All of lore.kernel.org
 help / color / mirror / Atom feed
From: Joerg Roedel <joro@8bytes.org>
To: Qian Cai <cai@lca.pw>
Cc: iommu@lists.linux-foundation.org, Joerg Roedel <jroedel@suse.de>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/2] iommu/amd: Use 'unsigned long' for domain->pt_root
Date: Fri, 26 Jun 2020 10:05:40 +0200	[thread overview]
Message-ID: <20200626080539.GP3701@8bytes.org> (raw)
In-Reply-To: <20200625153720.GA1127@lca.pw>

Hi Qian,

On Thu, Jun 25, 2020 at 11:37:20AM -0400, Qian Cai wrote:
> On Thu, Jun 25, 2020 at 04:52:27PM +0200, Joerg Roedel wrote:
> > -	u64 pt_root = atomic64_read(&domain->pt_root);
> > +	unsigned long pt_root = domain->pt_root;
> 
> The pt_root might be reload later in case of register pressure where the
> compiler decides to not store it as a stack variable, so it needs
> smp_rmb() here to match to the smp_wmb() in
> amd_iommu_domain_set_pt_root() to make the load visiable to all CPUs.
> 
> Then, smp_rmb/wmb() wouldn't be able to deal with data races, so it
> needs,
> 
> unsigned long pt_root = READ_ONCE(domain->pt_root);
> 
> >  
> >  	pgtable->root = (u64 *)(pt_root & PAGE_MASK);
> >  	pgtable->mode = pt_root & 7; /* lowest 3 bits encode pgtable mode */
> > @@ -164,7 +164,13 @@ static void amd_iommu_domain_get_pgtable(struct protection_domain *domain,
> >  
> >  static void amd_iommu_domain_set_pt_root(struct protection_domain *domain, u64 root)
> >  {
> > -	atomic64_set(&domain->pt_root, root);
> > +	domain->pt_root = root;
> 
> WRITE_ONCE(domain->pt_root, root);

Thanks for your review. I addressed your comments and will send an
updated version shortly.


Regards,

	Joerg

_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

WARNING: multiple messages have this Message-ID (diff)
From: Joerg Roedel <joro@8bytes.org>
To: Qian Cai <cai@lca.pw>
Cc: iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org,
	Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>,
	Joerg Roedel <jroedel@suse.de>
Subject: Re: [PATCH 2/2] iommu/amd: Use 'unsigned long' for domain->pt_root
Date: Fri, 26 Jun 2020 10:05:40 +0200	[thread overview]
Message-ID: <20200626080539.GP3701@8bytes.org> (raw)
In-Reply-To: <20200625153720.GA1127@lca.pw>

Hi Qian,

On Thu, Jun 25, 2020 at 11:37:20AM -0400, Qian Cai wrote:
> On Thu, Jun 25, 2020 at 04:52:27PM +0200, Joerg Roedel wrote:
> > -	u64 pt_root = atomic64_read(&domain->pt_root);
> > +	unsigned long pt_root = domain->pt_root;
> 
> The pt_root might be reload later in case of register pressure where the
> compiler decides to not store it as a stack variable, so it needs
> smp_rmb() here to match to the smp_wmb() in
> amd_iommu_domain_set_pt_root() to make the load visiable to all CPUs.
> 
> Then, smp_rmb/wmb() wouldn't be able to deal with data races, so it
> needs,
> 
> unsigned long pt_root = READ_ONCE(domain->pt_root);
> 
> >  
> >  	pgtable->root = (u64 *)(pt_root & PAGE_MASK);
> >  	pgtable->mode = pt_root & 7; /* lowest 3 bits encode pgtable mode */
> > @@ -164,7 +164,13 @@ static void amd_iommu_domain_get_pgtable(struct protection_domain *domain,
> >  
> >  static void amd_iommu_domain_set_pt_root(struct protection_domain *domain, u64 root)
> >  {
> > -	atomic64_set(&domain->pt_root, root);
> > +	domain->pt_root = root;
> 
> WRITE_ONCE(domain->pt_root, root);

Thanks for your review. I addressed your comments and will send an
updated version shortly.


Regards,

	Joerg


  reply	other threads:[~2020-06-26  8:05 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-25 14:52 [PATCH 0/2] iommu/amd: Don't use atomic64_t for domain->pt_root Joerg Roedel
2020-06-25 14:52 ` Joerg Roedel
2020-06-25 14:52 ` [PATCH 1/2] iommu/amd: Add helper functions to update domain->pt_root Joerg Roedel
2020-06-25 14:52   ` Joerg Roedel
2020-06-25 14:52 ` [PATCH 2/2] iommu/amd: Use 'unsigned long' for domain->pt_root Joerg Roedel
2020-06-25 14:52   ` Joerg Roedel
2020-06-25 15:37   ` Qian Cai
2020-06-25 15:37     ` Qian Cai
2020-06-26  8:05     ` Joerg Roedel [this message]
2020-06-26  8:05       ` Joerg Roedel

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=20200626080539.GP3701@8bytes.org \
    --to=joro@8bytes.org \
    --cc=cai@lca.pw \
    --cc=iommu@lists.linux-foundation.org \
    --cc=jroedel@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.