[PATCH v2 0/2] efi/arm*: wire up ESRT table

[PATCH v2 0/2] efi/arm*: wire up ESRT table

[Ard Biesheuvel]

This v2 is a followup to the series I sent out on Feb 15. These now
apply onto Matt's memmap refactoring work which is intended for inclusion
in v4.9, and the diffstat below reveals that enabling ESRT is now absolutely
trivial. The only things needed are replacing ioremap() with memremap(),
and adding the missing efi_esrt_init() invocation.

I dropped the Tested-by tags, simply because the underlying code is completely
different now. I kept Peter's ack on #1, though.

Ard Biesheuvel (2):
  efi: esrt: use memremap not ioremap to access ESRT table in memory
  efi/arm*: esrt: add missing call to efi_esrt_init()

 drivers/firmware/efi/arm-init.c | 1 +
 drivers/firmware/efi/esrt.c     | 5 +++--
 2 files changed, 4 insertions(+), 2 deletions(-)

-- 
1.9.1

[PATCH v2 0/2] efi/arm*: wire up ESRT table

[Ard Biesheuvel]

This v2 is a followup to the series I sent out on Feb 15. These now
apply onto Matt's memmap refactoring work which is intended for inclusion
in v4.9, and the diffstat below reveals that enabling ESRT is now absolutely
trivial. The only things needed are replacing ioremap() with memremap(),
and adding the missing efi_esrt_init() invocation.

I dropped the Tested-by tags, simply because the underlying code is completely
different now. I kept Peter's ack on #1, though.

Ard Biesheuvel (2):
  efi: esrt: use memremap not ioremap to access ESRT table in memory
  efi/arm*: esrt: add missing call to efi_esrt_init()

 drivers/firmware/efi/arm-init.c | 1 +
 drivers/firmware/efi/esrt.c     | 5 +++--
 2 files changed, 4 insertions(+), 2 deletions(-)

-- 
1.9.1

[PATCH v2 1/2] efi: esrt: use memremap not ioremap to access ESRT table in memory

[Ard Biesheuvel]

On ARM and arm64, ioremap() and memremap() are not interchangeable like
on x86, and the use of ioremap() on ordinary RAM is typically flagged
as an error if the memory region being mapped is also covered by the
linear mapping, since that would lead to aliases with conflicting
cacheability attributes.

Since what we are dealing with is not an I/O region with side effects,
using ioremap() here is arguably incorrect anyway, so let's replace
it with memremap() instead.

Acked-by: Peter Jones <pjones-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
 drivers/firmware/efi/esrt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/firmware/efi/esrt.c b/drivers/firmware/efi/esrt.c
index b93cd11f9bcc..14914074f716 100644
--- a/drivers/firmware/efi/esrt.c
+++ b/drivers/firmware/efi/esrt.c
@@ -16,6 +16,7 @@
 #include <linux/device.h>
 #include <linux/efi.h>
 #include <linux/init.h>
+#include <linux/io.h>
 #include <linux/kernel.h>
 #include <linux/kobject.h>
 #include <linux/list.h>
@@ -387,9 +388,9 @@ static int __init esrt_sysfs_init(void)
 	if (!esrt_data || !esrt_data_size)
 		return -ENOSYS;
 
-	esrt = ioremap(esrt_data, esrt_data_size);
+	esrt = memremap(esrt_data, esrt_data_size, MEMREMAP_WB);
 	if (!esrt) {
-		pr_err("ioremap(%pa, %zu) failed.\n", &esrt_data,
+		pr_err("memremap(%pa, %zu) failed.\n", &esrt_data,
 		       esrt_data_size);
 		return -ENOMEM;
 	}
-- 
1.9.1

[PATCH v2 1/2] efi: esrt: use memremap not ioremap to access ESRT table in memory

[Ard Biesheuvel]

On ARM and arm64, ioremap() and memremap() are not interchangeable like
on x86, and the use of ioremap() on ordinary RAM is typically flagged
as an error if the memory region being mapped is also covered by the
linear mapping, since that would lead to aliases with conflicting
cacheability attributes.

Since what we are dealing with is not an I/O region with side effects,
using ioremap() here is arguably incorrect anyway, so let's replace
it with memremap() instead.

Acked-by: Peter Jones <pjones@redhat.com>
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
---
 drivers/firmware/efi/esrt.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/firmware/efi/esrt.c b/drivers/firmware/efi/esrt.c
index b93cd11f9bcc..14914074f716 100644
--- a/drivers/firmware/efi/esrt.c
+++ b/drivers/firmware/efi/esrt.c
@@ -16,6 +16,7 @@
 #include <linux/device.h>
 #include <linux/efi.h>
 #include <linux/init.h>
+#include <linux/io.h>
 #include <linux/kernel.h>
 #include <linux/kobject.h>
 #include <linux/list.h>
@@ -387,9 +388,9 @@ static int __init esrt_sysfs_init(void)
 	if (!esrt_data || !esrt_data_size)
 		return -ENOSYS;
 
-	esrt = ioremap(esrt_data, esrt_data_size);
+	esrt = memremap(esrt_data, esrt_data_size, MEMREMAP_WB);
 	if (!esrt) {
-		pr_err("ioremap(%pa, %zu) failed.\n", &esrt_data,
+		pr_err("memremap(%pa, %zu) failed.\n", &esrt_data,
 		       esrt_data_size);
 		return -ENOMEM;
 	}
-- 
1.9.1

Re: [PATCH v2 1/2] efi: esrt: use memremap not ioremap to access ESRT table in memory

[Matt Fleming]

On Mon, 11 Jul, at 09:00:45PM, Ard Biesheuvel wrote:
> On ARM and arm64, ioremap() and memremap() are not interchangeable like
> on x86, and the use of ioremap() on ordinary RAM is typically flagged
> as an error if the memory region being mapped is also covered by the
> linear mapping, since that would lead to aliases with conflicting
> cacheability attributes.
> 
> Since what we are dealing with is not an I/O region with side effects,
> using ioremap() here is arguably incorrect anyway, so let's replace
> it with memremap() instead.
> 
> Acked-by: Peter Jones <pjones-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
> ---
>  drivers/firmware/efi/esrt.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)

Looks sensible, thanks Ard, applied.

[PATCH v2 1/2] efi: esrt: use memremap not ioremap to access ESRT table in memory

[Matt Fleming]

On Mon, 11 Jul, at 09:00:45PM, Ard Biesheuvel wrote:
> On ARM and arm64, ioremap() and memremap() are not interchangeable like
> on x86, and the use of ioremap() on ordinary RAM is typically flagged
> as an error if the memory region being mapped is also covered by the
> linear mapping, since that would lead to aliases with conflicting
> cacheability attributes.
> 
> Since what we are dealing with is not an I/O region with side effects,
> using ioremap() here is arguably incorrect anyway, so let's replace
> it with memremap() instead.
> 
> Acked-by: Peter Jones <pjones@redhat.com>
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> ---
>  drivers/firmware/efi/esrt.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)

Looks sensible, thanks Ard, applied.

[PATCH v2 2/2] efi/arm*: esrt: add missing call to efi_esrt_init()

[Ard Biesheuvel]

ESRT support is built by default for all architectures that define
CONFIG_EFI. However, this support was not wired up yet for ARM/arm64,
since efi_esrt_init() was never called. So add the missing call.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
---
 drivers/firmware/efi/arm-init.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/firmware/efi/arm-init.c b/drivers/firmware/efi/arm-init.c
index 5a2df3fefccc..e0a511d4074f 100644
--- a/drivers/firmware/efi/arm-init.c
+++ b/drivers/firmware/efi/arm-init.c
@@ -247,6 +247,7 @@ void __init efi_init(void)
 
 	reserve_regions();
 	efi_memattr_init();
+	efi_esrt_init();
 	efi_memmap_unmap();
 
 	memblock_reserve(params.mmap & PAGE_MASK,
-- 
1.9.1

[PATCH v2 2/2] efi/arm*: esrt: add missing call to efi_esrt_init()

[Ard Biesheuvel]

ESRT support is built by default for all architectures that define
CONFIG_EFI. However, this support was not wired up yet for ARM/arm64,
since efi_esrt_init() was never called. So add the missing call.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
---
 drivers/firmware/efi/arm-init.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/firmware/efi/arm-init.c b/drivers/firmware/efi/arm-init.c
index 5a2df3fefccc..e0a511d4074f 100644
--- a/drivers/firmware/efi/arm-init.c
+++ b/drivers/firmware/efi/arm-init.c
@@ -247,6 +247,7 @@ void __init efi_init(void)
 
 	reserve_regions();
 	efi_memattr_init();
+	efi_esrt_init();
 	efi_memmap_unmap();
 
 	memblock_reserve(params.mmap & PAGE_MASK,
-- 
1.9.1

Re: [PATCH v2 2/2] efi/arm*: esrt: add missing call to efi_esrt_init()

[Matt Fleming]

On Mon, 11 Jul, at 09:00:46PM, Ard Biesheuvel wrote:
> ESRT support is built by default for all architectures that define
> CONFIG_EFI. However, this support was not wired up yet for ARM/arm64,
> since efi_esrt_init() was never called. So add the missing call.
> 
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
> ---
>  drivers/firmware/efi/arm-init.c | 1 +
>  1 file changed, 1 insertion(+)

Thanks Ard, applied and queued up for v4.9 on top of the EFI memmap
refactoring.

[PATCH v2 2/2] efi/arm*: esrt: add missing call to efi_esrt_init()

[Matt Fleming]

On Mon, 11 Jul, at 09:00:46PM, Ard Biesheuvel wrote:
> ESRT support is built by default for all architectures that define
> CONFIG_EFI. However, this support was not wired up yet for ARM/arm64,
> since efi_esrt_init() was never called. So add the missing call.
> 
> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> ---
>  drivers/firmware/efi/arm-init.c | 1 +
>  1 file changed, 1 insertion(+)

Thanks Ard, applied and queued up for v4.9 on top of the EFI memmap
refactoring.