diff options
Diffstat (limited to 'arch/powerpc/platforms/powernv')
| -rw-r--r-- | arch/powerpc/platforms/powernv/Kconfig | 1 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/Makefile | 4 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/eeh-ioda.c | 109 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/opal-msglog.c | 6 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/opal-sysparam.c | 4 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/pci.c | 81 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/setup.c | 3 | ||||
| -rw-r--r-- | arch/powerpc/platforms/powernv/smp.c | 6 |
8 files changed, 139 insertions, 75 deletions
diff --git a/arch/powerpc/platforms/powernv/Kconfig b/arch/powerpc/platforms/powernv/Kconfig index c252ee95bddf..45a8ed0585cd 100644 --- a/arch/powerpc/platforms/powernv/Kconfig +++ b/arch/powerpc/platforms/powernv/Kconfig @@ -17,6 +17,7 @@ config PPC_POWERNV select CPU_FREQ_GOV_USERSPACE select CPU_FREQ_GOV_ONDEMAND select CPU_FREQ_GOV_CONSERVATIVE + select PPC_DOORBELL default y config PPC_POWERNV_RTAS diff --git a/arch/powerpc/platforms/powernv/Makefile b/arch/powerpc/platforms/powernv/Makefile index 4ad0d345bc96..d55891f89a2c 100644 --- a/arch/powerpc/platforms/powernv/Makefile +++ b/arch/powerpc/platforms/powernv/Makefile @@ -1,9 +1,9 @@ obj-y += setup.o opal-takeover.o opal-wrappers.o opal.o opal-async.o obj-y += opal-rtc.o opal-nvram.o opal-lpc.o opal-flash.o obj-y += rng.o opal-elog.o opal-dump.o opal-sysparam.o opal-sensor.o -obj-y += opal-msglog.o subcore.o subcore-asm.o +obj-y += opal-msglog.o -obj-$(CONFIG_SMP) += smp.o +obj-$(CONFIG_SMP) += smp.o subcore.o subcore-asm.o obj-$(CONFIG_PCI) += pci.o pci-p5ioc2.o pci-ioda.o obj-$(CONFIG_EEH) += eeh-ioda.o eeh-powernv.o obj-$(CONFIG_PPC_SCOM) += opal-xscom.o diff --git a/arch/powerpc/platforms/powernv/eeh-ioda.c b/arch/powerpc/platforms/powernv/eeh-ioda.c index 753f08e36dfa..8ad0c5b891f4 100644 --- a/arch/powerpc/platforms/powernv/eeh-ioda.c +++ b/arch/powerpc/platforms/powernv/eeh-ioda.c @@ -267,7 +267,7 @@ static int ioda_eeh_get_state(struct eeh_pe *pe) { s64 ret = 0; u8 fstate; - u16 pcierr; + __be16 pcierr; u32 pe_no; int result; struct pci_controller *hose = pe->phb; @@ -316,7 +316,7 @@ static int ioda_eeh_get_state(struct eeh_pe *pe) result = 0; result &= ~EEH_STATE_RESET_ACTIVE; - if (pcierr != OPAL_EEH_PHB_ERROR) { + if (be16_to_cpu(pcierr) != OPAL_EEH_PHB_ERROR) { result |= EEH_STATE_MMIO_ACTIVE; result |= EEH_STATE_DMA_ACTIVE; result |= EEH_STATE_MMIO_ENABLED; @@ -705,18 +705,19 @@ static int ioda_eeh_next_error(struct eeh_pe **pe) { struct pci_controller *hose; struct pnv_phb *phb; - struct eeh_pe *phb_pe; - u64 frozen_pe_no; - u16 err_type, severity; + struct eeh_pe *phb_pe, *parent_pe; + __be64 frozen_pe_no; + __be16 err_type, severity; + int active_flags = (EEH_STATE_MMIO_ACTIVE | EEH_STATE_DMA_ACTIVE); long rc; - int ret = EEH_NEXT_ERR_NONE; + int state, ret = EEH_NEXT_ERR_NONE; /* * While running here, it's safe to purge the event queue. * And we should keep the cached OPAL notifier event sychronized * between the kernel and firmware. */ - eeh_remove_event(NULL); + eeh_remove_event(NULL, false); opal_notifier_update_evt(OPAL_EVENT_PCI_ERROR, 0x0ul); list_for_each_entry(hose, &hose_list, list_node) { @@ -742,8 +743,8 @@ static int ioda_eeh_next_error(struct eeh_pe **pe) } /* If the PHB doesn't have error, stop processing */ - if (err_type == OPAL_EEH_NO_ERROR || - severity == OPAL_EEH_SEV_NO_ERROR) { + if (be16_to_cpu(err_type) == OPAL_EEH_NO_ERROR || + be16_to_cpu(severity) == OPAL_EEH_SEV_NO_ERROR) { pr_devel("%s: No error found on PHB#%x\n", __func__, hose->global_number); continue; @@ -755,14 +756,14 @@ static int ioda_eeh_next_error(struct eeh_pe **pe) * specific PHB. */ pr_devel("%s: Error (%d, %d, %llu) on PHB#%x\n", - __func__, err_type, severity, - frozen_pe_no, hose->global_number); - switch (err_type) { + __func__, be16_to_cpu(err_type), be16_to_cpu(severity), + be64_to_cpu(frozen_pe_no), hose->global_number); + switch (be16_to_cpu(err_type)) { case OPAL_EEH_IOC_ERROR: - if (severity == OPAL_EEH_SEV_IOC_DEAD) { + if (be16_to_cpu(severity) == OPAL_EEH_SEV_IOC_DEAD) { pr_err("EEH: dead IOC detected\n"); ret = EEH_NEXT_ERR_DEAD_IOC; - } else if (severity == OPAL_EEH_SEV_INF) { + } else if (be16_to_cpu(severity) == OPAL_EEH_SEV_INF) { pr_info("EEH: IOC informative error " "detected\n"); ioda_eeh_hub_diag(hose); @@ -771,20 +772,26 @@ static int ioda_eeh_next_error(struct eeh_pe **pe) break; case OPAL_EEH_PHB_ERROR: - if (severity == OPAL_EEH_SEV_PHB_DEAD) { + if (be16_to_cpu(severity) == OPAL_EEH_SEV_PHB_DEAD) { *pe = phb_pe; - pr_err("EEH: dead PHB#%x detected\n", - hose->global_number); + pr_err("EEH: dead PHB#%x detected, " + "location: %s\n", + hose->global_number, + eeh_pe_loc_get(phb_pe)); ret = EEH_NEXT_ERR_DEAD_PHB; - } else if (severity == OPAL_EEH_SEV_PHB_FENCED) { + } else if (be16_to_cpu(severity) == + OPAL_EEH_SEV_PHB_FENCED) { *pe = phb_pe; - pr_err("EEH: fenced PHB#%x detected\n", - hose->global_number); + pr_err("EEH: Fenced PHB#%x detected, " + "location: %s\n", + hose->global_number, + eeh_pe_loc_get(phb_pe)); ret = EEH_NEXT_ERR_FENCED_PHB; - } else if (severity == OPAL_EEH_SEV_INF) { + } else if (be16_to_cpu(severity) == OPAL_EEH_SEV_INF) { pr_info("EEH: PHB#%x informative error " - "detected\n", - hose->global_number); + "detected, location: %s\n", + hose->global_number, + eeh_pe_loc_get(phb_pe)); ioda_eeh_phb_diag(hose); ret = EEH_NEXT_ERR_NONE; } @@ -792,34 +799,33 @@ static int ioda_eeh_next_error(struct eeh_pe **pe) break; case OPAL_EEH_PE_ERROR: /* - * If we can't find the corresponding PE, the - * PEEV / PEST would be messy. So we force an - * fenced PHB so that it can be recovered. - * - * If the PE has been marked as isolated, that - * should have been removed permanently or in - * progress with recovery. We needn't report - * it again. + * If we can't find the corresponding PE, we + * just try to unfreeze. */ - if (ioda_eeh_get_pe(hose, frozen_pe_no, pe)) { - *pe = phb_pe; - pr_err("EEH: Escalated fenced PHB#%x " - "detected for PE#%llx\n", - hose->global_number, - frozen_pe_no); - ret = EEH_NEXT_ERR_FENCED_PHB; + if (ioda_eeh_get_pe(hose, + be64_to_cpu(frozen_pe_no), pe)) { + /* Try best to clear it */ + pr_info("EEH: Clear non-existing PHB#%x-PE#%llx\n", + hose->global_number, frozen_pe_no); + pr_info("EEH: PHB location: %s\n", + eeh_pe_loc_get(phb_pe)); + opal_pci_eeh_freeze_clear(phb->opal_id, frozen_pe_no, + OPAL_EEH_ACTION_CLEAR_FREEZE_ALL); + ret = EEH_NEXT_ERR_NONE; } else if ((*pe)->state & EEH_PE_ISOLATED) { ret = EEH_NEXT_ERR_NONE; } else { pr_err("EEH: Frozen PE#%x on PHB#%x detected\n", (*pe)->addr, (*pe)->phb->global_number); + pr_err("EEH: PE location: %s, PHB location: %s\n", + eeh_pe_loc_get(*pe), eeh_pe_loc_get(phb_pe)); ret = EEH_NEXT_ERR_FROZEN_PE; } break; default: pr_warn("%s: Unexpected error type %d\n", - __func__, err_type); + __func__, be16_to_cpu(err_type)); } /* @@ -837,6 +843,31 @@ static int ioda_eeh_next_error(struct eeh_pe **pe) } /* + * We probably have the frozen parent PE out there and + * we need have to handle frozen parent PE firstly. + */ + if (ret == EEH_NEXT_ERR_FROZEN_PE) { + parent_pe = (*pe)->parent; + while (parent_pe) { + /* Hit the ceiling ? */ + if (parent_pe->type & EEH_PE_PHB) + break; + + /* Frozen parent PE ? */ + state = ioda_eeh_get_state(parent_pe); + if (state > 0 && + (state & active_flags) != active_flags) + *pe = parent_pe; + + /* Next parent level */ + parent_pe = parent_pe->parent; + } + + /* We possibly migrate to another PE */ + eeh_pe_state_mark(*pe, EEH_PE_ISOLATED); + } + + /* * If we have no errors on the specific PHB or only * informative error there, we continue poking it. * Otherwise, we need actions to be taken by upper diff --git a/arch/powerpc/platforms/powernv/opal-msglog.c b/arch/powerpc/platforms/powernv/opal-msglog.c index 1bb25b952504..44ed78af1a0d 100644 --- a/arch/powerpc/platforms/powernv/opal-msglog.c +++ b/arch/powerpc/platforms/powernv/opal-msglog.c @@ -37,7 +37,8 @@ static ssize_t opal_msglog_read(struct file *file, struct kobject *kobj, { struct memcons *mc = bin_attr->private; const char *conbuf; - size_t ret, first_read = 0; + ssize_t ret; + size_t first_read = 0; uint32_t out_pos, avail; if (!mc) @@ -69,6 +70,9 @@ static ssize_t opal_msglog_read(struct file *file, struct kobject *kobj, to += first_read; count -= first_read; pos -= avail; + + if (count <= 0) + goto out; } /* Sanity check. The firmware should not do this to us. */ diff --git a/arch/powerpc/platforms/powernv/opal-sysparam.c b/arch/powerpc/platforms/powernv/opal-sysparam.c index d202f9bc3683..9d1acf22a099 100644 --- a/arch/powerpc/platforms/powernv/opal-sysparam.c +++ b/arch/powerpc/platforms/powernv/opal-sysparam.c @@ -260,10 +260,10 @@ void __init opal_sys_param_init(void) attr[i].kobj_attr.attr.mode = S_IRUGO; break; case OPAL_SYSPARAM_WRITE: - attr[i].kobj_attr.attr.mode = S_IWUGO; + attr[i].kobj_attr.attr.mode = S_IWUSR; break; case OPAL_SYSPARAM_RW: - attr[i].kobj_attr.attr.mode = S_IRUGO | S_IWUGO; + attr[i].kobj_attr.attr.mode = S_IRUGO | S_IWUSR; break; default: break; diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c index eefbfcc3fd8c..f91a4e5d872e 100644 --- a/arch/powerpc/platforms/powernv/pci.c +++ b/arch/powerpc/platforms/powernv/pci.c @@ -206,72 +206,91 @@ static void pnv_pci_dump_phb3_diag_data(struct pci_controller *hose, data = (struct OpalIoPhb3ErrorData*)common; pr_info("PHB3 PHB#%d Diag-data (Version: %d)\n", - hose->global_number, common->version); + hose->global_number, be32_to_cpu(common->version)); if (data->brdgCtl) pr_info("brdgCtl: %08x\n", - data->brdgCtl); + be32_to_cpu(data->brdgCtl)); if (data->portStatusReg || data->rootCmplxStatus || data->busAgentStatus) pr_info("UtlSts: %08x %08x %08x\n", - data->portStatusReg, data->rootCmplxStatus, - data->busAgentStatus); + be32_to_cpu(data->portStatusReg), + be32_to_cpu(data->rootCmplxStatus), + be32_to_cpu(data->busAgentStatus)); if (data->deviceStatus || data->slotStatus || data->linkStatus || data->devCmdStatus || data->devSecStatus) pr_info("RootSts: %08x %08x %08x %08x %08x\n", - data->deviceStatus, data->slotStatus, - data->linkStatus, data->devCmdStatus, - data->devSecStatus); + be32_to_cpu(data->deviceStatus), + be32_to_cpu(data->slotStatus), + be32_to_cpu(data->linkStatus), + be32_to_cpu(data->devCmdStatus), + be32_to_cpu(data->devSecStatus)); if (data->rootErrorStatus || data->uncorrErrorStatus || data->corrErrorStatus) pr_info("RootErrSts: %08x %08x %08x\n", - data->rootErrorStatus, data->uncorrErrorStatus, - data->corrErrorStatus); + be32_to_cpu(data->rootErrorStatus), + be32_to_cpu(data->uncorrErrorStatus), + be32_to_cpu(data->corrErrorStatus)); if (data->tlpHdr1 || data->tlpHdr2 || data->tlpHdr3 || data->tlpHdr4) pr_info("RootErrLog: %08x %08x %08x %08x\n", - data->tlpHdr1, data->tlpHdr2, - data->tlpHdr3, data->tlpHdr4); + be32_to_cpu(data->tlpHdr1), + be32_to_cpu(data->tlpHdr2), + be32_to_cpu(data->tlpHdr3), + be32_to_cpu(data->tlpHdr4)); if (data->sourceId || data->errorClass || data->correlator) pr_info("RootErrLog1: %08x %016llx %016llx\n", - data->sourceId, data->errorClass, - data->correlator); + be32_to_cpu(data->sourceId), + be64_to_cpu(data->errorClass), + be64_to_cpu(data->correlator)); if (data->nFir) pr_info("nFir: %016llx %016llx %016llx\n", - data->nFir, data->nFirMask, - data->nFirWOF); + be64_to_cpu(data->nFir), + be64_to_cpu(data->nFirMask), + be64_to_cpu(data->nFirWOF)); if (data->phbPlssr || data->phbCsr) pr_info("PhbSts: %016llx %016llx\n", - data->phbPlssr, data->phbCsr); + be64_to_cpu(data->phbPlssr), + be64_to_cpu(data->phbCsr)); if (data->lemFir) pr_info("Lem: %016llx %016llx %016llx\n", - data->lemFir, data->lemErrorMask, - data->lemWOF); + be64_to_cpu(data->lemFir), + be64_to_cpu(data->lemErrorMask), + be64_to_cpu(data->lemWOF)); if (data->phbErrorStatus) pr_info("PhbErr: %016llx %016llx %016llx %016llx\n", - data->phbErrorStatus, data->phbFirstErrorStatus, - data->phbErrorLog0, data->phbErrorLog1); + be64_to_cpu(data->phbErrorStatus), + be64_to_cpu(data->phbFirstErrorStatus), + be64_to_cpu(data->phbErrorLog0), + be64_to_cpu(data->phbErrorLog1)); if (data->mmioErrorStatus) pr_info("OutErr: %016llx %016llx %016llx %016llx\n", - data->mmioErrorStatus, data->mmioFirstErrorStatus, - data->mmioErrorLog0, data->mmioErrorLog1); + be64_to_cpu(data->mmioErrorStatus), + be64_to_cpu(data->mmioFirstErrorStatus), + be64_to_cpu(data->mmioErrorLog0), + be64_to_cpu(data->mmioErrorLog1)); if (data->dma0ErrorStatus) pr_info("InAErr: %016llx %016llx %016llx %016llx\n", - data->dma0ErrorStatus, data->dma0FirstErrorStatus, - data->dma0ErrorLog0, data->dma0ErrorLog1); + be64_to_cpu(data->dma0ErrorStatus), + be64_to_cpu(data->dma0FirstErrorStatus), + be64_to_cpu(data->dma0ErrorLog0), + be64_to_cpu(data->dma0ErrorLog1)); if (data->dma1ErrorStatus) pr_info("InBErr: %016llx %016llx %016llx %016llx\n", - data->dma1ErrorStatus, data->dma1FirstErrorStatus, - data->dma1ErrorLog0, data->dma1ErrorLog1); + be64_to_cpu(data->dma1ErrorStatus), + be64_to_cpu(data->dma1FirstErrorStatus), + be64_to_cpu(data->dma1ErrorLog0), + be64_to_cpu(data->dma1ErrorLog1)); for (i = 0; i < OPAL_PHB3_NUM_PEST_REGS; i++) { - if ((data->pestA[i] >> 63) == 0 && - (data->pestB[i] >> 63) == 0) + if ((be64_to_cpu(data->pestA[i]) >> 63) == 0 && + (be64_to_cpu(data->pestB[i]) >> 63) == 0) continue; pr_info("PE[%3d] A/B: %016llx %016llx\n", - i, data->pestA[i], data->pestB[i]); + i, be64_to_cpu(data->pestA[i]), + be64_to_cpu(data->pestB[i])); } } @@ -284,7 +303,7 @@ void pnv_pci_dump_phb_diag_data(struct pci_controller *hose, return; common = (struct OpalIoPhbErrorCommon *)log_buff; - switch (common->ioType) { + switch (be32_to_cpu(common->ioType)) { case OPAL_PHB_ERROR_DATA_TYPE_P7IOC: pnv_pci_dump_p7ioc_diag_data(hose, common); break; @@ -293,7 +312,7 @@ void pnv_pci_dump_phb_diag_data(struct pci_controller *hose, break; default: pr_warn("%s: Unrecognized ioType %d\n", - __func__, common->ioType); + __func__, be32_to_cpu(common->ioType)); } } diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c index 8c16a5f96728..d9b88fa7c5a3 100644 --- a/arch/powerpc/platforms/powernv/setup.c +++ b/arch/powerpc/platforms/powernv/setup.c @@ -35,11 +35,14 @@ #include <asm/rtas.h> #include <asm/opal.h> #include <asm/kexec.h> +#include <asm/smp.h> #include "powernv.h" static void __init pnv_setup_arch(void) { + set_arch_panic_timeout(10, ARCH_PANIC_TIMEOUT); + /* Initialize SMP */ pnv_smp_init(); diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c index 0062a43a2e0d..5fcfcf44e3a9 100644 --- a/arch/powerpc/platforms/powernv/smp.c +++ b/arch/powerpc/platforms/powernv/smp.c @@ -32,6 +32,7 @@ #include <asm/opal.h> #include <asm/runlatch.h> #include <asm/code-patching.h> +#include <asm/dbell.h> #include "powernv.h" @@ -46,6 +47,11 @@ static void pnv_smp_setup_cpu(int cpu) { if (cpu != boot_cpuid) xics_setup_cpu(); + +#ifdef CONFIG_PPC_DOORBELL + if (cpu_has_feature(CPU_FTR_DBELL)) + doorbell_setup_this_cpu(); +#endif } int pnv_smp_kick_cpu(int nr) |
