From: Greg KH <gregkh@linuxfoundation.org>
To: Tianfei Zhang <tianfei.zhang@intel.com>
Cc: bhelgaas@google.com, linux-pci@vger.kernel.org,
linux-fpga@vger.kernel.org, lukas@wunner.de, kabel@kernel.org,
mani@kernel.org, pali@kernel.org, mdf@kernel.org,
hao.wu@intel.com, yilun.xu@intel.com, trix@redhat.com,
jgg@ziepe.ca, ira.weiny@intel.com,
andriy.shevchenko@linux.intel.com, dan.j.williams@intel.com,
keescook@chromium.org, rafael@kernel.org,
russell.h.weight@intel.com, corbet@lwn.net,
linux-doc@vger.kernel.org, ilpo.jarvinen@linux.intel.com,
lee@kernel.org, matthew.gerlach@linux.intel.com
Subject: Re: [PATCH v1 01/12] PCI: hotplug: add new callbacks on hotplug_slot_ops
Date: Thu, 19 Jan 2023 14:31:07 +0100 [thread overview]
Message-ID: <Y8lGG39I0By5d5wh@kroah.com> (raw)
In-Reply-To: <20230119013602.607466-2-tianfei.zhang@intel.com>
On Wed, Jan 18, 2023 at 08:35:51PM -0500, Tianfei Zhang wrote:
> To reprogram an PCIe-based FPGA card, a new image is
> burned into FLASH on the card and then the card BMC is
> triggered to reboot the card and load the new image.
>
> Two new operation callbacks are defined in hotplug_slot_ops
> to trigger the reprogramming of an FPGA-based PCIe card:
>
> - available_images: Optional: available FPGA images
> - image_load: Optional: trigger the FPGA to load a new image
>
> Signed-off-by: Tianfei Zhang <tianfei.zhang@intel.com>
> Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
> ---
> drivers/pci/hotplug/pci_hotplug_core.c | 88 ++++++++++++++++++++++++++
> include/linux/pci_hotplug.h | 5 ++
> 2 files changed, 93 insertions(+)
>
> diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
> index 058d5937d8a9..2b14b6513a03 100644
> --- a/drivers/pci/hotplug/pci_hotplug_core.c
> +++ b/drivers/pci/hotplug/pci_hotplug_core.c
> @@ -231,6 +231,52 @@ static struct pci_slot_attribute hotplug_slot_attr_test = {
> .store = test_write_file
> };
>
> +static ssize_t available_images_read_file(struct pci_slot *pci_slot, char *buf)
> +{
> + struct hotplug_slot *slot = pci_slot->hotplug;
> + ssize_t count = 0;
> +
> + if (!try_module_get(slot->owner))
> + return -ENODEV;
> +
> + if (slot->ops->available_images(slot, buf))
> + count = slot->ops->available_images(slot, buf);
> +
> + module_put(slot->owner);
> +
> + return count;
> +}
> +
> +static struct pci_slot_attribute hotplug_slot_attr_available_images = {
> + .attr = { .name = "available_images", .mode = 0444 },
> + .show = available_images_read_file,
If you name things properly, you can use the correct macros and not have
to open-code any of this :(
> +static ssize_t image_load_write_file(struct pci_slot *pci_slot,
> + const char *buf, size_t count)
> +{
> + struct hotplug_slot *slot = pci_slot->hotplug;
> + int retval = 0;
> +
> + if (!try_module_get(slot->owner))
> + return -ENODEV;
> +
> + if (slot->ops->image_load)
> + retval = slot->ops->image_load(slot, buf);
> +
> + module_put(slot->owner);
> +
> + if (retval)
> + return retval;
> +
> + return count;
> +}
> +
> +static struct pci_slot_attribute hotplug_slot_attr_image_load = {
> + .attr = { .name = "image_load", .mode = 0644 },
> + .store = image_load_write_file,
Same here, don't open-code this.
> +};
> +
> static bool has_power_file(struct pci_slot *pci_slot)
> {
> struct hotplug_slot *slot = pci_slot->hotplug;
> @@ -289,6 +335,20 @@ static bool has_test_file(struct pci_slot *pci_slot)
> return false;
> }
>
> +static bool has_available_images_file(struct pci_slot *pci_slot)
> +{
> + struct hotplug_slot *slot = pci_slot->hotplug;
> +
> + return slot && slot->ops && slot->ops->available_images;
> +}
> +
> +static bool has_image_load_file(struct pci_slot *pci_slot)
> +{
> + struct hotplug_slot *slot = pci_slot->hotplug;
> +
> + return slot && slot->ops && slot->ops->image_load;
> +}
> +
> static int fs_add_slot(struct pci_slot *pci_slot)
> {
> int retval = 0;
> @@ -331,8 +391,30 @@ static int fs_add_slot(struct pci_slot *pci_slot)
> goto exit_test;
> }
>
> + if (has_available_images_file(pci_slot)) {
> + retval = sysfs_create_file(&pci_slot->kobj,
> + &hotplug_slot_attr_available_images.attr);
> + if (retval)
> + goto exit_available_images;
> + }
> +
> + if (has_image_load_file(pci_slot)) {
> + retval = sysfs_create_file(&pci_slot->kobj,
> + &hotplug_slot_attr_image_load.attr);
> + if (retval)
> + goto exit_image_load;
> + }
> +
> goto exit;
>
> +exit_image_load:
> + if (has_adapter_file(pci_slot))
> + sysfs_remove_file(&pci_slot->kobj,
> + &hotplug_slot_attr_available_images.attr);
> +exit_available_images:
> + if (has_adapter_file(pci_slot))
> + sysfs_remove_file(&pci_slot->kobj,
> + &hotplug_slot_attr_test.attr);
> exit_test:
> if (has_adapter_file(pci_slot))
> sysfs_remove_file(&pci_slot->kobj,
> @@ -372,6 +454,12 @@ static void fs_remove_slot(struct pci_slot *pci_slot)
> if (has_test_file(pci_slot))
> sysfs_remove_file(&pci_slot->kobj, &hotplug_slot_attr_test.attr);
>
> + if (has_available_images_file(pci_slot))
> + sysfs_remove_file(&pci_slot->kobj, &hotplug_slot_attr_available_images.attr);
> +
> + if (has_image_load_file(pci_slot))
> + sysfs_remove_file(&pci_slot->kobj, &hotplug_slot_attr_image_load.attr);
> +
Ick no, please just make this an attribute group that properly shows or
does not show, the attribute when created. Do not manually add
individual sysfs files.
Yes, I know the existing code does this, so it's not really your fault,
but let's not persist in making this code even messier. Convert to a
group first and then your new files will be added automagically without
having to care about anything here at all.
thanks,
greg k-h
next prev parent reply other threads:[~2023-01-19 13:32 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-19 1:35 [PATCH v1 00/12] add FPGA hotplug manager driver Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 01/12] PCI: hotplug: add new callbacks on hotplug_slot_ops Tianfei Zhang
2023-01-19 13:31 ` Greg KH [this message]
2023-01-19 1:35 ` [PATCH v1 02/12] PCI: hotplug: expose APIs from pciehp driver Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 03/12] PCI: hotplug: add and expose link disable API Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 04/12] PCI: hotplug: add FPGA PCI hotplug manager driver Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 05/12] fpga: dfl: register dfl-pci device into fpgahph driver Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 06/12] driver core: expose device_is_ancestor() API Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 07/12] PCI: hotplug: add register/unregister function for BMC device Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 08/12] fpga: m10bmc-sec: register BMC device into fpgahp driver Tianfei Zhang
2023-01-19 1:35 ` [PATCH v1 09/12] fpga: dfl: remove non-reserved devices Tianfei Zhang
2023-01-19 1:36 ` [PATCH v1 10/12] PCI: hotplug: implement the hotplug_slot_ops callback for fpgahp Tianfei Zhang
2023-01-19 13:28 ` Greg KH
2023-01-20 22:38 ` Russ Weight
2023-01-21 7:35 ` Greg KH
2023-01-19 1:36 ` [PATCH v1 11/12] fpga: m10bmc-sec: add m10bmc_sec_retimer_load callback Tianfei Zhang
2023-01-19 14:22 ` Lee Jones
2023-01-19 1:36 ` [PATCH v1 12/12] Documentation: fpga: add description of fpgahp driver Tianfei Zhang
2023-01-19 9:38 ` Bagas Sanjaya
2023-01-19 8:06 ` [PATCH v1 00/12] add FPGA hotplug manager driver Pali Rohár
2023-01-19 8:17 ` Zhang, Tianfei
2023-01-19 11:27 ` andriy.shevchenko
2023-01-19 12:09 ` Zhang, Tianfei
2023-01-19 13:33 ` Greg KH
2023-01-19 13:43 ` Rafael J. Wysocki
2023-01-19 15:33 ` Greg KH
2023-01-20 16:28 ` Russ Weight
2023-01-20 18:42 ` Lukas Wunner
2023-01-21 7:34 ` Greg KH
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=Y8lGG39I0By5d5wh@kroah.com \
--to=gregkh@linuxfoundation.org \
--cc=andriy.shevchenko@linux.intel.com \
--cc=bhelgaas@google.com \
--cc=corbet@lwn.net \
--cc=dan.j.williams@intel.com \
--cc=hao.wu@intel.com \
--cc=ilpo.jarvinen@linux.intel.com \
--cc=ira.weiny@intel.com \
--cc=jgg@ziepe.ca \
--cc=kabel@kernel.org \
--cc=keescook@chromium.org \
--cc=lee@kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-fpga@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=lukas@wunner.de \
--cc=mani@kernel.org \
--cc=matthew.gerlach@linux.intel.com \
--cc=mdf@kernel.org \
--cc=pali@kernel.org \
--cc=rafael@kernel.org \
--cc=russell.h.weight@intel.com \
--cc=tianfei.zhang@intel.com \
--cc=trix@redhat.com \
--cc=yilun.xu@intel.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 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).