Gcc report warning as follows:
arch/x86/boot/compressed/sev.c: In function ‘enforce_vmpl0’:
arch/x86/boot/compressed/sev.c:256:13: error: unused variable ‘err’
[-Werror=unused-variable]
Fix it by removing the unused variable "err".
Fixes: 81cc3df9a90e ("x86/sev: Check the VMPL level")
Signed-off-by: Li Zetao <[email protected]>
---
arch/x86/boot/compressed/sev.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/x86/boot/compressed/sev.c b/arch/x86/boot/compressed/sev.c
index c93930d5ccbd..b9451761a69a 100644
--- a/arch/x86/boot/compressed/sev.c
+++ b/arch/x86/boot/compressed/sev.c
@@ -253,7 +253,6 @@ void do_boot_stage2_vc(struct pt_regs *regs, unsigned long exit_code)
static void enforce_vmpl0(void)
{
u64 attrs;
- int err;
/*
* RMPADJUST modifies RMP permissions of a lesser-privileged (numerically
--
2.34.1
On Fri, Sep 23, 2022 at 11:32:09AM +0000, Li Zetao wrote:
> Gcc report warning as follows:
>
> arch/x86/boot/compressed/sev.c: In function ‘enforce_vmpl0’:
> arch/x86/boot/compressed/sev.c:256:13: error: unused variable ‘err’
> [-Werror=unused-variable]
How do you trigger this? Compiler version and .config?
I haven't seen this one in any of my builds until now...
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
On 2022/9/23 20:19, Borislav Petkov wrote:
> On Fri, Sep 23, 2022 at 11:32:09AM +0000, Li Zetao wrote:
>> Gcc report warning as follows:
>>
>> arch/x86/boot/compressed/sev.c: In function ‘enforce_vmpl0’:
>> arch/x86/boot/compressed/sev.c:256:13: error: unused variable ‘err’
>> [-Werror=unused-variable]
> How do you trigger this? Compiler version and .config?
>
> I haven't seen this one in any of my builds until now...
>
> Thx.
My Compiler version is 11.2.0. I used the default .config and compiled
with flags "-Wall -Werror", just as follows:
make allmodconfig
make EXTRA_CFLAGS="-Wall -Werror"
Best regards
Li Zetao
On Sat, Sep 24, 2022 at 06:18:19PM +0800, Li Zetao wrote:
> My Compiler version is 11.2.0. I used the default .config and compiled with
> flags "-Wall -Werror", just as follows:
>
> make allmodconfig
>
> make EXTRA_CFLAGS="-Wall -Werror"
Ah, compressed/Makefile doesn't have -Wall. That's the old problem of
x86 not sharing makefiles. I need to dust off my patches which do that.
Now, if you do:
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 35ce1a64068b..10abb7c45d04 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -34,7 +34,7 @@ targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma \
# be valid.
KBUILD_CFLAGS := -m$(BITS) -O2 $(CLANG_FLAGS)
KBUILD_CFLAGS += -fno-strict-aliasing -fPIE
-KBUILD_CFLAGS += -Wundef
+KBUILD_CFLAGS += -Wundef -Wall
KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING
cflags-$(CONFIG_X86_32) := -march=i386
cflags-$(CONFIG_X86_64) := -mcmodel=small -mno-red-zone
then allmodconfig triggers a bunch:
arch/x86/boot/compressed/efi.c: In function ‘efi_get_system_table’:
arch/x86/boot/compressed/efi.c:62:23: warning: unused variable ‘et’ [-Wunused-variable]
62 | enum efi_type et;
| ^~
arch/x86/boot/compressed/efi.c: In function ‘efi_get_conf_table’:
arch/x86/boot/compressed/efi.c:134:13: warning: unused variable ‘ret’ [-Wunused-variable]
134 | int ret;
| ^~~
arch/x86/boot/compressed/acpi.c: In function ‘__efi_get_rsdp_addr’:
arch/x86/boot/compressed/acpi.c:27:13: warning: unused variable ‘ret’ [-Wunused-variable]
27 | int ret;
| ^~~
arch/x86/boot/compressed/acpi.c: In function ‘efi_get_rsdp_addr’:
arch/x86/boot/compressed/acpi.c:55:22: warning: unused variable ‘nr_tables’ [-Wunused-variable]
55 | unsigned int nr_tables;
| ^~~~~~~~~
arch/x86/boot/compressed/sev.c: In function ‘enforce_vmpl0’:
arch/x86/boot/compressed/sev.c:256:13: warning: unused variable ‘err’ [-Wunused-variable]
256 | int err;
| ^~~
Would you like to do the above instead and fix them all in one go?
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
On September 26, 2022 7:48:59 AM PDT, Borislav Petkov <[email protected]> wrote:
>On Sat, Sep 24, 2022 at 06:18:19PM +0800, Li Zetao wrote:
>> My Compiler version is 11.2.0. I used the default .config and compiled with
>> flags "-Wall -Werror", just as follows:
>>
>> make allmodconfig
>>
>> make EXTRA_CFLAGS="-Wall -Werror"
>
>Ah, compressed/Makefile doesn't have -Wall. That's the old problem of
>x86 not sharing makefiles. I need to dust off my patches which do that.
>
>Now, if you do:
>
>diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
>index 35ce1a64068b..10abb7c45d04 100644
>--- a/arch/x86/boot/compressed/Makefile
>+++ b/arch/x86/boot/compressed/Makefile
>@@ -34,7 +34,7 @@ targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma \
> # be valid.
> KBUILD_CFLAGS := -m$(BITS) -O2 $(CLANG_FLAGS)
> KBUILD_CFLAGS += -fno-strict-aliasing -fPIE
>-KBUILD_CFLAGS += -Wundef
>+KBUILD_CFLAGS += -Wundef -Wall
> KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING
> cflags-$(CONFIG_X86_32) := -march=i386
> cflags-$(CONFIG_X86_64) := -mcmodel=small -mno-red-zone
>
>then allmodconfig triggers a bunch:
>
>arch/x86/boot/compressed/efi.c: In function ‘efi_get_system_table’:
>arch/x86/boot/compressed/efi.c:62:23: warning: unused variable ‘et’ [-Wunused-variable]
> 62 | enum efi_type et;
> | ^~
>arch/x86/boot/compressed/efi.c: In function ‘efi_get_conf_table’:
>arch/x86/boot/compressed/efi.c:134:13: warning: unused variable ‘ret’ [-Wunused-variable]
> 134 | int ret;
> | ^~~
>arch/x86/boot/compressed/acpi.c: In function ‘__efi_get_rsdp_addr’:
>arch/x86/boot/compressed/acpi.c:27:13: warning: unused variable ‘ret’ [-Wunused-variable]
> 27 | int ret;
> | ^~~
>arch/x86/boot/compressed/acpi.c: In function ‘efi_get_rsdp_addr’:
>arch/x86/boot/compressed/acpi.c:55:22: warning: unused variable ‘nr_tables’ [-Wunused-variable]
> 55 | unsigned int nr_tables;
> | ^~~~~~~~~
>arch/x86/boot/compressed/sev.c: In function ‘enforce_vmpl0’:
>arch/x86/boot/compressed/sev.c:256:13: warning: unused variable ‘err’ [-Wunused-variable]
> 256 | int err;
> | ^~~
>
>
>Would you like to do the above instead and fix them all in one go?
>
>Thx.
>
The boot code really needs to be refactored, especially that the real-mode BIOS code is built in the root of that tree.
On Mon, Sep 26, 2022 at 10:52:08AM -0700, H. Peter Anvin wrote:
> The boot code really needs to be refactored, especially that the
> real-mode BIOS code is built in the root of that tree.
You're going to have to expand on that a bit and give more details as to
what do you mean exactly. I think you mean arch/x86/realmode/ but I'm
not sure...
Thx.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette