From: Alistair Popple <alistair@popple.id.au>
To: Alexey Kardashevskiy <aik@ozlabs.ru>
Cc: "Jose Ricardo Ziviani" <joserz@linux.ibm.com>,
"Sam Bobroff" <sbobroff@linux.ibm.com>,
linuxppc-dev@lists.ozlabs.org,
"Alex Williamson" <alex.williamson@redhat.com>,
kvm-ppc@vger.kernel.org,
"Piotr Jaroszynski" <pjaroszynski@nvidia.com>,
"Oliver O'Halloran" <oohall@gmail.com>,
"Andrew Donnellan" <andrew.donnellan@au1.ibm.com>,
"Leonardo Augusto Guimarães Garcia" <lagarcia@br.ibm.com>,
"Reza Arbab" <arbab@linux.ibm.com>,
"David Gibson" <david@gibson.dropbear.id.au>
Subject: Re: [PATCH kernel v3 07/22] powerpc/powernv/npu: Move OPAL calls away from context manipulation
Date: Wed, 14 Nov 2018 15:57:37 +1100 [thread overview]
Message-ID: <2793658.kiX7AcRqs3@townsend> (raw)
In-Reply-To: <20181113082823.2440-8-aik@ozlabs.ru>
> - /*
> - * Setup the NPU context table for a particular GPU. These need to be
> - * per-GPU as we need the tables to filter ATSDs when there are no
> - * active contexts on a particular GPU. It is safe for these to be
> - * called concurrently with destroy as the OPAL call takes appropriate
> - * locks and refcounts on init/destroy.
> - */
> - rc = opal_npu_init_context(nphb->opal_id, mm->context.id, flags,
> - PCI_DEVID(gpdev->bus->number, gpdev->devfn));
> - if (rc < 0)
> - return ERR_PTR(-ENOSPC);
> -
This will prevent any drivers from setting up contexts with different MSR
values (which is what the flags argument is for) than a standard userspace
context (MSR_DR | MSR_PR | MSR_HV). In practice this currently never happens
and I'm unsure if that's ever likely to change.
We should at least return an error if flags != (MSR_DR | MSR_PR | MSR_HV).
> /*
> * We store the npu pci device so we can more easily get at the
> * associated npus.
> @@ -755,9 +738,6 @@ struct npu_context *pnv_npu2_init_context(struct pci_dev
> *gpdev, if (npu_context->release_cb != cb ||
> npu_context->priv != priv) {
> spin_unlock(&npu_context_lock);
> - opal_npu_destroy_context(nphb->opal_id, mm->context.id,
> - PCI_DEVID(gpdev->bus->number,
> - gpdev->devfn));
> return ERR_PTR(-EINVAL);
> }
>
> @@ -783,9 +763,6 @@ struct npu_context *pnv_npu2_init_context(struct pci_dev
> *gpdev,
>
> if (rc) {
> kfree(npu_context);
> - opal_npu_destroy_context(nphb->opal_id, mm->context.id,
> - PCI_DEVID(gpdev->bus->number,
> - gpdev->devfn));
> return ERR_PTR(rc);
> }
>
> @@ -838,7 +815,6 @@ void pnv_npu2_destroy_context(struct npu_context
> *npu_context, struct pci_dev *gpdev)
> {
> int removed;
> - struct pnv_phb *nphb;
> struct npu *npu;
> struct pci_dev *npdev = pnv_pci_get_npu_dev(gpdev, 0);
> struct device_node *nvlink_dn;
> @@ -847,10 +823,6 @@ void pnv_npu2_destroy_context(struct npu_context
> *npu_context, if (WARN_ON(!npdev))
> return;
>
> - if (!firmware_has_feature(FW_FEATURE_OPAL))
> - return;
> -
> - nphb = pci_bus_to_host(npdev->bus)->private_data;
> npu = npdev_to_npu(npdev);
> if (!npu)
> return;
> @@ -859,8 +831,6 @@ void pnv_npu2_destroy_context(struct npu_context
> *npu_context, &nvlink_index)))
> return;
> WRITE_ONCE(npu_context->npdev[npu->index][nvlink_index], NULL);
> - opal_npu_destroy_context(nphb->opal_id, npu_context->mm->context.id,
> - PCI_DEVID(gpdev->bus->number, gpdev->devfn));
> spin_lock(&npu_context_lock);
> removed = kref_put(&npu_context->kref, pnv_npu2_release_context);
> spin_unlock(&npu_context_lock);
> @@ -892,9 +862,6 @@ int pnv_npu2_handle_fault(struct npu_context *context,
> uintptr_t *ea, /* mmap_sem should be held so the struct_mm must be present
> */
> struct mm_struct *mm = context->mm;
>
> - if (!firmware_has_feature(FW_FEATURE_OPAL))
> - return -ENODEV;
> -
> WARN_ON(!rwsem_is_locked(&mm->mmap_sem));
>
> for (i = 0; i < count; i++) {
> @@ -923,15 +890,11 @@ int pnv_npu2_handle_fault(struct npu_context *context,
> uintptr_t *ea, }
> EXPORT_SYMBOL(pnv_npu2_handle_fault);
>
> -int pnv_npu2_init(struct pnv_phb *phb)
> +int pnv_npu2_init(struct pci_controller *hose)
> {
> unsigned int i;
> u64 mmio_atsd;
> - struct device_node *dn;
> - struct pci_dev *gpdev;
> static int npu_index;
> - uint64_t rc = 0;
> - struct pci_controller *hose = phb->hose;
> struct npu *npu;
> int ret;
>
> @@ -940,18 +903,6 @@ int pnv_npu2_init(struct pnv_phb *phb)
> return -ENOMEM;
>
> npu->nmmu_flush = of_property_read_bool(hose->dn, "ibm,nmmu-flush");
> - for_each_child_of_node(phb->hose->dn, dn) {
> - gpdev = pnv_pci_get_gpu_dev(get_pci_dev(dn));
> - if (gpdev) {
> - rc = opal_npu_map_lpar(phb->opal_id,
> - PCI_DEVID(gpdev->bus->number, gpdev->devfn),
> - 0, 0);
> - if (rc)
> - dev_err(&gpdev->dev,
> - "Error %lld mapping device to LPAR\n",
> - rc);
> - }
> - }
>
> for (i = 0; !of_property_read_u64_index(hose->dn, "ibm,mmio-atsd",
> i, &mmio_atsd); i++)
> @@ -981,3 +932,57 @@ int pnv_npu2_init(struct pnv_phb *phb)
>
> return ret;
> }
> +
> +int pnv_npu2_map_lpar_dev(struct pci_dev *gpdev, unsigned int lparid,
> + unsigned long msr)
> +{
> + int ret;
> + struct pci_dev *npdev = pnv_pci_get_npu_dev(gpdev, 0);
> + struct pci_controller *hose;
> + struct pnv_phb *nphb;
> +
> + if (!npdev)
> + return -ENODEV;
> +
> + hose = pci_bus_to_host(npdev->bus);
> + nphb = hose->private_data;
> +
> + dev_dbg(&gpdev->dev, "Map LPAR opalid=%llu lparid=%u\n",
> + nphb->opal_id, lparid);
> + /*
> + * Currently we only support radix and non-zero LPCR only makes sense
> + * for hash tables so skiboot expects the LPCR parameter to be a zero.
> + */
> + ret = opal_npu_map_lpar(nphb->opal_id,
> + PCI_DEVID(gpdev->bus->number, gpdev->devfn), lparid,
> + 0 /* LPCR bits */);
> + if (ret) {
> + dev_err(&gpdev->dev, "Error %d mapping device to LPAR\n", ret);
> + return ret;
> + }
> +
> + dev_dbg(&gpdev->dev, "init context opalid=%llu msr=%lx\n",
> + nphb->opal_id, msr);
> + ret = opal_npu_init_context(nphb->opal_id, 0/*__unused*/, msr,
> + PCI_DEVID(gpdev->bus->number, gpdev->devfn));
> + if (ret < 0)
> + dev_err(&gpdev->dev, "Failed to init context: %d\n", ret);
> + else
> + ret = 0;
> +
> + return 0;
> +}
> +EXPORT_SYMBOL_GPL(pnv_npu2_map_lpar_dev);
> +
> +void pnv_npu2_map_lpar(struct pnv_ioda_pe *gpe, unsigned long msr)
> +{
> + int ret;
> + struct pci_dev *gpdev;
> +
> + list_for_each_entry(gpdev, &gpe->pbus->devices, bus_list) {
> + ret = pnv_npu2_map_lpar_dev(gpdev, 0, msr);
> + if (ret < 0)
> + dev_err(&gpdev->dev, "Failed to init context: %d\n",
> + ret);
> + }
> +}
> diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c
> b/arch/powerpc/platforms/powernv/pci-ioda.c index c78c204..ec235ca 100644
> --- a/arch/powerpc/platforms/powernv/pci-ioda.c
> +++ b/arch/powerpc/platforms/powernv/pci-ioda.c
> @@ -1271,19 +1271,20 @@ static void pnv_ioda_setup_npu_PEs(struct pci_bus
> *bus)
>
> static void pnv_pci_ioda_setup_PEs(void)
> {
> - struct pci_controller *hose, *tmp;
> + struct pci_controller *hose;
> struct pnv_phb *phb;
> struct pci_bus *bus;
> struct pci_dev *pdev;
> + struct pnv_ioda_pe *pe;
>
> - list_for_each_entry_safe(hose, tmp, &hose_list, list_node) {
> + list_for_each_entry(hose, &hose_list, list_node) {
> phb = hose->private_data;
> if (phb->type == PNV_PHB_NPU_NVLINK) {
> /* PE#0 is needed for error reporting */
> pnv_ioda_reserve_pe(phb, 0);
> pnv_ioda_setup_npu_PEs(hose->bus);
> if (phb->model == PNV_PHB_MODEL_NPU2)
> - pnv_npu2_init(phb);
> + pnv_npu2_init(hose);
> }
> if (phb->type == PNV_PHB_NPU_OCAPI) {
> bus = hose->bus;
> @@ -1291,6 +1292,14 @@ static void pnv_pci_ioda_setup_PEs(void)
> pnv_ioda_setup_dev_PE(pdev);
> }
> }
> + list_for_each_entry(hose, &hose_list, list_node) {
> + phb = hose->private_data;
> + if (phb->type != PNV_PHB_IODA2)
> + continue;
> +
> + list_for_each_entry(pe, &phb->ioda.pe_list, list)
> + pnv_npu2_map_lpar(pe, MSR_DR | MSR_PR | MSR_HV);
> + }
> }
>
> #ifdef CONFIG_PCI_IOV
next prev parent reply other threads:[~2018-11-14 4:59 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-13 8:28 [PATCH kernel v3 00/22] powerpc/powernv/npu, vfio: NVIDIA V100 + P9 passthrough Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 01/22] powerpc/ioda/npu: Call skiboot's hot reset hook when disabling NPU2 Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 02/22] powerpc/mm/iommu/vfio_spapr_tce: Change mm_iommu_get to reference a region Alexey Kardashevskiy
2018-11-15 5:32 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 03/22] powerpc/mm/iommu: Make mm_iommu_new() fail on existing regions Alexey Kardashevskiy
2018-11-15 5:38 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 04/22] powerpc/vfio/iommu/kvm: Do not pin device memory Alexey Kardashevskiy
2018-11-16 3:11 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 05/22] powerpc/powernv/npu: Add helper to access struct npu for NPU device Alexey Kardashevskiy
2018-11-14 3:42 ` Alistair Popple
2018-11-13 8:28 ` [PATCH kernel v3 06/22] powerpc/powernv: Detach npu struct from pnv_phb Alexey Kardashevskiy
2018-11-14 4:28 ` Alistair Popple
2018-11-19 7:18 ` Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 07/22] powerpc/powernv/npu: Move OPAL calls away from context manipulation Alexey Kardashevskiy
2018-11-14 4:57 ` Alistair Popple [this message]
2018-11-13 8:28 ` [PATCH kernel v3 08/22] powerpc/pseries/iommu: Allow dynamic window to start from zero Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 09/22] powerpc/pseries/iommu: Force default DMA window removal Alexey Kardashevskiy
2018-11-16 4:54 ` David Gibson
2018-11-19 7:28 ` Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 10/22] powerpc/pseries/iommu: Use memory@ nodes in max RAM address calculation Alexey Kardashevskiy
2018-11-16 5:23 ` David Gibson
2018-11-19 7:43 ` Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 11/22] powerpc/pseries/npu: Enable platform support Alexey Kardashevskiy
2018-11-16 5:25 ` David Gibson
2018-11-19 7:50 ` Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 12/22] powerpc/pseries: Remove IOMMU API support for non-LPAR systems Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 13/22] powerpc/powernv/pseries: Rework device adding to IOMMU groups Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 14/22] powerpc/iommu_api: Move IOMMU groups setup to a single place Alexey Kardashevskiy
2018-11-19 0:15 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 15/22] powerpc/powernv: Reference iommu_table while it is linked to a group Alexey Kardashevskiy
2018-11-19 0:20 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 16/22] powerpc/powernv: Add purge cache OPAL call Alexey Kardashevskiy
2018-11-19 0:21 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 17/22] powerpc/powernv/npu: Convert NPU IOMMU helpers to iommu_table_group_ops Alexey Kardashevskiy
2018-11-19 0:24 ` David Gibson
2018-11-13 8:28 ` [PATCH kernel v3 18/22] powerpc/powernv/npu: Add compound IOMMU groups Alexey Kardashevskiy
2018-11-19 1:12 ` David Gibson
2018-11-19 2:29 ` Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 19/22] powerpc/powernv/npu: Add release_ownership hook Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 20/22] vfio_pci: Allow mapping extra regions Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 21/22] vfio_pci: Allow regions to add own capabilities Alexey Kardashevskiy
2018-11-13 8:28 ` [PATCH kernel v3 22/22] vfio_pci: Add NVIDIA GV100GL [Tesla V100 SXM2] [10de:1db1] subdriver Alexey Kardashevskiy
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2793658.kiX7AcRqs3@townsend \
--to=alistair@popple.id.au \
--cc=aik@ozlabs.ru \
--cc=alex.williamson@redhat.com \
--cc=andrew.donnellan@au1.ibm.com \
--cc=arbab@linux.ibm.com \
--cc=david@gibson.dropbear.id.au \
--cc=joserz@linux.ibm.com \
--cc=kvm-ppc@vger.kernel.org \
--cc=lagarcia@br.ibm.com \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=oohall@gmail.com \
--cc=pjaroszynski@nvidia.com \
--cc=sbobroff@linux.ibm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).