From: Bjorn Helgaas <helgaas@kernel.org>
To: Dan Williams <dan.j.williams@intel.com>
Cc: linux-pci@vger.kernel.org,
"Kan Liang" <kan.liang@linux.intel.com>,
"Tony Luck" <tony.luck@intel.com>,
"David E Box" <david.e.box@intel.com>,
"Yunying Sun" <yunying.sun@intel.com>,
"Dave Jiang" <dave.jiang@intel.com>,
"Mika Westerberg" <mika.westerberg@linux.intel.com>,
"Giovanni Cabiddu" <giovanni.cabiddu@intel.com>,
"Herbert Xu" <herbert@gondor.apana.org.au>,
"Hans de Goede" <hdegoede@redhat.com>,
"Florent DELAHAYE" <linuxkernelml@undead.fr>,
"Konrad J Hambrick" <kjhambrick@gmail.com>,
"Matt Hansen" <2lprbe78@duck.com>,
"Nicholas Johnson" <nicholas.johnson-opensource@outlook.com.au>,
"Benoit Grégoire" <benoitg@coeus.ca>,
"Werner Sembach" <wse@tuxedocomputers.com>,
mumblingdrunkard@protonmail.com, linux-kernel@vger.kernel.org,
"Bjorn Helgaas" <bhelgaas@google.com>
Subject: Re: [PATCH 2/2] x86/pci: Treat EfiMemoryMappedIO as reservation of ECAM space
Date: Tue, 10 Jan 2023 13:30:43 -0600 [thread overview]
Message-ID: <20230110193043.GA1603645@bhelgaas> (raw)
In-Reply-To: <63bdae72740db_5178e294bd@dwillia2-xfh.jf.intel.com.notmuch>
On Tue, Jan 10, 2023 at 10:29:06AM -0800, Dan Williams wrote:
> Bjorn Helgaas wrote:
> > From: Bjorn Helgaas <bhelgaas@google.com>
> >
> > Normally we reject ECAM space unless it is reported as reserved in the E820
> > table or via a PNP0C02 _CRS method (PCI Firmware, r3.3, sec 4.1.2). This
> > means PCI extended config space (offsets 0x100-0xfff) may not be accessible.
> >
> > Some firmware doesn't report ECAM space via PNP0C02 _CRS methods, but does
> > mention it as an EfiMemoryMappedIO region via EFI GetMemoryMap(), which is
> > normally converted to an E820 entry by a bootloader or EFI stub.
> >
> > 07eab0901ede ("efi/x86: Remove EfiMemoryMappedIO from E820 map"), removes
> > E820 entries that correspond to EfiMemoryMappedIO regions because some
> > other firmware uses EfiMemoryMappedIO for PCI host bridge windows, and the
> > E820 entries prevent Linux from allocating BAR space for hot-added devices.
> >
> > Allow use of ECAM for extended config space when the region is covered by
> > an EfiMemoryMappedIO region, even if it's not included in E820 or PNP0C02
> > _CRS.
> >
> > Reported by Kan Liang, Tony Luck, and Giovanni Cabiddu.
> >
> > Fixes: 07eab0901ede ("efi/x86: Remove EfiMemoryMappedIO from E820 map")
> > Link: https://lore.kernel.org/r/ac2693d8-8ba3-72e0-5b66-b3ae008d539d@linux.intel.com
> > Reported-by: Kan Liang <kan.liang@linux.intel.com>
> > Reported-by: Tony Luck <tony.luck@intel.com>
> > Reported-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
> > Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
> > ---
> > arch/x86/pci/mmconfig-shared.c | 31 +++++++++++++++++++++++++++++++
> > 1 file changed, 31 insertions(+)
> >
> > diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c
> > index cd16bef5f2d9..da4b6e8e9df0 100644
> > --- a/arch/x86/pci/mmconfig-shared.c
> > +++ b/arch/x86/pci/mmconfig-shared.c
> > @@ -12,6 +12,7 @@
> > */
> >
> > #include <linux/acpi.h>
> > +#include <linux/efi.h>
> > #include <linux/pci.h>
> > #include <linux/init.h>
> > #include <linux/bitmap.h>
> > @@ -442,6 +443,32 @@ static bool is_acpi_reserved(u64 start, u64 end, enum e820_type not_used)
> > return mcfg_res.flags;
> > }
> >
> > +static bool is_efi_mmio(u64 start, u64 end, enum e820_type not_used)
> > +{
> > +#ifdef CONFIG_EFI
> > + efi_memory_desc_t *md;
> > + u64 size, mmio_start, mmio_end;
> > +
> > + for_each_efi_memory_desc(md) {
> > + if (md->type == EFI_MEMORY_MAPPED_IO) {
> > + size = md->num_pages << EFI_PAGE_SHIFT;
> > + mmio_start = md->phys_addr;
> > + mmio_end = mmio_start + size;
> > +
> > + /*
> > + * N.B. Caller supplies (start, start + size),
> > + * so to match, mmio_end is the first address
> > + * *past* the EFI_MEMORY_MAPPED_IO area.
> > + */
> > + if (mmio_start <= start && end <= mmio_end)
> > + return true;
> > + }
> > + }
> > +#endif
>
> Perhaps the following trick (compile tested), but either way:
>
> Reviewed-by: Dan Williams <dan.j.williams@intel.com>
That's a great trick, and I wish I'd thought of it. I have some
follow-on patches I'm considering for v6.3, so in the interest of
streamlining the path of this one to v6.2-rc4, I think I'll wait on
this until v6.3.
> diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c
> index da4b6e8e9df0..ae95d1b073c6 100644
> --- a/arch/x86/pci/mmconfig-shared.c
> +++ b/arch/x86/pci/mmconfig-shared.c
> @@ -445,7 +445,6 @@ static bool is_acpi_reserved(u64 start, u64 end, enum e820_type not_used)
>
> static bool is_efi_mmio(u64 start, u64 end, enum e820_type not_used)
> {
> -#ifdef CONFIG_EFI
> efi_memory_desc_t *md;
> u64 size, mmio_start, mmio_end;
>
> @@ -464,7 +463,6 @@ static bool is_efi_mmio(u64 start, u64 end, enum e820_type not_used)
> return true;
> }
> }
> -#endif
>
> return false;
> }
> diff --git a/include/linux/efi.h b/include/linux/efi.h
> index 4b27519143f5..3ab0c255b791 100644
> --- a/include/linux/efi.h
> +++ b/include/linux/efi.h
> @@ -790,8 +790,12 @@ extern int efi_memattr_apply_permissions(struct mm_struct *mm,
> *
> * Once the loop finishes @md must not be accessed.
> */
> +#ifdef CONFIG_EFI
> #define for_each_efi_memory_desc(md) \
> for_each_efi_memory_desc_in_map(&efi.memmap, md)
> +#else
> +#define for_each_efi_memory_desc(md) for (; 0;)
> +#endif
>
> /*
> * Format an EFI memory descriptor's type and attributes to a user-provided
next prev parent reply other threads:[~2023-01-10 19:30 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-10 18:02 [PATCH 0/2] PCI: Fix extended config space regression Bjorn Helgaas
2023-01-10 18:02 ` [PATCH 1/2] x86/pci: Simplify is_mmconf_reserved() messages Bjorn Helgaas
2023-01-10 18:07 ` Dan Williams
2023-01-10 18:12 ` Bjorn Helgaas
2023-01-10 18:02 ` [PATCH 2/2] x86/pci: Treat EfiMemoryMappedIO as reservation of ECAM space Bjorn Helgaas
2023-01-10 18:15 ` Rafael J. Wysocki
2023-01-10 18:29 ` Dan Williams
2023-01-10 19:30 ` Bjorn Helgaas [this message]
2023-10-12 15:33 ` Tomasz Pala
2023-10-16 17:31 ` Tomasz Pala
2023-10-26 20:53 ` Bjorn Helgaas
2023-11-03 19:18 ` Bjorn Helgaas
2023-11-09 18:44 ` Bjorn Helgaas
2023-11-18 14:21 ` Tomasz Pala
2023-11-20 16:29 ` Bjorn Helgaas
2023-11-21 15:24 ` Tomasz Pala
2023-11-21 18:19 ` Bjorn Helgaas
2023-11-08 17:47 ` Bjorn Helgaas
2023-11-08 20:20 ` Bjorn Helgaas
2023-12-06 11:54 ` Linux regression tracking #update (Thorsten Leemhuis)
2023-01-10 20:38 ` [PATCH 0/2] PCI: Fix extended config space regression Liang, Kan
2023-01-10 22:57 ` Luck, Tony
2023-01-11 12:10 ` Giovanni Cabiddu
2023-01-11 22:30 ` Bjorn Helgaas
2023-01-12 5:50 ` Sun, Yunying
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=20230110193043.GA1603645@bhelgaas \
--to=helgaas@kernel.org \
--cc=2lprbe78@duck.com \
--cc=benoitg@coeus.ca \
--cc=bhelgaas@google.com \
--cc=dan.j.williams@intel.com \
--cc=dave.jiang@intel.com \
--cc=david.e.box@intel.com \
--cc=giovanni.cabiddu@intel.com \
--cc=hdegoede@redhat.com \
--cc=herbert@gondor.apana.org.au \
--cc=kan.liang@linux.intel.com \
--cc=kjhambrick@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=linuxkernelml@undead.fr \
--cc=mika.westerberg@linux.intel.com \
--cc=mumblingdrunkard@protonmail.com \
--cc=nicholas.johnson-opensource@outlook.com.au \
--cc=tony.luck@intel.com \
--cc=wse@tuxedocomputers.com \
--cc=yunying.sun@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).