From: Tom Lendacky <thomas.lendacky@amd.com>
To: Borislav Petkov <bp@alien8.de>
Cc: linux-arch@vger.kernel.org, linux-efi@vger.kernel.org,
kvm@vger.kernel.org, linux-doc@vger.kernel.org, x86@kernel.org,
kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
kasan-dev@googlegroups.com, linux-mm@kvack.org,
iommu@lists.linux-foundation.org,
"Rik van Riel" <riel@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>,
"Toshimitsu Kani" <toshi.kani@hpe.com>,
"Arnd Bergmann" <arnd@arndb.de>,
"Jonathan Corbet" <corbet@lwn.net>,
"Matt Fleming" <matt@codeblueprint.co.uk>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Joerg Roedel" <joro@8bytes.org>,
"Konrad Rzeszutek Wilk" <konrad.wilk@oracle.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Larry Woodman" <lwoodman@redhat.com>,
"Brijesh Singh" <brijesh.singh@amd.com>,
"Ingo Molnar" <mingo@redhat.com>,
"Andy Lutomirski" <luto@kernel.org>,
"H. Peter Anvin" <hpa@zytor.com>,
"Andrey Ryabinin" <aryabinin@virtuozzo.com>,
"Alexander Potapenko" <glider@google.com>,
"Dave Young" <dyoung@redhat.com>,
"Thomas Gleixner" <tglx@linutronix.de>,
"Dmitry Vyukov" <dvyukov@google.com>
Subject: Re: [PATCH v5 28/32] x86/mm, kexec: Allow kexec to be used with SME
Date: Fri, 19 May 2017 15:45:28 -0500 [thread overview]
Message-ID: <1b74e0e6-3dda-f638-461b-f73af9904360@amd.com> (raw)
In-Reply-To: <20170517191755.h2xluopk2p6suw32@pd.tnic>
On 5/17/2017 2:17 PM, Borislav Petkov wrote:
> On Tue, Apr 18, 2017 at 04:21:21PM -0500, Tom Lendacky wrote:
>> Provide support so that kexec can be used to boot a kernel when SME is
>> enabled.
>>
>> Support is needed to allocate pages for kexec without encryption. This
>> is needed in order to be able to reboot in the kernel in the same manner
>> as originally booted.
>>
>> Additionally, when shutting down all of the CPUs we need to be sure to
>> flush the caches and then halt. This is needed when booting from a state
>> where SME was not active into a state where SME is active (or vice-versa).
>> Without these steps, it is possible for cache lines to exist for the same
>> physical location but tagged both with and without the encryption bit. This
>> can cause random memory corruption when caches are flushed depending on
>> which cacheline is written last.
>>
>> Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
>> ---
>> arch/x86/include/asm/init.h | 1 +
>> arch/x86/include/asm/irqflags.h | 5 +++++
>> arch/x86/include/asm/kexec.h | 8 ++++++++
>> arch/x86/include/asm/pgtable_types.h | 1 +
>> arch/x86/kernel/machine_kexec_64.c | 35 +++++++++++++++++++++++++++++++++-
>> arch/x86/kernel/process.c | 26 +++++++++++++++++++++++--
>> arch/x86/mm/ident_map.c | 11 +++++++----
>> include/linux/kexec.h | 14 ++++++++++++++
>> kernel/kexec_core.c | 7 +++++++
>> 9 files changed, 101 insertions(+), 7 deletions(-)
>
> ...
>
>> @@ -86,7 +86,7 @@ static int init_transition_pgtable(struct kimage *image, pgd_t *pgd)
>> set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE));
>> }
>> pte = pte_offset_kernel(pmd, vaddr);
>> - set_pte(pte, pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL_EXEC));
>> + set_pte(pte, pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL_EXEC_NOENC));
>> return 0;
>> err:
>> free_transition_pgtable(image);
>> @@ -114,6 +114,7 @@ static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
>> .alloc_pgt_page = alloc_pgt_page,
>> .context = image,
>> .pmd_flag = __PAGE_KERNEL_LARGE_EXEC,
>> + .kernpg_flag = _KERNPG_TABLE_NOENC,
>> };
>> unsigned long mstart, mend;
>> pgd_t *level4p;
>> @@ -597,3 +598,35 @@ void arch_kexec_unprotect_crashkres(void)
>> {
>> kexec_mark_crashkres(false);
>> }
>> +
>> +int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
>> +{
>> + int ret;
>> +
>> + if (sme_active()) {
>
> if (!sme_active())
> return 0;
>
> /*
> * If SME...
>
Ok.
>
>> + /*
>> + * If SME is active we need to be sure that kexec pages are
>> + * not encrypted because when we boot to the new kernel the
>> + * pages won't be accessed encrypted (initially).
>> + */
>> + ret = set_memory_decrypted((unsigned long)vaddr, pages);
>> + if (ret)
>> + return ret;
>> +
>> + if (gfp & __GFP_ZERO)
>> + memset(vaddr, 0, pages * PAGE_SIZE);
>
> This function is called after alloc_pages() which already zeroes memory
> when __GFP_ZERO is supplied.
>
> If you need to clear the memory *after* set_memory_encrypted() happens,
> then you should probably mask out __GFP_ZERO before the alloc_pages()
> call so as not to do it twice.
I'll look into that. I could put the memset() at the end of this
function so that it is done here no matter what. And update the
default arch_kexec_post_alloc_pages() to also do the memset(). It
just hides the clearing of the pages a bit though by doing that.
>
>> + }
>> +
>> + return 0;
>> +}
>> +
>> +void arch_kexec_pre_free_pages(void *vaddr, unsigned int pages)
>> +{
>> + if (sme_active()) {
>> + /*
>> + * If SME is active we need to reset the pages back to being
>> + * an encrypted mapping before freeing them.
>> + */
>> + set_memory_encrypted((unsigned long)vaddr, pages);
>> + }
>> +}
>> diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
>> index 0bb8842..f4e5de6 100644
>> --- a/arch/x86/kernel/process.c
>> +++ b/arch/x86/kernel/process.c
>> @@ -24,6 +24,7 @@
>> #include <linux/cpuidle.h>
>> #include <trace/events/power.h>
>> #include <linux/hw_breakpoint.h>
>> +#include <linux/kexec.h>
>> #include <asm/cpu.h>
>> #include <asm/apic.h>
>> #include <asm/syscalls.h>
>> @@ -355,8 +356,25 @@ bool xen_set_default_idle(void)
>> return ret;
>> }
>> #endif
>> +
>> void stop_this_cpu(void *dummy)
>> {
>> + bool do_wbinvd_halt = false;
>> +
>> + if (kexec_in_progress && boot_cpu_has(X86_FEATURE_SME)) {
>> + /*
>> + * If we are performing a kexec and the processor supports
>> + * SME then we need to clear out cache information before
>> + * halting. With kexec, going from SME inactive to SME active
>> + * requires clearing cache entries so that addresses without
>> + * the encryption bit set don't corrupt the same physical
>> + * address that has the encryption bit set when caches are
>> + * flushed. Perform a wbinvd followed by a halt to achieve
>> + * this.
>> + */
>> + do_wbinvd_halt = true;
>> + }
>> +
>> local_irq_disable();
>> /*
>> * Remove this CPU:
>> @@ -365,8 +383,12 @@ void stop_this_cpu(void *dummy)
>> disable_local_APIC();
>> mcheck_cpu_clear(this_cpu_ptr(&cpu_info));
>>
>> - for (;;)
>> - halt();
>> + for (;;) {
>> + if (do_wbinvd_halt)
>> + native_wbinvd_halt();
>
> No need for that native_wbinvd_halt() thing:
>
> for (;;) {
> if (do_wbinvd)
> wbinvd();
>
> halt();
> }
>
Actually there is. The above will result in data in the cache because
halt() turns into a function call if CONFIG_PARAVIRT is defined (refer
to the comment above where do_wbinvd_halt is set to true). I could make
this a native_wbinvd() and native_halt() as long as those are
guaranteed to never turn into function calls. But never say never, so
that's why I created native_wbinvd_halt().
Thanks,
Tom
>> /*
>> diff --git a/arch/x86/mm/ident_map.c b/arch/x86/mm/ident_map.c
>> index 04210a2..2c9fd3e 100644
>> --- a/arch/x86/mm/ident_map.c
>> +++ b/arch/x86/mm/ident_map.c
>> @@ -20,6 +20,7 @@ static void ident_pmd_init(struct x86_mapping_info *info, pmd_t *pmd_page,
>> static int ident_pud_init(struct x86_mapping_info *info, pud_t *pud_page,
>> unsigned long addr, unsigned long end)
>> {
>> + unsigned long kernpg_flag = info->kernpg_flag ? : _KERNPG_TABLE;
>
> You're already supplying a x86_mapping_info and thus you can init
> kernpg_flag to default _KERNPG_TABLE and override it in the SME+kexec
> case, as you already do. And this way you can simply do:
>
> set_pud(pud, __pud(__pa(pmd) | info->kernpg_flag));
>
> here and in the other pagetable functions I've snipped below, and save
> yourself some lines.
Ok, I'll check into that.
Thanks,
Tom
>
> ...
>
next prev parent reply other threads:[~2017-05-19 20:45 UTC|newest]
Thread overview: 126+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-18 21:16 [PATCH v5 00/32] x86: Secure Memory Encryption (AMD) Tom Lendacky
2017-04-18 21:16 ` [PATCH v5 01/32] x86: Documentation for AMD Secure Memory Encryption (SME) Tom Lendacky
2017-04-19 9:02 ` Borislav Petkov
2017-04-19 14:23 ` Tom Lendacky
2017-04-19 15:38 ` Borislav Petkov
2017-04-19 9:52 ` David Howells
2017-04-18 21:16 ` [PATCH v5 02/32] x86/mm/pat: Set write-protect cache mode for full PAT support Tom Lendacky
2017-04-18 21:16 ` [PATCH v5 03/32] x86, mpparse, x86/acpi, x86/PCI, SFI: Use memremap for RAM mappings Tom Lendacky
2017-04-18 21:17 ` [PATCH v5 04/32] x86/CPU/AMD: Add the Secure Memory Encryption CPU feature Tom Lendacky
2017-04-18 21:17 ` [PATCH v5 05/32] x86/CPU/AMD: Handle SME reduction in physical address size Tom Lendacky
2017-04-20 16:59 ` Borislav Petkov
2017-04-20 17:29 ` Tom Lendacky
2017-04-20 18:52 ` Borislav Petkov
2017-04-18 21:17 ` [PATCH v5 06/32] x86/mm: Add Secure Memory Encryption (SME) support Tom Lendacky
2017-04-27 15:46 ` Borislav Petkov
2017-05-04 14:24 ` Tom Lendacky
2017-05-04 14:36 ` Borislav Petkov
2017-05-16 19:28 ` Tom Lendacky
2017-05-17 7:05 ` Borislav Petkov
2017-04-18 21:17 ` [PATCH v5 07/32] x86/mm: Add support to enable SME in early boot processing Tom Lendacky
2017-04-21 14:55 ` Borislav Petkov
2017-04-21 21:40 ` Tom Lendacky
2017-04-18 21:17 ` [PATCH v5 08/32] x86/mm: Simplify p[g4um]d_page() macros Tom Lendacky
2017-04-18 21:17 ` [PATCH v5 09/32] x86/mm: Provide general kernel support for memory encryption Tom Lendacky
2017-04-21 21:52 ` Dave Hansen
2017-04-24 15:53 ` Tom Lendacky
2017-04-24 15:57 ` Dave Hansen
2017-04-24 16:10 ` Tom Lendacky
2017-04-27 16:12 ` Borislav Petkov
2017-05-04 14:34 ` Tom Lendacky
2017-05-04 17:01 ` Borislav Petkov
2017-04-18 21:18 ` [PATCH v5 10/32] x86/mm: Extend early_memremap() support with additional attrs Tom Lendacky
2017-04-18 21:18 ` [PATCH v5 11/32] x86/mm: Add support for early encrypt/decrypt of memory Tom Lendacky
2017-04-18 21:18 ` [PATCH v5 12/32] x86/mm: Insure that boot memory areas are mapped properly Tom Lendacky
2017-05-04 10:16 ` Borislav Petkov
2017-05-04 14:39 ` Tom Lendacky
2017-04-18 21:18 ` [PATCH v5 13/32] x86/boot/e820: Add support to determine the E820 type of an address Tom Lendacky
2017-05-05 17:11 ` Borislav Petkov
2017-05-06 7:48 ` Ard Biesheuvel
2017-04-18 21:18 ` [PATCH v5 14/32] efi: Add an EFI table address match function Tom Lendacky
2017-05-15 18:09 ` Borislav Petkov
2017-05-16 21:53 ` Tom Lendacky
2017-04-18 21:19 ` [PATCH v5 15/32] efi: Update efi_mem_type() to return an error rather than 0 Tom Lendacky
2017-05-07 17:18 ` Borislav Petkov
2017-05-08 13:20 ` Tom Lendacky
2017-04-18 21:19 ` [PATCH v5 16/32] x86/efi: Update EFI pagetable creation to work with SME Tom Lendacky
2017-04-18 21:19 ` [PATCH v5 17/32] x86/mm: Add support to access boot related data in the clear Tom Lendacky
2017-05-15 18:35 ` Borislav Petkov
2017-05-17 18:54 ` Tom Lendacky
2017-05-18 9:02 ` Borislav Petkov
2017-05-19 20:50 ` Tom Lendacky
2017-05-21 7:16 ` Borislav Petkov
2017-05-30 16:46 ` Tom Lendacky
2017-05-31 11:31 ` Borislav Petkov
2017-05-18 19:50 ` Matt Fleming
2017-05-26 16:22 ` Tom Lendacky
2017-05-26 16:35 ` Borislav Petkov
2017-05-30 17:47 ` Tom Lendacky
2017-04-18 21:19 ` [PATCH v5 18/32] x86, mpparse: Use memremap to map the mpf and mpc data Tom Lendacky
2017-05-16 8:36 ` Borislav Petkov
2017-05-17 20:26 ` Tom Lendacky
2017-05-18 9:03 ` Borislav Petkov
2017-04-18 21:19 ` [PATCH v5 19/32] x86/mm: Add support to access persistent memory in the clear Tom Lendacky
2017-05-16 14:04 ` Borislav Petkov
2017-05-19 19:52 ` Tom Lendacky
2017-04-18 21:19 ` [PATCH v5 20/32] x86/mm: Add support for changing the memory encryption attribute Tom Lendacky
2017-04-18 21:19 ` [PATCH v5 21/32] x86, realmode: Decrypt trampoline area if memory encryption is active Tom Lendacky
2017-04-18 21:20 ` [PATCH v5 22/32] x86, swiotlb: DMA support for memory encryption Tom Lendacky
2017-05-16 14:27 ` Borislav Petkov
2017-05-19 19:54 ` Tom Lendacky
2017-04-18 21:20 ` [PATCH v5 23/32] swiotlb: Add warnings for use of bounce buffers with SME Tom Lendacky
2017-05-16 14:52 ` Borislav Petkov
2017-05-19 19:55 ` Tom Lendacky
2017-04-18 21:20 ` [PATCH v5 24/32] iommu/amd: Disable AMD IOMMU if memory encryption is active Tom Lendacky
2017-04-18 21:20 ` [PATCH v5 25/32] x86, realmode: Check for memory encryption on the APs Tom Lendacky
2017-04-18 21:20 ` [PATCH v5 26/32] x86, drm, fbdev: Do not specify encrypted memory for video mappings Tom Lendacky
2017-05-16 17:35 ` Borislav Petkov
2017-05-30 20:07 ` Tom Lendacky
2017-04-18 21:21 ` [PATCH v5 27/32] kvm: x86: svm: Enable Secure Memory Encryption within KVM Tom Lendacky
2017-04-18 21:21 ` [PATCH v5 28/32] x86/mm, kexec: Allow kexec to be used with SME Tom Lendacky
2017-05-17 19:17 ` Borislav Petkov
2017-05-19 20:45 ` Tom Lendacky [this message]
2017-05-19 20:58 ` Borislav Petkov
2017-05-19 21:07 ` Tom Lendacky
2017-05-19 21:28 ` Borislav Petkov
2017-05-19 21:38 ` Tom Lendacky
2017-05-26 4:17 ` Xunlei Pang
2017-05-27 2:17 ` Dave Young
2017-05-30 17:46 ` Tom Lendacky
2017-05-31 10:01 ` Borislav Petkov
2017-05-31 15:03 ` Xunlei Pang
2017-05-31 15:48 ` Borislav Petkov
2017-04-18 21:21 ` [PATCH v5 29/32] x86/mm: Add support to encrypt the kernel in-place Tom Lendacky
2017-05-18 12:46 ` Borislav Petkov
2017-05-25 22:24 ` Tom Lendacky
2017-05-26 16:25 ` Borislav Petkov
2017-05-30 16:39 ` Tom Lendacky
2017-05-31 9:51 ` Borislav Petkov
2017-05-31 13:12 ` Tom Lendacky
2017-04-18 21:22 ` [PATCH v5 30/32] x86/boot: Add early cmdline parsing for options with arguments Tom Lendacky
2017-04-18 21:22 ` [PATCH v5 31/32] x86: Add sysfs support for Secure Memory Encryption Tom Lendacky
2017-04-21 21:55 ` Dave Hansen
2017-04-27 7:25 ` Dave Young
2017-04-27 15:52 ` Dave Hansen
2017-04-28 5:32 ` Dave Young
2017-05-04 14:17 ` Tom Lendacky
2017-05-04 14:13 ` Tom Lendacky
2017-05-18 17:01 ` Borislav Petkov
2017-05-26 2:49 ` Dave Young
2017-05-26 5:04 ` Xunlei Pang
2017-05-26 15:47 ` Tom Lendacky
2017-04-18 21:22 ` [PATCH v5 32/32] x86/mm: Add support to make use of " Tom Lendacky
2017-04-21 18:56 ` Tom Lendacky
2017-05-19 11:30 ` Borislav Petkov
2017-05-19 20:16 ` Josh Poimboeuf
2017-05-19 20:29 ` Borislav Petkov
2017-05-30 15:48 ` Tom Lendacky
2017-05-31 9:15 ` Borislav Petkov
2017-05-30 15:46 ` Tom Lendacky
2017-05-19 11:27 ` Borislav Petkov
2017-05-30 14:38 ` Tom Lendacky
2017-05-30 14:55 ` Borislav Petkov
2017-05-30 15:37 ` Tom Lendacky
2017-05-31 8:49 ` Borislav Petkov
2017-05-31 13:37 ` Tom Lendacky
2017-05-31 14:12 ` Borislav Petkov
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=1b74e0e6-3dda-f638-461b-f73af9904360@amd.com \
--to=thomas.lendacky@amd.com \
--cc=arnd@arndb.de \
--cc=aryabinin@virtuozzo.com \
--cc=bp@alien8.de \
--cc=brijesh.singh@amd.com \
--cc=corbet@lwn.net \
--cc=dvyukov@google.com \
--cc=dyoung@redhat.com \
--cc=glider@google.com \
--cc=hpa@zytor.com \
--cc=iommu@lists.linux-foundation.org \
--cc=joro@8bytes.org \
--cc=kasan-dev@googlegroups.com \
--cc=kexec@lists.infradead.org \
--cc=konrad.wilk@oracle.com \
--cc=kvm@vger.kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-efi@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=lwoodman@redhat.com \
--cc=matt@codeblueprint.co.uk \
--cc=mingo@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=riel@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=tglx@linutronix.de \
--cc=toshi.kani@hpe.com \
--cc=x86@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).