Message-ID: <2648aef32cd5a2272cd3ce8cd7ed5b29b2d21cad.camel@redhat.com>
Date:   Mon, 04 Dec 2023 14:39:58 +0100
From:   Philipp Stanner <pstanner@...hat.com>
To:     Bjorn Helgaas <bhelgaas@...gle.com>, Arnd Bergmann <arnd@...db.de>,
        Hanjun Guo <guohanjun@...wei.com>, NeilBrown <neilb@...e.de>,
        Kent Overstreet <kent.overstreet@...il.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Niklas Schnelle <schnelle@...ux.ibm.com>,
        Uladzislau Koshchanka <koshchanka@...il.com>,
        John Sanpe <sanpeqf@...il.com>,
        Dave Jiang <dave.jiang@...el.com>,
        "Masami Hiramatsu (Google)" <mhiramat@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        David Gow <davidgow@...gle.com>,
        Herbert Xu <herbert@...dor.apana.org.au>,
        Shuah Khan <skhan@...uxfoundation.org>,
        "wuqiang.matt" <wuqiang.matt@...edance.com>,
        Yury Norov <yury.norov@...il.com>,
        Jason Baron <jbaron@...mai.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Ben Dooks <ben.dooks@...ethink.co.uk>, dakr@...hat.com
Cc:     linux-kernel@...r.kernel.org, linux-pci@...r.kernel.org,
        linux-arch@...r.kernel.org, stable@...r.kernel.org,
        Arnd Bergmann <arnd@...nel.org>
Subject: Re: [PATCH v3 5/5] lib, pci: unify generic pci_iounmap()

On Mon, 2023-12-04 at 13:38 +0100, Philipp Stanner wrote:
> The implementation of pci_iounmap() is currently scattered over two
> files, drivers/pci/iomap.c and lib/iomap.c. Additionally,
> architectures can define their own version.
> 
> To have only one version, it's necessary to create a helper function,
> iomem_is_ioport(), that tells pci_iounmap() whether the passed
> address
> points to an ioport or normal memory.
> 
> iomem_is_ioport() can be provided through two different ways:
>   1. The architecture itself provides it. As of today, the version
>      coming from lib/iomap.c de facto is the x86-specific version and
>      comes into play when CONFIG_GENERIC_IOMAP is selected. This
> rather
>      confusing naming is an artifact left by the removal of IA64.
>   2. As a default version in include/asm-generic/io.h for those
>      architectures that don't use CONFIG_GENERIC_IOMAP, but also
> don't
>      provide their own version of iomem_is_ioport().
> 
> Once all architectures that support ports provide iomem_is_ioport(),
> the
> arch-specific definitions for pci_iounmap() can be removed and the
> archs
> can use the generic implementation, instead.
> 
> Create a unified version of pci_iounmap() in drivers/pci/iomap.c.
> Provide the function iomem_is_ioport() in include/asm-generic/io.h
> (generic) and lib/iomap.c ("pseudo-generic" for x86).
> 
> Remove the CONFIG_GENERIC_IOMAP guard around
> ARCH_WANTS_GENERIC_PCI_IOUNMAP so that configs that set
> CONFIG_GENERIC_PCI_IOMAP without CONFIG_GENERIC_IOMAP still get the
> function.
> 
> Add TODOs for follow-up work on the "generic is not generic but
> x86-spcific"-Problem.
> 
> Suggested-by: Arnd Bergmann <arnd@...nel.org>
> Signed-off-by: Philipp Stanner <pstanner@...hat.com>
> ---
>  drivers/pci/iomap.c         | 46 ++++++++++++-----------------------
> --
>  include/asm-generic/io.h    | 27 ++++++++++++++++++++--
>  include/asm-generic/iomap.h | 21 +++++++++++++++++
>  lib/iomap.c                 | 28 ++++++++++++++++------
>  4 files changed, 82 insertions(+), 40 deletions(-)
> 
> diff --git a/drivers/pci/iomap.c b/drivers/pci/iomap.c
> index 91285fcff1ba..439ba2e9710f 100644
> --- a/drivers/pci/iomap.c
> +++ b/drivers/pci/iomap.c
> @@ -135,44 +135,28 @@ void __iomem *pci_iomap_wc(struct pci_dev *dev,
> int bar, unsigned long maxlen)
>  EXPORT_SYMBOL_GPL(pci_iomap_wc);
>  
>  /*
> - * pci_iounmap() somewhat illogically comes from lib/iomap.c for the
> - * CONFIG_GENERIC_IOMAP case, because that's the code that knows
> about
> - * the different IOMAP ranges.
> + * This check is still necessary due to legacy reasons.
>   *
> - * But if the architecture does not use the generic iomap code, and
> if
> - * it has _not_ defined it's own private pci_iounmap function, we
> define
> - * it here.
> - *
> - * NOTE! This default implementation assumes that if the
> architecture
> - * support ioport mapping (HAS_IOPORT_MAP), the ioport mapping will
> - * be fixed to the range [ PCI_IOBASE, PCI_IOBASE+IO_SPACE_LIMIT [,
> - * and does not need unmapping with 'ioport_unmap()'.
> - *
> - * If you have different rules for your architecture, you need to
> - * implement your own pci_iounmap() that knows the rules for where
> - * and how IO vs MEM get mapped.
> - *
> - * This code is odd, and the ARCH_HAS/ARCH_WANTS #define logic comes
> - * from legacy <asm-generic/io.h> header file behavior. In
> particular,
> - * it would seem to make sense to do the iounmap(p) for the non-IO-
> space
> - * case here regardless, but that's not what the old header file
> code
> - * did. Probably incorrectly, but this is meant to be bug-for-bug
> - * compatible.
> + * TODO: Have all architectures that provide their own pci_iounmap()
> provide
> + * iomem_is_ioport() instead. Remove this #if afterwards.
>   */
>  #if defined(ARCH_WANTS_GENERIC_PCI_IOUNMAP)
>  
> -void pci_iounmap(struct pci_dev *dev, void __iomem *p)
> +/**
> + * pci_iounmap - Unmapp a mapping
> + * @dev: PCI device the mapping belongs to
> + * @addr: start address of the mapping
> + *
> + * Unmapp a PIO or MMIO mapping.
> + */
> +void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
>  {
> -#ifdef ARCH_HAS_GENERIC_IOPORT_MAP
> -       uintptr_t start = (uintptr_t) PCI_IOBASE;
> -       uintptr_t addr = (uintptr_t) p;
> -
> -       if (addr >= start && addr < start + IO_SPACE_LIMIT) {
> -               ioport_unmap(p);
> +       if (iomem_is_ioport(addr)) {
> +               ioport_unmap(addr);
>                 return;
>         }
> -#endif
> -       iounmap(p);
> +
> +       iounmap(addr);
>  }
>  EXPORT_SYMBOL(pci_iounmap);
>  
> diff --git a/include/asm-generic/io.h b/include/asm-generic/io.h
> index bac63e874c7b..58c7bf4080da 100644
> --- a/include/asm-generic/io.h
> +++ b/include/asm-generic/io.h
> @@ -1129,11 +1129,34 @@ extern void ioport_unmap(void __iomem *p);
>  #endif /* CONFIG_GENERIC_IOMAP */
>  #endif /* CONFIG_HAS_IOPORT_MAP */
>  
> -#ifndef CONFIG_GENERIC_IOMAP
> +/*
> + * TODO:
> + * remove this once all architectures replaced their pci_iounmap()
> with
> + * a custom implementation of iomem_is_ioport().
> + */
>  #ifndef pci_iounmap
> +#define pci_iounmap pci_iounmap
>  #define ARCH_WANTS_GENERIC_PCI_IOUNMAP
> +#endif /* pci_iounmap */
> +
> +/*
> + * This function is a helper only needed for the generic
> pci_iounmap().
> + * It's provided here if the architecture does not provide its own
> version.
> + */
> +#ifndef iomem_is_ioport
> +#define iomem_is_ioport iomem_is_ioport
> +static inline bool iomem_is_ioport(void __iomem *addr_raw)
> +{
> +#ifdef CONFIG_HAS_IOPORT
> +       uintptr_t start = (uintptr_t)PCI_IOBASE;
> +       uintptr_t addr = (uintptr_t)addr_raw;
> +
> +       if (addr >= start && addr < start + IO_SPACE_LIMIT)
> +               return true;
>  #endif
> -#endif
> +       return false;
> +}
> +#endif /* iomem_is_ioport */
>  
>  #ifndef xlate_dev_mem_ptr
>  #define xlate_dev_mem_ptr xlate_dev_mem_ptr
> diff --git a/include/asm-generic/iomap.h b/include/asm-
> generic/iomap.h
> index 196087a8126e..2cdc6988a102 100644
> --- a/include/asm-generic/iomap.h
> +++ b/include/asm-generic/iomap.h
> @@ -110,6 +110,27 @@ static inline void __iomem
> *ioremap_np(phys_addr_t offset, size_t size)
>  }
>  #endif
>  
> +/*
> + * If CONFIG_GENERIC_IOMAP is selected and the architecture does NOT
> provide its
> + * own version, ARCH_WANTS_GENERIC_IOMEM_IS_IOPORT makes sure that
> the generic
> + * version from asm-generic/io.h is NOT used and instead the second
> "generic"
> + * version from lib/iomap.c is used.
> + *
> + * There are currently two generic versions because of a difficult
> cleanup
> + * process. Namely, the version in lib/iomap.c once was really
> generic when IA64
> + * still existed. Today, it's only really used by x86.
> + *
> + * TODO: Move the version from lib/iomap.c to x86 specific code.
> Then, remove
> + * this ARCH_WANTS_GENERIC_IOMEM_IS_IOPORT-mechanism.
> + */
> +#ifdef CONFIG_GENERIC_IOMAP
> +#ifndef iomem_is_ioport
> +#define iomem_is_ioport iomem_is_ioport
> +bool iomem_is_ioport(void __iomem *addr);
> +#define ARCH_WANTS_GENERIC_IOMEM_IS_IOPORT
> +#endif /* iomem_is_ioport */
> +#endif /* CONFIG_GENERIC_IOMAP */
> +
>  #include <asm-generic/pci_iomap.h>
>  
>  #endif
> diff --git a/lib/iomap.c b/lib/iomap.c
> index 4f8b31baa575..eb9a879ebf42 100644
> --- a/lib/iomap.c
> +++ b/lib/iomap.c
> @@ -418,12 +418,26 @@ EXPORT_SYMBOL(ioport_map);
>  EXPORT_SYMBOL(ioport_unmap);
>  #endif /* CONFIG_HAS_IOPORT_MAP */
>  
> -#ifdef CONFIG_PCI
> -/* Hide the details if this is a MMIO or PIO address space and just
> do what
> - * you expect in the correct way. */
> -void pci_iounmap(struct pci_dev *dev, void __iomem * addr)
> +/*
> + * If CONFIG_GENERIC_IOMAP is selected and the architecture does NOT
> provide its
> + * own version, ARCH_WANTS_GENERIC_IOMEM_IS_IOPORT makes sure that
> the generic
> + * version from asm-generic/io.h is NOT used and instead the second
> "generic"
> + * version from this file here is used.
> + *
> + * There are currently two generic versions because of a difficult
> cleanup
> + * process. Namely, the version in lib/iomap.c once was really
> generic when IA64
> + * still existed. Today, it's only really used by x86.
> + *
> + * TODO: Move this function to x86-specific code.
> + */
> +#if defined(ARCH_WANTS_GENERIC_IOMEM_IS_IOPORT)
> +bool iomem_is_ioport(void __iomem *addr)
>  {
> -       IO_COND(addr, /* nothing */, iounmap(addr));
> +       unsigned long port = (unsigned long __force)addr;
> +
> +       if (port > PIO_OFFSET && port < PIO_RESERVED)

by the way:
Reading that again my instinctive feeling would be that it should be
port >= PIO_OFFSET.

This is, however, the exactly copied logic from the IO_COND macro in
lib/iomap.c. Is it possible that this macro contains a bug here?

P.

> +               return true;
> +
> +       return false;
>  }
> -EXPORT_SYMBOL(pci_iounmap);
> -#endif /* CONFIG_PCI */
> +#endif /* ARCH_WANTS_GENERIC_IOMEM_IS_IOPORT */

Powered by blists - more mailing lists