Re: [PATCH v3] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
by Lorenzo Pieralisi
On Thu, Sep 23, 2021 at 11:35:57AM +0800, Jia He wrote:
> This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
>
> The memory semantics added in commit 437b38c51162 causes SystemMemory
> Operation region, whose address range is not described in the EFI memory
> map to be mapped as NormalNC memory on arm64 platforms (through
> acpi_os_map_memory() in acpi_ex_system_memory_space_handler()).
>
> This triggers the following abort on an ARM64 Ampere eMAG machine,
> because presumably the physical address range area backing the Opregion
> does not support NormalNC memory attributes driven on the bus.
>
> Internal error: synchronous external abort: 96000410 [#1] SMP
> Modules linked in:
> CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
> Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
> pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [...snip...]
> Call trace:
> acpi_ex_system_memory_space_handler+0x26c/0x2c8
> acpi_ev_address_space_dispatch+0x228/0x2c4
> acpi_ex_access_region+0x114/0x268
> acpi_ex_field_datum_io+0x128/0x1b8
> acpi_ex_extract_from_field+0x14c/0x2ac
> acpi_ex_read_data_from_field+0x190/0x1b8
> acpi_ex_resolve_node_to_value+0x1ec/0x288
> acpi_ex_resolve_to_value+0x250/0x274
> acpi_ds_evaluate_name_path+0xac/0x124
> acpi_ds_exec_end_op+0x90/0x410
> acpi_ps_parse_loop+0x4ac/0x5d8
> acpi_ps_parse_aml+0xe0/0x2c8
> acpi_ps_execute_method+0x19c/0x1ac
> acpi_ns_evaluate+0x1f8/0x26c
> acpi_ns_init_one_device+0x104/0x140
> acpi_ns_walk_namespace+0x158/0x1d0
> acpi_ns_initialize_devices+0x194/0x218
> acpi_initialize_objects+0x48/0x50
> acpi_init+0xe0/0x498
>
> If the Opregion address range is not present in the EFI memory map there
> is no way for us to determine the memory attributes to use to map it -
> defaulting to NormalNC does not work (and it is not correct on a memory
> region that may have read side-effects) and therefore commit
> 437b38c51162 should be reverted, which means reverting back to the
> original behavior whereby address ranges that are mapped using
> acpi_os_map_memory() default to the safe devicenGnRnE attributes on
> ARM64 if the mapped address range is not defined in the EFI memory map.
>
> Fixes: 437b38c51162 ("ACPI: Add memory semantics to acpi_os_map_memory()")
> Signed-off-by: Jia He <justin.he(a)arm.com>
> Cc: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
> Cc: Ard Biesheuvel <ardb(a)kernel.org>
> Cc: Hanjun Guo <guohanjun(a)huawei.com>
> Cc: Catalin Marinas <catalin.marinas(a)arm.com>
> Cc: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com>
> Cc: Harb Abdulhamid <harb(a)amperecomputing.com>
> ---
> V3: refine the commit log which came from Lorenzo Pieralisi
>
> arch/arm64/include/asm/acpi.h | 3 ---
> arch/arm64/kernel/acpi.c | 19 +++----------------
> drivers/acpi/osl.c | 23 +++++++----------------
> include/acpi/acpi_io.h | 8 --------
> 4 files changed, 10 insertions(+), 43 deletions(-)
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
Rafael if possible please apply it, thank you very much.
> diff --git a/arch/arm64/include/asm/acpi.h b/arch/arm64/include/asm/acpi.h
> index 7535dc7cc5aa..bd68e1b7f29f 100644
> --- a/arch/arm64/include/asm/acpi.h
> +++ b/arch/arm64/include/asm/acpi.h
> @@ -50,9 +50,6 @@ pgprot_t __acpi_get_mem_attribute(phys_addr_t addr);
> void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size);
> #define acpi_os_ioremap acpi_os_ioremap
>
> -void __iomem *acpi_os_memmap(acpi_physical_address phys, acpi_size size);
> -#define acpi_os_memmap acpi_os_memmap
> -
> typedef u64 phys_cpuid_t;
> #define PHYS_CPUID_INVALID INVALID_HWID
>
> diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c
> index 1c9c2f7a1c04..f3851724fe35 100644
> --- a/arch/arm64/kernel/acpi.c
> +++ b/arch/arm64/kernel/acpi.c
> @@ -273,8 +273,7 @@ pgprot_t __acpi_get_mem_attribute(phys_addr_t addr)
> return __pgprot(PROT_DEVICE_nGnRnE);
> }
>
> -static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
> - acpi_size size, bool memory)
> +void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
> {
> efi_memory_desc_t *md, *region = NULL;
> pgprot_t prot;
> @@ -300,11 +299,9 @@ static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
> * It is fine for AML to remap regions that are not represented in the
> * EFI memory map at all, as it only describes normal memory, and MMIO
> * regions that require a virtual mapping to make them accessible to
> - * the EFI runtime services. Determine the region default
> - * attributes by checking the requested memory semantics.
> + * the EFI runtime services.
> */
> - prot = memory ? __pgprot(PROT_NORMAL_NC) :
> - __pgprot(PROT_DEVICE_nGnRnE);
> + prot = __pgprot(PROT_DEVICE_nGnRnE);
> if (region) {
> switch (region->type) {
> case EFI_LOADER_CODE:
> @@ -364,16 +361,6 @@ static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
> return __ioremap(phys, size, prot);
> }
>
> -void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
> -{
> - return __acpi_os_ioremap(phys, size, false);
> -}
> -
> -void __iomem *acpi_os_memmap(acpi_physical_address phys, acpi_size size)
> -{
> - return __acpi_os_ioremap(phys, size, true);
> -}
> -
> /*
> * Claim Synchronous External Aborts as a firmware first notification.
> *
> diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
> index a43f1521efe6..45c5c0e45e33 100644
> --- a/drivers/acpi/osl.c
> +++ b/drivers/acpi/osl.c
> @@ -284,8 +284,7 @@ acpi_map_lookup_virt(void __iomem *virt, acpi_size size)
> #define should_use_kmap(pfn) page_is_ram(pfn)
> #endif
>
> -static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz,
> - bool memory)
> +static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz)
> {
> unsigned long pfn;
>
> @@ -295,8 +294,7 @@ static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz,
> return NULL;
> return (void __iomem __force *)kmap(pfn_to_page(pfn));
> } else
> - return memory ? acpi_os_memmap(pg_off, pg_sz) :
> - acpi_os_ioremap(pg_off, pg_sz);
> + return acpi_os_ioremap(pg_off, pg_sz);
> }
>
> static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
> @@ -311,10 +309,9 @@ static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
> }
>
> /**
> - * __acpi_os_map_iomem - Get a virtual address for a given physical address range.
> + * acpi_os_map_iomem - Get a virtual address for a given physical address range.
> * @phys: Start of the physical address range to map.
> * @size: Size of the physical address range to map.
> - * @memory: true if remapping memory, false if IO
> *
> * Look up the given physical address range in the list of existing ACPI memory
> * mappings. If found, get a reference to it and return a pointer to it (its
> @@ -324,8 +321,8 @@ static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
> * During early init (when acpi_permanent_mmap has not been set yet) this
> * routine simply calls __acpi_map_table() to get the job done.
> */
> -static void __iomem __ref
> -*__acpi_os_map_iomem(acpi_physical_address phys, acpi_size size, bool memory)
> +void __iomem __ref
> +*acpi_os_map_iomem(acpi_physical_address phys, acpi_size size)
> {
> struct acpi_ioremap *map;
> void __iomem *virt;
> @@ -356,7 +353,7 @@ static void __iomem __ref
>
> pg_off = round_down(phys, PAGE_SIZE);
> pg_sz = round_up(phys + size, PAGE_SIZE) - pg_off;
> - virt = acpi_map(phys, size, memory);
> + virt = acpi_map(phys, size);
> if (!virt) {
> mutex_unlock(&acpi_ioremap_lock);
> kfree(map);
> @@ -375,17 +372,11 @@ static void __iomem __ref
> mutex_unlock(&acpi_ioremap_lock);
> return map->virt + (phys - map->phys);
> }
> -
> -void __iomem *__ref
> -acpi_os_map_iomem(acpi_physical_address phys, acpi_size size)
> -{
> - return __acpi_os_map_iomem(phys, size, false);
> -}
> EXPORT_SYMBOL_GPL(acpi_os_map_iomem);
>
> void *__ref acpi_os_map_memory(acpi_physical_address phys, acpi_size size)
> {
> - return (void *)__acpi_os_map_iomem(phys, size, true);
> + return (void *)acpi_os_map_iomem(phys, size);
> }
> EXPORT_SYMBOL_GPL(acpi_os_map_memory);
>
> diff --git a/include/acpi/acpi_io.h b/include/acpi/acpi_io.h
> index a0212e67d6f4..027faa8883aa 100644
> --- a/include/acpi/acpi_io.h
> +++ b/include/acpi/acpi_io.h
> @@ -14,14 +14,6 @@ static inline void __iomem *acpi_os_ioremap(acpi_physical_address phys,
> }
> #endif
>
> -#ifndef acpi_os_memmap
> -static inline void __iomem *acpi_os_memmap(acpi_physical_address phys,
> - acpi_size size)
> -{
> - return ioremap_cache(phys, size);
> -}
> -#endif
> -
> extern bool acpi_permanent_mmap;
>
> void __iomem __ref
> --
> 2.17.1
>
10 months, 2 weeks
Re: [PATCH v2] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
by Lorenzo Pieralisi
On Thu, Sep 23, 2021 at 01:09:58AM +0200, Mark Kettenis wrote:
> > Date: Wed, 22 Sep 2021 17:33:36 +0100
> > From: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
> >
> > On Fri, Sep 10, 2021 at 10:32:23PM +0800, Jia He wrote:
> > > This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
> > >
> > > After this commit, a boot panic is alway hit on an Ampere EMAG server
> > > with call trace as follows:
> > > Internal error: synchronous external abort: 96000410 [#1] SMP
> > > Modules linked in:
> > > CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
> > > Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
> > > pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > > [...snip...]
> > > Call trace:
> > > acpi_ex_system_memory_space_handler+0x26c/0x2c8
> > > acpi_ev_address_space_dispatch+0x228/0x2c4
> > > acpi_ex_access_region+0x114/0x268
> > > acpi_ex_field_datum_io+0x128/0x1b8
> > > acpi_ex_extract_from_field+0x14c/0x2ac
> > > acpi_ex_read_data_from_field+0x190/0x1b8
> > > acpi_ex_resolve_node_to_value+0x1ec/0x288
> > > acpi_ex_resolve_to_value+0x250/0x274
> > > acpi_ds_evaluate_name_path+0xac/0x124
> > > acpi_ds_exec_end_op+0x90/0x410
> > > acpi_ps_parse_loop+0x4ac/0x5d8
> > > acpi_ps_parse_aml+0xe0/0x2c8
> > > acpi_ps_execute_method+0x19c/0x1ac
> > > acpi_ns_evaluate+0x1f8/0x26c
> > > acpi_ns_init_one_device+0x104/0x140
> > > acpi_ns_walk_namespace+0x158/0x1d0
> > > acpi_ns_initialize_devices+0x194/0x218
> > > acpi_initialize_objects+0x48/0x50
> > > acpi_init+0xe0/0x498
> > >
> > > As mentioned by Lorenzo:
> > > "We are forcing memory semantics mappings to PROT_NORMAL_NC, which
> > > eMAG does not like at all and I'd need to understand why. It looks
> > > like the issue happen in SystemMemory Opregion handler."
> > >
> > > Hence just revert it before everything is clear.
> > >
> > > Fixes: 437b38c51162 ("ACPI: Add memory semantics to acpi_os_map_memory()")
> > > Cc: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
> > > Cc: Ard Biesheuvel <ardb(a)kernel.org>
> > > Cc: Hanjun Guo <guohanjun(a)huawei.com>
> > > Cc: Catalin Marinas <catalin.marinas(a)arm.com>
> > > Cc: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com>
> > > Cc: Harb Abdulhamid <harb(a)amperecomputing.com>
> > >
> > > Signed-off-by: Jia He <justin.he(a)arm.com>
> >
> > Rewrote the commit log, please take the patch below and repost
> > it as a v3.
> >
> > It would still be great if Ampere can help us understand why
> > the NormalNC attributes trigger a sync abort on the opregion
> > before merging it.
>
> To be honest, I don't think you really need an explanation from Ampere
> here. Mapping a part of the address space that doesn't provide memory
> semantics with NormalNC attributes is wrong and triggering a sync
> abort in that case is way better than silently ignoring the access.
That's understood and that's what I explained in the revert commit
log, no question about it.
I was just asking to confirm if that's what's actually happening.
> Putting my OpenBSD hat on (where we have our own ACPI OSPM
> implementation) I must say that we always interpreted SystemMemory as
> memory mapped IO and I think that is a logical choice as SystemIO is
> used for (non-memory mapped) IO. And I'd say that the ACPI OSPM code
> should make sure that it uses properly aligned access to any Field
> object that doesn't use AnyAcc as its access type. Even on x86! And
> I'd say that AML that uses AnyAcc fields for SystemMemory OpRegions on
> arm64 is buggy.
>
> But maybe relaxing this when the EFI memory map indicates that the
> address space in question does provide memory semantics does make
> sense. That should defenitely be documented in the ACPI standard
> though.
Mapping SystemMemory Opregions as "memory" does not make sense
at all to me. Still, that's what Linux ACPICA code does (*if*
that's what acpi_os_map_memory() is supposed to mean).
https://lore.kernel.org/linux-acpi/[email protected]
Where do we go from here, to be defined, we still have a bug
to fix after the revert is applied.
drivers/acpi/sysfs.c
maps BERT error regions with acpi_os_map_memory(). If the BERT error
region is not in the EFI memory map, we map that memory as device-nGnRnE
and we execute memory semantics operation on it.
https://lore.kernel.org/linux-acpi/[email protected]
I could change that code to map those regions as ioremap_wc() because
supposedly we *know* that's memory but this is becoming a slippery
slope to follow IMO.
> > -- >8 --
> > Subject: [PATCH] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
> >
> > This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
> >
> > The memory semantics added in commit 437b38c51162 causes SystemMemory
> > Operation region, whose address range is not described in the EFI memory
> > map to be mapped as NormalNC memory on arm64 platforms (through
> > acpi_os_map_memory() in acpi_ex_system_memory_space_handler()).
> >
> > This triggers the following abort on an ARM64 Ampere eMAG machine,
> > because presumably the physical address range area backing the Opregion
> > does not support NormalNC memory attributes driven on the bus.
> >
> > Internal error: synchronous external abort: 96000410 [#1] SMP
> > Modules linked in:
> > CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
> > Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
> > pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > [...snip...]
> > Call trace:
> > acpi_ex_system_memory_space_handler+0x26c/0x2c8
> > acpi_ev_address_space_dispatch+0x228/0x2c4
> > acpi_ex_access_region+0x114/0x268
> > acpi_ex_field_datum_io+0x128/0x1b8
> > acpi_ex_extract_from_field+0x14c/0x2ac
> > acpi_ex_read_data_from_field+0x190/0x1b8
> > acpi_ex_resolve_node_to_value+0x1ec/0x288
> > acpi_ex_resolve_to_value+0x250/0x274
> > acpi_ds_evaluate_name_path+0xac/0x124
> > acpi_ds_exec_end_op+0x90/0x410
> > acpi_ps_parse_loop+0x4ac/0x5d8
> > acpi_ps_parse_aml+0xe0/0x2c8
> > acpi_ps_execute_method+0x19c/0x1ac
> > acpi_ns_evaluate+0x1f8/0x26c
> > acpi_ns_init_one_device+0x104/0x140
> > acpi_ns_walk_namespace+0x158/0x1d0
> > acpi_ns_initialize_devices+0x194/0x218
> > acpi_initialize_objects+0x48/0x50
> > acpi_init+0xe0/0x498
> >
> > If the Opregion address range is not present in the EFI memory map there
> > is no way for us to determine the memory attributes to use to map it -
> > defaulting to NormalNC does not work (and it is not correct on a memory
> > region that may have read side-effects) and therefore commit
> > 437b38c51162 should be reverted, which means reverting back to the
> > original behavior whereby address ranges that are mapped using
> > acpi_os_map_memory() default to the safe devicenGnRnE attributes on
> > ARM64 if the mapped address range is not defined in the EFI memory map.
> >
> > Fixes: 437b38c51162 ("ACPI: Add memory semantics to acpi_os_map_memory()")
> > Signed-off-by: Jia He <justin.he(a)arm.com>
> > Cc: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
> > Cc: Ard Biesheuvel <ardb(a)kernel.org>
> > Cc: Hanjun Guo <guohanjun(a)huawei.com>
> > Cc: Catalin Marinas <catalin.marinas(a)arm.com>
> > Cc: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com>
> > Cc: Harb Abdulhamid <harb(a)amperecomputing.com>
> > ---
> > arch/arm64/include/asm/acpi.h | 3 ---
> > arch/arm64/kernel/acpi.c | 19 +++----------------
> > drivers/acpi/osl.c | 23 +++++++----------------
> > include/acpi/acpi_io.h | 8 --------
> > 4 files changed, 10 insertions(+), 43 deletions(-)
> >
> > diff --git a/arch/arm64/include/asm/acpi.h b/arch/arm64/include/asm/acpi.h
> > index 7535dc7cc5aa..bd68e1b7f29f 100644
> > --- a/arch/arm64/include/asm/acpi.h
> > +++ b/arch/arm64/include/asm/acpi.h
> > @@ -50,9 +50,6 @@ pgprot_t __acpi_get_mem_attribute(phys_addr_t addr);
> > void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size);
> > #define acpi_os_ioremap acpi_os_ioremap
> >
> > -void __iomem *acpi_os_memmap(acpi_physical_address phys, acpi_size size);
> > -#define acpi_os_memmap acpi_os_memmap
> > -
> > typedef u64 phys_cpuid_t;
> > #define PHYS_CPUID_INVALID INVALID_HWID
> >
> > diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c
> > index 1c9c2f7a1c04..f3851724fe35 100644
> > --- a/arch/arm64/kernel/acpi.c
> > +++ b/arch/arm64/kernel/acpi.c
> > @@ -273,8 +273,7 @@ pgprot_t __acpi_get_mem_attribute(phys_addr_t addr)
> > return __pgprot(PROT_DEVICE_nGnRnE);
> > }
> >
> > -static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
> > - acpi_size size, bool memory)
> > +void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
> > {
> > efi_memory_desc_t *md, *region = NULL;
> > pgprot_t prot;
> > @@ -300,11 +299,9 @@ static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
> > * It is fine for AML to remap regions that are not represented in the
> > * EFI memory map at all, as it only describes normal memory, and MMIO
> > * regions that require a virtual mapping to make them accessible to
> > - * the EFI runtime services. Determine the region default
> > - * attributes by checking the requested memory semantics.
> > + * the EFI runtime services.
> > */
> > - prot = memory ? __pgprot(PROT_NORMAL_NC) :
> > - __pgprot(PROT_DEVICE_nGnRnE);
> > + prot = __pgprot(PROT_DEVICE_nGnRnE);
> > if (region) {
> > switch (region->type) {
> > case EFI_LOADER_CODE:
> > @@ -364,16 +361,6 @@ static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
> > return __ioremap(phys, size, prot);
> > }
> >
> > -void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
> > -{
> > - return __acpi_os_ioremap(phys, size, false);
> > -}
> > -
> > -void __iomem *acpi_os_memmap(acpi_physical_address phys, acpi_size size)
> > -{
> > - return __acpi_os_ioremap(phys, size, true);
> > -}
> > -
> > /*
> > * Claim Synchronous External Aborts as a firmware first notification.
> > *
> > diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
> > index a43f1521efe6..45c5c0e45e33 100644
> > --- a/drivers/acpi/osl.c
> > +++ b/drivers/acpi/osl.c
> > @@ -284,8 +284,7 @@ acpi_map_lookup_virt(void __iomem *virt, acpi_size size)
> > #define should_use_kmap(pfn) page_is_ram(pfn)
> > #endif
> >
> > -static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz,
> > - bool memory)
> > +static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz)
> > {
> > unsigned long pfn;
> >
> > @@ -295,8 +294,7 @@ static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz,
> > return NULL;
> > return (void __iomem __force *)kmap(pfn_to_page(pfn));
> > } else
> > - return memory ? acpi_os_memmap(pg_off, pg_sz) :
> > - acpi_os_ioremap(pg_off, pg_sz);
> > + return acpi_os_ioremap(pg_off, pg_sz);
> > }
> >
> > static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
> > @@ -311,10 +309,9 @@ static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
> > }
> >
> > /**
> > - * __acpi_os_map_iomem - Get a virtual address for a given physical address range.
> > + * acpi_os_map_iomem - Get a virtual address for a given physical address range.
> > * @phys: Start of the physical address range to map.
> > * @size: Size of the physical address range to map.
> > - * @memory: true if remapping memory, false if IO
> > *
> > * Look up the given physical address range in the list of existing ACPI memory
> > * mappings. If found, get a reference to it and return a pointer to it (its
> > @@ -324,8 +321,8 @@ static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
> > * During early init (when acpi_permanent_mmap has not been set yet) this
> > * routine simply calls __acpi_map_table() to get the job done.
> > */
> > -static void __iomem __ref
> > -*__acpi_os_map_iomem(acpi_physical_address phys, acpi_size size, bool memory)
> > +void __iomem __ref
> > +*acpi_os_map_iomem(acpi_physical_address phys, acpi_size size)
> > {
> > struct acpi_ioremap *map;
> > void __iomem *virt;
> > @@ -356,7 +353,7 @@ static void __iomem __ref
> >
> > pg_off = round_down(phys, PAGE_SIZE);
> > pg_sz = round_up(phys + size, PAGE_SIZE) - pg_off;
> > - virt = acpi_map(phys, size, memory);
> > + virt = acpi_map(phys, size);
> > if (!virt) {
> > mutex_unlock(&acpi_ioremap_lock);
> > kfree(map);
> > @@ -375,17 +372,11 @@ static void __iomem __ref
> > mutex_unlock(&acpi_ioremap_lock);
> > return map->virt + (phys - map->phys);
> > }
> > -
> > -void __iomem *__ref
> > -acpi_os_map_iomem(acpi_physical_address phys, acpi_size size)
> > -{
> > - return __acpi_os_map_iomem(phys, size, false);
> > -}
> > EXPORT_SYMBOL_GPL(acpi_os_map_iomem);
> >
> > void *__ref acpi_os_map_memory(acpi_physical_address phys, acpi_size size)
> > {
> > - return (void *)__acpi_os_map_iomem(phys, size, true);
> > + return (void *)acpi_os_map_iomem(phys, size);
> > }
> > EXPORT_SYMBOL_GPL(acpi_os_map_memory);
> >
> > diff --git a/include/acpi/acpi_io.h b/include/acpi/acpi_io.h
> > index a0212e67d6f4..027faa8883aa 100644
> > --- a/include/acpi/acpi_io.h
> > +++ b/include/acpi/acpi_io.h
> > @@ -14,14 +14,6 @@ static inline void __iomem *acpi_os_ioremap(acpi_physical_address phys,
> > }
> > #endif
> >
> > -#ifndef acpi_os_memmap
> > -static inline void __iomem *acpi_os_memmap(acpi_physical_address phys,
> > - acpi_size size)
> > -{
> > - return ioremap_cache(phys, size);
> > -}
> > -#endif
> > -
> > extern bool acpi_permanent_mmap;
> >
> > void __iomem __ref
> > --
> > 2.31.0
> >
> > _______________________________________________
> > linux-arm-kernel mailing list
> > linux-arm-kernel(a)lists.infradead.org
> > http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
> >
10 months, 2 weeks
Re: [PATCH v2] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
by Jeremy Linton
Hi,
On 9/22/21 6:11 AM, Ard Biesheuvel wrote:
> On Tue, 21 Sept 2021 at 12:05, Lorenzo Pieralisi
> <lorenzo.pieralisi(a)arm.com> wrote:
>>
>> On Mon, Sep 20, 2021 at 07:32:56PM +0200, Rafael J. Wysocki wrote:
>>> On Mon, Sep 20, 2021 at 7:03 PM Lorenzo Pieralisi
>>> <lorenzo.pieralisi(a)arm.com> wrote:
>>>>
>>>> On Thu, Sep 16, 2021 at 05:08:27PM +0100, Lorenzo Pieralisi wrote:
>>>>> On Fri, Sep 10, 2021 at 07:28:49PM +0200, Ard Biesheuvel wrote:
>>>>>> On Fri, 10 Sept 2021 at 16:32, Jia He <justin.he(a)arm.com> wrote:
>>>>>>>
>>>>>>> This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
>>>>>>>
>>>>>>> After this commit, a boot panic is alway hit on an Ampere EMAG server
>>>>>>> with call trace as follows:
>>>>>>> Internal error: synchronous external abort: 96000410 [#1] SMP
>>>>>>> Modules linked in:
>>>>>>> CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
>>>>>>> Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
>>>>>>> pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
>>>>>>> [...snip...]
>>>>>>> Call trace:
>>>>>>> acpi_ex_system_memory_space_handler+0x26c/0x2c8
>>>>>>> acpi_ev_address_space_dispatch+0x228/0x2c4
>>>>>>> acpi_ex_access_region+0x114/0x268
>>>>>>> acpi_ex_field_datum_io+0x128/0x1b8
>>>>>>> acpi_ex_extract_from_field+0x14c/0x2ac
>>>>>>> acpi_ex_read_data_from_field+0x190/0x1b8
>>>>>>> acpi_ex_resolve_node_to_value+0x1ec/0x288
>>>>>>> acpi_ex_resolve_to_value+0x250/0x274
>>>>>>> acpi_ds_evaluate_name_path+0xac/0x124
>>>>>>> acpi_ds_exec_end_op+0x90/0x410
>>>>>>> acpi_ps_parse_loop+0x4ac/0x5d8
>>>>>>> acpi_ps_parse_aml+0xe0/0x2c8
>>>>>>> acpi_ps_execute_method+0x19c/0x1ac
>>>>>>> acpi_ns_evaluate+0x1f8/0x26c
>>>>>>> acpi_ns_init_one_device+0x104/0x140
>>>>>>> acpi_ns_walk_namespace+0x158/0x1d0
>>>>>>> acpi_ns_initialize_devices+0x194/0x218
>>>>>>> acpi_initialize_objects+0x48/0x50
>>>>>>> acpi_init+0xe0/0x498
>>>>>>>
>>>>>>> As mentioned by Lorenzo:
>>>>>>> "We are forcing memory semantics mappings to PROT_NORMAL_NC, which
>>>>>>> eMAG does not like at all and I'd need to understand why. It looks
>>>>>>> like the issue happen in SystemMemory Opregion handler."
>>>>>>>
>>>>>>> Hence just revert it before everything is clear.
>>>>>>>
>>>>>>
>>>>>> Can we try to find the root cause first? -rc1 is not even out yet, and
>>>>>> reverting it now means we can not resubmit it until the next merge
>>>>>> window.
>>>>>
>>>>> I am waiting to debug this on an eMAG but I noticed something that
>>>>> I wanted to bring up.
>>>>>
>>>>> SystemMemory Operation region handler - ie
>>>>>
>>>>> acpi_ex_system_memory_space_handler()
>>>>>
>>>>> maps the Operation Region (that AFAICS is MMIO, it is _not_ memory)
>>>>> with acpi_os_map_memory() and I believe that's what is causing this
>>>>> bug.
>>>>>
>>>>> On the other hand, acpi_os_map_generic_address(), to handle spaceid
>>>>> ACPI_ADR_SPACE_SYSTEM_MEMORY, uses acpi_os_map_iomem() that is more
>>>>> in line with my expectations.
>>>>
>>>> Hi Rafael,
>>>>
>>>> I wanted to ask please if you have any insights on why
>>>>
>>>> (1) acpi_ex_system_memory_space_handler()
>>>> (2) acpi_os_map_generic_address()
>>>>
>>>> Use two different calls to map memory for the _same_ address space ID
>>>> (SystemMemory).
>>>>
>>>> (3) acpi_os_map_memory()
>>>> vs
>>>> (4) acpi_os_map_iomem()
>>>
>>> I don't really have a good answer here.
>>>
>>> On x86 this doesn't really matter and that's where
>>> acpi_ex_system_memory_space_handler() was first introduced. It is not
>>> only used for IOMEM (there are SystemMemory operation regions in RAM),
>>> but since it may be in IOMEM, it should assume so.
>>>
>>>> I am struggling to understand why (1) uses (3) ("memory semantics") when
>>>> (2) uses (4) - it is actually unclear how the distinction between
>>>> the two mapping APIs is to be drawn and on what basis one should
>>>> choose which one to use.
>>>>
>>>> I am still waiting to grab some HW to debug this report but the issue
>>>> here is that we are mapping an OpRegion SystemMemory with (3) in the
>>>> memory space handler and given the patch we are reverting we end up
>>>> mapping the operation region with normal non-cacheable memory attributes
>>>> that probably the physical address range behind the OpRegion does not
>>>> support.
>>>
>>> If that is the case, there needs to be a mechanism to decide what kind
>>> of mapping to use for SystemMemory operation regions based on the type
>>> of physical memory the address range in question is located in.
>>
>> Thank you Rafael. The mechanism we are currently relying on is the EFI
>> memory map but if the Opregion address is not described there then we
>> are left with a default choice to make (theoretically I may also parse
>> all _CRS in the namespace to find whether a resource include the
>> Opregion and I may infer attributes from the _CRS resource entry).
>>
>
> I'm not sure that would help, as I would expected the memory described
> by _CRS to be mostly mutually exclusive from memory used by OpRegions.
>
>> Maybe we should update the ACPI specs to enforce it; with current
>> firmware the idea of using the OS expected *usage* of memory (ie
>> memory vs IO) described by the mapping function prototype can't work
>> as this revert shows (even though it would be better if I manage
>> to find what the precise issue is).
>>
>> We can't map something with specific attributes if we don't know
>> whether the physical address space backing the region supports it.
>>
>
> We don't have a a safe default in either direction, so I agree this is
> a hole in the specs.
I just debugged down to this patch because of boot failures with the
rpi4. Then I found this thread.
I had always assumed that SystemIO was x86/PIO, and SystemMemory was
defined as MMIO but on arm we would have to determine if the memory
region was described in the uefi memory map as actual system ram, and if
not assume device memory. I was looking at tweaking acpi_map() to check
that similar to what is happening on !arm64.
Gotta run, but throwing that out there since it seems a possible fix
without creating a new OpregionType to differentiate actual memory vs MMIO.
>
>> I am left with little choice: I assume the best thing I could do
>> to fix the original bug is to use ioremap_* in acpi_data_show()
>> instead of acpi_os_map/unmap_memory() to map that memory with
>> specific attributes (for BERT error regions, they must be RAM
>> so, _hopefully_, we know it can be mapped with eg normal memory
>> mappings).
>>
>> Thoughts ?
>>
>
> One thing I just realized is that the EFI memory map is not a complete
> solution to begin with, as it may not cover hot/coldplugged memory
> regions that are only described via ACPI.
>
> Did you make any progress with the eMAG?
>
10 months, 2 weeks
Re: [PATCH v2] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
by Lorenzo Pieralisi
On Fri, Sep 10, 2021 at 10:32:23PM +0800, Jia He wrote:
> This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
>
> After this commit, a boot panic is alway hit on an Ampere EMAG server
> with call trace as follows:
> Internal error: synchronous external abort: 96000410 [#1] SMP
> Modules linked in:
> CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
> Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
> pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> [...snip...]
> Call trace:
> acpi_ex_system_memory_space_handler+0x26c/0x2c8
> acpi_ev_address_space_dispatch+0x228/0x2c4
> acpi_ex_access_region+0x114/0x268
> acpi_ex_field_datum_io+0x128/0x1b8
> acpi_ex_extract_from_field+0x14c/0x2ac
> acpi_ex_read_data_from_field+0x190/0x1b8
> acpi_ex_resolve_node_to_value+0x1ec/0x288
> acpi_ex_resolve_to_value+0x250/0x274
> acpi_ds_evaluate_name_path+0xac/0x124
> acpi_ds_exec_end_op+0x90/0x410
> acpi_ps_parse_loop+0x4ac/0x5d8
> acpi_ps_parse_aml+0xe0/0x2c8
> acpi_ps_execute_method+0x19c/0x1ac
> acpi_ns_evaluate+0x1f8/0x26c
> acpi_ns_init_one_device+0x104/0x140
> acpi_ns_walk_namespace+0x158/0x1d0
> acpi_ns_initialize_devices+0x194/0x218
> acpi_initialize_objects+0x48/0x50
> acpi_init+0xe0/0x498
>
> As mentioned by Lorenzo:
> "We are forcing memory semantics mappings to PROT_NORMAL_NC, which
> eMAG does not like at all and I'd need to understand why. It looks
> like the issue happen in SystemMemory Opregion handler."
>
> Hence just revert it before everything is clear.
>
> Fixes: 437b38c51162 ("ACPI: Add memory semantics to acpi_os_map_memory()")
> Cc: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
> Cc: Ard Biesheuvel <ardb(a)kernel.org>
> Cc: Hanjun Guo <guohanjun(a)huawei.com>
> Cc: Catalin Marinas <catalin.marinas(a)arm.com>
> Cc: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com>
> Cc: Harb Abdulhamid <harb(a)amperecomputing.com>
>
> Signed-off-by: Jia He <justin.he(a)arm.com>
Rewrote the commit log, please take the patch below and repost
it as a v3.
It would still be great if Ampere can help us understand why
the NormalNC attributes trigger a sync abort on the opregion
before merging it.
-- >8 --
Subject: [PATCH] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
The memory semantics added in commit 437b38c51162 causes SystemMemory
Operation region, whose address range is not described in the EFI memory
map to be mapped as NormalNC memory on arm64 platforms (through
acpi_os_map_memory() in acpi_ex_system_memory_space_handler()).
This triggers the following abort on an ARM64 Ampere eMAG machine,
because presumably the physical address range area backing the Opregion
does not support NormalNC memory attributes driven on the bus.
Internal error: synchronous external abort: 96000410 [#1] SMP
Modules linked in:
CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[...snip...]
Call trace:
acpi_ex_system_memory_space_handler+0x26c/0x2c8
acpi_ev_address_space_dispatch+0x228/0x2c4
acpi_ex_access_region+0x114/0x268
acpi_ex_field_datum_io+0x128/0x1b8
acpi_ex_extract_from_field+0x14c/0x2ac
acpi_ex_read_data_from_field+0x190/0x1b8
acpi_ex_resolve_node_to_value+0x1ec/0x288
acpi_ex_resolve_to_value+0x250/0x274
acpi_ds_evaluate_name_path+0xac/0x124
acpi_ds_exec_end_op+0x90/0x410
acpi_ps_parse_loop+0x4ac/0x5d8
acpi_ps_parse_aml+0xe0/0x2c8
acpi_ps_execute_method+0x19c/0x1ac
acpi_ns_evaluate+0x1f8/0x26c
acpi_ns_init_one_device+0x104/0x140
acpi_ns_walk_namespace+0x158/0x1d0
acpi_ns_initialize_devices+0x194/0x218
acpi_initialize_objects+0x48/0x50
acpi_init+0xe0/0x498
If the Opregion address range is not present in the EFI memory map there
is no way for us to determine the memory attributes to use to map it -
defaulting to NormalNC does not work (and it is not correct on a memory
region that may have read side-effects) and therefore commit
437b38c51162 should be reverted, which means reverting back to the
original behavior whereby address ranges that are mapped using
acpi_os_map_memory() default to the safe devicenGnRnE attributes on
ARM64 if the mapped address range is not defined in the EFI memory map.
Fixes: 437b38c51162 ("ACPI: Add memory semantics to acpi_os_map_memory()")
Signed-off-by: Jia He <justin.he(a)arm.com>
Cc: Lorenzo Pieralisi <lorenzo.pieralisi(a)arm.com>
Cc: Ard Biesheuvel <ardb(a)kernel.org>
Cc: Hanjun Guo <guohanjun(a)huawei.com>
Cc: Catalin Marinas <catalin.marinas(a)arm.com>
Cc: Rafael J. Wysocki <rafael.j.wysocki(a)intel.com>
Cc: Harb Abdulhamid <harb(a)amperecomputing.com>
---
arch/arm64/include/asm/acpi.h | 3 ---
arch/arm64/kernel/acpi.c | 19 +++----------------
drivers/acpi/osl.c | 23 +++++++----------------
include/acpi/acpi_io.h | 8 --------
4 files changed, 10 insertions(+), 43 deletions(-)
diff --git a/arch/arm64/include/asm/acpi.h b/arch/arm64/include/asm/acpi.h
index 7535dc7cc5aa..bd68e1b7f29f 100644
--- a/arch/arm64/include/asm/acpi.h
+++ b/arch/arm64/include/asm/acpi.h
@@ -50,9 +50,6 @@ pgprot_t __acpi_get_mem_attribute(phys_addr_t addr);
void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size);
#define acpi_os_ioremap acpi_os_ioremap
-void __iomem *acpi_os_memmap(acpi_physical_address phys, acpi_size size);
-#define acpi_os_memmap acpi_os_memmap
-
typedef u64 phys_cpuid_t;
#define PHYS_CPUID_INVALID INVALID_HWID
diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c
index 1c9c2f7a1c04..f3851724fe35 100644
--- a/arch/arm64/kernel/acpi.c
+++ b/arch/arm64/kernel/acpi.c
@@ -273,8 +273,7 @@ pgprot_t __acpi_get_mem_attribute(phys_addr_t addr)
return __pgprot(PROT_DEVICE_nGnRnE);
}
-static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
- acpi_size size, bool memory)
+void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
{
efi_memory_desc_t *md, *region = NULL;
pgprot_t prot;
@@ -300,11 +299,9 @@ static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
* It is fine for AML to remap regions that are not represented in the
* EFI memory map at all, as it only describes normal memory, and MMIO
* regions that require a virtual mapping to make them accessible to
- * the EFI runtime services. Determine the region default
- * attributes by checking the requested memory semantics.
+ * the EFI runtime services.
*/
- prot = memory ? __pgprot(PROT_NORMAL_NC) :
- __pgprot(PROT_DEVICE_nGnRnE);
+ prot = __pgprot(PROT_DEVICE_nGnRnE);
if (region) {
switch (region->type) {
case EFI_LOADER_CODE:
@@ -364,16 +361,6 @@ static void __iomem *__acpi_os_ioremap(acpi_physical_address phys,
return __ioremap(phys, size, prot);
}
-void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size)
-{
- return __acpi_os_ioremap(phys, size, false);
-}
-
-void __iomem *acpi_os_memmap(acpi_physical_address phys, acpi_size size)
-{
- return __acpi_os_ioremap(phys, size, true);
-}
-
/*
* Claim Synchronous External Aborts as a firmware first notification.
*
diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index a43f1521efe6..45c5c0e45e33 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -284,8 +284,7 @@ acpi_map_lookup_virt(void __iomem *virt, acpi_size size)
#define should_use_kmap(pfn) page_is_ram(pfn)
#endif
-static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz,
- bool memory)
+static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz)
{
unsigned long pfn;
@@ -295,8 +294,7 @@ static void __iomem *acpi_map(acpi_physical_address pg_off, unsigned long pg_sz,
return NULL;
return (void __iomem __force *)kmap(pfn_to_page(pfn));
} else
- return memory ? acpi_os_memmap(pg_off, pg_sz) :
- acpi_os_ioremap(pg_off, pg_sz);
+ return acpi_os_ioremap(pg_off, pg_sz);
}
static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
@@ -311,10 +309,9 @@ static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
}
/**
- * __acpi_os_map_iomem - Get a virtual address for a given physical address range.
+ * acpi_os_map_iomem - Get a virtual address for a given physical address range.
* @phys: Start of the physical address range to map.
* @size: Size of the physical address range to map.
- * @memory: true if remapping memory, false if IO
*
* Look up the given physical address range in the list of existing ACPI memory
* mappings. If found, get a reference to it and return a pointer to it (its
@@ -324,8 +321,8 @@ static void acpi_unmap(acpi_physical_address pg_off, void __iomem *vaddr)
* During early init (when acpi_permanent_mmap has not been set yet) this
* routine simply calls __acpi_map_table() to get the job done.
*/
-static void __iomem __ref
-*__acpi_os_map_iomem(acpi_physical_address phys, acpi_size size, bool memory)
+void __iomem __ref
+*acpi_os_map_iomem(acpi_physical_address phys, acpi_size size)
{
struct acpi_ioremap *map;
void __iomem *virt;
@@ -356,7 +353,7 @@ static void __iomem __ref
pg_off = round_down(phys, PAGE_SIZE);
pg_sz = round_up(phys + size, PAGE_SIZE) - pg_off;
- virt = acpi_map(phys, size, memory);
+ virt = acpi_map(phys, size);
if (!virt) {
mutex_unlock(&acpi_ioremap_lock);
kfree(map);
@@ -375,17 +372,11 @@ static void __iomem __ref
mutex_unlock(&acpi_ioremap_lock);
return map->virt + (phys - map->phys);
}
-
-void __iomem *__ref
-acpi_os_map_iomem(acpi_physical_address phys, acpi_size size)
-{
- return __acpi_os_map_iomem(phys, size, false);
-}
EXPORT_SYMBOL_GPL(acpi_os_map_iomem);
void *__ref acpi_os_map_memory(acpi_physical_address phys, acpi_size size)
{
- return (void *)__acpi_os_map_iomem(phys, size, true);
+ return (void *)acpi_os_map_iomem(phys, size);
}
EXPORT_SYMBOL_GPL(acpi_os_map_memory);
diff --git a/include/acpi/acpi_io.h b/include/acpi/acpi_io.h
index a0212e67d6f4..027faa8883aa 100644
--- a/include/acpi/acpi_io.h
+++ b/include/acpi/acpi_io.h
@@ -14,14 +14,6 @@ static inline void __iomem *acpi_os_ioremap(acpi_physical_address phys,
}
#endif
-#ifndef acpi_os_memmap
-static inline void __iomem *acpi_os_memmap(acpi_physical_address phys,
- acpi_size size)
-{
- return ioremap_cache(phys, size);
-}
-#endif
-
extern bool acpi_permanent_mmap;
void __iomem __ref
--
2.31.0
10 months, 2 weeks
Re: [PATCH v2] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
by Lorenzo Pieralisi
On Wed, Sep 22, 2021 at 01:11:26PM +0200, Ard Biesheuvel wrote:
> On Tue, 21 Sept 2021 at 12:05, Lorenzo Pieralisi
> <lorenzo.pieralisi(a)arm.com> wrote:
> >
> > On Mon, Sep 20, 2021 at 07:32:56PM +0200, Rafael J. Wysocki wrote:
> > > On Mon, Sep 20, 2021 at 7:03 PM Lorenzo Pieralisi
> > > <lorenzo.pieralisi(a)arm.com> wrote:
> > > >
> > > > On Thu, Sep 16, 2021 at 05:08:27PM +0100, Lorenzo Pieralisi wrote:
> > > > > On Fri, Sep 10, 2021 at 07:28:49PM +0200, Ard Biesheuvel wrote:
> > > > > > On Fri, 10 Sept 2021 at 16:32, Jia He <justin.he(a)arm.com> wrote:
> > > > > > >
> > > > > > > This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
> > > > > > >
> > > > > > > After this commit, a boot panic is alway hit on an Ampere EMAG server
> > > > > > > with call trace as follows:
> > > > > > > Internal error: synchronous external abort: 96000410 [#1] SMP
> > > > > > > Modules linked in:
> > > > > > > CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
> > > > > > > Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
> > > > > > > pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > > > > > > [...snip...]
> > > > > > > Call trace:
> > > > > > > acpi_ex_system_memory_space_handler+0x26c/0x2c8
> > > > > > > acpi_ev_address_space_dispatch+0x228/0x2c4
> > > > > > > acpi_ex_access_region+0x114/0x268
> > > > > > > acpi_ex_field_datum_io+0x128/0x1b8
> > > > > > > acpi_ex_extract_from_field+0x14c/0x2ac
> > > > > > > acpi_ex_read_data_from_field+0x190/0x1b8
> > > > > > > acpi_ex_resolve_node_to_value+0x1ec/0x288
> > > > > > > acpi_ex_resolve_to_value+0x250/0x274
> > > > > > > acpi_ds_evaluate_name_path+0xac/0x124
> > > > > > > acpi_ds_exec_end_op+0x90/0x410
> > > > > > > acpi_ps_parse_loop+0x4ac/0x5d8
> > > > > > > acpi_ps_parse_aml+0xe0/0x2c8
> > > > > > > acpi_ps_execute_method+0x19c/0x1ac
> > > > > > > acpi_ns_evaluate+0x1f8/0x26c
> > > > > > > acpi_ns_init_one_device+0x104/0x140
> > > > > > > acpi_ns_walk_namespace+0x158/0x1d0
> > > > > > > acpi_ns_initialize_devices+0x194/0x218
> > > > > > > acpi_initialize_objects+0x48/0x50
> > > > > > > acpi_init+0xe0/0x498
> > > > > > >
> > > > > > > As mentioned by Lorenzo:
> > > > > > > "We are forcing memory semantics mappings to PROT_NORMAL_NC, which
> > > > > > > eMAG does not like at all and I'd need to understand why. It looks
> > > > > > > like the issue happen in SystemMemory Opregion handler."
> > > > > > >
> > > > > > > Hence just revert it before everything is clear.
> > > > > > >
> > > > > >
> > > > > > Can we try to find the root cause first? -rc1 is not even out yet, and
> > > > > > reverting it now means we can not resubmit it until the next merge
> > > > > > window.
> > > > >
> > > > > I am waiting to debug this on an eMAG but I noticed something that
> > > > > I wanted to bring up.
> > > > >
> > > > > SystemMemory Operation region handler - ie
> > > > >
> > > > > acpi_ex_system_memory_space_handler()
> > > > >
> > > > > maps the Operation Region (that AFAICS is MMIO, it is _not_ memory)
> > > > > with acpi_os_map_memory() and I believe that's what is causing this
> > > > > bug.
> > > > >
> > > > > On the other hand, acpi_os_map_generic_address(), to handle spaceid
> > > > > ACPI_ADR_SPACE_SYSTEM_MEMORY, uses acpi_os_map_iomem() that is more
> > > > > in line with my expectations.
> > > >
> > > > Hi Rafael,
> > > >
> > > > I wanted to ask please if you have any insights on why
> > > >
> > > > (1) acpi_ex_system_memory_space_handler()
> > > > (2) acpi_os_map_generic_address()
> > > >
> > > > Use two different calls to map memory for the _same_ address space ID
> > > > (SystemMemory).
> > > >
> > > > (3) acpi_os_map_memory()
> > > > vs
> > > > (4) acpi_os_map_iomem()
> > >
> > > I don't really have a good answer here.
> > >
> > > On x86 this doesn't really matter and that's where
> > > acpi_ex_system_memory_space_handler() was first introduced. It is not
> > > only used for IOMEM (there are SystemMemory operation regions in RAM),
> > > but since it may be in IOMEM, it should assume so.
> > >
> > > > I am struggling to understand why (1) uses (3) ("memory semantics") when
> > > > (2) uses (4) - it is actually unclear how the distinction between
> > > > the two mapping APIs is to be drawn and on what basis one should
> > > > choose which one to use.
> > > >
> > > > I am still waiting to grab some HW to debug this report but the issue
> > > > here is that we are mapping an OpRegion SystemMemory with (3) in the
> > > > memory space handler and given the patch we are reverting we end up
> > > > mapping the operation region with normal non-cacheable memory attributes
> > > > that probably the physical address range behind the OpRegion does not
> > > > support.
> > >
> > > If that is the case, there needs to be a mechanism to decide what kind
> > > of mapping to use for SystemMemory operation regions based on the type
> > > of physical memory the address range in question is located in.
> >
> > Thank you Rafael. The mechanism we are currently relying on is the EFI
> > memory map but if the Opregion address is not described there then we
> > are left with a default choice to make (theoretically I may also parse
> > all _CRS in the namespace to find whether a resource include the
> > Opregion and I may infer attributes from the _CRS resource entry).
> >
>
> I'm not sure that would help, as I would expected the memory described
> by _CRS to be mostly mutually exclusive from memory used by OpRegions.
>
> > Maybe we should update the ACPI specs to enforce it; with current
> > firmware the idea of using the OS expected *usage* of memory (ie
> > memory vs IO) described by the mapping function prototype can't work
> > as this revert shows (even though it would be better if I manage
> > to find what the precise issue is).
> >
> > We can't map something with specific attributes if we don't know
> > whether the physical address space backing the region supports it.
> >
>
> We don't have a a safe default in either direction, so I agree this is
> a hole in the specs.
>
> > I am left with little choice: I assume the best thing I could do
> > to fix the original bug is to use ioremap_* in acpi_data_show()
> > instead of acpi_os_map/unmap_memory() to map that memory with
> > specific attributes (for BERT error regions, they must be RAM
> > so, _hopefully_, we know it can be mapped with eg normal memory
> > mappings).
> >
> > Thoughts ?
> >
>
> One thing I just realized is that the EFI memory map is not a complete
> solution to begin with, as it may not cover hot/coldplugged memory
> regions that are only described via ACPI.
>
> Did you make any progress with the eMAG?
I manage to get the ACPI tables dump. The fault is triggered on
a SystemMemory OPregion access (FYI - should be a reset register),
probably (but on this only Ampere can help us) because the MMIO
range in question does not support the AXI attributes assigned
by the NormalNC mapping.
I believe mapping SystemMemory Opregions as NormalNC does not make
much sense anyway.
The UEFI specs seem to hint that the ACPI Op-region cacheability
attributes must be determined through the UEFI memory map, not
sure whether that means that the OpRegion itself _must_ be in
the EFI memory map.
I believe we need to go on with the revert and find a way to fix the
BERT error region mappings, to make them NormalNC so that we can do
unaligned accesses on them.
What to do specs side - to be debated, we have to do something because
it is impossible to handle it sensibly otherwise.
Lorenzo
10 months, 2 weeks
Re: [PATCH v2] Revert "ACPI: Add memory semantics to acpi_os_map_memory()"
by Lorenzo Pieralisi
On Fri, Sep 10, 2021 at 07:28:49PM +0200, Ard Biesheuvel wrote:
> On Fri, 10 Sept 2021 at 16:32, Jia He <justin.he(a)arm.com> wrote:
> >
> > This reverts commit 437b38c51162f8b87beb28a833c4d5dc85fa864e.
> >
> > After this commit, a boot panic is alway hit on an Ampere EMAG server
> > with call trace as follows:
> > Internal error: synchronous external abort: 96000410 [#1] SMP
> > Modules linked in:
> > CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.14.0+ #462
> > Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
> > pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
> > [...snip...]
> > Call trace:
> > acpi_ex_system_memory_space_handler+0x26c/0x2c8
> > acpi_ev_address_space_dispatch+0x228/0x2c4
> > acpi_ex_access_region+0x114/0x268
> > acpi_ex_field_datum_io+0x128/0x1b8
> > acpi_ex_extract_from_field+0x14c/0x2ac
> > acpi_ex_read_data_from_field+0x190/0x1b8
> > acpi_ex_resolve_node_to_value+0x1ec/0x288
> > acpi_ex_resolve_to_value+0x250/0x274
> > acpi_ds_evaluate_name_path+0xac/0x124
> > acpi_ds_exec_end_op+0x90/0x410
> > acpi_ps_parse_loop+0x4ac/0x5d8
> > acpi_ps_parse_aml+0xe0/0x2c8
> > acpi_ps_execute_method+0x19c/0x1ac
> > acpi_ns_evaluate+0x1f8/0x26c
> > acpi_ns_init_one_device+0x104/0x140
> > acpi_ns_walk_namespace+0x158/0x1d0
> > acpi_ns_initialize_devices+0x194/0x218
> > acpi_initialize_objects+0x48/0x50
> > acpi_init+0xe0/0x498
> >
> > As mentioned by Lorenzo:
> > "We are forcing memory semantics mappings to PROT_NORMAL_NC, which
> > eMAG does not like at all and I'd need to understand why. It looks
> > like the issue happen in SystemMemory Opregion handler."
> >
> > Hence just revert it before everything is clear.
> >
>
> Can we try to find the root cause first? -rc1 is not even out yet, and
> reverting it now means we can not resubmit it until the next merge
> window.
I am waiting to debug this on an eMAG but I noticed something that
I wanted to bring up.
SystemMemory Operation region handler - ie
acpi_ex_system_memory_space_handler()
maps the Operation Region (that AFAICS is MMIO, it is _not_ memory)
with acpi_os_map_memory() and I believe that's what is causing this
bug.
On the other hand, acpi_os_map_generic_address(), to handle spaceid
ACPI_ADR_SPACE_SYSTEM_MEMORY, uses acpi_os_map_iomem() that is more
in line with my expectations.
Question is: is the mapping in acpi_ex_system_memory_space_handler()
wrong (and should be patched with acpi_os_map_iomem() ?)
On x86 this should not change a thing, on ARM it would.
I don't think it is right to map SystemMemory Operation regions with
memory semantics but on the other hand, other than the EFI memory map,
there is nothing we can do to determine what a SystemMemory Operation
region address space actually represents.
Thoughts ? Before embarking on patching
acpi_ex_system_memory_space_handler()
I want to make sure my understanding of the SystemMemory space is
correct, comments welcome.
I will pinpoint the trigger for this bug shortly and before doing
anything else.
Thanks,
Lorenzo
10 months, 2 weeks
Re: [PATCH AUTOSEL 4.4 24/35] ACPICA: iASL: Fix for WPBT table with no command-line arguments
by Moore, Robert
This structure is used by the iASL data table disassembler.
Bob
-----Original Message-----
From: Pavel Machek <pavel(a)denx.de>
Sent: Friday, September 10, 2021 12:46 AM
To: Sasha Levin <sashal(a)kernel.org>
Cc: linux-kernel(a)vger.kernel.org; stable(a)vger.kernel.org; Moore, Robert <robert.moore(a)intel.com>; Wysocki, Rafael J <rafael.j.wysocki(a)intel.com>; linux-acpi(a)vger.kernel.org; devel(a)acpica.org
Subject: Re: [PATCH AUTOSEL 4.4 24/35] ACPICA: iASL: Fix for WPBT table with no command-line arguments
Hi!
> Handle the case where the Command-line Arguments table field does not
> exist.
>
> ACPICA commit d6487164497fda170a1b1453c5d58f2be7c873d6
I'm not sure what is going on here, but adding unused definition will not make any difference for 4.4 users, so we don't need this in -stable...?
Best regards,
Pavel
> +++ b/include/acpi/actbl3.h
> @@ -738,6 +738,10 @@ struct acpi_table_wpbt {
> u16 arguments_length;
> };
>
> +struct acpi_wpbt_unicode {
> + u16 *unicode_string;
> +};
> +
> /*******************************************************************************
> *
> * XENV - Xen Environment Table (ACPI 6.0)
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
10 months, 3 weeks
[rafael-pm:bleeding-edge] BUILD SUCCESS 668a3514b8a3dccf147a042777d1e569badca8fa
by kernel test robot
tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git bleeding-edge
branch HEAD: 668a3514b8a3dccf147a042777d1e569badca8fa Merge branch 'devprop' into bleeding-edge
elapsed time: 1321m
configs tested: 105
configs skipped: 4
The following configs have been built successfully.
More configs may be tested in the coming days.
gcc tested configs:
arm defconfig
arm64 allyesconfig
arm64 defconfig
arm allyesconfig
arm allmodconfig
i386 randconfig-c001-20210916
mips loongson1b_defconfig
sh se7712_defconfig
powerpc warp_defconfig
arm integrator_defconfig
powerpc mpc837x_rdb_defconfig
s390 zfcpdump_defconfig
powerpc arches_defconfig
m68k m5275evb_defconfig
mips workpad_defconfig
arm eseries_pxa_defconfig
powerpc klondike_defconfig
sh se7343_defconfig
powerpc mpc837x_mds_defconfig
mips loongson2k_defconfig
powerpc tqm8560_defconfig
xtensa cadence_csp_defconfig
arm spear3xx_defconfig
x86_64 randconfig-c001-20210916
arm randconfig-c002-20210916
ia64 allmodconfig
ia64 defconfig
ia64 allyesconfig
m68k allmodconfig
m68k defconfig
m68k allyesconfig
nios2 defconfig
arc allyesconfig
nds32 allnoconfig
nds32 defconfig
nios2 allyesconfig
csky defconfig
alpha defconfig
alpha allyesconfig
xtensa allyesconfig
h8300 allyesconfig
arc defconfig
sh allmodconfig
parisc defconfig
s390 allyesconfig
s390 allmodconfig
parisc allyesconfig
s390 defconfig
i386 allyesconfig
sparc allyesconfig
sparc defconfig
i386 defconfig
mips allyesconfig
mips allmodconfig
powerpc allyesconfig
powerpc allmodconfig
powerpc allnoconfig
x86_64 randconfig-a016-20210916
x86_64 randconfig-a013-20210916
x86_64 randconfig-a012-20210916
x86_64 randconfig-a011-20210916
x86_64 randconfig-a014-20210916
x86_64 randconfig-a015-20210916
i386 randconfig-a016-20210916
i386 randconfig-a015-20210916
i386 randconfig-a011-20210916
i386 randconfig-a012-20210916
i386 randconfig-a013-20210916
i386 randconfig-a014-20210916
riscv randconfig-r042-20210916
s390 randconfig-r044-20210916
arc randconfig-r043-20210916
riscv nommu_k210_defconfig
riscv allyesconfig
riscv nommu_virt_defconfig
riscv allnoconfig
riscv defconfig
riscv rv32_defconfig
riscv allmodconfig
x86_64 rhel-8.3-kselftests
um x86_64_defconfig
um i386_defconfig
x86_64 allyesconfig
x86_64 defconfig
x86_64 rhel-8.3
x86_64 kexec
clang tested configs:
riscv randconfig-c006-20210916
x86_64 randconfig-c007-20210916
mips randconfig-c004-20210916
powerpc randconfig-c003-20210916
arm randconfig-c002-20210916
i386 randconfig-c001-20210916
s390 randconfig-c005-20210916
x86_64 randconfig-a002-20210916
x86_64 randconfig-a003-20210916
x86_64 randconfig-a006-20210916
x86_64 randconfig-a004-20210916
x86_64 randconfig-a005-20210916
x86_64 randconfig-a001-20210916
i386 randconfig-a004-20210916
i386 randconfig-a005-20210916
i386 randconfig-a006-20210916
i386 randconfig-a002-20210916
i386 randconfig-a003-20210916
i386 randconfig-a001-20210916
hexagon randconfig-r045-20210916
hexagon randconfig-r041-20210916
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected].org
10 months, 3 weeks
Re: [rafael-pm:bleeding-edge 6/11] drivers/base/swnode.c:1120:17: error: too many arguments to function 'software_node_notify'
by Rafael J. Wysocki
On Thu, Sep 16, 2021 at 8:58 AM Heikki Krogerus
<heikki.krogerus(a)linux.intel.com> wrote:
>
> On Thu, Sep 16, 2021 at 08:13:24AM +0800, kernel test robot wrote:
> > tree: https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git bleeding-edge
> > head: 9808623e1bba7d929b50865d7e2d7042cd4e5e7b
> > commit: 810fc8571f8b70a4263266d979f97edec2804574 [6/11] software node: balance refcount for managed sw nodes
> > config: arc-randconfig-r001-20210916 (attached as .config)
> > compiler: arc-elf-gcc (GCC) 11.2.0
> > reproduce (this is a W=1 build):
> > wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
> > chmod +x ~/bin/make.cross
> > # https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git/commi...
> > git remote add rafael-pm https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git
> > git fetch --no-tags rafael-pm bleeding-edge
> > git checkout 810fc8571f8b70a4263266d979f97edec2804574
> > # save the attached .config to linux build tree
> > mkdir build_dir
> > COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.2.0 make.cross O=build_dir ARCH=arc SHELL=/bin/bash drivers/
> >
> > If you fix the issue, kindly add following tag as appropriate
> > Reported-by: kernel test robot <lkp(a)intel.com>
> >
> > All errors (new ones prefixed by >>):
> >
> > drivers/base/swnode.c: In function 'device_create_managed_software_node':
> > >> drivers/base/swnode.c:1120:17: error: too many arguments to function 'software_node_notify'
> > 1120 | software_node_notify(dev, KOBJ_ADD);
> > | ^~~~~~~~~~~~~~~~~~~~
> > In file included from drivers/base/swnode.c:14:
> > drivers/base/base.h:206:6: note: declared here
> > 206 | void software_node_notify(struct device *dev);
> > | ^~~~~~~~~~~~~~~~~~~~
>
> Ah, that was modified in commit 384f5a857bae ("software nodes: Split
> software_node_notify()").
>
> So the patch is OK for stable, but for v5.15-rc1 it has to be changed:
>
> diff --git a/drivers/base/swnode.c b/drivers/base/swnode.c
> index 7bd0f3cfb7eb4..c46f6a8e14d23 100644
> --- a/drivers/base/swnode.c
> +++ b/drivers/base/swnode.c
> @@ -1116,6 +1116,9 @@ int device_create_managed_software_node(struct device *dev,
> to_swnode(fwnode)->managed = true;
> set_secondary_fwnode(dev, fwnode);
>
> + if (device_is_registered(dev))
> + software_node_notify(dev);
> +
> return 0;
> }
> EXPORT_SYMBOL_GPL(device_create_managed_software_node);
>
>
> Rafael, how do want to handle this?
I'll fix up the commit, thanks!
10 months, 3 weeks
[rafael-pm:bleeding-edge 6/11] drivers/base/swnode.c:1120:29: error: too many arguments to function call, expected single argument 'dev', have 2 arguments
by kernel test robot
tree: https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git bleeding-edge
head: 9808623e1bba7d929b50865d7e2d7042cd4e5e7b
commit: 810fc8571f8b70a4263266d979f97edec2804574 [6/11] software node: balance refcount for managed sw nodes
config: x86_64-randconfig-a002-20210916 (attached as .config)
compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project 8cbbd7e0b2aa21ce7e416cfb63d9965518948c35)
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git/commi...
git remote add rafael-pm https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git
git fetch --no-tags rafael-pm bleeding-edge
git checkout 810fc8571f8b70a4263266d979f97edec2804574
# save the attached .config to linux build tree
mkdir build_dir
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=x86_64 SHELL=/bin/bash
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp(a)intel.com>
All errors (new ones prefixed by >>):
>> drivers/base/swnode.c:1120:29: error: too many arguments to function call, expected single argument 'dev', have 2 arguments
software_node_notify(dev, KOBJ_ADD);
~~~~~~~~~~~~~~~~~~~~ ^~~~~~~~
drivers/base/base.h:206:6: note: 'software_node_notify' declared here
void software_node_notify(struct device *dev);
^
1 error generated.
vim +/dev +1120 drivers/base/swnode.c
1085
1086 /**
1087 * device_create_managed_software_node - Create a software node for a device
1088 * @dev: The device the software node is assigned to.
1089 * @properties: Device properties for the software node.
1090 * @parent: Parent of the software node.
1091 *
1092 * Creates a software node as a managed resource for @dev, which means the
1093 * lifetime of the newly created software node is tied to the lifetime of @dev.
1094 * Software nodes created with this function should not be reused or shared
1095 * because of that. The function takes a deep copy of @properties for the
1096 * software node.
1097 *
1098 * Since the new software node is assigned directly to @dev, and since it should
1099 * not be shared, it is not returned to the caller. The function returns 0 on
1100 * success, and errno in case of an error.
1101 */
1102 int device_create_managed_software_node(struct device *dev,
1103 const struct property_entry *properties,
1104 const struct software_node *parent)
1105 {
1106 struct fwnode_handle *p = software_node_fwnode(parent);
1107 struct fwnode_handle *fwnode;
1108
1109 if (parent && !p)
1110 return -EINVAL;
1111
1112 fwnode = fwnode_create_software_node(properties, p);
1113 if (IS_ERR(fwnode))
1114 return PTR_ERR(fwnode);
1115
1116 to_swnode(fwnode)->managed = true;
1117 set_secondary_fwnode(dev, fwnode);
1118
1119 if (device_is_registered(dev))
> 1120 software_node_notify(dev, KOBJ_ADD);
1121
1122 return 0;
1123 }
1124 EXPORT_SYMBOL_GPL(device_create_managed_software_node);
1125
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]
10 months, 3 weeks