Re: [PATCH] ACPI: OSL: Fix the memory mapping of an ACPI GAS that addresses a data structure

[James Liu <james.liu@hpe.com>]

O Tue, May 17, 2022 at 08:25:55PM +0200, Rafael J. Wysocki wrote:
> On Tue, May 17, 2022 at 7:56 PM James Liu <james.liu@hpe.com> wrote:
> >
> > Hi Rafael and all,
> > Could you take a look into this patches? The mentioned bug blocks EINJ
> > testing whenever a system firmware can correctly support GAS according
> > to ACPI 6.4.
> 
> The kernel test robot reported an issue with it.  Have you seen that report?
>
Yes, I saw that the issue is warnings about the initialization of "u64 addr".
Will you suggest fixing it in another patch or just merge the change
(i.e., u64 addr = 0) into this one? Thanks for reviewing this.

> > On Fri, Apr 01, 2022 at 05:28:40PM +0000, james.liu@hpe.com wrote:
> > > From: James Liu <james.liu@hpe.com>
> > >
> > >     Modify acpi_os_map_generic_address and acpi_os_unmap_generic_address
> > >     to handle a case that a GAS table (i.e., Table 5.1 in ACPI 6.4) is used
> > >     to address a data structure; in this case, the GAS has the field of
> > >     "Register Bit Width" equal to 0.
> > >
> > >     For example, "Injection Instruction Entry" (Table 18.25 in ACPI 6.4)
> > >     has a RegisterRegion field that is a GAS that points to a data
> > >     structure SET_ERROR_TYPE_WITH_ADDRESS (Table 18.30), which is required
> > >     when using EINJ (Error Injection module).
> > >
> > >     This fix preserves a fairly sufficient memory space (i.e., page size)
> > >     to store the data structure so as to prevent EINJ module from loading
> > >     failure if platform firmware can support Injection Instruction Entry in
> > >     an EINJ table.
> > >
> > > Signed-off-by: James Liu <james.liu@hpe.com>
> > > ---
> > >  drivers/acpi/osl.c | 23 +++++++++++++++++++++--
> > >  1 file changed, 21 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
> > > index 45c5c0e45..ab2f584b1 100644
> > > --- a/drivers/acpi/osl.c
> > > +++ b/drivers/acpi/osl.c
> > > @@ -457,9 +457,15 @@ void __iomem *acpi_os_map_generic_address(struct acpi_generic_address *gas)
> > >       if (gas->space_id != ACPI_ADR_SPACE_SYSTEM_MEMORY)
> > >               return NULL;
> > >
> > > +     /* Handle a case that GAS is used to address an ACPI data structure */
> > > +     if (!gas->bit_width) {
> > > +             pr_info("An ACPI data structure at 0x%llx is mapped\n", addr);
> > > +             return  acpi_os_map_iomem(addr, PAGE_SIZE);
> > > +     }
> > > +
> > >       /* Handle possible alignment issues */
> > >       memcpy(&addr, &gas->address, sizeof(addr));
> > > -     if (!addr || !gas->bit_width)
> > > +     if (!addr)
> > >               return NULL;
> > >
> > >       return acpi_os_map_iomem(addr, gas->bit_width / 8);
> > > @@ -474,9 +480,22 @@ void acpi_os_unmap_generic_address(struct acpi_generic_address *gas)
> > >       if (gas->space_id != ACPI_ADR_SPACE_SYSTEM_MEMORY)
> > >               return;
> > >
> > > +     /* Handle a case that GAS is used to address an ACPI data structure */
> > > +     if (!gas->bit_width) {
> > > +             pr_info("An ACPI data structure at 0x%llx is unmapped\n", addr);
> > > +             mutex_lock(&acpi_ioremap_lock);
> > > +             map = acpi_map_lookup(addr, PAGE_SIZE);
> > > +             if (!map) {
> > > +                     mutex_unlock(&acpi_ioremap_lock);
> > > +                     return;
> > > +             }
> > > +             acpi_os_drop_map_ref(map);
> > > +             mutex_unlock(&acpi_ioremap_lock);
> > > +     }
> > > +
> > >       /* Handle possible alignment issues */
> > >       memcpy(&addr, &gas->address, sizeof(addr));
> > > -     if (!addr || !gas->bit_width)
> > > +     if (!addr)
> > >               return;
> > >
> > >       mutex_lock(&acpi_ioremap_lock);
> > > --
> > > 2.25.1
> > >