public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <levinsasha928@gmail.com>
To: Avi Kivity <avi@redhat.com>
Cc: kvm@vger.kernel.org, Marcelo Tosatti <mtosatti@redhat.com>
Subject: Re: [PATCH] KVM: Refactor and simplify kvm_dev_ioctl_get_supported_cpuid
Date: Sun, 27 Nov 2011 16:55:14 +0200	[thread overview]
Message-ID: <1322405714.8871.3.camel@lappy> (raw)
In-Reply-To: <4ED24CDD.7090504@redhat.com>

On Sun, 2011-11-27 at 16:44 +0200, Avi Kivity wrote:
> On 11/25/2011 10:12 AM, Sasha Levin wrote:
> > This patch cleans and simplifies kvm_dev_ioctl_get_supported_cpuid by using a table
> > instead of duplicating code as Avi suggested.
> >
> > This patch also fixes a bug where kvm_dev_ioctl_get_supported_cpuid would return
> > -E2BIG when amount of entries passed was just right.
> >
> > Cc: Avi Kivity <avi@redhat.com>
> > Cc: Marcelo Tosatti <mtosatti@redhat.com>
> > Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
> > ---
> >  arch/x86/kvm/cpuid.c |   78 ++++++++++++++++++++++---------------------------
> >  1 files changed, 35 insertions(+), 43 deletions(-)
> >
> > diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
> > index bbaa6d8..c9ede4d 100644
> > --- a/arch/x86/kvm/cpuid.c
> > +++ b/arch/x86/kvm/cpuid.c
> > @@ -425,12 +425,31 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
> >  
> >  #undef F
> >  
> > +struct kvm_cpuid_param {
> > +	u32 func;
> > +	u32 idx;
> > +	u32 leaves;
> 
> What does 'leaves' mean?  That it has a leaf count in eax?  if so it
> want a better name and to be a bool.
> 
> > +	int (*qualifier)(struct kvm_cpuid_param *param);
> 
> bool
> 
> > +};
> > +
> > +int is_centaur_cpu(struct kvm_cpuid_param *param)
> > +{
> > +	return boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR;
> > +}
> > +
> static
> 
> >  int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
> >  				      struct kvm_cpuid_entry2 __user *entries)
> >  {
> >  	struct kvm_cpuid_entry2 *cpuid_entries;
> > -	int limit, nent = 0, r = -E2BIG;
> > +	int limit, nent = 0, r = -E2BIG, i;
> >  	u32 func;
> > +	struct kvm_cpuid_param param[] = {
> > +		{ .func = 0, .leaves = 1 },
> > +		{ .func = 0x80000000, .leaves = 1 },
> > +		{ .func = 0xC0000000, .qualifier = is_centaur_cpu, .leaves = 1 },
> > +		{ .func = KVM_CPUID_SIGNATURE },
> > +		{ .func = KVM_CPUID_FEATURES },
> > +	};
> 
> static
> 
> >  
> >  	if (cpuid->nent < 1)
> >  		goto out;
> > @@ -441,59 +460,32 @@ int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid,
> >  	if (!cpuid_entries)
> >  		goto out;
> >  
> > -	do_cpuid_ent(&cpuid_entries[0], 0, 0, &nent, cpuid->nent);
> > -	limit = cpuid_entries[0].eax;
> > -	for (func = 1; func <= limit && nent < cpuid->nent; ++func)
> > -		do_cpuid_ent(&cpuid_entries[nent], func, 0,
> > -			     &nent, cpuid->nent);
> > -	r = -E2BIG;
> > -	if (nent >= cpuid->nent)
> > -		goto out_free;
> > -
> > -	do_cpuid_ent(&cpuid_entries[nent], 0x80000000, 0, &nent, cpuid->nent);
> > -	limit = cpuid_entries[nent - 1].eax;
> > -	for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func)
> > -		do_cpuid_ent(&cpuid_entries[nent], func, 0,
> > -			     &nent, cpuid->nent);
> > -
> > +	for (i = 0; i < ARRAY_SIZE(param); i++) {
> > +		struct kvm_cpuid_param *ent = &param[i];
> >  
> > +		if (ent->qualifier && !ent->qualifier(ent))
> > +			continue;
> >  
> > -	r = -E2BIG;
> > -	if (nent >= cpuid->nent)
> > -		goto out_free;
> > +		if (nent >= cpuid->nent)
> > +			break;
> >  
> > -	/* Add support for Centaur's CPUID instruction. */
> > -	if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR) {
> > -		do_cpuid_ent(&cpuid_entries[nent], 0xC0000000, 0,
> > +		do_cpuid_ent(&cpuid_entries[nent], ent->func, ent->idx,
> >  				&nent, cpuid->nent);
> >  
> > -		r = -E2BIG;
> > -		if (nent >= cpuid->nent)
> > -			goto out_free;
> > +		if (!ent->leaves)
> > +			continue;
> >  
> >  		limit = cpuid_entries[nent - 1].eax;
> > -		for (func = 0xC0000001;
> > -			func <= limit && nent < cpuid->nent; ++func)
> > -			do_cpuid_ent(&cpuid_entries[nent], func, 0,
> > -					&nent, cpuid->nent);
> > +		for (func = ent->func + 1; func <= limit && nent < cpuid->nent; ++func)
> > +			do_cpuid_ent(&cpuid_entries[nent], func, ent->idx,
> > +				     &nent, cpuid->nent);
> >  
> > -		r = -E2BIG;
> > -		if (nent >= cpuid->nent)
> > -			goto out_free;
> > +		if (func <= limit)
> > +			break;
> >  	}
> >  
> > -	do_cpuid_ent(&cpuid_entries[nent], KVM_CPUID_SIGNATURE, 0, &nent,
> > -		     cpuid->nent);
> > -
> > -	r = -E2BIG;
> > -	if (nent >= cpuid->nent)
> > -		goto out_free;
> > -
> > -	do_cpuid_ent(&cpuid_entries[nent], KVM_CPUID_FEATURES, 0, &nent,
> > -		     cpuid->nent);
> > -
> >  	r = -E2BIG;
> > -	if (nent >= cpuid->nent)
> > +	if (i < ARRAY_SIZE(param) || nent > cpuid->nent)
> >  		goto out_free;
> >  
> 
> Isn't the first condition sufficient?

Probably, but with all the possible cases of termination from that loop
I felt it was better going for safe than sorry.

> 
> The code may be more readable if do_cpuid_ent() (and do_cpuid_1_ent)
> returned -E2BIG on overflow instead of all this cleverness.
> 

I'll send a new version with fixes.

-- 

Sasha.


      reply	other threads:[~2011-11-27 14:55 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-25  8:12 [PATCH] KVM: Refactor and simplify kvm_dev_ioctl_get_supported_cpuid Sasha Levin
2011-11-27 14:44 ` Avi Kivity
2011-11-27 14:55   ` Sasha Levin [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1322405714.8871.3.camel@lappy \
    --to=levinsasha928@gmail.com \
    --cc=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox