xen/x86: drop redundant zeroing from cpu_initialize_context()

[PATCH] xen/x86: drop redundant zeroing from cpu_initialize_context()

Posted by Jan Beulich 2 years, 7 months ago

Just after having obtained the pointer from kzalloc() there's no reason
at all to set part of the area to all zero yet another time. Similarly
there's no point explicitly clearing "ldt_ents".

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/arch/x86/xen/smp_pv.c
+++ b/arch/x86/xen/smp_pv.c
@@ -290,8 +290,6 @@ cpu_initialize_context(unsigned int cpu,
 
 	gdt = get_cpu_gdt_rw(cpu);
 
-	memset(&ctxt->fpu_ctxt, 0, sizeof(ctxt->fpu_ctxt));
-
 	/*
 	 * Bring up the CPU in cpu_bringup_and_idle() with the stack
 	 * pointing just below where pt_regs would be if it were a normal
@@ -308,8 +306,6 @@ cpu_initialize_context(unsigned int cpu,
 
 	xen_copy_trap_info(ctxt->trap_ctxt);
 
-	ctxt->ldt_ents = 0;
-
 	BUG_ON((unsigned long)gdt & ~PAGE_MASK);
 
 	gdt_mfn = arbitrary_virt_to_mfn(gdt);

Re: [PATCH] xen/x86: drop redundant zeroing from cpu_initialize_context()

Posted by Boris Ostrovsky 2 years, 7 months ago

On 9/16/21 11:05 AM, Jan Beulich wrote:
> Just after having obtained the pointer from kzalloc() there's no reason
> at all to set part of the area to all zero yet another time. Similarly
> there's no point explicitly clearing "ldt_ents".
>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>


Reviewed-by: Boris Ostrovsky <boris.ostrvsky@oracle.com>

Re: [PATCH] xen/x86: drop redundant zeroing from cpu_initialize_context()

Posted by Juergen Gross 2 years, 7 months ago

On 16.09.21 17:05, Jan Beulich wrote:
> Just after having obtained the pointer from kzalloc() there's no reason
> at all to set part of the area to all zero yet another time. Similarly
> there's no point explicitly clearing "ldt_ents".
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Pushed to xen/tip.git for-linus-5.15b


Juergen