Re: [stable-4.4.y] compile error with KAISER and no vsyscall emulation

Re: [stable-4.4.y] compile error with KAISER and no vsyscall emulation

[Hugh Dickins]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 2634 bytes --]

On Sat, 10 Feb 2018, Tobias Jakobi wrote:
> Hello,
> 
> I noticed a compile error with a recent version (4.4.115) of the 4.4.y branch:
> arch/x86/mm/kaiser.c: In function ‘kaiser_init’:
> arch/x86/mm/kaiser.c:348:8: error: ‘vsyscall_pgprot’ undeclared (first use in
> this function)
> 
> It seems like my combination of kernel options doesn't work for KAISER.
> X86_VSYSCALL_EMULATION is not set on my system, while LEGACY_VSYSCALL is set to
> NONE (LEGACY_VSYSCALL_NONE=y). I have managed to get things compiling again, by
> moving the 'extern unsigned long vsyscall_pgprot' outside of the preprocessor
> statement. This works for me, I guess because the code in question is never
> called during runtime anyway (vsyscall_enabled() always returns false).
> 
> With best wishes,
> Tobias

Yes, sorry about that, thanks Tobias. Same error on 4.9 too.
Here's the patch that you correctly suggest, and it's good for both:


[PATCH 4.4 and 4.9] kaiser: fix compile error without vsyscall

Tobias noticed a compile error on 4.4.115, and it's the same on 4.9.80:
arch/x86/mm/kaiser.c: In function ‘kaiser_init’:
arch/x86/mm/kaiser.c:348:8: error: ‘vsyscall_pgprot’ undeclared
                                   (first use in this function)

It seems like his combination of kernel options doesn't work for KAISER.
X86_VSYSCALL_EMULATION is not set on his system, while LEGACY_VSYSCALL
is set to NONE (LEGACY_VSYSCALL_NONE=y). He managed to get things
compiling again, by moving the 'extern unsigned long vsyscall_pgprot'
outside of the preprocessor statement. This works because the optimizer
removes that code (vsyscall_enabled() is always false) - and that's how
it was done in some older backports.

Reported-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de>
Signed-off-by: Hugh Dickins <hughd@google.com>
---

 arch/x86/include/asm/vsyscall.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/arch/x86/include/asm/vsyscall.h
+++ b/arch/x86/include/asm/vsyscall.h
@@ -13,7 +13,6 @@ extern void map_vsyscall(void);
  */
 extern bool emulate_vsyscall(struct pt_regs *regs, unsigned long address);
 extern bool vsyscall_enabled(void);
-extern unsigned long vsyscall_pgprot;
 #else
 static inline void map_vsyscall(void) {}
 static inline bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
@@ -22,5 +21,6 @@ static inline bool emulate_vsyscall(struct pt_regs *regs, unsigned long address)
 }
 static inline bool vsyscall_enabled(void) { return false; }
 #endif
+extern unsigned long vsyscall_pgprot;
 
 #endif /* _ASM_X86_VSYSCALL_H */

Re: [stable-4.4.y] compile error with KAISER and no vsyscall emulation

[Greg Kroah-Hartman]

On Sat, Feb 10, 2018 at 04:24:10PM -0800, Hugh Dickins wrote:
> On Sat, 10 Feb 2018, Tobias Jakobi wrote:
> > Hello,
> > 
> > I noticed a compile error with a recent version (4.4.115) of the 4.4.y branch:
> > arch/x86/mm/kaiser.c: In function ‘kaiser_init’:
> > arch/x86/mm/kaiser.c:348:8: error: ‘vsyscall_pgprot’ undeclared (first use in
> > this function)
> > 
> > It seems like my combination of kernel options doesn't work for KAISER.
> > X86_VSYSCALL_EMULATION is not set on my system, while LEGACY_VSYSCALL is set to
> > NONE (LEGACY_VSYSCALL_NONE=y). I have managed to get things compiling again, by
> > moving the 'extern unsigned long vsyscall_pgprot' outside of the preprocessor
> > statement. This works for me, I guess because the code in question is never
> > called during runtime anyway (vsyscall_enabled() always returns false).
> > 
> > With best wishes,
> > Tobias
> 
> Yes, sorry about that, thanks Tobias. Same error on 4.9 too.
> Here's the patch that you correctly suggest, and it's good for both:
> 
> 
> [PATCH 4.4 and 4.9] kaiser: fix compile error without vsyscall

Thanks, now queued up.

greg k-h