qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Isaku Yamahata <yamahata@valinux.co.jp>
Cc: qemu-devel@nongnu.org
Subject: [Qemu-devel] Re: [PATCH 21/25] pci: make bar update function aware of pci bridge.
Date: Sun, 4 Oct 2009 12:53:45 +0200	[thread overview]
Message-ID: <20091004105345.GO16887@redhat.com> (raw)
In-Reply-To: <1254514577-11896-22-git-send-email-yamahata@valinux.co.jp>

On Sat, Oct 03, 2009 at 05:16:13AM +0900, Isaku Yamahata wrote:
> header type of 01 has differenct BAR to type 00.
> It has only BAR0,1 and expantion rom whose offset address
> is different from type 00 one.

There are some typos in the comment

> Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
> ---
>  hw/pci.c |   66 +++++++++++++++++++++++++++++++++++++++++++++++++++----------
>  hw/pci.h |    1 +
>  2 files changed, 56 insertions(+), 11 deletions(-)
> 
> diff --git a/hw/pci.c b/hw/pci.c
> index ed9b785..b708d71 100644
> --- a/hw/pci.c
> +++ b/hw/pci.c
> @@ -498,6 +498,35 @@ int pci_unregister_device(PCIDevice *pci_dev)
>      return 0;
>  }
>  
> +#define PCI_BAR_INVALID  UINT32_MAX

Just use -1 to report error.

> +static uint32_t pci_bar_config_offset(PCIDevice *d, int region_num)

Why uint32_t? I think it should be just int.
Please add small comment documenting what function does.

> +{

This functions does too much sanity checking, which is where
all the complexity comes from.  It's not the best place to do
sanity checking: do it when adding region, if you must.

So this should be simply:

static int pci_bar_config_offset(PCIDevice *d, int region_num)
{
	if (region_num != PCI_ROM_ADDRESS)
		return PCI_BASE_ADDRESS_0 + region_num * 4;

	return (d->config[PCI_HEADER_TYPE] & PCI_HEADER_TYPE_BRIDGE) ?
		PCI_ROM_ADDRESS : PCI_BRIDGE_ROM_ADDRESS;
}


> +    switch (d->config[PCI_HEADER_TYPE] & ~PCI_HEADER_TYPE_MULTI_FUNCTION) {
> +    case PCI_HEADER_TYPE_NORMAL:
> +        /* BAR 0-5 and Expantion ROM*/

Nit: space before */

> +        if (region_num < PCI_ROM_SLOT) {
> +            return PCI_BASE_ADDRESS_0 + region_num * 4;
> +        } else if (region_num == PCI_ROM_SLOT) {
> +            return PCI_ROM_ADDRESS;
> +        }
> +        break;
> +    case PCI_HEADER_TYPE_BRIDGE:
> +        /* BAR 0-1 and Expantion ROM */

typo

> +        if (region_num < 2) {

what's 2?

> +            return PCI_BASE_ADDRESS_0 + region_num * 4;
> +        } else if (region_num == PCI_ROM_SLOT) {
> +            return PCI_ROM_ADDRESS1;
> +        }
> +        break;
> +    case PCI_HEADER_TYPE_CARDBUS:

the last line is useless

> +    default:
> +        break;
> +    }
> +    fprintf(stderr, "ERROR: %s: unknow PCI config header type %d or bar %d\n",
> +            __func__, d->config[PCI_HEADER_TYPE], region_num);
> +    return PCI_BAR_INVALID;

Caller does not seem to check PCI_BAR_INVALID.

> +}
> +
>  void pci_register_bar(PCIDevice *pci_dev, int region_num,
>                              pcibus_t size, int type,
>                              PCIMapIORegionFunc *map_func)
> @@ -521,13 +550,11 @@ void pci_register_bar(PCIDevice *pci_dev, int region_num,
>      r->type = type;
>      r->map_func = map_func;
>  
> +    addr = pci_bar_config_offset(pci_dev, region_num);
>      wmask = ~(size - 1);
>      if (region_num == PCI_ROM_SLOT) {
> -        addr = 0x30;
>          /* ROM enable bit is writeable */
>          wmask |= PCI_ROM_ADDRESS_ENABLE;
> -    } else {
> -        addr = 0x10 + region_num * 4;
>      }
>      pci_set_long(pci_dev->config + addr, type);
>      if (pci_bar_is_64bit(r)) {
> @@ -549,11 +576,9 @@ static void pci_update_mappings(PCIDevice *d)
>      cmd = pci_get_word(d->config + PCI_COMMAND);
>      for(i = 0; i < PCI_NUM_REGIONS; i++) {
>          r = &d->io_regions[i];
> -        if (i == PCI_ROM_SLOT) {
> -            config_ofs = 0x30;
> -        } else {
> -            config_ofs = 0x10 + i * 4;
> -        }
> +

kill extra empty line

> +        config_ofs = pci_bar_config_offset(d, i);
> +
>          if (r->size != 0) {
>              if (r->type & PCI_ADDRESS_SPACE_IO) {
>                  if (cmd & PCI_COMMAND_IO) {
> @@ -1133,10 +1158,29 @@ static void pci_bridge_write_config(PCIDevice *d,
>                               uint32_t address, uint32_t val, int len)
>  {
>      PCIBridge *s = (PCIBridge *)d;
> +    PCIBus *bus = s->bus;
> +    uint8_t *orig = pci_write_config_init(d, address, len);
> +
> +    pci_default_write_config_common(d, address, val, len);
>  
> -    pci_default_write_config(d, address, val, len);
> -    s->bus->bus_num = d->config[PCI_SECONDARY_BUS];
> -    s->bus->sub_bus = d->config[PCI_SUBORDINATE_BUS];
> +    if (pci_config_changed(orig, d->config, address, len,
> +                           PCI_BASE_ADDRESS_0, PCI_BASE_ADDRESS_2 + 4) ||
> +        pci_config_changed_with_size(orig, d->config, address, len,
> +                                     PCI_ROM_ADDRESS1, 4) ||
> +        pci_config_changed_with_size(orig, d->config, address, len,
> +                                     PCI_COMMAND, 1)) {
> +        pci_update_mappings(d);
> +    }
> +    if (pci_config_changed_with_size(orig, d->config, address, len,
> +                                     PCI_SECONDARY_BUS, 1)) {
> +        bus->bus_num = d->config[PCI_SECONDARY_BUS];
> +    }
> +    if (pci_config_changed_with_size(orig, d->config, address, len,
> +                                     PCI_SECONDARY_BUS, 1)) {
> +        bus->sub_bus = d->config[PCI_SUBORDINATE_BUS];
> +    }

So simply
d->config[PCI_SUBORDINATE_BUS] != orig[PCI_SUBORDINATE_BUS]
would be enough if we keep orig and config in sync
as I suggested in my previous post.

> +
> +    pci_write_config_done(orig);
>  }
>  
>  PCIBus *pci_find_bus(PCIBus *bus, int bus_num)
> diff --git a/hw/pci.h b/hw/pci.h
> index f19b7a8..645dacd 100644
> --- a/hw/pci.h
> +++ b/hw/pci.h
> @@ -174,6 +174,7 @@ static inline int pci_bar_is_64bit(const PCIIORegion *r)
>  
>  /* Header type 1 (PCI-to-PCI bridges) */
>  #define PCI_SUBORDINATE_BUS     0x1a    /* Highest bus number behind the bridge */
> +#define PCI_ROM_ADDRESS1        0x38    /* Same as PCI_ROM_ADDRESS, but for htype 1 */

PCI_BRIDGE_ROM_ADDRESS would be clearer.
Also don't mix your own macros and ones imported from pci_regs.h

>  
>  /* Size of the standard PCI config header */
>  #define PCI_CONFIG_HEADER_SIZE 0x40
> -- 
> 1.6.0.2
> 
> 

  reply	other threads:[~2009-10-04 10:55 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-10-02 20:15 [Qemu-devel] [PATCH 00/25] pci: various pci clean up and pci express support. V3 Isaku Yamahata
2009-10-02 20:15 ` [Qemu-devel] [PATCH 01/25] pci: fix PCI_DPRINTF() wrt variadic macro Isaku Yamahata
2009-10-02 20:15 ` [Qemu-devel] [PATCH 02/25] pci: use appropriate PRIs in PCI_DPRINTF() for portability Isaku Yamahata
2009-10-04  9:51   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-05  9:30     ` Isaku Yamahata
2009-10-05  9:56       ` Michael S. Tsirkin
2009-10-02 20:15 ` [Qemu-devel] [PATCH 03/25] pci: introduce constant PCI_NUM_PINS for the number of interrupt pins, 4 Isaku Yamahata
2009-10-04 10:04   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-05  9:32     ` Isaku Yamahata
2009-10-02 20:15 ` [Qemu-devel] [PATCH 04/25] pci: use the symbolic constant, PCI_ROM_ADDRESS_ENABLE instead of 1 Isaku Yamahata
2009-10-04 10:06   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:15 ` [Qemu-devel] [PATCH 05/25] pci: use PCI_SLOT() and PCI_FUNC() Isaku Yamahata
2009-10-04 10:02   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:15 ` [Qemu-devel] [PATCH 06/25] pci: define a constant to represent a unmapped bar and use it Isaku Yamahata
2009-10-04 10:04   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:15 ` [Qemu-devel] [PATCH 07/25] pci: helper functions to access PCIDevice::config Isaku Yamahata
2009-10-04  9:48   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 08/25] pci: use helper functions to access pci config space Isaku Yamahata
2009-10-04  9:47   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 09/25] pci: introduce pcibus_t to represent pci bus address/size instead of uint32_t Isaku Yamahata
2009-10-04 10:06   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 10/25] pci: introduce FMT_pcibus for printf format for pcibus_t Isaku Yamahata
2009-10-04  9:46   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 11/25] pci: typedef pcibus_t as uint64_t instead of uint32_t Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 12/25] pci: 64bit bar support Isaku Yamahata
2009-10-04 10:26   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-05  9:45     ` Isaku Yamahata
2009-10-05 10:08       ` Michael S. Tsirkin
2009-10-05 10:26         ` Isaku Yamahata
2009-10-05 11:04           ` Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 13/25] pci: make pci configuration transaction more accurate Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 14/25] pci: factor out the logic to get pci device from address Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 15/25] pci_host.h: split non-inline static function in pci_host.h into pci_host.c Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 16/25] pci: pcie host and mmcfg support Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 17/25] pci: fix pci_default_write_config() Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 18/25] pci: add helper functions for pci config write function Isaku Yamahata
2009-10-04 10:30   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 19/25] pci: use helper function in pci_default_write_config() Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 20/25] pci: factor out config update logic Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 21/25] pci: make bar update function aware of pci bridge Isaku Yamahata
2009-10-04 10:53   ` Michael S. Tsirkin [this message]
2009-10-05  9:47     ` [Qemu-devel] " Isaku Yamahata
2009-10-05 10:10       ` Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 22/25] pci/brdige: qdevfy and initialize secondary bus and subordinate bus Isaku Yamahata
2009-10-04 11:04   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-05  9:51     ` Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 23/25] pci: add helper function to initialize wmask Isaku Yamahata
2009-10-04 11:05   ` [Qemu-devel] " Michael S. Tsirkin
2009-10-02 20:16 ` [Qemu-devel] [PATCH 24/25] pci: initialize wmask according to pci header type Isaku Yamahata
2009-10-02 20:16 ` [Qemu-devel] [PATCH 25/25] pci/monitor: print out bridge's filtering values and so on Isaku Yamahata
2009-10-04 11:10   ` [Qemu-devel] " 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=20091004105345.GO16887@redhat.com \
    --to=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --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 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).