All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: qemu-devel <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH v2 8/9] msix: Align MSI-X constants to libpci definitions and extend them
Date: Wed, 8 Jun 2011 22:53:16 +0300	[thread overview]
Message-ID: <20110608195316.GD30805@redhat.com> (raw)
In-Reply-To: <219d3206ad6eafed76d4f108a40912fe72884e19.1307550106.git.jan.kiszka@siemens.com>

On Wed, Jun 08, 2011 at 06:21:51PM +0200, Jan Kiszka wrote:
> Add PCI_MSIX_TABLE and PCI_MSIX_PBA, align other MSIX related constant
> names to libpci style. Will be used for device assignment code in
> qemu-kvm.
> 
> Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>

Besides keeping pci_regs.h aligned with the original,
I also think ideally pci register banging should stay
within the pci subsystem.

Could we add high-level APIs to help with that,
instead of having kvm look at config space directly?


> ---
>  hw/msix.c     |   24 +++++++++++-------------
>  hw/pci_regs.h |   14 ++++++++------
>  2 files changed, 19 insertions(+), 19 deletions(-)
> 
> diff --git a/hw/msix.c b/hw/msix.c
> index 600f5fb..b20cf7c 100644
> --- a/hw/msix.c
> +++ b/hw/msix.c
> @@ -16,15 +16,12 @@
>  #include "pci.h"
>  #include "range.h"
>  
> -/* MSI-X capability structure */
> -#define MSIX_TABLE_OFFSET 4
> -#define MSIX_PBA_OFFSET 8
>  #define MSIX_CAP_LENGTH 12
>  
> -/* MSI enable bit and maskall bit are in byte 1 in FLAGS register */
> -#define MSIX_CONTROL_OFFSET (PCI_MSIX_FLAGS + 1)
> -#define MSIX_ENABLE_MASK (PCI_MSIX_FLAGS_ENABLE >> 8)
> -#define MSIX_MASKALL_MASK (PCI_MSIX_FLAGS_MASKALL >> 8)
> +/* MSI enable bit and maskall bit are in byte 1 in control register */
> +#define MSIX_CONTROL_OFFSET (PCI_MSIX_CTRL + 1)
> +#define MSIX_ENABLE_MASK (PCI_MSIX_ENABLE >> 8)
> +#define MSIX_MASKALL_MASK (PCI_MSIX_MASK >> 8)
>  
>  /* MSI-X table format */
>  #define MSIX_MSG_ADDR 0
> @@ -58,8 +55,9 @@ static int msix_add_config(struct PCIDevice *pdev, unsigned short nentries,
>      uint8_t *config;
>      uint32_t new_size;
>  
> -    if (nentries < 1 || nentries > PCI_MSIX_FLAGS_QSIZE + 1)
> +    if (nentries < 1 || nentries > PCI_MSIX_TABSIZE + 1) {
>          return -EINVAL;
> +    }
>      if (bar_size > 0x80000000)
>          return -ENOSPC;
>  
> @@ -80,11 +78,11 @@ static int msix_add_config(struct PCIDevice *pdev, unsigned short nentries,
>          return config_offset;
>      config = pdev->config + config_offset;
>  
> -    pci_set_word(config + PCI_MSIX_FLAGS, nentries - 1);
> +    pci_set_word(config + PCI_MSIX_CTRL, nentries - 1);
>      /* Table on top of BAR */
> -    pci_set_long(config + MSIX_TABLE_OFFSET, bar_size | bar_nr);
> +    pci_set_long(config + PCI_MSIX_TABLE, bar_size | bar_nr);
>      /* Pending bits on top of that */
> -    pci_set_long(config + MSIX_PBA_OFFSET, (bar_size + MSIX_PAGE_PENDING) |
> +    pci_set_long(config + PCI_MSIX_PBA, (bar_size + MSIX_PAGE_PENDING) |
>                   bar_nr);
>      pdev->msix_cap = config_offset;
>      /* Make flags bit writable. */
> @@ -208,11 +206,11 @@ void msix_mmio_map(PCIDevice *d, int region_num,
>                     pcibus_t addr, pcibus_t size, int type)
>  {
>      uint8_t *config = d->config + d->msix_cap;
> -    uint32_t table = pci_get_long(config + MSIX_TABLE_OFFSET);
> +    uint32_t table = pci_get_long(config + PCI_MSIX_TABLE);
>      uint32_t offset = table & ~(MSIX_PAGE_SIZE - 1);
>      /* TODO: for assigned devices, we'll want to make it possible to map
>       * pending bits separately in case they are in a separate bar. */
> -    int table_bir = table & PCI_MSIX_FLAGS_BIRMASK;
> +    int table_bir = table & PCI_MSIX_BIR;
>  
>      if (table_bir != region_num)
>          return;
> diff --git a/hw/pci_regs.h b/hw/pci_regs.h
> index 5a5ab89..c17c22f 100644
> --- a/hw/pci_regs.h
> +++ b/hw/pci_regs.h
> @@ -300,12 +300,14 @@
>  #define PCI_MSI_DATA_64		12	/* 16 bits of data for 64-bit devices */
>  #define PCI_MSI_MASK_64		16	/* Mask bits register for 64-bit devices */
>  
> -/* MSI-X registers (these are at offset PCI_MSIX_FLAGS) */
> -#define PCI_MSIX_FLAGS		2
> -#define  PCI_MSIX_FLAGS_QSIZE	0x7FF
> -#define  PCI_MSIX_FLAGS_ENABLE	(1 << 15)
> -#define  PCI_MSIX_FLAGS_MASKALL	(1 << 14)
> -#define PCI_MSIX_FLAGS_BIRMASK	(7 << 0)
> +/* MSI-X registers */
> +#define PCI_MSIX_CTRL           2       /* Message control */
> +#define  PCI_MSIX_TABSIZE       0x7FF   /* Table size - 1 */
> +#define  PCI_MSIX_MASK          0x4000  /* Mask all vectors */
> +#define  PCI_MSIX_ENABLE        0x8000  /* Enable MSI-X */
> +#define PCI_MSIX_TABLE          4       /* MSI-X table */
> +#define PCI_MSIX_PBA            8       /* Pending bit array */
> +#define  PCI_MSIX_BIR           0x7     /* BAR indication register */
>  
>  /* CompactPCI Hotswap Register */
>  
> -- 
> 1.7.1

  parent reply	other threads:[~2011-06-08 19:53 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-08 16:21 [Qemu-devel] [PATCH v2 0/9] msi: Small cleanups and fixes Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 1/9] msi: Fix copy&paste mistake in msi_uninit Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 2/9] msi: Guard msi/msix_write_config with msi_present Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 3/9] msi: Guard msi_reset " Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 4/9] msi: Use msi/msix_present more consistently Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 5/9] msi: Invoke msi/msix_reset from PCI core Jan Kiszka
2011-06-08 19:59   ` Michael S. Tsirkin
2011-06-08 20:47     ` Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 6/9] msi: Invoke msi/msix_write_config " Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 7/9] msi: Invoke msi/msix_uninit " Jan Kiszka
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 8/9] msix: Align MSI-X constants to libpci definitions and extend them Jan Kiszka
2011-06-08 19:46   ` Michael S. Tsirkin
2011-06-08 19:53   ` Michael S. Tsirkin [this message]
2011-06-08 20:48     ` Jan Kiszka
2011-06-08 21:00       ` Michael S. Tsirkin
2011-06-08 21:02         ` Jan Kiszka
2011-06-08 21:09           ` Michael S. Tsirkin
2011-06-08 21:11             ` Jan Kiszka
2011-06-08 21:15               ` Michael S. Tsirkin
2011-06-08 16:21 ` [Qemu-devel] [PATCH v2 9/9] msi: Move PCI_MSI_PENDING_32/64 into pci_regs.h Jan Kiszka
2011-06-08 19:48   ` Michael S. Tsirkin
2011-06-08 20:44     ` Jan Kiszka
2011-06-08 20:56       ` Michael S. Tsirkin
2011-06-08 20:57         ` Jan Kiszka
2011-06-08 21:01           ` Michael S. Tsirkin
2011-06-08 21:03             ` Jan Kiszka
2011-06-08 21:14               ` Michael S. Tsirkin
2011-06-08 21:18                 ` Jan Kiszka
2011-06-08 21:24                   ` Jan Kiszka
2011-06-08 21:30                   ` 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=20110608195316.GD30805@redhat.com \
    --to=mst@redhat.com \
    --cc=jan.kiszka@siemens.com \
    --cc=qemu-devel@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.