Received-SPF: pass (google.com: domain of linux-kernel+bounces-16094-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Wed, 3 Jan 2024 17:01:47 -0600
From: Bjorn Helgaas <helgaas@kernel.org>
To: Dan Williams <dan.j.williams@intel.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>, Ira Weiny <ira.weiny@intel.com>,
	Jonathan Cameron <jonathan.cameron@huawei.com>,
	Smita Koralahalli <Smita.KoralahalliChannabasappa@amd.com>,
	Shiju Jose <shiju.jose@huawei.com>,
	Yazen Ghannam <yazen.ghannam@amd.com>,
	Davidlohr Bueso <dave@stgolabs.net>,
	Dave Jiang <dave.jiang@intel.com>,
	Alison Schofield <alison.schofield@intel.com>,
	Vishal Verma <vishal.l.verma@intel.com>,
	Ard Biesheuvel <ardb@kernel.org>, linux-efi@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-cxl@vger.kernel.org,
	linux-pci@vger.kernel.org
Subject: Re: [PATCH v5 8/9] PCI: Define scoped based management functions
Message-ID: <20240103230147.GA1800245@bhelgaas>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <6595e201beb4_be7022944d@dwillia2-xfh.jf.intel.com.notmuch>

On Wed, Jan 03, 2024 at 02:38:57PM -0800, Dan Williams wrote:
> Ira Weiny wrote:
> > Users of pci_dev_get() can benefit from a scoped based put.  Also,
> > locking a PCI device is often done within a single scope.
> > 
> > Define a pci_dev_put() free function and a PCI device lock guard.  These
> > will initially be used in new CXL event processing code but is defined
> > in a separate patch for others to pickup and use/backport easier.
> 
> Any heartburn if I take this through cxl.git with the rest in this
> series? Patch 9 has a dependency on this one.

No real heartburn.  I was trying to figure out what this does
since I'm not familiar with the "scoped based put" idea and
'git grep -i "scope.*base"' wasn't much help.

I would kind of like the commit log to say a little more about what
the "scoped based put" (does that have too many past tenses in it?)
means and how users of pci_dev_get() will benefit.

I don't know what "locking a PCI device is often done within a single
scope" is trying to tell me either.  What if it's *not* done within a
single scope?

Does this change anything for callers of pci_dev_get() and
pci_dev_put()?

Does this avoid a common programming error?  I just don't know what
the benefit of this is yet.

I'm sure this is really cool stuff, but there's little documentation,
few existing users, and I don't know what I need to look for when
reviewing things.

> > --- a/include/linux/pci.h
> > +++ b/include/linux/pci.h
> > @@ -1170,6 +1170,7 @@ int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge);
> >  u8 pci_common_swizzle(struct pci_dev *dev, u8 *pinp);
> >  struct pci_dev *pci_dev_get(struct pci_dev *dev);
> >  void pci_dev_put(struct pci_dev *dev);
> > +DEFINE_FREE(pci_dev_put, struct pci_dev *, if (_T) pci_dev_put(_T))
> >  void pci_remove_bus(struct pci_bus *b);
> >  void pci_stop_and_remove_bus_device(struct pci_dev *dev);
> >  void pci_stop_and_remove_bus_device_locked(struct pci_dev *dev);
> > @@ -1871,6 +1872,7 @@ void pci_cfg_access_unlock(struct pci_dev *dev);
> >  void pci_dev_lock(struct pci_dev *dev);
> >  int pci_dev_trylock(struct pci_dev *dev);
> >  void pci_dev_unlock(struct pci_dev *dev);
> > +DEFINE_GUARD(pci_dev, struct pci_dev *, pci_dev_lock(_T), pci_dev_unlock(_T))
> >  
> >  /*
> >   * PCI domain support.  Sometimes called PCI segment (eg by ACPI),
> > 
> > -- 
> > 2.43.0
> > 
> 
>