search for: exc_vmm_communication

On Tue, Jul 14, 2020 at 02:09:05PM +0200, Joerg Roedel wrote: > @@ -1028,6 +1036,16 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) > struct ghcb *ghcb; > > lockdep_assert_irqs_disabled(); > + > + /* > + * #DB is special and needs to be handled outside of the intrumentation_begin()/end(). > + * Otherwise the #VC handler could be raised recursivly. > + */ > + if (error_code == SVM_EXIT_EXCP_B...

On Tue, Jul 14, 2020 at 02:09:05PM +0200, Joerg Roedel wrote: > @@ -1028,6 +1036,16 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) > struct ghcb *ghcb; > > lockdep_assert_irqs_disabled(); > + > + /* > + * #DB is special and needs to be handled outside of the intrumentation_begin()/end(). > + * Otherwise the #VC handler could be raised recursivly. > + */ > + if (error_code == SVM_EXIT_EXCP_B...

..._handle_trap_db(struct pt_regs *regs) +{ + if (user_mode(regs)) + noist_exc_debug(regs); + else + exc_debug(regs); +} + static enum es_result vc_handle_exitcode(struct es_em_ctxt *ctxt, struct ghcb *ghcb, unsigned long exit_code) @@ -1028,6 +1036,16 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) struct ghcb *ghcb; lockdep_assert_irqs_disabled(); + + /* + * #DB is special and needs to be handled outside of the intrumentation_begin()/end(). + * Otherwise the #VC handler could be raised recursivly. + */ + if (error_code == SVM_EXIT_EXCP_BASE + X86_TRAP_DB) { + vc_handle_trap_db(reg...

On Wed, Jul 15, 2020 at 10:47:52AM +0200, Peter Zijlstra wrote: > On Tue, Jul 14, 2020 at 02:09:05PM +0200, Joerg Roedel wrote: > > > @@ -1028,6 +1036,16 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) > > struct ghcb *ghcb; > > > > lockdep_assert_irqs_disabled(); > > + > > + /* > > + * #DB is special and needs to be handled outside of the intrumentation_begin()/end(). > > + * Otherwise the #VC handler could be raised recursivly. > > + */...

..._handle_trap_db(struct pt_regs *regs) +{ + if (user_mode(regs)) + noist_exc_debug(regs); + else + exc_debug(regs); +} + static enum es_result vc_handle_exitcode(struct es_em_ctxt *ctxt, struct ghcb *ghcb, unsigned long exit_code) @@ -1033,6 +1041,15 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) struct ghcb *ghcb; lockdep_assert_irqs_disabled(); + + /* + * Handle #DB before calling into !noinstr code to avoid recursive #DB. + */ + if (error_code == SVM_EXIT_EXCP_BASE + X86_TRAP_DB) { + vc_handle_trap_db(regs); + return; + } + instrumentation_begin(); /* -- 2.28.0

On Wed, Jul 15, 2020 at 11:13:37AM +0200, Joerg Roedel wrote: > On Wed, Jul 15, 2020 at 10:47:52AM +0200, Peter Zijlstra wrote: > > On Tue, Jul 14, 2020 at 02:09:05PM +0200, Joerg Roedel wrote: > > > > > @@ -1028,6 +1036,16 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) > > > struct ghcb *ghcb; > > > > > > lockdep_assert_irqs_disabled(); > > > + > > > + /* > > > + * #DB is special and needs to be handled outside of the intrumentation_begin()/end(). > > > + * Otherwise the #VC handler could be...

On Wed, Jul 15, 2020 at 11:13:37AM +0200, Joerg Roedel wrote: > On Wed, Jul 15, 2020 at 10:47:52AM +0200, Peter Zijlstra wrote: > > On Tue, Jul 14, 2020 at 02:09:05PM +0200, Joerg Roedel wrote: > > > > > @@ -1028,6 +1036,16 @@ DEFINE_IDTENTRY_VC_SAFE_STACK(exc_vmm_communication) > > > struct ghcb *ghcb; > > > > > > lockdep_assert_irqs_disabled(); > > > + > > > + /* > > > + * #DB is special and needs to be handled outside of the intrumentation_begin()/end(). > > > + * Otherwise the #VC handler could be...

From: Joerg Roedel <jroedel at suse.de> Hi, here is the fourth version of the SEV-ES Guest Support patches. I addressed the review comments sent to me for the previous version and rebased the code v5.8-rc5. The biggest change in this version is the IST handling code for the #VC handler. I adapted the entry code for the #VC handler to the big pile of entry code changes merged into

From: Joerg Roedel <jroedel at suse.de> Hi, here is the fourth version of the SEV-ES Guest Support patches. I addressed the review comments sent to me for the previous version and rebased the code v5.8-rc5. The biggest change in this version is the IST handling code for the #VC handler. I adapted the entry code for the #VC handler to the big pile of entry code changes merged into

From: Joerg Roedel <jroedel at suse.de> Hi, here is a rebased version of the latest SEV-ES patches. They are now based on latest tip/master instead of upstream Linux and include the necessary changes. Changes to v4 are in particular: - Moved early IDT setup code to idt.c, because the idt_descr and the idt_table are now static - This required to make stack protector work early (or

From: Joerg Roedel <jroedel at suse.de> Hi, here is the new version of the SEV-ES client enabling patch-set. It is based on the latest tip/master branch and contains the necessary changes. In particular those ar: - Enabling CR4.FSGSBASE early on supported processors so that early #VC exceptions on APs can be handled. - Add another patch (patch 1) to fix a KVM frame-size build

From: Joerg Roedel <jroedel at suse.de> Hi, here is a new version of the SEV-ES Guest Support patches for x86. The previous versions can be found as a linked list starting here: https://lore.kernel.org/lkml/20200824085511.7553-1-joro at 8bytes.org/ I updated the patch-set based on ther review comments I got and the discussions around it. Another important change is that the early IDT

From: Joerg Roedel <jroedel at suse.de> Hi, here is a new version of the SEV-ES Guest Support patches for x86. The previous versions can be found as a linked list starting here: https://lore.kernel.org/lkml/20200824085511.7553-1-joro at 8bytes.org/ I updated the patch-set based on ther review comments I got and the discussions around it. Another important change is that the early IDT