From: Niklas Schnelle <schnelle@linux.ibm.com>
To: Lukas Wunner <lukas@wunner.de>,
Gerald Schaefer <gerald.schaefer@linux.ibm.com>,
Heiko Carstens <hca@linux.ibm.com>,
Vasily Gorbik <gor@linux.ibm.com>,
Alexander Gordeev <agordeev@linux.ibm.com>,
Christian Borntraeger <borntraeger@linux.ibm.com>,
Sven Schnelle <svens@linux.ibm.com>,
Bjorn Helgaas <helgaas@kernel.org>
Cc: linux-s390@vger.kernel.org, linux-pci@vger.kernel.org,
Alistair Francis <alistair23@gmail.com>,
Jonathan.Cameron@huawei.com, alex.williamson@redhat.com,
christian.koenig@amd.com, kch@nvidia.com,
gregkh@linuxfoundation.org, logang@deltatee.com,
chaitanyak@nvidia.com, rdunlap@infradead.org,
Alistair Francis <alistair.francis@wdc.com>,
Sebastian Ott <sebott@redhat.com>
Subject: Re: [PATCH] s390/pci: Stop usurping pdev->dev.groups
Date: Wed, 07 Aug 2024 17:44:25 +0200 [thread overview]
Message-ID: <ff738a270fc5ad18310e10d4624955430c86e7a2.camel@linux.ibm.com> (raw)
In-Reply-To: <7b970f7923e373d1b23784721208f93418720485.1722870934.git.lukas@wunner.de>
On Mon, 2024-08-05 at 17:24 +0200, Lukas Wunner wrote:
> Bjorn suggests using pdev->dev.groups for attribute_groups constructed
> on PCI device enumeration:
>
> "Is it feasible to build an attribute group in pci_doe_init() and
> add it to dev->groups so device_add() will automatically add them?"
> https://msgid.link/20231019165829.GA1381099@bhelgaas
>
> Unfortunately on s390, pcibios_device_add() usurps pdev->dev.groups for
> arch-specific attribute_groups, preventing its use for anything else.
>
> Introduce an ARCH_PCI_DEV_GROUPS macro which arches can define in
> <asm/pci.h>. The macro is visible in drivers/pci/pci-sysfs.c through
> the inclusion of <linux/pci.h>, which in turn includes <asm/pci.h>.
>
> On s390, define the macro to the three attribute_groups previously
> assigned to pdev->dev.groups. Thereby pdev->dev.groups is made
> available for use by the PCI core.
>
> As a side effect, arch/s390/pci/pci_sysfs.c no longer needs to be
> compiled into the kernel if CONFIG_SYSFS=n.
>
> Signed-off-by: Lukas Wunner <lukas@wunner.de>
> ---
> arch/s390/include/asm/pci.h | 9 ++++++++-
> arch/s390/pci/Makefile | 3 ++-
> arch/s390/pci/pci.c | 1 -
> arch/s390/pci/pci_sysfs.c | 14 ++++----------
> drivers/pci/pci-sysfs.c | 5 +++++
> 5 files changed, 19 insertions(+), 13 deletions(-)
>
> diff --git a/arch/s390/include/asm/pci.h b/arch/s390/include/asm/pci.h
> index 30820a6..9d920ce 100644
> --- a/arch/s390/include/asm/pci.h
> +++ b/arch/s390/include/asm/pci.h
> @@ -191,7 +191,14 @@ static inline bool zdev_enabled(struct zpci_dev *zdev)
> return (zdev->fh & (1UL << 31)) ? true : false;
> }
>
> -extern const struct attribute_group *zpci_attr_groups[];
> +extern const struct attribute_group zpci_attr_group;
> +extern const struct attribute_group pfip_attr_group;
> +extern const struct attribute_group zpci_ident_attr_group;
> +
> +#define ARCH_PCI_DEV_GROUPS &zpci_attr_group, \
> + &pfip_attr_group, \
> + &zpci_ident_attr_group,
> +
> extern unsigned int s390_pci_force_floating __initdata;
> extern unsigned int s390_pci_no_rid;
>
> diff --git a/arch/s390/pci/Makefile b/arch/s390/pci/Makefile
> index 0547a10..2c21f03 100644
> --- a/arch/s390/pci/Makefile
> +++ b/arch/s390/pci/Makefile
> @@ -3,7 +3,8 @@
> # Makefile for the s390 PCI subsystem.
> #
>
> -obj-$(CONFIG_PCI) += pci.o pci_irq.o pci_clp.o pci_sysfs.o \
> +obj-$(CONFIG_PCI) += pci.o pci_irq.o pci_clp.o \
> pci_event.o pci_debug.o pci_insn.o pci_mmio.o \
> pci_bus.o pci_kvm_hook.o
> obj-$(CONFIG_PCI_IOV) += pci_iov.o
> +obj-$(CONFIG_SYSFS) += pci_sysfs.o
> diff --git a/arch/s390/pci/pci.c b/arch/s390/pci/pci.c
> index cff4838..bd9624c 100644
> --- a/arch/s390/pci/pci.c
> +++ b/arch/s390/pci/pci.c
> @@ -587,7 +587,6 @@ int pcibios_device_add(struct pci_dev *pdev)
> if (pdev->is_physfn)
> pdev->no_vf_scan = 1;
>
> - pdev->dev.groups = zpci_attr_groups;
> zpci_map_resources(pdev);
>
> for (i = 0; i < PCI_STD_NUM_BARS; i++) {
> diff --git a/arch/s390/pci/pci_sysfs.c b/arch/s390/pci/pci_sysfs.c
> index 0f4f1e8..1f81f6f 100644
> --- a/arch/s390/pci/pci_sysfs.c
> +++ b/arch/s390/pci/pci_sysfs.c
> @@ -197,7 +197,7 @@ static umode_t zpci_index_is_visible(struct kobject *kobj,
> NULL,
> };
>
> -static struct attribute_group zpci_ident_attr_group = {
> +const struct attribute_group zpci_ident_attr_group = {
> .attrs = zpci_ident_attrs,
> .is_visible = zpci_index_is_visible,
> };
> @@ -223,7 +223,7 @@ static umode_t zpci_index_is_visible(struct kobject *kobj,
> NULL,
> };
>
> -static struct attribute_group zpci_attr_group = {
> +const struct attribute_group zpci_attr_group = {
> .attrs = zpci_dev_attrs,
> .bin_attrs = zpci_bin_attrs,
> };
> @@ -235,14 +235,8 @@ static umode_t zpci_index_is_visible(struct kobject *kobj,
> &dev_attr_segment3.attr,
> NULL,
> };
> -static struct attribute_group pfip_attr_group = {
> +
> +const struct attribute_group pfip_attr_group = {
> .name = "pfip",
> .attrs = pfip_attrs,
> };
> -
> -const struct attribute_group *zpci_attr_groups[] = {
> - &zpci_attr_group,
> - &pfip_attr_group,
> - &zpci_ident_attr_group,
> - NULL,
> -};
> diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
> index 40cfa71..5d0f4db 100644
> --- a/drivers/pci/pci-sysfs.c
> +++ b/drivers/pci/pci-sysfs.c
> @@ -31,6 +31,10 @@
> #include <linux/aperture.h>
> #include "pci.h"
>
> +#ifndef ARCH_PCI_DEV_GROUPS
> +#define ARCH_PCI_DEV_GROUPS
> +#endif
> +
> static int sysfs_initialized; /* = 0 */
>
> /* show configuration fields */
> @@ -1624,6 +1628,7 @@ static umode_t pcie_dev_attrs_are_visible(struct kobject *kobj,
> &pci_dev_acpi_attr_group,
> #endif
> &pci_dev_resource_resize_group,
> + ARCH_PCI_DEV_GROUPS
> NULL,
> };
>
Thanks for the patch and sorry for the delay I'll be on vacation
starting tomorrow and it's a bit busy. This makes sense to me and I
agree with Bjorn that this can go via the PCI tree, especially if it is
already useful there. I also gave this a compile and boot test on and
checked that the s390 specific PCI device attributes are still there.
Acked-by: Niklas Schnelle <schnelle@linux.ibm.com>
next prev parent reply other threads:[~2024-08-07 15:45 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-05 15:24 [PATCH] s390/pci: Stop usurping pdev->dev.groups Lukas Wunner
2024-08-06 19:30 ` Bjorn Helgaas
2024-08-06 19:55 ` Lukas Wunner
2024-08-07 18:08 ` Konstantin Ryabitsev
2024-08-07 15:44 ` Niklas Schnelle [this message]
2024-08-09 20:02 ` Bjorn Helgaas
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=ff738a270fc5ad18310e10d4624955430c86e7a2.camel@linux.ibm.com \
--to=schnelle@linux.ibm.com \
--cc=Jonathan.Cameron@huawei.com \
--cc=agordeev@linux.ibm.com \
--cc=alex.williamson@redhat.com \
--cc=alistair.francis@wdc.com \
--cc=alistair23@gmail.com \
--cc=borntraeger@linux.ibm.com \
--cc=chaitanyak@nvidia.com \
--cc=christian.koenig@amd.com \
--cc=gerald.schaefer@linux.ibm.com \
--cc=gor@linux.ibm.com \
--cc=gregkh@linuxfoundation.org \
--cc=hca@linux.ibm.com \
--cc=helgaas@kernel.org \
--cc=kch@nvidia.com \
--cc=linux-pci@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=logang@deltatee.com \
--cc=lukas@wunner.de \
--cc=rdunlap@infradead.org \
--cc=sebott@redhat.com \
--cc=svens@linux.ibm.com \
/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.