linux-kernel - Re: swiotlb detection should be memory hotplug aware ?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <1279732414.10874.6.camel@ank32.eng.vmware.com>
Date:	Wed, 21 Jul 2010 10:13:34 -0700
From:	Alok Kataria <akataria@...are.com>
To:	FUJITA Tomonori <fujita.tomonori@....ntt.co.jp>
Cc:	"konrad.wilk@...cle.com" <konrad.wilk@...cle.com>,
	"ak@...ux.intel.com" <ak@...ux.intel.com>,
	"lenb@...nel.org" <lenb@...nel.org>,
	"x86@...nel.org" <x86@...nel.org>,
	"linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Petr Vandrovec <petr@...are.com>
Subject: Re: swiotlb detection should be memory hotplug aware ?


On Tue, 2010-07-20 at 21:58 -0700, FUJITA Tomonori wrote:
> On Tue, 20 Jul 2010 15:14:57 -0700
> Alok Kataria <akataria@...are.com> wrote:
> 
> > Reviving a 4 month old thread. 
> > I am still waiting for any clues on this question below. 
> 
> Basically, you want to add hot-plug memory and enable swiotlb, right?

Not really, I am planning to do something like this, 

@@ -52,7 +52,7 @@ int __init pci_swiotlb_detect(void)
 
 	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
 #ifdef CONFIG_X86_64
-	if (!no_iommu && max_pfn > MAX_DMA32_PFN)
+	if (!no_iommu && (max_pfn > MAX_DMA32_PFN || hotplug_possible()))
 		swiotlb = 1;
 #endif
 	if (swiotlb_force)

BUT, I don't know how that hotplug_possible function will look like or
if such an interface already exists in the kernel (my search didn't turn
up any) ?

IMO, it should be possible to go read the SRAT to see if this system has
support for hotplug memory and then enable swiotlb if it does.

Sounds right ?

Thanks,
Alok

> 
> We can't start swiotlb reliably after a system starts.
> 
> See dma32_reserve_boatmen() and dma32_free_bootmem(). What we do is
> reserving huge memory in DMA32 zone for swiotlb and releasing it if we
> find that we don't need swiotlb. We can't find enough memory for
> swiotlb in dma32 after a system starts.

--
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/