From: Bjorn Helgaas <helgaas@kernel.org>
To: Sui Jingfeng <sui.jingfeng@linux.dev>
Cc: Bjorn Helgaas <bhelgaas@google.com>,
Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
Maxime Ripard <mripard@kernel.org>,
Thomas Zimmermann <tzimmermann@suse.de>,
David Airlie <airlied@gmail.com>, Daniel Vetter <daniel@ffwll.ch>,
Sui@freedesktop.org, Jingfeng@loongson.cn,
Sui Jingfeng <suijingfeng@loongson.cn>,
linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
dri-devel@lists.freedesktop.org,
loongson-kernel@lists.loongnix.cn,
Mario Limonciello <mario.limonciello@amd.com>
Subject: Re: [PATCH 2/6] PCI/VGA: Deal with PCI VGA compatible devices only
Date: Wed, 19 Jul 2023 13:26:17 -0500 [thread overview]
Message-ID: <20230719182617.GA509912@bhelgaas> (raw)
In-Reply-To: <20230711134354.755966-3-sui.jingfeng@linux.dev>
On Tue, Jul 11, 2023 at 09:43:50PM +0800, Sui Jingfeng wrote:
> From: Sui Jingfeng <suijingfeng@loongson.cn>
>
> Currently, vgaarb only cares about PCI VGA-compatible class devices.
>
> While vga_arbiter_del_pci_device() gets called unbalanced when some PCI
> device is about to be removed. This happens even during the boot process.
The previous code calls vga_arbiter_add_pci_device() for every device
(every device present at boot and also every hot-added device). It
only allocates a vga_device if pdev->class is 0x0300XX.
It calls vga_arbiter_del_pci_device() for every device removal. It
does nothing unless it finds a vga_device.
This seems symmetric and reasonable to me. Did you observe a problem
with it?
> Another reason is that the vga_arb_device_init() function is not efficient.
> Since we only care about VGA-compatible devices (pdev->class == 0x030000),
> We could filter the unqualified devices out in the vga_arb_device_init()
> function. While the current implementation is to search all PCI devices
> in a system, this is not necessary.
Optimization is fine, but the most important thing here is to be clear
about what functional change this patch makes. As I mentioned at [1],
if this patch affects the class codes accepted, please make that clear
here.
> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
> Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
I do not see Mario's Reviewed-by on the list. I do see Mario's
Reviewed-by [2] for a previous version, but that version added this in
pci_notify():
+ if (pdev->class != PCI_CLASS_DISPLAY_VGA << 8)
+ return 0;
while this version adds:
+ if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
+ return 0;
It's OK to carry a review to future versions if there are
insignificant changes, but this is a functional change that seems
significant to me. The first matches only 0x030000, while the second
discards the low eight bits so it matches 0x0300XX.
[1] https://lore.kernel.org/r/20230718231400.GA496927@bhelgaas
[2] https://lore.kernel.org/all/5b6fdf65-b354-94a9-f883-be820157efad@amd.com/
> ---
> drivers/pci/vgaarb.c | 25 +++++++++++++------------
> 1 file changed, 13 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/pci/vgaarb.c b/drivers/pci/vgaarb.c
> index c1bc6c983932..021116ed61cb 100644
> --- a/drivers/pci/vgaarb.c
> +++ b/drivers/pci/vgaarb.c
> @@ -754,10 +754,6 @@ static bool vga_arbiter_add_pci_device(struct pci_dev *pdev)
> struct pci_dev *bridge;
> u16 cmd;
>
> - /* Only deal with VGA class devices */
> - if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
> - return false;
> -
> /* Allocate structure */
> vgadev = kzalloc(sizeof(struct vga_device), GFP_KERNEL);
> if (vgadev == NULL) {
> @@ -1502,6 +1498,10 @@ static int pci_notify(struct notifier_block *nb, unsigned long action,
>
> vgaarb_dbg(dev, "%s\n", __func__);
>
> + /* Deal with VGA compatible devices only */
> + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
> + return 0;
> +
> /* For now we're only intereted in devices added and removed. I didn't
> * test this thing here, so someone needs to double check for the
> * cases of hotplugable vga cards. */
> @@ -1534,8 +1534,8 @@ static struct miscdevice vga_arb_device = {
>
> static int __init vga_arb_device_init(void)
> {
> + struct pci_dev *pdev = NULL;
> int rc;
> - struct pci_dev *pdev;
>
> rc = misc_register(&vga_arb_device);
> if (rc < 0)
> @@ -1543,13 +1543,14 @@ static int __init vga_arb_device_init(void)
>
> bus_register_notifier(&pci_bus_type, &pci_notifier);
>
> - /* We add all PCI devices satisfying VGA class in the arbiter by
> - * default */
> - pdev = NULL;
> - while ((pdev =
> - pci_get_subsys(PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
> - PCI_ANY_ID, pdev)) != NULL)
> - vga_arbiter_add_pci_device(pdev);
> + /*
> + * We add all PCI VGA compatible devices in the arbiter by default
> + */
> + do {
> + pdev = pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, pdev);
> + if (pdev)
> + vga_arbiter_add_pci_device(pdev);
> + } while (pdev);
>
> pr_info("loaded\n");
> return rc;
> --
> 2.25.1
>
WARNING: multiple messages have this Message-ID (diff)
From: Bjorn Helgaas <helgaas@kernel.org>
To: Sui Jingfeng <sui.jingfeng@linux.dev>
Cc: loongson-kernel@lists.loongnix.cn, Jingfeng@loongson.cn,
Sui Jingfeng <suijingfeng@loongson.cn>,
Thomas Zimmermann <tzimmermann@suse.de>,
linux-pci@vger.kernel.org, linux-kernel@vger.kernel.org,
Maxime Ripard <mripard@kernel.org>,
Sui@freedesktop.org, dri-devel@lists.freedesktop.org,
Bjorn Helgaas <bhelgaas@google.com>,
Mario Limonciello <mario.limonciello@amd.com>
Subject: Re: [PATCH 2/6] PCI/VGA: Deal with PCI VGA compatible devices only
Date: Wed, 19 Jul 2023 13:26:17 -0500 [thread overview]
Message-ID: <20230719182617.GA509912@bhelgaas> (raw)
In-Reply-To: <20230711134354.755966-3-sui.jingfeng@linux.dev>
On Tue, Jul 11, 2023 at 09:43:50PM +0800, Sui Jingfeng wrote:
> From: Sui Jingfeng <suijingfeng@loongson.cn>
>
> Currently, vgaarb only cares about PCI VGA-compatible class devices.
>
> While vga_arbiter_del_pci_device() gets called unbalanced when some PCI
> device is about to be removed. This happens even during the boot process.
The previous code calls vga_arbiter_add_pci_device() for every device
(every device present at boot and also every hot-added device). It
only allocates a vga_device if pdev->class is 0x0300XX.
It calls vga_arbiter_del_pci_device() for every device removal. It
does nothing unless it finds a vga_device.
This seems symmetric and reasonable to me. Did you observe a problem
with it?
> Another reason is that the vga_arb_device_init() function is not efficient.
> Since we only care about VGA-compatible devices (pdev->class == 0x030000),
> We could filter the unqualified devices out in the vga_arb_device_init()
> function. While the current implementation is to search all PCI devices
> in a system, this is not necessary.
Optimization is fine, but the most important thing here is to be clear
about what functional change this patch makes. As I mentioned at [1],
if this patch affects the class codes accepted, please make that clear
here.
> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
> Signed-off-by: Sui Jingfeng <suijingfeng@loongson.cn>
I do not see Mario's Reviewed-by on the list. I do see Mario's
Reviewed-by [2] for a previous version, but that version added this in
pci_notify():
+ if (pdev->class != PCI_CLASS_DISPLAY_VGA << 8)
+ return 0;
while this version adds:
+ if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
+ return 0;
It's OK to carry a review to future versions if there are
insignificant changes, but this is a functional change that seems
significant to me. The first matches only 0x030000, while the second
discards the low eight bits so it matches 0x0300XX.
[1] https://lore.kernel.org/r/20230718231400.GA496927@bhelgaas
[2] https://lore.kernel.org/all/5b6fdf65-b354-94a9-f883-be820157efad@amd.com/
> ---
> drivers/pci/vgaarb.c | 25 +++++++++++++------------
> 1 file changed, 13 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/pci/vgaarb.c b/drivers/pci/vgaarb.c
> index c1bc6c983932..021116ed61cb 100644
> --- a/drivers/pci/vgaarb.c
> +++ b/drivers/pci/vgaarb.c
> @@ -754,10 +754,6 @@ static bool vga_arbiter_add_pci_device(struct pci_dev *pdev)
> struct pci_dev *bridge;
> u16 cmd;
>
> - /* Only deal with VGA class devices */
> - if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
> - return false;
> -
> /* Allocate structure */
> vgadev = kzalloc(sizeof(struct vga_device), GFP_KERNEL);
> if (vgadev == NULL) {
> @@ -1502,6 +1498,10 @@ static int pci_notify(struct notifier_block *nb, unsigned long action,
>
> vgaarb_dbg(dev, "%s\n", __func__);
>
> + /* Deal with VGA compatible devices only */
> + if ((pdev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
> + return 0;
> +
> /* For now we're only intereted in devices added and removed. I didn't
> * test this thing here, so someone needs to double check for the
> * cases of hotplugable vga cards. */
> @@ -1534,8 +1534,8 @@ static struct miscdevice vga_arb_device = {
>
> static int __init vga_arb_device_init(void)
> {
> + struct pci_dev *pdev = NULL;
> int rc;
> - struct pci_dev *pdev;
>
> rc = misc_register(&vga_arb_device);
> if (rc < 0)
> @@ -1543,13 +1543,14 @@ static int __init vga_arb_device_init(void)
>
> bus_register_notifier(&pci_bus_type, &pci_notifier);
>
> - /* We add all PCI devices satisfying VGA class in the arbiter by
> - * default */
> - pdev = NULL;
> - while ((pdev =
> - pci_get_subsys(PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
> - PCI_ANY_ID, pdev)) != NULL)
> - vga_arbiter_add_pci_device(pdev);
> + /*
> + * We add all PCI VGA compatible devices in the arbiter by default
> + */
> + do {
> + pdev = pci_get_class(PCI_CLASS_DISPLAY_VGA << 8, pdev);
> + if (pdev)
> + vga_arbiter_add_pci_device(pdev);
> + } while (pdev);
>
> pr_info("loaded\n");
> return rc;
> --
> 2.25.1
>
next prev parent reply other threads:[~2023-07-19 18:26 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-11 13:43 [PATCH 0/6] PCI/VGA: Fix typos, comments and copyright Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
2023-07-11 13:43 ` [PATCH 1/6] PCI/VGA: Use unsigned type for the io_state variable Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
2023-07-11 13:43 ` [PATCH 2/6] PCI/VGA: Deal with PCI VGA compatible devices only Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
2023-07-17 9:51 ` suijingfeng
2023-07-17 13:17 ` Sui Jingfeng
2023-07-17 13:28 ` suijingfeng
2023-07-19 18:26 ` Bjorn Helgaas [this message]
2023-07-19 18:26 ` Bjorn Helgaas
2023-07-19 19:58 ` Sui Jingfeng
2023-07-19 19:58 ` Sui Jingfeng
2023-07-19 20:06 ` suijingfeng
2023-07-19 20:06 ` suijingfeng
2023-07-19 20:08 ` suijingfeng
2023-07-19 20:08 ` suijingfeng
2023-07-19 20:16 ` suijingfeng
2023-07-19 20:16 ` suijingfeng
2023-07-19 21:13 ` Sui Jingfeng
2023-07-19 21:13 ` Sui Jingfeng
2023-07-19 21:27 ` suijingfeng
2023-07-19 21:27 ` suijingfeng
2023-07-22 8:11 ` suijingfeng
2023-07-22 8:11 ` suijingfeng
2023-07-25 21:49 ` Bjorn Helgaas
2023-07-25 21:49 ` Bjorn Helgaas
2023-08-01 7:17 ` Sui Jingfeng
2023-08-01 7:17 ` Sui Jingfeng
2023-07-11 13:43 ` [PATCH 3/6] PCI/VGA: drop the inline of vga_update_device_decodes() function Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
2023-07-24 13:02 ` suijingfeng
2023-07-24 13:02 ` suijingfeng
2023-07-11 13:43 ` [PATCH 4/6] PCI/VGA: Move the new_state assignment out the loop Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
2023-07-24 13:02 ` suijingfeng
2023-07-24 13:02 ` suijingfeng
2023-07-25 21:51 ` Bjorn Helgaas
2023-07-25 21:51 ` Bjorn Helgaas
2023-07-11 13:43 ` [PATCH 5/6] PCI/VGA: Tidy up the code and comment format Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
2023-07-11 13:43 ` [PATCH 6/6] PCI/VGA: Replace full MIT license text with SPDX identifier Sui Jingfeng
2023-07-11 13:43 ` Sui Jingfeng
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=20230719182617.GA509912@bhelgaas \
--to=helgaas@kernel.org \
--cc=Jingfeng@loongson.cn \
--cc=Sui@freedesktop.org \
--cc=airlied@gmail.com \
--cc=bhelgaas@google.com \
--cc=daniel@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=loongson-kernel@lists.loongnix.cn \
--cc=maarten.lankhorst@linux.intel.com \
--cc=mario.limonciello@amd.com \
--cc=mripard@kernel.org \
--cc=sui.jingfeng@linux.dev \
--cc=suijingfeng@loongson.cn \
--cc=tzimmermann@suse.de \
/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.