Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753254AbbDFP31 (ORCPT ); Mon, 6 Apr 2015 11:29:27 -0400 Received: from mail-lb0-f182.google.com ([209.85.217.182]:36843 "EHLO mail-lb0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752960AbbDFP30 (ORCPT ); Mon, 6 Apr 2015 11:29:26 -0400 MIME-Version: 1.0 In-Reply-To: <20150406141051.GF12596@l.oracle.com> References: <8a03355698fe5b94194e9e7360f19f91c1b2cf1f.1428100853.git.luto@kernel.org> <20150406141051.GF12596@l.oracle.com> From: Andy Lutomirski Date: Mon, 6 Apr 2015 08:29:04 -0700 Message-ID: Subject: Re: [PATCH] x86, paravirt, xen: Remove the 64-bit irq_enable_sysexit pvop To: Konrad Rzeszutek Wilk Cc: Andy Lutomirski , Boris Ostrovsky , Linux Kernel Mailing List , X86 ML , Ingo Molnar , Denys Vlasenko , Linus Torvalds , Denys Vlasenko , "H. Peter Anvin" , Borislav Petkov , "xen-devel@lists.xenproject.org" Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 8309 Lines: 215 On Mon, Apr 6, 2015 at 7:10 AM, Konrad Rzeszutek Wilk wrote: > On Fri, Apr 03, 2015 at 03:52:30PM -0700, Andy Lutomirski wrote: >> [cc: Boris and Konrad. Whoops] >> >> On Fri, Apr 3, 2015 at 3:51 PM, Andy Lutomirski wrote: >> > We don't use irq_enable_sysexit on 64-bit kernels any more. Remove > > Is there an commit (or name of patch) that explains why 32-bit-user-space-on-64-bit > kernels is unsavory? sysexit never tasted very good :-p We're (hopefully) not breaking 32-bit-user-space-on-64-bit, but we're trying an unconventional approach to making the code faster and less scary. As a result, 64-bit kernels won't use sysexit any more. Hopefully Xen is okay with the slightly sneaky thing we're doing. AFAICT Xen thinks of sysretl and sysexit as slightly funny irets, so I don't expect there to be any problem. See: https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=x86/asm&id=4214a16b02971c60960afd675d03544e109e0d75 and https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/commit/?h=x86/asm&id=47091e3c5b072daca29a15d2a3caf40359b0d140 --Andy > > Thank you! >> > all the paravirt and Xen machinery to support it on 64-bit kernels. >> > >> > Signed-off-by: Andy Lutomirski >> > --- >> > >> > I haven't actually tested this on Xen, but it builds for me. >> > >> > arch/x86/ia32/ia32entry.S | 6 ------ >> > arch/x86/include/asm/paravirt_types.h | 7 ++++--- >> > arch/x86/kernel/asm-offsets.c | 2 ++ >> > arch/x86/kernel/paravirt.c | 4 +++- >> > arch/x86/kernel/paravirt_patch_64.c | 1 - >> > arch/x86/xen/enlighten.c | 3 ++- >> > arch/x86/xen/xen-asm_64.S | 16 ---------------- >> > arch/x86/xen/xen-ops.h | 2 ++ >> > 8 files changed, 13 insertions(+), 28 deletions(-) >> > >> > diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S >> > index 5d8f987a340d..eb1eb7b70f4b 100644 >> > --- a/arch/x86/ia32/ia32entry.S >> > +++ b/arch/x86/ia32/ia32entry.S >> > @@ -77,12 +77,6 @@ ENTRY(native_usergs_sysret32) >> > swapgs >> > sysretl >> > ENDPROC(native_usergs_sysret32) >> > - >> > -ENTRY(native_irq_enable_sysexit) >> > - swapgs >> > - sti >> > - sysexit >> > -ENDPROC(native_irq_enable_sysexit) >> > #endif >> > >> > /* >> > diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h >> > index 7549b8b369e4..38a0ff9ef06e 100644 >> > --- a/arch/x86/include/asm/paravirt_types.h >> > +++ b/arch/x86/include/asm/paravirt_types.h >> > @@ -160,13 +160,14 @@ struct pv_cpu_ops { >> > u64 (*read_pmc)(int counter); >> > unsigned long long (*read_tscp)(unsigned int *aux); >> > >> > +#ifdef CONFIG_X86_32 >> > /* >> > * Atomically enable interrupts and return to userspace. This >> > - * is only ever used to return to 32-bit processes; in a >> > - * 64-bit kernel, it's used for 32-on-64 compat processes, but >> > - * never native 64-bit processes. (Jump, not call.) >> > + * is only used in 32-bit kernels. 64-bit kernels use >> > + * usergs_sysret32 instead. >> > */ >> > void (*irq_enable_sysexit)(void); >> > +#endif >> > >> > /* >> > * Switch to usermode gs and return to 64-bit usermode using >> > diff --git a/arch/x86/kernel/asm-offsets.c b/arch/x86/kernel/asm-offsets.c >> > index 9f6b9341950f..2d27ebf0aed8 100644 >> > --- a/arch/x86/kernel/asm-offsets.c >> > +++ b/arch/x86/kernel/asm-offsets.c >> > @@ -49,7 +49,9 @@ void common(void) { >> > OFFSET(PV_IRQ_irq_disable, pv_irq_ops, irq_disable); >> > OFFSET(PV_IRQ_irq_enable, pv_irq_ops, irq_enable); >> > OFFSET(PV_CPU_iret, pv_cpu_ops, iret); >> > +#ifdef CONFIG_X86_32 >> > OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit); >> > +#endif >> > OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0); >> > OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2); >> > #endif >> > diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c >> > index 548d25f00c90..7563114d9c3a 100644 >> > --- a/arch/x86/kernel/paravirt.c >> > +++ b/arch/x86/kernel/paravirt.c >> > @@ -154,7 +154,9 @@ unsigned paravirt_patch_default(u8 type, u16 clobbers, void *insnbuf, >> > ret = paravirt_patch_ident_64(insnbuf, len); >> > >> > else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) || >> > +#ifdef CONFIG_X86_32 >> > type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) || >> > +#endif >> > type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret32) || >> > type == PARAVIRT_PATCH(pv_cpu_ops.usergs_sysret64)) >> > /* If operation requires a jmp, then jmp */ >> > @@ -371,7 +373,7 @@ __visible struct pv_cpu_ops pv_cpu_ops = { >> > >> > .load_sp0 = native_load_sp0, >> > >> > -#if defined(CONFIG_X86_32) || defined(CONFIG_IA32_EMULATION) >> > +#if defined(CONFIG_X86_32) >> > .irq_enable_sysexit = native_irq_enable_sysexit, >> > #endif >> > #ifdef CONFIG_X86_64 >> > diff --git a/arch/x86/kernel/paravirt_patch_64.c b/arch/x86/kernel/paravirt_patch_64.c >> > index a1da6737ba5b..0de21c62c348 100644 >> > --- a/arch/x86/kernel/paravirt_patch_64.c >> > +++ b/arch/x86/kernel/paravirt_patch_64.c >> > @@ -49,7 +49,6 @@ unsigned native_patch(u8 type, u16 clobbers, void *ibuf, >> > PATCH_SITE(pv_irq_ops, save_fl); >> > PATCH_SITE(pv_irq_ops, irq_enable); >> > PATCH_SITE(pv_irq_ops, irq_disable); >> > - PATCH_SITE(pv_cpu_ops, irq_enable_sysexit); >> > PATCH_SITE(pv_cpu_ops, usergs_sysret32); >> > PATCH_SITE(pv_cpu_ops, usergs_sysret64); >> > PATCH_SITE(pv_cpu_ops, swapgs); >> > diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c >> > index 81665c9f2132..3797b6b31f95 100644 >> > --- a/arch/x86/xen/enlighten.c >> > +++ b/arch/x86/xen/enlighten.c >> > @@ -1267,10 +1267,11 @@ static const struct pv_cpu_ops xen_cpu_ops __initconst = { >> > .read_tscp = native_read_tscp, >> > >> > .iret = xen_iret, >> > - .irq_enable_sysexit = xen_sysexit, >> > #ifdef CONFIG_X86_64 >> > .usergs_sysret32 = xen_sysret32, >> > .usergs_sysret64 = xen_sysret64, >> > +#else >> > + .irq_enable_sysexit = xen_sysexit, >> > #endif >> > >> > .load_tr_desc = paravirt_nop, >> > diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S >> > index 985fc3ee0973..a2cabb8bd6bf 100644 >> > --- a/arch/x86/xen/xen-asm_64.S >> > +++ b/arch/x86/xen/xen-asm_64.S >> > @@ -47,22 +47,6 @@ ENTRY(xen_iret) >> > ENDPATCH(xen_iret) >> > RELOC(xen_iret, 1b+1) >> > >> > -/* >> > - * sysexit is not used for 64-bit processes, so it's only ever used to >> > - * return to 32-bit compat userspace. >> > - */ >> > -ENTRY(xen_sysexit) >> > - pushq $__USER32_DS >> > - pushq %rcx >> > - pushq $X86_EFLAGS_IF >> > - pushq $__USER32_CS >> > - pushq %rdx >> > - >> > - pushq $0 >> > -1: jmp hypercall_iret >> > -ENDPATCH(xen_sysexit) >> > -RELOC(xen_sysexit, 1b+1) >> > - >> > ENTRY(xen_sysret64) >> > /* >> > * We're already on the usermode stack at this point, but >> > diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h >> > index 9e195c683549..c20fe29e65f4 100644 >> > --- a/arch/x86/xen/xen-ops.h >> > +++ b/arch/x86/xen/xen-ops.h >> > @@ -134,7 +134,9 @@ DECL_ASM(void, xen_restore_fl_direct, unsigned long); >> > >> > /* These are not functions, and cannot be called normally */ >> > __visible void xen_iret(void); >> > +#ifdef CONFIG_X86_32 >> > __visible void xen_sysexit(void); >> > +#endif >> > __visible void xen_sysret32(void); >> > __visible void xen_sysret64(void); >> > __visible void xen_adjust_exception_frame(void); >> > -- >> > 2.3.0 >> > >> >> >> >> -- >> Andy Lutomirski >> AMA Capital Management, LLC -- Andy Lutomirski AMA Capital Management, LLC -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/