tools: testing: fix phys_addr_t size on 64-bit systems

[PATCH] tools: testing: fix phys_addr_t size on 64-bit systems

Posted by Lorenzo Stoakes 1 year, 3 months ago

The phys_addr_t size is predicated on whether CONFIG_PHYS_ADDR_T_64BIT is
set or not.

In the VMA tests, virt_to_phys() from tools/include/linux casts a volatile
void * pointer to phys_addr_t, if CONFIG_PHYS_ADDR_T_64BIT is not set, this
will be 32-bit and trigger a warning.

Obviously this might also lead to truncation, which we would rather avoid.

Fix this by adjusting the generation of generated/bit-length.h to generate
a CONFIG_PHYS_ADDR_T{bits}BIT define.

This does result in the generation of the useless CONFIG_PHYS_ADDR_T_32BIT
define for 32-bit systems, but this should have no effect, and makes
implementation of this easier.

This resolves the issue and the warning.

Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
---
 tools/testing/shared/shared.mk | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/shared/shared.mk b/tools/testing/shared/shared.mk
index a6bc51d0b0bf..b37362224a73 100644
--- a/tools/testing/shared/shared.mk
+++ b/tools/testing/shared/shared.mk
@@ -69,6 +69,7 @@ generated/bit-length.h: FORCE
 	@if ! grep -qws CONFIG_$(LONG_BIT)BIT generated/bit-length.h; then   \
 		echo "Generating $@";                                        \
 		echo "#define CONFIG_$(LONG_BIT)BIT 1" > $@;                 \
+		echo "#define CONFIG_PHYS_ADDR_T_$(LONG_BIT)BIT 1" > $@;     \
 	fi
 
 FORCE: ;
-- 
2.46.2

Re: [PATCH] tools: testing: fix phys_addr_t size on 64-bit systems

Posted by Lorenzo Stoakes 1 year, 3 months ago

On Thu, Oct 17, 2024 at 05:56:38PM +0100, Lorenzo Stoakes wrote:
> The phys_addr_t size is predicated on whether CONFIG_PHYS_ADDR_T_64BIT is
> set or not.
>
> In the VMA tests, virt_to_phys() from tools/include/linux casts a volatile
> void * pointer to phys_addr_t, if CONFIG_PHYS_ADDR_T_64BIT is not set, this
> will be 32-bit and trigger a warning.
>
> Obviously this might also lead to truncation, which we would rather avoid.
>
> Fix this by adjusting the generation of generated/bit-length.h to generate
> a CONFIG_PHYS_ADDR_T{bits}BIT define.
>
> This does result in the generation of the useless CONFIG_PHYS_ADDR_T_32BIT
> define for 32-bit systems, but this should have no effect, and makes
> implementation of this easier.
>
> This resolves the issue and the warning.
>
> Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
> ---
>  tools/testing/shared/shared.mk | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/tools/testing/shared/shared.mk b/tools/testing/shared/shared.mk
> index a6bc51d0b0bf..b37362224a73 100644
> --- a/tools/testing/shared/shared.mk
> +++ b/tools/testing/shared/shared.mk
> @@ -69,6 +69,7 @@ generated/bit-length.h: FORCE
>  	@if ! grep -qws CONFIG_$(LONG_BIT)BIT generated/bit-length.h; then   \
>  		echo "Generating $@";                                        \
>  		echo "#define CONFIG_$(LONG_BIT)BIT 1" > $@;                 \
> +		echo "#define CONFIG_PHYS_ADDR_T_$(LONG_BIT)BIT 1" > $@;     \
>  	fi
>
>  FORCE: ;
> --
> 2.46.2
>

Andrew - apologies, I made a mistake here, VMA tests are not properly
importing the bit-length.h header (the radix tree tests do appear to be)
and we are overwriting that constant (ahem)...

Could you apply the following fix-patch? Thanks!

----8<----
From 92c247fe99b57f14b91c5e7479cb29616f0c8281 Mon Sep 17 00:00:00 2001
From: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Date: Tue, 22 Oct 2024 15:01:58 +0100
Subject: [PATCH] correctly set bit length in VMA and radix tree tests

---
 tools/testing/shared/shared.mk | 2 +-
 tools/testing/vma/vma.c        | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/testing/shared/shared.mk b/tools/testing/shared/shared.mk
index b37362224a73..923ee2492256 100644
--- a/tools/testing/shared/shared.mk
+++ b/tools/testing/shared/shared.mk
@@ -69,7 +69,7 @@ generated/bit-length.h: FORCE
 	@if ! grep -qws CONFIG_$(LONG_BIT)BIT generated/bit-length.h; then   \
 		echo "Generating $@";                                        \
 		echo "#define CONFIG_$(LONG_BIT)BIT 1" > $@;                 \
-		echo "#define CONFIG_PHYS_ADDR_T_$(LONG_BIT)BIT 1" > $@;     \
+		echo "#define CONFIG_PHYS_ADDR_T_$(LONG_BIT)BIT 1" >> $@;    \
 	fi

 FORCE: ;
diff --git a/tools/testing/vma/vma.c b/tools/testing/vma/vma.c
index b33b47342d41..8fab5e13c7c3 100644
--- a/tools/testing/vma/vma.c
+++ b/tools/testing/vma/vma.c
@@ -4,6 +4,8 @@
 #include <stdio.h>
 #include <stdlib.h>

+#include "generated/bit-length.h"
+
 #include "maple-shared.h"
 #include "vma_internal.h"

--
2.47.0

Re: [PATCH] tools: testing: fix phys_addr_t size on 64-bit systems

Posted by Liam R. Howlett 1 year, 3 months ago

* Lorenzo Stoakes <lorenzo.stoakes@oracle.com> [241017 12:56]:
> The phys_addr_t size is predicated on whether CONFIG_PHYS_ADDR_T_64BIT is
> set or not.
> 
> In the VMA tests, virt_to_phys() from tools/include/linux casts a volatile
> void * pointer to phys_addr_t, if CONFIG_PHYS_ADDR_T_64BIT is not set, this
> will be 32-bit and trigger a warning.
> 
> Obviously this might also lead to truncation, which we would rather avoid.
> 
> Fix this by adjusting the generation of generated/bit-length.h to generate
> a CONFIG_PHYS_ADDR_T{bits}BIT define.
> 
> This does result in the generation of the useless CONFIG_PHYS_ADDR_T_32BIT
> define for 32-bit systems, but this should have no effect, and makes
> implementation of this easier.
> 
> This resolves the issue and the warning.
> 
> Signed-off-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>

Thanks, I just saw this, but your change fixes the vma test code for me.

Tested-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@Oracle.com>

(both tags since I don't see the double tag in the submitting-patches
doc)

> ---
>  tools/testing/shared/shared.mk | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/tools/testing/shared/shared.mk b/tools/testing/shared/shared.mk
> index a6bc51d0b0bf..b37362224a73 100644
> --- a/tools/testing/shared/shared.mk
> +++ b/tools/testing/shared/shared.mk
> @@ -69,6 +69,7 @@ generated/bit-length.h: FORCE
>  	@if ! grep -qws CONFIG_$(LONG_BIT)BIT generated/bit-length.h; then   \
>  		echo "Generating $@";                                        \
>  		echo "#define CONFIG_$(LONG_BIT)BIT 1" > $@;                 \
> +		echo "#define CONFIG_PHYS_ADDR_T_$(LONG_BIT)BIT 1" > $@;     \
>  	fi
>  
>  FORCE: ;
> -- 
> 2.46.2
>