Message-ID: <20131218212150.GE11717@phenom.dumpdata.com>
Date:	Wed, 18 Dec 2013 16:21:50 -0500
From:	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
To:	Stefano Stabellini <stefano.stabellini@...citrix.com>
Cc:	xen-devel@...ts.xenproject.org, linux-kernel@...r.kernel.org,
	boris.ostrovsky@...cle.com, david.vrabel@...rix.com,
	mukesh.rathor@...cle.com, jbeulich@...e.com
Subject: Re: [Xen-devel] [PATCH v11 10/12] xen/pvh: Piggyback on PVHVM for
 grant driver.

On Wed, Dec 18, 2013 at 06:46:56PM +0000, Stefano Stabellini wrote:
> On Tue, 17 Dec 2013, Konrad Rzeszutek Wilk wrote:
> > In PVH the shared grant frame is the PFN and not MFN,
> > hence its mapped via the same code path as HVM.
> > 
> > The allocation of the grant frame is done differently - we
> > do not use the early platform-pci driver and have an
> > ioremap area - instead we use balloon memory and stitch
> > all of the non-contingous pages in a virtualized area.
> > 
> > That means when we call the hypervisor to replace the GMFN
> > with a XENMAPSPACE_grant_table type, we need to lookup the
> > old PFN for every iteration instead of assuming a flat
> > contingous PFN allocation.
> > 
> > Lastly, we only use v1 for grants. This is because PVHVM
> > is not able to use v2 due to no XENMEM_add_to_physmap
> > calls on the error status page (see commit
> > 69e8f430e243d657c2053f097efebc2e2cd559f0
> >  xen/granttable: Disable grant v2 for HVM domains.)
> > 
> > Until that is implemented this workaround has to
> > be in place.
> > 
> > Also per suggestions by Stefano utilize the PVHVM paths
> > as they share common functionality.
> > 
> > Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
> > ---
> >  drivers/xen/gntdev.c       |  2 +-
> >  drivers/xen/grant-table.c  | 80 ++++++++++++++++++++++++++++++++++++++++++----
> >  drivers/xen/platform-pci.c |  2 +-
> >  include/xen/grant_table.h  |  2 +-
> >  4 files changed, 76 insertions(+), 10 deletions(-)
> > 
> > diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
> > index e41c79c..073b4a1 100644
> > --- a/drivers/xen/gntdev.c
> > +++ b/drivers/xen/gntdev.c
> > @@ -846,7 +846,7 @@ static int __init gntdev_init(void)
> >  	if (!xen_domain())
> >  		return -ENODEV;
> >  
> > -	use_ptemod = xen_pv_domain();
> > +	use_ptemod = !xen_feature(XENFEAT_auto_translated_physmap);
> >  
> >  	err = misc_register(&gntdev_miscdev);
> >  	if (err != 0) {
> > diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
> > index aa846a4..c0ded9f 100644
> > --- a/drivers/xen/grant-table.c
> > +++ b/drivers/xen/grant-table.c
> > @@ -47,6 +47,7 @@
> >  #include <xen/interface/xen.h>
> >  #include <xen/page.h>
> >  #include <xen/grant_table.h>
> > +#include <xen/balloon.h>
> >  #include <xen/interface/memory.h>
> >  #include <xen/hvc-console.h>
> >  #include <xen/swiotlb-xen.h>
> > @@ -66,8 +67,8 @@ static unsigned int boot_max_nr_grant_frames;
> >  static int gnttab_free_count;
> >  static grant_ref_t gnttab_free_head;
> >  static DEFINE_SPINLOCK(gnttab_list_lock);
> > -unsigned long xen_hvm_resume_frames;
> > -EXPORT_SYMBOL_GPL(xen_hvm_resume_frames);
> > +unsigned long xen_auto_xlat_grant_frames;
> > +EXPORT_SYMBOL_GPL(xen_auto_xlat_grant_frames);
> >  
> >  static union {
> >  	struct grant_entry_v1 *v1;
> > @@ -1060,7 +1061,7 @@ static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
> >  	unsigned int nr_gframes = end_idx + 1;
> >  	int rc;
> >  
> > -	if (xen_hvm_domain()) {
> > +	if (xen_feature(XENFEAT_auto_translated_physmap)) {
> >  		struct xen_add_to_physmap xatp;
> >  		unsigned int i = end_idx;
> >  		rc = 0;
> > @@ -1069,10 +1070,24 @@ static int gnttab_map(unsigned int start_idx, unsigned int end_idx)
> >  		 * index, ensuring that the table will grow only once.
> >  		 */
> >  		do {
> > +			unsigned long vaddr;
> > +			unsigned int level;
> > +			pte_t *pte;
> > +
> >  			xatp.domid = DOMID_SELF;
> >  			xatp.idx = i;
> >  			xatp.space = XENMAPSPACE_grant_table;
> > -			xatp.gpfn = (xen_hvm_resume_frames >> PAGE_SHIFT) + i;
> > +
> > +			/*
> > +			 * Don't assume the memory is contingous. Lookup each.
> > +			 */
> > +			vaddr = xen_auto_xlat_grant_frames + (i * PAGE_SIZE);
> > +			if (xen_hvm_domain())
> > +				xatp.gpfn = vaddr >> PAGE_SHIFT;
> > +			else {
> > +				pte = lookup_address(vaddr, &level);
> 
> lookup_address is x86 only. I added an empty implementation under
> arch/arm to be able to compile stuff under drivers/xen but I would like
> to get rid of it.
> If you can't find a way to replace lookup_address with something that
> works on arm and arm64 too, you could always turn
> xen_auto_xlat_grant_frames into an array and store the pfns there.

Yeah, that would work nicely too. Thought we would have to store the
size somewhere. Maybe a struct then.

struct xlat_grant_frames {
	int	nr;
	xen_pfn_t	*array;
	struct page	*page;
};

> 
> 
> > +				xatp.gpfn = pte_mfn(*pte);
> 
> Do you actually need the mfn here? Don't you just need the pfn?

Just (pte.pte & ~PAGE_FLAGS) >> PAGE_SHIFT.

> 
> 
> 
> > +			}
> >  			rc = HYPERVISOR_memory_op(XENMEM_add_to_physmap, &xatp);
> >  			if (rc != 0) {
> >  				pr_warn("grant table add_to_physmap failed, err=%d\n",
> > @@ -1135,7 +1150,7 @@ static void gnttab_request_version(void)
> >  	int rc;
> >  	struct gnttab_set_version gsv;
> >  
> > -	if (xen_hvm_domain())
> > +	if (xen_hvm_domain() || xen_feature(XENFEAT_auto_translated_physmap))
> 
> just turn this to
> 
> if (xen_feature(XENFEAT_auto_translated_physmap))

Hmm, I would rather actually keep that until I fix the underlaying issue
of why v2 can't be used. And then we can get rid of this workaround.

> 
> 
> >  		gsv.version = 1;
> >  	else
> >  		gsv.version = 2;
> > @@ -1161,6 +1176,46 @@ static void gnttab_request_version(void)
> >  	pr_info("Grant tables using version %d layout\n", grant_table_version);
> >  }
> >  
> > +static int xlated_setup_gnttab_pages(unsigned long nr_grant_frames,
> > +				     unsigned long max, void *addr)
> > +{
> > +	struct page **pages;
> > +	unsigned long *pfns;
> > +	int rc, i;
> > +
> > +	pages = kcalloc(nr_grant_frames, sizeof(pages[0]), GFP_KERNEL);
> > +	if (!pages)
> > +		return -ENOMEM;
> > +
> > +	pfns = kcalloc(nr_grant_frames, sizeof(pfns[0]), GFP_KERNEL);
> > +	if (!pfns) {
> > +		kfree(pages);
> > +		return -ENOMEM;
> > +	}
> > +	rc = alloc_xenballooned_pages(nr_grant_frames, pages, 0 /* lowmem */);
> > +	if (rc) {
> > +		pr_warn("%s Couldn't balloon alloc %ld pfns rc:%d\n", __func__,
> > +			nr_grant_frames, rc);
> > +		kfree(pages);
> > +		kfree(pfns);
> > +		return rc;
> > +	}
> > +	for (i = 0; i < nr_grant_frames; i++)
> > +		pfns[i] = page_to_pfn(pages[i]);
> > +
> > +	rc = arch_gnttab_map_shared(pfns, nr_grant_frames, max, addr);
> > +
> > +	kfree(pages);
> > +	kfree(pfns);
> > +	if (rc) {
> > +		pr_warn("%s Couldn't map %ld pfns rc:%d\n", __func__,
> > +			nr_grant_frames, rc);
> > +		free_xenballooned_pages(nr_grant_frames, pages);
> > +		return rc;
> > +	}
> > +	return rc;
> > +}
> 
> Please move this function somewhere under arch/x86/xen and call it from
> xen_start_kernel, it doesn't belong to common code.

<nods>

Actually, looking at how the xen-platform-pci does it. I am wondering
if that can be just made in one function that will do the right
thing.

Can the xen-platform-pci use balloon pages instead of the MMIO BARs?
Or would that be a waste of memory?

> 
> 
> >  static int gnttab_setup(void)
> >  {
> >  	unsigned int max_nr_gframes;
> > @@ -1169,15 +1224,26 @@ static int gnttab_setup(void)
> >  	if (max_nr_gframes < nr_grant_frames)
> >  		return -ENOSYS;
> >  
> > +	if (xen_feature(XENFEAT_auto_translated_physmap) && !xen_auto_xlat_grant_frames) {
> > +		/*
> > +		 * xen_auto_xlat_grant_frames is setup for PVHVM by
> > +		 * alloc_xen_mmio by the time this is called.
> > +		 */
> > +		int rc = xlated_setup_gnttab_pages(max_nr_gframes, max_nr_gframes,
> > +						   &gnttab_shared.addr);
> > +		if (rc)
> > +			return rc;
> > +		xen_auto_xlat_grant_frames = (unsigned long)gnttab_shared.addr;
> > +	}
> 
> I think you should do this from arch/x86/xen. When gnttab_setup is
> called in your guest, xen_auto_xlat_grant_frames should have already
> been setup correctly.

<nods> And have some early init that sets this up.

> 
> 
> >  	if (xen_pv_domain())
> >  		return gnttab_map(0, nr_grant_frames - 1);
> >  
> >  	if (gnttab_shared.addr == NULL) {
> > -		gnttab_shared.addr = xen_remap(xen_hvm_resume_frames,
> > +		gnttab_shared.addr = xen_remap(xen_auto_xlat_grant_frames,
> >  						PAGE_SIZE * max_nr_gframes);
> >  		if (gnttab_shared.addr == NULL) {
> >  			pr_warn("Failed to ioremap gnttab share frames (addr=0x%08lx)!\n",
> > -					xen_hvm_resume_frames);
> > +					xen_auto_xlat_grant_frames);
> >  			return -ENOMEM;
> >  		}
> >  	}
> > diff --git a/drivers/xen/platform-pci.c b/drivers/xen/platform-pci.c
> > index 2f3528e..44bc5a6 100644
> > --- a/drivers/xen/platform-pci.c
> > +++ b/drivers/xen/platform-pci.c
> > @@ -154,7 +154,7 @@ static int platform_pci_init(struct pci_dev *pdev,
> >  	}
> >  
> >  	max_nr_gframes = gnttab_max_grant_frames();
> > -	xen_hvm_resume_frames = alloc_xen_mmio(PAGE_SIZE * max_nr_gframes);
> > +	xen_auto_xlat_grant_frames = alloc_xen_mmio(PAGE_SIZE * max_nr_gframes);
> >  	ret = gnttab_init();
> >  	if (ret)
> >  		goto out;
> > diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h
> > index 694dcaf..24280ac 100644
> > --- a/include/xen/grant_table.h
> > +++ b/include/xen/grant_table.h
> > @@ -178,7 +178,7 @@ int arch_gnttab_map_status(uint64_t *frames, unsigned long nr_gframes,
> >  			   grant_status_t **__shared);
> >  void arch_gnttab_unmap(void *shared, unsigned long nr_gframes);
> >  
> > -extern unsigned long xen_hvm_resume_frames;
> > +extern unsigned long xen_auto_xlat_grant_frames;
> >  unsigned int gnttab_max_grant_frames(void);
> >  
> >  #define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr))
> 
> You need to s/xen_hvm_resume_frames/xen_auto_xlat_grant_frames under
> arch/arm/xen too.

Right. Thanks for reminding me!
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists