All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Cédric Le Goater" <clg@kaod.org>
To: Daniel Henrique Barboza <danielhb413@gmail.com>, <qemu-devel@nongnu.org>
Cc: qemu-ppc@nongnu.org, david@gibson.dropbear.id.au
Subject: Re: [PATCH v2 17/18] ppc/pnv: Introduce user creatable pnv-phb4 devices
Date: Thu, 6 Jan 2022 15:49:51 +0100	[thread overview]
Message-ID: <4c3eab1f-e435-b468-96e3-9f3d78430534@kaod.org> (raw)
In-Reply-To: <20220105212338.49899-18-danielhb413@gmail.com>

On 1/5/22 22:23, Daniel Henrique Barboza wrote:
> This patch introduces pnv-phb4 user creatable devices that are created
> in a similar manner as pnv-phb3 devices, allowing the user to interact
> with the PHBs directly instead of creating PCI Express Controllers that
> will create a certain amount of PHBs per controller index.
> 
> First thing we need is to discover which stack will host the created
> PHB, which is done by the new pnv_phb4_get_stack() function. During
> pnv_phb4_realize() we'll inspect phb->stack to see if we're dealing with
> an user creatable device or not. When using default settings, the
> automatically created PHB4 devices will be realized with phb->stack
> already assigned beforehand during PEC realize. In case we're dealing
> with an user device, find its stack, set the PHB attributes based on the
> stack it belongs and assign the PHB to the stack.
> 
> The xscom stack initialization takes place in pnv_pec_stk_realize() when
> using default settings, but at that point we aren't aware of any user
> PHB4 devices that will belong to the stack. In that case we'll postpone
> xscom stack init until the the end of pnv_phb4_realize(), after all the
> memory mappings of the PHB are done.
> 
> Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
> ---
>   hw/pci-host/pnv_phb4.c     | 84 +++++++++++++++++++++++++++++++++++++-
>   hw/pci-host/pnv_phb4_pec.c | 12 +++---
>   hw/ppc/pnv.c               |  2 +
>   3 files changed, 90 insertions(+), 8 deletions(-)
> 
> diff --git a/hw/pci-host/pnv_phb4.c b/hw/pci-host/pnv_phb4.c
> index 430a5c10f4..1c2334d491 100644
> --- a/hw/pci-host/pnv_phb4.c
> +++ b/hw/pci-host/pnv_phb4.c
> @@ -1236,6 +1236,41 @@ static void pnv_phb4_instance_init(Object *obj)
>       object_initialize_child(obj, "source", &phb->xsrc, TYPE_XIVE_SOURCE);
>   }
>   
> +static PnvPhb4PecStack *pnv_phb4_get_stack(PnvChip *chip, PnvPHB4 *phb,
> +                                           Error **errp)
> +{
> +    Pnv9Chip *chip9 = NULL;
> +    int chip_id = phb->chip_id;
> +    int index = phb->phb_id;
> +    int i, j;
> +
> +    if (chip->num_pecs == 0) {

assert() maybe ?


> +        /* Something weird happened. Bail out */
> +        error_setg(errp, "chip id %d has no PCIE controllers", chip_id);
> +        return NULL;
> +    }
> +
> +    chip9 = PNV9_CHIP(chip);
> +
> +    for (i = 0; i < chip->num_pecs; i++) {
> +        /*
> +         * For each PEC, check the amount of stacks it supports
> +         * and see if the given phb4 index matches a stack.
> +         */
> +        PnvPhb4PecState *pec = &chip9->pecs[i];
> +
> +        for (j = 0; j < pec->num_stacks; j++) {
> +            if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
> +                return &pec->stacks[j];
> +            }
> +        }
> +    }
> +
> +    error_setg(errp, "pnv-phb4 index %d didn't match any existing PEC",
> +               chip_id);
> +    return NULL;
> +}
> +
>   static void pnv_phb4_realize(DeviceState *dev, Error **errp)
>   {
>       PnvPHB4 *phb = PNV_PHB4(dev);
> @@ -1243,8 +1278,49 @@ static void pnv_phb4_realize(DeviceState *dev, Error **errp)
>       XiveSource *xsrc = &phb->xsrc;
>       int nr_irqs;
>       char name[32];
> +    PnvPhb4PecStack *stack = NULL;
> +    bool stack_init_xscom = false;
> +    Error *local_err = NULL;
>   
> -    assert(phb->stack);
> +    /* User created PHB */
> +    if (!phb->stack) {
> +        PnvMachineState *pnv = PNV_MACHINE(qdev_get_machine());
> +        PnvChip *chip = pnv_get_chip(pnv, phb->chip_id);
> +        BusState *s;
> +
> +        if (!chip) {
> +            error_setg(errp, "invalid chip id: %d", phb->chip_id);
> +            return;
> +        }
> +
> +        stack = pnv_phb4_get_stack(chip, phb, &local_err);
> +        if (local_err) {
> +            error_propagate(errp, local_err);
> +            return;
> +        }
> +
> +        object_property_set_int(OBJECT(phb), "index",
> +                                phb->phb_id, &error_abort);
> +
> +        pnv_phb4_set_stack_phb_props(stack, phb);
> +
> +        /* Assign the phb to the stack */
> +        stack->phb = phb;

The stack object should check the validity of the stack->phb pointer always.

> +
> +        /*
> +         * Reparent user created devices to the chip to build
> +         * correctly the device tree.
> +         */
> +        pnv_chip_parent_fixup(chip, OBJECT(phb), phb->phb_id);
> +
> +        s = qdev_get_parent_bus(DEVICE(chip));
> +        if (!qdev_set_parent_bus(DEVICE(phb), s, &local_err)) {
> +            error_propagate(errp, local_err);
> +            return;
> +        }
> +
> +        stack_init_xscom = true;

This looks wrong.

> +    }
>   
>       /* Set the "big_phb" flag */
>       phb->big_phb = phb->phb_id == 0 || phb->phb_id == 3;
> @@ -1298,6 +1374,10 @@ static void pnv_phb4_realize(DeviceState *dev, Error **errp)
>       pnv_phb4_update_xsrc(phb);
>   
>       phb->qirqs = qemu_allocate_irqs(xive_source_set_irq, xsrc, xsrc->nr_irqs);
> +
> +    if (stack_init_xscom) {
> +        pnv_pec_init_stack_xscom(stack);
> +    }

Isn't it done under the pnv_pec_stk_realize() routine already ?

>   }
>   
>   static const char *pnv_phb4_root_bus_path(PCIHostState *host_bridge,
> @@ -1347,7 +1427,7 @@ static void pnv_phb4_class_init(ObjectClass *klass, void *data)
>       dc->realize         = pnv_phb4_realize;
>       device_class_set_props(dc, pnv_phb4_properties);
>       set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
> -    dc->user_creatable  = false;
> +    dc->user_creatable  = true;
>   
>       xfc->notify         = pnv_phb4_xive_notify;
>   }
> diff --git a/hw/pci-host/pnv_phb4_pec.c b/hw/pci-host/pnv_phb4_pec.c
> index 41c79d24c4..4417beb97d 100644
> --- a/hw/pci-host/pnv_phb4_pec.c
> +++ b/hw/pci-host/pnv_phb4_pec.c
> @@ -573,13 +573,13 @@ static void pnv_pec_stk_realize(DeviceState *dev, Error **errp)
>                             &pnv_pec_stk_pci_xscom_ops, stack, name,
>                             PHB4_PEC_PCI_STK_REGS_COUNT);
>   
> -    /* PHB pass-through */
> -    pnv_phb4_set_stack_phb_props(stack, stack->phb);
> -    if (!sysbus_realize(SYS_BUS_DEVICE(&stack->phb), errp)) {
> -        return;
> +    /*
> +     * There is no guarantee that stack->phb will be available
> +     * at this point.
> +     */
> +    if (stack->phb) {
> +        pnv_pec_init_stack_xscom(stack);
>       }
> -
> -    pnv_pec_init_stack_xscom(stack);


This looks wrong also. I don't understand why.

Thanks,

C.


>   }
>   
>   static Property pnv_pec_stk_properties[] = {
> diff --git a/hw/ppc/pnv.c b/hw/ppc/pnv.c
> index fe7e67e73a..837146a2fb 100644
> --- a/hw/ppc/pnv.c
> +++ b/hw/ppc/pnv.c
> @@ -1960,6 +1960,8 @@ static void pnv_machine_power9_class_init(ObjectClass *oc, void *data)
>       pmc->compat = compat;
>       pmc->compat_size = sizeof(compat);
>       pmc->dt_power_mgt = pnv_dt_power_mgt;
> +
> +    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB4);
>   }
>   
>   static void pnv_machine_power10_class_init(ObjectClass *oc, void *data)
> 



  reply	other threads:[~2022-01-06 14:52 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-05 21:23 [PATCH v2 00/18] user creatable pnv-phb3/pnv-phb4 devices Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 01/18] pnv_phb3.c: add unique chassis and slot for pnv_phb3_root_port Daniel Henrique Barboza
2022-01-06  7:39   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 02/18] pnv_phb4.c: add unique chassis and slot for pnv_phb4_root_port Daniel Henrique Barboza
2022-01-06  7:39   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 03/18] ppc/pnv: Attach PHB3 root port device when defaults are enabled Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 04/18] pnv_phb4.c: make pnv-phb4-root-port user creatable Daniel Henrique Barboza
2022-01-06  7:40   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 05/18] pnv_phb4.c: check if root port exists in rc_config functions Daniel Henrique Barboza
2022-01-06  7:40   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 06/18] ppc/pnv: Introduce support for user created PHB3 devices Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 07/18] ppc/pnv: Reparent user created PHB3 devices to the PnvChip Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 08/18] ppc/pnv: Complete user created PHB3 devices Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 09/18] ppc/pnv: Move num_phbs under Pnv8Chip Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 10/18] pnv_phb3.h: change TYPE_PNV_PHB3_ROOT_BUS name Daniel Henrique Barboza
2022-01-06  7:41   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 11/18] pnv_phb4.c: introduce pnv_phb4_set_stack_phb_props() Daniel Henrique Barboza
2022-01-06 14:18   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 12/18] pnv_phb4_pec.c: move pnv_pec_phb_offset() to pnv_phb4.c Daniel Henrique Barboza
2022-01-06 14:18   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 13/18] pnv_phb4_pec: use pnv_phb4_pec_get_phb_id() in pnv_pec_dt_xscom() Daniel Henrique Barboza
2022-01-06 14:19   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 14/18] pnv_phb4.h: turn phb into a pointer in struct PnvPhb4PecStack Daniel Henrique Barboza
2022-01-06 14:24   ` Cédric Le Goater
2022-01-06 14:36   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 15/18] pnv_phb4_pec.c: use 'default_enabled()' to init stack->phb Daniel Henrique Barboza
2022-01-06 14:33   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 16/18] pnv_phb4.c: introduce pnv_pec_init_stack_xscom() Daniel Henrique Barboza
2022-01-06 14:38   ` Cédric Le Goater
2022-01-05 21:23 ` [PATCH v2 17/18] ppc/pnv: Introduce user creatable pnv-phb4 devices Daniel Henrique Barboza
2022-01-06 14:49   ` Cédric Le Goater [this message]
2022-01-07 21:17     ` Daniel Henrique Barboza
2022-01-08 11:11       ` Cédric Le Goater
2022-01-08 12:58         ` Daniel Henrique Barboza
2022-01-05 21:23 ` [PATCH v2 18/18] pnv_phb4.c: change TYPE_PNV_PHB4_ROOT_BUS name Daniel Henrique Barboza
2022-01-06 14:19   ` Cédric Le Goater
2022-01-06  8:18 ` [PATCH v2 00/18] user creatable pnv-phb3/pnv-phb4 devices Cédric Le Goater
2022-01-06 12:36   ` Daniel Henrique Barboza
2022-01-06 17:35     ` Cédric Le Goater

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=4c3eab1f-e435-b468-96e3-9f3d78430534@kaod.org \
    --to=clg@kaod.org \
    --cc=danielhb413@gmail.com \
    --cc=david@gibson.dropbear.id.au \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.