All of lore.kernel.org
 help / color / mirror / Atom feed
From: Cornelia Huck <cohuck@redhat.com>
To: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Janosch Frank <frankja@linux.vnet.ibm.com>,
	KVM <kvm@vger.kernel.org>, David Hildenbrand <david@redhat.com>,
	Thomas Huth <thuth@redhat.com>,
	Ulrich Weigand <Ulrich.Weigand@de.ibm.com>,
	Claudio Imbrenda <imbrenda@linux.ibm.com>,
	Andrea Arcangeli <aarcange@redhat.com>
Subject: Re: [RFCv2 06/37] s390: add (non)secure page access exceptions handlers
Date: Tue, 4 Feb 2020 12:18:17 +0100	[thread overview]
Message-ID: <20200204121817.10a5f271.cohuck@redhat.com> (raw)
In-Reply-To: <20200203131957.383915-7-borntraeger@de.ibm.com>

On Mon,  3 Feb 2020 08:19:26 -0500
Christian Borntraeger <borntraeger@de.ibm.com> wrote:

> From: Vasily Gorbik <gor@linux.ibm.com>
> 
> Add exceptions handlers performing transparent transition of non-secure
> pages to secure (import) upon guest access and secure pages to
> non-secure (export) upon hypervisor access.
> 
> Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
> Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
> [adding checks for failures]
> Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
> [further changes like adding a check for gmap fault]
> ---
>  arch/s390/kernel/pgm_check.S |  4 +-
>  arch/s390/mm/fault.c         | 87 ++++++++++++++++++++++++++++++++++++
>  2 files changed, 89 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/s390/kernel/pgm_check.S b/arch/s390/kernel/pgm_check.S
> index 59dee9d3bebf..27ac4f324c70 100644
> --- a/arch/s390/kernel/pgm_check.S
> +++ b/arch/s390/kernel/pgm_check.S
> @@ -78,8 +78,8 @@ PGM_CHECK(do_dat_exception)		/* 39 */
>  PGM_CHECK(do_dat_exception)		/* 3a */
>  PGM_CHECK(do_dat_exception)		/* 3b */
>  PGM_CHECK_DEFAULT			/* 3c */
> -PGM_CHECK_DEFAULT			/* 3d */
> -PGM_CHECK_DEFAULT			/* 3e */
> +PGM_CHECK(do_secure_storage_access)	/* 3d */
> +PGM_CHECK(do_non_secure_storage_access)	/* 3e */

I suppose that these two can only happen when we actually run a
protected virt guest...

>  PGM_CHECK_DEFAULT			/* 3f */
>  PGM_CHECK_DEFAULT			/* 40 */
>  PGM_CHECK_DEFAULT			/* 41 */
> diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
> index 7b0bb475c166..bd75b0765cf1 100644
> --- a/arch/s390/mm/fault.c
> +++ b/arch/s390/mm/fault.c
> @@ -38,6 +38,7 @@
>  #include <asm/irq.h>
>  #include <asm/mmu_context.h>
>  #include <asm/facility.h>
> +#include <asm/uv.h>
>  #include "../kernel/entry.h"
>  
>  #define __FAIL_ADDR_MASK -4096L
> @@ -816,3 +817,89 @@ static int __init pfault_irq_init(void)
>  early_initcall(pfault_irq_init);
>  
>  #endif /* CONFIG_PFAULT */
> +
> +#ifdef CONFIG_KVM_S390_PROTECTED_VIRTUALIZATION_HOST
> +
> +void do_secure_storage_access(struct pt_regs *regs)
> +{
> +	unsigned long addr = regs->int_parm_long & __FAIL_ADDR_MASK;
> +	struct vm_area_struct *vma;
> +	struct mm_struct *mm;
> +	struct page *page;
> +	int rc;

...so this rather complex function will never be called if we don't act
as a host for protected virt guests?

> +
> +	switch (get_fault_type(regs)) {
> +	case USER_FAULT:
> +		mm = current->mm;
> +		down_read(&mm->mmap_sem);
> +		vma = find_vma(mm, addr);
> +		if (!vma) {
> +			up_read(&mm->mmap_sem);
> +			do_fault_error(regs, VM_READ | VM_WRITE, VM_FAULT_BADMAP);
> +			break;
> +		}
> +		page = follow_page(vma, addr, FOLL_WRITE | FOLL_GET);
> +		if (IS_ERR_OR_NULL(page)) {
> +			up_read(&mm->mmap_sem);
> +			break;
> +		}
> +		if (arch_make_page_accessible(page))
> +			send_sig(SIGSEGV, current, 0);
> +		put_page(page);
> +		up_read(&mm->mmap_sem);
> +		break;
> +	case KERNEL_FAULT:
> +		page = phys_to_page(addr);
> +		if (unlikely(!try_get_page(page)))
> +			break;
> +		rc = arch_make_page_accessible(page);
> +		put_page(page);
> +		if (rc)
> +			BUG();
> +		break;
> +	case VDSO_FAULT:
> +		/* fallthrough */
> +	case GMAP_FAULT:
> +		/* fallthrough */
> +	default:
> +		do_fault_error(regs, VM_READ | VM_WRITE, VM_FAULT_BADMAP);
> +		WARN_ON_ONCE(1);
> +	}
> +}
> +NOKPROBE_SYMBOL(do_secure_storage_access);
> +
> +void do_non_secure_storage_access(struct pt_regs *regs)
> +{
> +	unsigned long gaddr = regs->int_parm_long & __FAIL_ADDR_MASK;
> +	struct gmap *gmap = (struct gmap *)S390_lowcore.gmap;
> +	struct uv_cb_cts uvcb = {
> +		.header.cmd = UVC_CMD_CONV_TO_SEC_STOR,
> +		.header.len = sizeof(uvcb),
> +		.guest_handle = gmap->se_handle,
> +		.gaddr = gaddr,
> +	};
> +	int rc;

Same for this function, of course.

> +
> +	if (get_fault_type(regs) != GMAP_FAULT) {
> +		do_fault_error(regs, VM_READ | VM_WRITE, VM_FAULT_BADMAP);
> +		WARN_ON_ONCE(1);
> +		return;
> +	}
> +
> +	rc = uv_make_secure(gmap, gaddr, &uvcb, 0);
> +	if (rc == -EINVAL && uvcb.header.rc != 0x104)
> +		send_sig(SIGSEGV, current, 0);
> +}
> +NOKPROBE_SYMBOL(do_non_secure_storage_access);
> +
> +#else
> +void do_secure_storage_access(struct pt_regs *regs)
> +{
> +	default_trap_handler(regs);
> +}
> +
> +void do_non_secure_storage_access(struct pt_regs *regs)
> +{
> +	default_trap_handler(regs);
> +}

And these should not really be called at all, I guess?

> +#endif

IOW, we don't really introduce complex code paths for systems not
acting as protected virt hosts, even if we switch on the config option
or ditch it completely?


  parent reply	other threads:[~2020-02-04 11:18 UTC|newest]

Thread overview: 200+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-03 13:19 [RFCv2 00/37] KVM: s390: Add support for protected VMs Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 01/37] DOCUMENTATION: protvirt: Protected virtual machine introduction Christian Borntraeger
2020-02-03 15:42   ` Cornelia Huck
2020-02-03 21:41     ` Christian Borntraeger
2020-02-04  9:19       ` Cornelia Huck
2020-02-04  9:32         ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 02/37] s390/protvirt: introduce host side setup Christian Borntraeger
2020-02-03 17:12   ` Cornelia Huck
2020-02-03 22:03     ` Christian Borntraeger
2020-02-04  9:28       ` Cornelia Huck
2020-02-04  9:38         ` Christian Borntraeger
2020-02-04  9:49           ` Cornelia Huck
2020-02-04  8:40   ` Thomas Huth
2020-02-04  9:12     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 03/37] s390/protvirt: add ultravisor initialization Christian Borntraeger
2020-02-04 10:03   ` Thomas Huth
2020-02-04 10:23   ` Cornelia Huck
2020-02-04 11:22     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 04/37] mm:gup/writeback: add callbacks for inaccessible pages Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 05/37] s390/mm: provide memory management functions for protected KVM guests Christian Borntraeger
2020-02-04 10:57   ` Cornelia Huck
2020-02-04 11:48     ` Claudio Imbrenda
2020-02-04 11:56     ` Christian Borntraeger
2020-02-04 12:15       ` Cornelia Huck
2020-02-05  9:52   ` Janosch Frank
2020-02-03 13:19 ` [RFCv2 06/37] s390: add (non)secure page access exceptions handlers Christian Borntraeger
2020-02-04 10:37   ` Thomas Huth
2020-02-04 11:41     ` Claudio Imbrenda
2020-02-04 12:48       ` Thomas Huth
2020-02-04 13:08         ` Claudio Imbrenda
2020-02-04 13:22           ` Thomas Huth
2020-02-04 12:03     ` Christian Borntraeger
2020-02-04 11:18   ` Cornelia Huck [this message]
2020-02-04 12:04     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 07/37] KVM: s390: add new variants of UV CALL Christian Borntraeger
2020-02-04 12:11   ` Cornelia Huck
2020-02-04 12:30     ` Claudio Imbrenda
2020-02-04 12:34       ` Cornelia Huck
2020-02-04 13:33     ` Christian Borntraeger
2020-02-04 13:27   ` Thomas Huth
2020-02-04 13:55     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 08/37] KVM: s390: protvirt: Add initial lifecycle handling Christian Borntraeger
2020-02-04 12:13   ` David Hildenbrand
2020-02-04 12:34     ` Christian Borntraeger
2020-02-05 10:01       ` Janosch Frank
2020-02-04 16:27   ` Cornelia Huck
2020-02-04 21:13     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 09/37] KVM: s390: protvirt: Add KVM api documentation Christian Borntraeger
2020-02-04 20:19   ` Thomas Huth
2020-02-04 21:40     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 10/37] KVM: s390: protvirt: Secure memory is not mergeable Christian Borntraeger
2020-02-04 12:15   ` David Hildenbrand
2020-02-04 12:53     ` Christian Borntraeger
2020-02-04 12:54       ` David Hildenbrand
2020-02-04 13:04         ` [PATCH v2 " Christian Borntraeger
2020-02-04 13:05           ` David Hildenbrand
2020-02-03 13:19 ` [RFCv2 11/37] KVM: s390/mm: Make pages accessible before destroying the guest Christian Borntraeger
2020-02-04 20:52   ` Thomas Huth
2020-02-05  9:03     ` Christian Borntraeger
2020-02-05 10:35   ` Cornelia Huck
2020-02-05 18:29     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 12/37] KVM: s390: protvirt: Handle SE notification interceptions Christian Borntraeger
2020-02-04 12:16   ` David Hildenbrand
2020-02-04 12:23     ` Christian Borntraeger
2020-02-04 21:16   ` Thomas Huth
2020-02-04 21:46     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 13/37] KVM: s390: protvirt: Instruction emulation Christian Borntraeger
2020-02-04 12:20   ` David Hildenbrand
2020-02-04 12:29     ` Christian Borntraeger
2020-02-04 17:07       ` Cornelia Huck
2020-02-04 21:21         ` Thomas Huth
2020-02-03 13:19 ` [RFCv2 14/37] KVM: s390: protvirt: Add interruption injection controls Christian Borntraeger
2020-02-05  6:59   ` Thomas Huth
2020-02-05  8:46     ` Christian Borntraeger
2020-02-05  8:56       ` Michael Mueller
2020-02-05  8:54     ` Michael Mueller
2020-02-05 10:45       ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 15/37] KVM: s390: protvirt: Implement interruption injection Christian Borntraeger
2020-02-05  9:51   ` Thomas Huth
2020-02-05 10:48     ` Christian Borntraeger
2020-02-05 11:25       ` Cornelia Huck
2020-02-05 17:49         ` Christian Borntraeger
2020-02-06  8:21           ` Cornelia Huck
2020-02-05 17:56     ` Christian Borntraeger
2020-02-05 11:31   ` Cornelia Huck
2020-02-05 11:46     ` Michael Mueller
2020-02-05 12:11       ` Cornelia Huck
2020-02-05 12:26         ` Michael Mueller
2020-02-05 18:00           ` Christian Borntraeger
2020-02-06  8:23             ` Cornelia Huck
2020-02-05 15:30     ` Michael Mueller
2020-02-03 13:19 ` [RFCv2 16/37] KVM: s390: protvirt: Add SCLP interrupt handling Christian Borntraeger
2020-02-05 12:22   ` Cornelia Huck
2020-02-05 18:14     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 17/37] KVM: s390: protvirt: Add machine-check interruption injection controls Christian Borntraeger
2020-02-05  7:06   ` Thomas Huth
2020-02-03 13:19 ` [RFCv2 18/37] KVM: s390: protvirt: Implement machine-check interruption injection Christian Borntraeger
2020-02-05  7:10   ` Thomas Huth
2020-02-05 13:47   ` Cornelia Huck
2020-02-05 18:18     ` Christian Borntraeger
2020-02-06  8:25       ` Cornelia Huck
2020-02-06  9:01         ` Christian Borntraeger
2020-02-06  9:11           ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 19/37] KVM: s390: protvirt: Handle spec exception loops Christian Borntraeger
2020-02-05 13:50   ` Cornelia Huck
2020-02-05 18:21     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 20/37] KVM: s390: protvirt: Add new gprs location handling Christian Borntraeger
2020-02-05 11:18   ` Thomas Huth
2020-02-05 18:45     ` Christian Borntraeger
2020-02-05 13:52   ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 21/37] KVM: S390: protvirt: Introduce instruction data area bounce buffer Christian Borntraeger
2020-02-05 11:43   ` David Hildenbrand
2020-02-06  8:43     ` Christian Borntraeger
2020-02-05 12:02   ` Thomas Huth
2020-02-05 12:16     ` Janosch Frank
2020-02-05 17:00       ` Thomas Huth
2020-02-06  9:07         ` Christian Borntraeger
2020-02-06  9:18           ` Thomas Huth
2020-02-06  9:21             ` Christian Borntraeger
2020-02-06  9:20           ` Christian Borntraeger
2020-02-06  9:32             ` Janosch Frank
2020-02-06  9:29           ` David Hildenbrand
2020-02-06  9:39             ` [RFCv2.1] " Christian Borntraeger
2020-02-06  9:54               ` David Hildenbrand
2020-02-06 10:02                 ` Christian Borntraeger
2020-02-06 10:08                   ` David Hildenbrand
2020-02-06 10:10                     ` Christian Borntraeger
2020-02-06 12:46                 ` Christian Borntraeger
2020-02-06 10:32               ` Thomas Huth
2020-02-06 11:07                 ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 22/37] KVM: s390: protvirt: handle secure guest prefix pages Christian Borntraeger
2020-02-05 11:51   ` David Hildenbrand
2020-02-05 19:38     ` Christian Borntraeger
2020-02-05 11:52   ` David Hildenbrand
2020-02-05 19:39     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 23/37] KVM: s390/mm: handle guest unpin events Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 24/37] KVM: s390: protvirt: Write sthyi data to instruction data area Christian Borntraeger
2020-02-05 12:09   ` Thomas Huth
2020-02-05 18:52     ` Christian Borntraeger
2020-02-05 14:27   ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 25/37] KVM: s390: protvirt: STSI handling Christian Borntraeger
2020-02-05 12:13   ` Thomas Huth
2020-02-05 19:22     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 26/37] KVM: s390: protvirt: disallow one_reg Christian Borntraeger
2020-02-05 12:16   ` Thomas Huth
2020-02-05 19:25     ` Christian Borntraeger
2020-02-05 14:42   ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 27/37] KVM: s390: protvirt: Only sync fmt4 registers Christian Borntraeger
2020-02-04 17:15   ` Cornelia Huck
2020-02-04 22:08     ` Christian Borntraeger
2020-02-05 16:29   ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 28/37] KVM: s390: protvirt: Add program exception injection Christian Borntraeger
2020-02-05 16:38   ` Cornelia Huck
2020-02-05 19:31     ` Christian Borntraeger
2020-02-05 17:59   ` Thomas Huth
2020-02-05 19:32     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 29/37] DOCUMENTATION: protvirt: Diag 308 IPL Christian Borntraeger
2020-02-03 16:13   ` Cornelia Huck
2020-02-04  8:13     ` Christian Borntraeger
2020-02-04  9:35       ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 30/37] KVM: s390: protvirt: Add diag 308 subcode 8 - 10 handling Christian Borntraeger
2020-02-04 18:51   ` Cornelia Huck
2020-02-04 22:19     ` Christian Borntraeger
2020-02-05 18:21   ` Thomas Huth
2020-02-06 10:59     ` Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 31/37] KVM: s390: protvirt: UV calls diag308 0, 1 Christian Borntraeger
2020-02-03 13:19 ` [RFCv2 32/37] KVM: s390: protvirt: Report CPU state to Ultravisor Christian Borntraeger
2020-02-06 11:12   ` Thomas Huth
2020-02-06 11:17     ` Christian Borntraeger
2020-02-06 14:29   ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 33/37] KVM: s390: protvirt: Support cmd 5 operation state Christian Borntraeger
2020-02-06 11:21   ` Thomas Huth
2020-02-06 15:17   ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 34/37] KVM: s390: protvirt: Add UV debug trace Christian Borntraeger
2020-02-06  9:41   ` Cornelia Huck
2020-02-07  8:05     ` Christian Borntraeger
2020-02-07  8:27       ` Cornelia Huck
2020-02-14  8:32     ` Christian Borntraeger
2020-02-14  8:53       ` Cornelia Huck
2020-02-03 13:19 ` [RFCv2 35/37] KVM: s390: protvirt: Mask PSW interrupt bits for interception 104 and 112 Christian Borntraeger
2020-02-06 10:10   ` Cornelia Huck
2020-02-06 11:11     ` Christian Borntraeger
2020-02-06 12:03   ` Thomas Huth
2020-02-03 13:19 ` [RFCv2 36/37] KVM: s390: protvirt: do not inject interrupts after start Christian Borntraeger
2020-02-06 10:13   ` Cornelia Huck
2020-02-06 14:24     ` Christian Borntraeger
2020-02-06 12:03   ` Thomas Huth
2020-02-03 13:19 ` [RFCv2 37/37] KVM: s390: protvirt: Add UV cpu reset calls Christian Borntraeger
2020-02-05 18:27   ` Thomas Huth
2020-02-05 20:24     ` Christian Borntraeger
2020-02-06 10:45   ` Cornelia Huck
2020-02-06 14:27     ` Christian Borntraeger
2020-02-06 14:39   ` David Hildenbrand
2020-02-06 15:07     ` Thomas Huth
2020-02-03 13:23 ` [RFCv2 00/37] KVM: s390: Add support for protected VMs Cornelia Huck
2020-02-03 13:32   ` Christian Borntraeger
2020-02-03 14:42     ` Cornelia Huck
2020-02-05 11:34 ` David Hildenbrand
2020-02-05 11:38   ` Christian Borntraeger
2020-02-05 11:40     ` Cornelia Huck

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=20200204121817.10a5f271.cohuck@redhat.com \
    --to=cohuck@redhat.com \
    --cc=Ulrich.Weigand@de.ibm.com \
    --cc=aarcange@redhat.com \
    --cc=borntraeger@de.ibm.com \
    --cc=david@redhat.com \
    --cc=frankja@linux.vnet.ibm.com \
    --cc=imbrenda@linux.ibm.com \
    --cc=kvm@vger.kernel.org \
    --cc=thuth@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 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.