From: "Michael S. Tsirkin" <mst@redhat.com>
To: Isaku Yamahata <yamahata@valinux.co.jp>
Cc: skandasa@cisco.com, adnan@khaleel.us, etmartin@cisco.com,
qemu-devel@nongnu.org, wexu2@cisco.com
Subject: [Qemu-devel] Re: [PATCH v5 07/14] pcie: helper functions for pcie capability and extended capability
Date: Tue, 19 Oct 2010 13:53:56 +0200 [thread overview]
Message-ID: <20101019115356.GB6752@redhat.com> (raw)
In-Reply-To: <e69a67a5be51dd8ab4137cb1fb884a1f9868e97d.1287478251.git.yamahata@valinux.co.jp>
On Tue, Oct 19, 2010 at 06:06:34PM +0900, Isaku Yamahata wrote:
> This patch implements helper functions for pci express capability
> and pci express extended capability allocation.
> NOTE: presence detection depends on pci_qdev_init() change.
>
> Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
> ---
> Changes v4 -> v5:
> - dropped FLR related members. This will be addressed at the next phase.
> - use pci_xxx_test_and_xxx_mask().
> - drop PCIDevice::written bits. and made related registers writable.
> - eliminated pcie_cap_slot_notify()
> - introduced PCIExpressDevice::hpev_intx
Please also add a comment saying that hpev_intx field defaults to 0,
and needs to be changed by devices that want to use another
interrupt for hotplug events.
> Changes v3 -> v4:
> - various clean up
> - dropped pcie_notify(), pcie_del_capability()
> - use pci_{clear_set, clear}_bit_xxx() helper functions.
> - dropped pci_exp_cap()
> ---
> Makefile.objs | 1 +
> hw/pci.h | 5 +
> hw/pcie.c | 540 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> hw/pcie.h | 107 ++++++++++++
> qemu-common.h | 1 +
> 5 files changed, 654 insertions(+), 0 deletions(-)
> create mode 100644 hw/pcie.c
> create mode 100644 hw/pcie.h
>
> diff --git a/Makefile.objs b/Makefile.objs
> index 5f5a4c5..eeb5134 100644
> --- a/Makefile.objs
> +++ b/Makefile.objs
> @@ -186,6 +186,7 @@ hw-obj-$(CONFIG_PIIX4) += piix4.o
> # PCI watchdog devices
> hw-obj-y += wdt_i6300esb.o
>
> +hw-obj-y += pcie.o
> hw-obj-y += msix.o msi.o
>
> # PCI network cards
> diff --git a/hw/pci.h b/hw/pci.h
> index 9e2f27d..d6c522b 100644
> --- a/hw/pci.h
> +++ b/hw/pci.h
> @@ -9,6 +9,8 @@
> /* PCI includes legacy ISA access. */
> #include "isa.h"
>
> +#include "pcie.h"
> +
> /* PCI bus */
>
> #define PCI_DEVFN(slot, func) ((((slot) & 0x1f) << 3) | ((func) & 0x07))
> @@ -175,6 +177,9 @@ struct PCIDevice {
> /* Offset of MSI capability in config space */
> uint8_t msi_cap;
>
> + /* PCI Express */
> + PCIExpressDevice exp;
> +
> /* Location of option rom */
> char *romfile;
> ram_addr_t rom_offset;
> diff --git a/hw/pcie.c b/hw/pcie.c
> new file mode 100644
> index 0000000..53d1fce
> --- /dev/null
> +++ b/hw/pcie.c
> @@ -0,0 +1,540 @@
> +/*
> + * pcie.c
> + *
> + * Copyright (c) 2010 Isaku Yamahata <yamahata at valinux co jp>
> + * VA Linux Systems Japan K.K.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#include "sysemu.h"
> +#include "pci_bridge.h"
> +#include "pcie.h"
> +#include "msix.h"
> +#include "msi.h"
> +#include "pci_internals.h"
> +#include "pcie_regs.h"
> +
> +//#define DEBUG_PCIE
> +#ifdef DEBUG_PCIE
> +# define PCIE_DPRINTF(fmt, ...) \
> + fprintf(stderr, "%s:%d " fmt, __func__, __LINE__, ## __VA_ARGS__)
> +#else
> +# define PCIE_DPRINTF(fmt, ...) do {} while (0)
> +#endif
> +#define PCIE_DEV_PRINTF(dev, fmt, ...) \
> + PCIE_DPRINTF("%s:%x "fmt, (dev)->name, (dev)->devfn, ## __VA_ARGS__)
> +
> +
> +/***************************************************************************
> + * pci express capability helper functions
> + */
> +int pcie_cap_init(PCIDevice *dev, uint8_t offset, uint8_t type, uint8_t port)
> +{
> + int pos;
> + uint8_t *exp_cap;
> +
> + assert(pci_is_express(dev));
> +
> + pos = pci_add_capability(dev, PCI_CAP_ID_EXP, offset,
> + PCI_EXP_VER2_SIZEOF);
> + if (pos < 0) {
> + return pos;
> + }
> + dev->exp.exp_cap = pos;
> + exp_cap = dev->config + pos;
> +
> + /* capability register
> + interrupt message number defaults to 0 */
> + pci_set_word(exp_cap + PCI_EXP_FLAGS,
> + ((type << PCI_EXP_FLAGS_TYPE_SHIFT) & PCI_EXP_FLAGS_TYPE) |
> + PCI_EXP_FLAGS_VER2);
> +
> + /* device capability register
> + * table 7-12:
> + * roll based error reporting bit must be set by all
> + * Functions conforming to the ECN, PCI Express Base
> + * Specification, Revision 1.1., or subsequent PCI Express Base
> + * Specification revisions.
> + */
> + pci_set_long(exp_cap + PCI_EXP_DEVCAP, PCI_EXP_DEVCAP_RBER);
> +
> + pci_set_long(exp_cap + PCI_EXP_LNKCAP,
> + (port << PCI_EXP_LNKCAP_PN_SHIFT) |
> + PCI_EXP_LNKCAP_ASPMS_0S |
> + PCI_EXP_LNK_MLW_1 |
> + PCI_EXP_LNK_LS_25);
> +
> + pci_set_word(exp_cap + PCI_EXP_LNKSTA,
> + PCI_EXP_LNK_MLW_1 | PCI_EXP_LNK_LS_25);
> +
> + pci_set_long(exp_cap + PCI_EXP_DEVCAP2,
> + PCI_EXP_DEVCAP2_EFF | PCI_EXP_DEVCAP2_EETLPP);
> +
> + pci_set_word(dev->wmask + pos, PCI_EXP_DEVCTL2_EETLPPB);
> + return pos;
> +}
> +
> +void pcie_cap_exit(PCIDevice *dev)
> +{
> + pci_del_capability(dev, PCI_CAP_ID_EXP, PCI_EXP_VER2_SIZEOF);
> +}
> +
> +uint8_t pcie_cap_get_type(const PCIDevice *dev)
> +{
> + uint32_t pos = dev->exp.exp_cap;
> + assert(pos > 0);
> + return (pci_get_word(dev->config + pos + PCI_EXP_FLAGS) &
> + PCI_EXP_FLAGS_TYPE) >> PCI_EXP_FLAGS_TYPE_SHIFT;
> +}
> +
> +/* MSI/MSI-X */
> +/* pci express interrupt message number */
> +/* 7.8.2 PCI Express Capabilities Register: Interrupt Message Number */
> +void pcie_cap_flags_set_vector(PCIDevice *dev, uint8_t vector)
> +{
> + uint8_t *exp_cap = dev->config + dev->exp.exp_cap;
> + assert(vector < 32);
> + pci_word_test_and_clear_mask(exp_cap + PCI_EXP_FLAGS, PCI_EXP_FLAGS_IRQ);
> + pci_word_test_and_set_mask(exp_cap + PCI_EXP_FLAGS,
> + vector << PCI_EXP_FLAGS_IRQ_SHIFT);
> +}
> +
> +uint8_t pcie_cap_flags_get_vector(PCIDevice *dev)
> +{
> + return (pci_get_word(dev->config + dev->exp.exp_cap + PCI_EXP_FLAGS) &
> + PCI_EXP_FLAGS_IRQ) >> PCI_EXP_FLAGS_IRQ_SHIFT;
> +}
> +
> +void pcie_cap_deverr_init(PCIDevice *dev)
> +{
> + uint32_t pos = dev->exp.exp_cap;
> + pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_DEVCAP,
> + PCI_EXP_DEVCAP_RBER);
> + pci_long_test_and_set_mask(dev->wmask + pos + PCI_EXP_DEVCTL,
> + PCI_EXP_DEVCTL_CERE | PCI_EXP_DEVCTL_NFERE |
> + PCI_EXP_DEVCTL_FERE | PCI_EXP_DEVCTL_URRE);
> + pci_long_test_and_set_mask(dev->w1cmask + pos + PCI_EXP_DEVSTA,
> + PCI_EXP_DEVSTA_CED | PCI_EXP_DEVSTA_NFED |
> + PCI_EXP_DEVSTA_URD | PCI_EXP_DEVSTA_URD);
> +}
> +
> +void pcie_cap_deverr_reset(PCIDevice *dev)
> +{
> + uint8_t *devctl = dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL;
> + pci_long_test_and_clear_mask(devctl,
> + PCI_EXP_DEVCTL_CERE | PCI_EXP_DEVCTL_NFERE |
> + PCI_EXP_DEVCTL_FERE | PCI_EXP_DEVCTL_URRE);
> +}
> +
> +/*
> + * A PCI Express Hot-Plug Event has occured, so update slot status register
> + * and notify OS of the event if necessary.
> + *
> + * 6.7.3 PCI Express Hot-Plug Events
> + * 6.7.3.4 Software Notification of Hot-Plug Events
> + */
> +static void pcie_cap_slot_event(PCIDevice *dev, PCIExpressHotPlugEvent event)
> +{
> + uint8_t *exp_cap = dev->config + dev->exp.exp_cap;
> + uint16_t sltctl = pci_get_word(exp_cap + PCI_EXP_SLTCTL);
> + uint16_t sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);
> +
> + PCIE_DEV_PRINTF(dev,
> + "sltctl: 0x%02"PRIx16" sltsta: 0x%02"PRIx16" event: %x\n",
> + sltctl, sltsta, event);
> +
> + if (pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA, event)) {
> + return;
> + }
> + sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);
> + PCIE_DEV_PRINTF(dev, "sltsta -> %02"PRIx16"\n", sltsta);
> +
> + if ((sltctl & PCI_EXP_SLTCTL_HPIE) &&
> + (sltctl & event & PCI_EXP_HP_EV_SUPPORTED)) {
> + if (pci_msi_enabled(dev)) {
> + pci_msi_notify(dev, pcie_cap_flags_get_vector(dev));
> + } else {
> + qemu_set_irq(dev->irq[dev->exp.hpev_intx], 1);
> + }
> + }
> +}
> +
> +static int pcie_cap_slot_hotplug(DeviceState *qdev,
> + PCIDevice *pci_dev, int state)
> +{
> + PCIDevice *d = DO_UPCAST(PCIDevice, qdev, qdev);
> + uint8_t *exp_cap = d->config + d->exp.exp_cap;
> + uint16_t sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);
> +
> + if (!pci_dev->qdev.hotplugged) {
> + assert(state); /* this case only happens at machine creation. */
> + pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA,
> + PCI_EXP_SLTSTA_PDS);
> + return 0;
> + }
> +
> + PCIE_DEV_PRINTF(pci_dev, "hotplug state: %d\n", state);
> + if (sltsta & PCI_EXP_SLTSTA_EIS) {
> + /* the slot is electromechanically locked.
> + * This error is propagated up to qdev and then to HMP/QMP.
> + */
> + return -EBUSY;
> + }
> +
> + /* TODO: multifunction hot-plug.
> + * Right now, only a device of function = 0 is allowed to be
> + * hot plugged/unplugged.
> + */
> + assert(PCI_FUNC(pci_dev->devfn) == 0);
> +
> + if (state) {
> + pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA,
> + PCI_EXP_SLTSTA_PDS);
> + pcie_cap_slot_event(d, PCI_EXP_HP_EV_PDC);
> + } else {
> + qdev_free(&pci_dev->qdev);
> + pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTSTA,
> + PCI_EXP_SLTSTA_PDS);
> + pcie_cap_slot_event(d, PCI_EXP_HP_EV_PDC);
> + }
> + return 0;
> +}
> +
> +/* pci express slot for pci express root/downstream port
> + PCI express capability slot registers */
> +void pcie_cap_slot_init(PCIDevice *dev, uint16_t slot)
> +{
> + uint32_t pos = dev->exp.exp_cap;
> +
> + pci_word_test_and_set_mask(dev->config + pos + PCI_EXP_FLAGS,
> + PCI_EXP_FLAGS_SLOT);
> +
> + pci_long_test_and_clear_mask(dev->config + pos + PCI_EXP_SLTCAP,
> + ~PCI_EXP_SLTCAP_PSN);
> + pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_SLTCAP,
> + (slot << PCI_EXP_SLTCAP_PSN_SHIFT) |
> + PCI_EXP_SLTCAP_EIP |
> + PCI_EXP_SLTCAP_HPS |
> + PCI_EXP_SLTCAP_HPC |
> + PCI_EXP_SLTCAP_PIP |
> + PCI_EXP_SLTCAP_AIP |
> + PCI_EXP_SLTCAP_ABP);
> +
> + pci_word_test_and_clear_mask(dev->config + pos + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_PIC |
> + PCI_EXP_SLTCTL_AIC);
> + pci_word_test_and_set_mask(dev->config + pos + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_PIC_OFF |
> + PCI_EXP_SLTCTL_AIC_OFF);
> + pci_word_test_and_set_mask(dev->wmask + pos + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_PIC |
> + PCI_EXP_SLTCTL_AIC |
> + PCI_EXP_SLTCTL_HPIE |
> + PCI_EXP_SLTCTL_CCIE |
> + PCI_EXP_SLTCTL_PDCE |
> + PCI_EXP_SLTCTL_ABPE);
> + /* Although reading PCI_EXP_SLTCTL_EIC returns always 0,
> + * make the bit writable here in order to detect 1b is written.
> + * pcie_cap_slot_write_config() test-and-clear the bit, so
> + * this bit always returns 0 to the guest.
> + */
> + pci_word_test_and_set_mask(dev->wmask + pos + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_EIC);
> +
> + pci_word_test_and_set_mask(dev->w1cmask + pos + PCI_EXP_SLTSTA,
> + PCI_EXP_HP_EV_SUPPORTED);
> +
> + pci_bus_hotplug(pci_bridge_get_sec_bus(DO_UPCAST(PCIBridge, dev, dev)),
> + pcie_cap_slot_hotplug, &dev->qdev);
> +}
> +
> +void pcie_cap_slot_reset(PCIDevice *dev)
> +{
> + uint8_t *exp_cap = dev->config + dev->exp.exp_cap;
> +
> + PCIE_DEV_PRINTF(dev, "reset\n");
> +
> + pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_EIC |
> + PCI_EXP_SLTCTL_PIC |
> + PCI_EXP_SLTCTL_AIC |
> + PCI_EXP_SLTCTL_HPIE |
> + PCI_EXP_SLTCTL_CCIE |
> + PCI_EXP_SLTCTL_PDCE |
> + PCI_EXP_SLTCTL_ABPE);
> + pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_PIC_OFF |
> + PCI_EXP_SLTCTL_AIC_OFF);
> +
> + pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTSTA,
> + PCI_EXP_SLTSTA_EIS |/* on reset,
> + the lock is released */
> + PCI_EXP_SLTSTA_CC |
> + PCI_EXP_SLTSTA_PDC |
> + PCI_EXP_SLTSTA_ABP);
> +}
> +
> +void pcie_cap_slot_write_config(PCIDevice *dev,
> + uint32_t addr, uint32_t val, int len,
> + uint16_t sltctl_prev)
> +{
> + uint32_t pos = dev->exp.exp_cap;
> + uint8_t *exp_cap = dev->config + pos;
> + uint16_t sltctl = pci_get_word(exp_cap + PCI_EXP_SLTCTL);
> + uint16_t sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);
> +
> + PCIE_DEV_PRINTF(dev,
> + "addr: 0x%"PRIx32" val: 0x%"PRIx32" len: %d\n"
> + "\tsltctl_prev: 0x%02"PRIx16" sltctl: 0x%02"PRIx16
> + " sltsta: 0x%02"PRIx16"\n",
> + addr, val, len, sltctl_prev, sltctl, sltsta);
> +
> + /* SLTCTL */
> + if (ranges_overlap(addr, len, pos + PCI_EXP_SLTCTL, 2)) {
> + PCIE_DEV_PRINTF(dev, "sltctl: 0x%02"PRIx16" -> 0x%02"PRIx16"\n",
> + sltctl_prev, sltctl);
> + if (pci_word_test_and_clear_mask(exp_cap + PCI_EXP_SLTCTL,
> + PCI_EXP_SLTCTL_EIC)) {
> + sltsta ^= PCI_EXP_SLTSTA_EIS; /* toggle PCI_EXP_SLTSTA_EIS bit */
> + pci_set_word(exp_cap + PCI_EXP_SLTSTA, sltsta);
> + PCIE_DEV_PRINTF(dev, "PCI_EXP_SLTCTL_EIC: "
> + "sltsta -> 0x%02"PRIx16"\n",
> + sltsta);
> + }
> +
> + /*
> + * The events control bits might be enabled or disabled,
> + * Check if the software notificastion condition is satisfied
> + * or disatisfied.
> + *
> + * 6.7.3.4 Software Notification of Hot-plug events
> + */
> + if (pci_msi_enabled(dev)) {
> + bool msi_trigger =
> + (sltctl & PCI_EXP_SLTCTL_HPIE) &&
> + ((sltctl_prev ^ sltctl) & sltctl & /* stlctl: 0 -> 1 */
> + sltsta & PCI_EXP_HP_EV_SUPPORTED);
> + if (msi_trigger) {
> + pci_msi_notify(dev, pcie_cap_flags_get_vector(dev));
> + }
> + } else {
> + int int_level =
> + (sltctl & PCI_EXP_SLTCTL_HPIE) &&
> + (sltctl & sltsta & PCI_EXP_HP_EV_SUPPORTED);
> + qemu_set_irq(dev->irq[dev->exp.hpev_intx], int_level);
> + }
> +
> + if (!((sltctl_prev ^ sltctl) & PCI_EXP_SLTCTL_SUPPORTED)) {
> + PCIE_DEV_PRINTF(dev,
> + "sprious command completion slctl "
> + "0x%"PRIx16" -> 0x%"PRIx16"\n",
> + sltctl_prev, sltctl);
> + }
> +
> + /* command completion.
> + * Real hardware might take a while to complete
> + * requested command because physical movement would be involved
> + * like locking the electromechanical lock.
> + * However in our case, command is completed instantaneously above,
> + * so send a command completion event right now.
> + *
> + * 6.7.3.2 Command Completed Events
> + */
> + /* set command completed bit */
> + pcie_cap_slot_event(dev, PCI_EXP_HP_EV_CCI);
> + }
> +}
> +
> +void pcie_cap_slot_push_attention_button(PCIDevice *dev)
> +{
> + pcie_cap_slot_event(dev, PCI_EXP_HP_EV_ABP);
> +}
> +
> +/* root control/capabilities/status. PME isn't emulated for now */
> +void pcie_cap_root_init(PCIDevice *dev)
> +{
> + pci_set_word(dev->wmask + dev->exp.exp_cap + PCI_EXP_RTCTL,
> + PCI_EXP_RTCTL_SECEE | PCI_EXP_RTCTL_SENFEE |
> + PCI_EXP_RTCTL_SEFEE);
> +}
> +
> +void pcie_cap_root_reset(PCIDevice *dev)
> +{
> + pci_set_word(dev->config + dev->exp.exp_cap + PCI_EXP_RTCTL, 0);
> +}
> +
> +/*
> + * TODO: implement FLR:
> + * Right now sets the bit which indicates FLR is supported.
> + */
> +/* function level reset(FLR) */
> +void pcie_cap_flr_init(PCIDevice *dev)
> +{
> + pci_long_test_and_set_mask(dev->config + dev->exp.exp_cap + PCI_EXP_DEVCAP,
> + PCI_EXP_DEVCAP_FLR);
> +
> + /* Although reading BCR_FLR returns always 0,
> + * the bit is made writable here in order to detect the 1b is written
> + * pcie_cap_flr_write_config() test-and-clear the bit, so
> + * this bit always returns 0 to the guest.
> + */
> + pci_word_test_and_set_mask(dev->wmask + dev->exp.exp_cap + PCI_EXP_DEVCTL,
> + PCI_EXP_DEVCTL_BCR_FLR);
> +}
> +
> +void pcie_cap_flr_write_config(PCIDevice *dev,
> + uint32_t addr, uint32_t val, int len)
> +{
> + uint8_t *devctl = dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL;
> + if (pci_word_test_and_clear_mask(devctl, PCI_EXP_DEVCTL_BCR_FLR)) {
> + /* TODO: implement FLR */
> + }
> +}
> +
> +/* Alternative Routing-ID Interpretation (ARI) */
> +/* ari forwarding support for down stream port */
> +void pcie_cap_ari_init(PCIDevice *dev)
> +{
> + uint32_t pos = dev->exp.exp_cap;
> + pci_long_test_and_set_mask(dev->config + pos + PCI_EXP_DEVCAP2,
> + PCI_EXP_DEVCAP2_ARI);
> + pci_long_test_and_set_mask(dev->wmask + pos + PCI_EXP_DEVCTL2,
> + PCI_EXP_DEVCTL2_ARI);
> +}
> +
> +void pcie_cap_ari_reset(PCIDevice *dev)
> +{
> + uint8_t *devctl2 = dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL2;
> + pci_long_test_and_clear_mask(devctl2, PCI_EXP_DEVCTL2_ARI);
> +}
> +
> +bool pcie_cap_is_ari_enabled(const PCIDevice *dev)
> +{
> + if (!pci_is_express(dev)) {
> + return false;
> + }
> + if (!dev->exp.exp_cap) {
> + return false;
> + }
> +
> + return pci_get_long(dev->config + dev->exp.exp_cap + PCI_EXP_DEVCTL2) &
> + PCI_EXP_DEVCTL2_ARI;
> +}
> +
> +/**************************************************************************
> + * pci express extended capability allocation functions
> + * uint16_t ext_cap_id (16 bit)
> + * uint8_t cap_ver (4 bit)
> + * uint16_t cap_offset (12 bit)
> + * uint16_t ext_cap_size
> + */
> +
> +static uint16_t pcie_find_capability_list(PCIDevice *dev, uint16_t cap_id,
> + uint16_t *prev_p)
> +{
> + uint16_t prev = 0;
> + uint16_t next;
> + uint32_t header = pci_get_long(dev->config + PCI_CONFIG_SPACE_SIZE);
> +
> + if (!header) {
> + /* no extended capability */
> + next = 0;
> + goto out;
> + }
> + for (next = PCI_CONFIG_SPACE_SIZE; next;
> + prev = next, next = PCI_EXT_CAP_NEXT(header)) {
> +
> + assert(next >= PCI_CONFIG_SPACE_SIZE);
> + assert(next <= PCIE_CONFIG_SPACE_SIZE - 8);
> +
> + header = pci_get_long(dev->config + next);
> + if (PCI_EXT_CAP_ID(header) == cap_id) {
> + break;
> + }
> + }
> +
> +out:
> + if (prev_p) {
> + *prev_p = prev;
> + }
> + return next;
> +}
> +
> +uint16_t pcie_find_capability(PCIDevice *dev, uint16_t cap_id)
> +{
> + return pcie_find_capability_list(dev, cap_id, NULL);
> +}
> +
> +static void pcie_ext_cap_set_next(PCIDevice *dev, uint16_t pos, uint16_t next)
> +{
> + uint16_t header = pci_get_long(dev->config + pos);
> + assert(!(next & (PCI_EXT_CAP_ALIGN - 1)));
> + header = (header & ~PCI_EXT_CAP_NEXT_MASK) |
> + ((next << PCI_EXT_CAP_NEXT_SHIFT) & PCI_EXT_CAP_NEXT_MASK);
> + pci_set_long(dev->config + pos, header);
> +}
> +
> +/*
> + * caller must supply valid (offset, size) * such that the range shouldn't
> + * overlap with other capability or other registers.
> + * This function doesn't check it.
> + */
> +void pcie_add_capability(PCIDevice *dev,
> + uint16_t cap_id, uint8_t cap_ver,
> + uint16_t offset, uint16_t size)
> +{
> + uint32_t header;
> + uint16_t next;
> +
> + assert(offset >= PCI_CONFIG_SPACE_SIZE);
> + assert(offset < offset + size);
> + assert(offset + size < PCIE_CONFIG_SPACE_SIZE);
> + assert(size >= 8);
> + assert(pci_is_express(dev));
> +
> + if (offset == PCI_CONFIG_SPACE_SIZE) {
> + header = pci_get_long(dev->config + offset);
> + next = PCI_EXT_CAP_NEXT(header);
> + } else {
> + uint16_t prev;
> +
> + /* 0 is reserved cap id. use internally to find the last capability
> + in the linked list */
> + next = pcie_find_capability_list(dev, 0, &prev);
> +
> + assert(prev >= PCI_CONFIG_SPACE_SIZE);
> + assert(next == 0);
> + pcie_ext_cap_set_next(dev, prev, offset);
> + }
> + pci_set_long(dev->config + offset, PCI_EXT_CAP(cap_id, cap_ver, next));
> +
> + /* Make capability read-only by default */
> + memset(dev->wmask + offset, 0, size);
> + memset(dev->w1cmask + offset, 0, size);
> + /* Check capability by default */
> + memset(dev->cmask + offset, 0xFF, size);
> +}
> +
> +/**************************************************************************
> + * pci express extended capability helper functions
> + */
> +
> +/* ARI */
> +void pcie_ari_init(PCIDevice *dev, uint16_t offset, uint16_t nextfn)
> +{
> + pcie_add_capability(dev, PCI_EXT_CAP_ID_ARI, PCI_ARI_VER,
> + offset, PCI_ARI_SIZEOF);
> + pci_set_long(dev->config + offset + PCI_ARI_CAP, PCI_ARI_CAP_NFN(nextfn));
> +}
> diff --git a/hw/pcie.h b/hw/pcie.h
> new file mode 100644
> index 0000000..68327d8
> --- /dev/null
> +++ b/hw/pcie.h
> @@ -0,0 +1,107 @@
> +/*
> + * pcie.h
> + *
> + * Copyright (c) 2010 Isaku Yamahata <yamahata at valinux co jp>
> + * VA Linux Systems Japan K.K.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option) any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License along
> + * with this program; if not, see <http://www.gnu.org/licenses/>.
> + */
> +
> +#ifndef QEMU_PCIE_H
> +#define QEMU_PCIE_H
> +
> +#include "hw.h"
> +#include "pci_regs.h"
> +#include "pcie_regs.h"
> +
> +typedef enum {
> + /* for attention and power indicator */
> + PCI_EXP_HP_IND_RESERVED = PCI_EXP_SLTCTL_IND_RESERVED,
> + PCI_EXP_HP_IND_ON = PCI_EXP_SLTCTL_IND_ON,
> + PCI_EXP_HP_IND_BLINK = PCI_EXP_SLTCTL_IND_BLINK,
> + PCI_EXP_HP_IND_OFF = PCI_EXP_SLTCTL_IND_OFF,
> +} PCIExpressIndicator;
> +
> +typedef enum {
> + /* these bits must match the bits in Slot Control/Status registers.
> + * PCI_EXP_HP_EV_xxx = PCI_EXP_SLTCTL_xxxE = PCI_EXP_SLTSTA_xxx
> + *
> + * Not all the bits of slot control register match with the ones of
> + * slot status. Not some bits of slot status register is used to
> + * show status, not to report event occurence.
> + * So such bits must be masked out when checking the software
> + * notification condition.
> + */
> + PCI_EXP_HP_EV_ABP = PCI_EXP_SLTCTL_ABPE,
> + /* attention button pressed */
> + PCI_EXP_HP_EV_PDC = PCI_EXP_SLTCTL_PDCE,
> + /* presence detect changed */
> + PCI_EXP_HP_EV_CCI = PCI_EXP_SLTCTL_CCIE,
> + /* command completed */
> +
> + PCI_EXP_HP_EV_SUPPORTED = PCI_EXP_HP_EV_ABP |
> + PCI_EXP_HP_EV_PDC |
> + PCI_EXP_HP_EV_CCI,
> + /* supported event mask */
> +
> + /* events not listed aren't supported */
> +} PCIExpressHotPlugEvent;
> +
> +struct PCIExpressDevice {
> + /* Offset of express capability in config space */
> + uint8_t exp_cap;
> +
> + /* TODO FLR */
> +
> + /* SLOT */
> + unsigned int hpev_intx; /* INTx for hot plug event */
> +};
> +
> +/* PCI express capability helper functions */
> +int pcie_cap_init(PCIDevice *dev, uint8_t offset, uint8_t type, uint8_t port);
> +void pcie_cap_exit(PCIDevice *dev);
> +uint8_t pcie_cap_get_type(const PCIDevice *dev);
> +void pcie_cap_flags_set_vector(PCIDevice *dev, uint8_t vector);
> +uint8_t pcie_cap_flags_get_vector(PCIDevice *dev);
> +
> +void pcie_cap_deverr_init(PCIDevice *dev);
> +void pcie_cap_deverr_reset(PCIDevice *dev);
> +
> +void pcie_cap_slot_init(PCIDevice *dev, uint16_t slot);
> +void pcie_cap_slot_reset(PCIDevice *dev);
> +void pcie_cap_slot_write_config(PCIDevice *dev,
> + uint32_t addr, uint32_t val, int len,
> + uint16_t sltctl_prev);
> +void pcie_cap_slot_push_attention_button(PCIDevice *dev);
> +
> +void pcie_cap_root_init(PCIDevice *dev);
> +void pcie_cap_root_reset(PCIDevice *dev);
> +
> +void pcie_cap_flr_init(PCIDevice *dev);
> +void pcie_cap_flr_write_config(PCIDevice *dev,
> + uint32_t addr, uint32_t val, int len);
> +
> +void pcie_cap_ari_init(PCIDevice *dev);
> +void pcie_cap_ari_reset(PCIDevice *dev);
> +bool pcie_cap_is_ari_enabled(const PCIDevice *dev);
> +
> +/* PCI express extended capability helper functions */
> +uint16_t pcie_find_capability(PCIDevice *dev, uint16_t cap_id);
> +void pcie_add_capability(PCIDevice *dev,
> + uint16_t cap_id, uint8_t cap_ver,
> + uint16_t offset, uint16_t size);
> +
> +void pcie_ari_init(PCIDevice *dev, uint16_t offset, uint16_t nextfn);
> +
> +#endif /* QEMU_PCIE_H */
> diff --git a/qemu-common.h b/qemu-common.h
> index d735235..6d9ee26 100644
> --- a/qemu-common.h
> +++ b/qemu-common.h
> @@ -219,6 +219,7 @@ typedef struct PCIHostState PCIHostState;
> typedef struct PCIExpressHost PCIExpressHost;
> typedef struct PCIBus PCIBus;
> typedef struct PCIDevice PCIDevice;
> +typedef struct PCIExpressDevice PCIExpressDevice;
> typedef struct PCIBridge PCIBridge;
> typedef struct SerialState SerialState;
> typedef struct IRQState *qemu_irq;
> --
> 1.7.1.1
next prev parent reply other threads:[~2010-10-19 12:00 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-19 9:06 [Qemu-devel] [PATCH v5 00/14] pcie port switch emulators Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 01/14] pci: introduce helper functions to test-and-{clear, set} mask in configuration space Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 02/14] pci: introduce helper function to handle msi-x and msi Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 03/14] pci: use pci_word_test_and_clear_mask() in pci_device_reset() Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 04/14] pci/bridge: fix pci_bridge_reset() Isaku Yamahata
2010-10-19 12:07 ` [Qemu-devel] " Michael S. Tsirkin
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 05/14] msi: implements msi Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 06/14] pcie: add pcie constants to pcie_regs.h Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 07/14] pcie: helper functions for pcie capability and extended capability Isaku Yamahata
2010-10-19 11:53 ` Michael S. Tsirkin [this message]
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 08/14] pcie/aer: helper functions for pcie aer capability Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 09/14] pcie port: define struct PCIEPort/PCIESlot and helper functions Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 10/14] ioh3420: pcie root port in X58 ioh Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 11/14] x3130: pcie upstream port Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 12/14] x3130: pcie downstream port Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 13/14] pcie/hotplug: introduce pushing attention button command Isaku Yamahata
2010-10-19 9:06 ` [Qemu-devel] [PATCH v5 14/14] pcie/aer: glue aer error injection into qemu monitor Isaku Yamahata
2010-10-19 11:51 ` [Qemu-devel] Re: [PATCH v5 00/14] pcie port switch emulators Michael S. Tsirkin
2010-10-19 15:19 ` Isaku Yamahata
2010-10-19 17:06 ` Michael S. Tsirkin
2010-10-19 22:36 ` Isaku Yamahata
2010-10-19 22:40 ` Michael S. Tsirkin
2010-10-19 22:55 ` Isaku Yamahata
2010-10-19 23:02 ` Michael S. Tsirkin
2010-10-19 11:56 ` Michael S. Tsirkin
2010-10-19 13:55 ` Isaku Yamahata
2010-10-19 13:00 ` Michael S. Tsirkin
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=20101019115356.GB6752@redhat.com \
--to=mst@redhat.com \
--cc=adnan@khaleel.us \
--cc=etmartin@cisco.com \
--cc=qemu-devel@nongnu.org \
--cc=skandasa@cisco.com \
--cc=wexu2@cisco.com \
--cc=yamahata@valinux.co.jp \
/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.