b Signed-off-by: Kyle McMartin <kmcmarti@redhat.com> Acked-by: Krishna Gudipati <kgudipat@brocade.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] qla2xxx: remove unnecessary reads of PCI_CAP_ID_EXPJon Mason2012-07-203-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The PCIE capability offset is saved during PCI bus walking. It will remove an unnecessary search in the PCI configuration space if this value is referenced instead of reacquiring it. Also, pci_is_pcie is a better way of determining if the device is PCIE or not (as it uses the same saved PCIE capability offset). Signed-off-by: Jon Mason <jdmason@kudzu.us> Acked-by: Chad Dupuis <chad.dupuis@qlogic.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
| * | [SCSI] qla4xxx: remove unnecessary read of PCI_CAP_ID_EXPJon Mason2012-07-201-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The PCIE capability offset is saved during PCI bus walking. It will remove an unnecessary search in the PCI configuration space if this value is referenced instead of reacquiring it. Signed-off-by: Jon Mason <jdmason@kudzu.us> Acked-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
OpenPOWER on IntegriCloud