From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Keir Fraser <keir@xen.org>, Jan Beulich <JBeulich@suse.com>,
Xen-devel <xen-devel@lists.xen.org>
Subject: Re: [PATCH RFC 1/9] x86/traps: Names for system descriptor types
Date: Thu, 15 May 2014 10:56:20 +0100 [thread overview]
Message-ID: <53748F44.2050801@citrix.com> (raw)
In-Reply-To: <1400147299-31772-2-git-send-email-andrew.cooper3@citrix.com>
On 15/05/14 10:48, Andrew Cooper wrote:
> Avoids some particularly obscure magic numbers
>
> Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
> CC: Keir Fraser <keir@xen.org>
> CC: Jan Beulich <JBeulich@suse.com>
> ---
Something I missed from the --- section.
I was toying with the naming scheme SYS_DESC_xxx instead as perhaps
being slightly more informative in context.
~Andrew
> xen/arch/x86/crash.c | 2 +-
> xen/arch/x86/traps.c | 8 ++++----
> xen/arch/x86/x86_64/traps.c | 4 ++--
> xen/include/asm-x86/desc.h | 8 ++++++++
> xen/include/asm-x86/ldt.h | 2 +-
> 5 files changed, 16 insertions(+), 8 deletions(-)
>
> diff --git a/xen/arch/x86/crash.c b/xen/arch/x86/crash.c
> index aed3b3e..97adf02 100644
> --- a/xen/arch/x86/crash.c
> +++ b/xen/arch/x86/crash.c
> @@ -149,7 +149,7 @@ static void nmi_shootdown_cpus(void)
> * This update is safe from a security point of view, as this pcpu
> * is never going to try to sysret back to a PV vcpu.
> */
> - _set_gate_lower(&idt_tables[i][TRAP_nmi], 14, 0, &trap_nop);
> + _set_gate_lower(&idt_tables[i][TRAP_nmi], DESC_TYPE_irq_gate, 0, &trap_nop);
> set_ist(&idt_tables[i][TRAP_machine_check], IST_NONE);
> }
> else
> diff --git a/xen/arch/x86/traps.c b/xen/arch/x86/traps.c
> index 40366f1..1de70af 100644
> --- a/xen/arch/x86/traps.c
> +++ b/xen/arch/x86/traps.c
> @@ -3431,8 +3431,8 @@ static void __set_intr_gate(unsigned int n, uint32_t dpl, void *addr)
> /* Keep secondary tables in sync with IRQ updates. */
> for ( i = 1; i < nr_cpu_ids; i++ )
> if ( idt_tables[i] != NULL )
> - _set_gate(&idt_tables[i][n], 14, dpl, addr);
> - _set_gate(&idt_table[n], 14, dpl, addr);
> + _set_gate(&idt_tables[i][n], DESC_TYPE_irq_gate, dpl, addr);
> + _set_gate(&idt_table[n], DESC_TYPE_irq_gate, dpl, addr);
> }
>
> static void set_swint_gate(unsigned int n, void *addr)
> @@ -3457,12 +3457,12 @@ void load_TR(void)
> this_cpu(gdt_table) + TSS_ENTRY - FIRST_RESERVED_GDT_ENTRY,
> (unsigned long)tss,
> offsetof(struct tss_struct, __cacheline_filler) - 1,
> - 9);
> + DESC_TYPE_tss_avail);
> _set_tssldt_desc(
> this_cpu(compat_gdt_table) + TSS_ENTRY - FIRST_RESERVED_GDT_ENTRY,
> (unsigned long)tss,
> offsetof(struct tss_struct, __cacheline_filler) - 1,
> - 11);
> + DESC_TYPE_tss_busy);
>
> /* Switch to non-compat GDT (which has B bit clear) to execute LTR. */
> asm volatile (
> diff --git a/xen/arch/x86/x86_64/traps.c b/xen/arch/x86/x86_64/traps.c
> index 3a48478..cdaf1e6 100644
> --- a/xen/arch/x86/x86_64/traps.c
> +++ b/xen/arch/x86/x86_64/traps.c
> @@ -394,10 +394,10 @@ void __devinit subarch_percpu_traps_init(void)
> * The 32-on-64 hypercall entry vector is only accessible from ring 1.
> * Also note that this is a trap gate, not an interrupt gate.
> */
> - _set_gate(idt_table+HYPERCALL_VECTOR, 15, 1, &compat_hypercall);
> + _set_gate(idt_table+HYPERCALL_VECTOR, DESC_TYPE_trap_gate, 1, &compat_hypercall);
>
> /* Fast trap for int80 (faster than taking the #GP-fixup path). */
> - _set_gate(idt_table+0x80, 15, 3, &int80_direct_trap);
> + _set_gate(idt_table+0x80, DESC_TYPE_trap_gate, 3, &int80_direct_trap);
> }
>
> stack_bottom = (char *)get_stack_bottom();
> diff --git a/xen/include/asm-x86/desc.h b/xen/include/asm-x86/desc.h
> index 4edb834..6b2094f 100644
> --- a/xen/include/asm-x86/desc.h
> +++ b/xen/include/asm-x86/desc.h
> @@ -98,6 +98,14 @@
>
> #ifndef __ASSEMBLY__
>
> +/* System Descriptor types in the GDT */
> +#define DESC_TYPE_ldt (2)
> +#define DESC_TYPE_tss_avail (9)
> +#define DESC_TYPE_tss_busy (11)
> +#define DESC_TYPE_call_gate (12)
> +#define DESC_TYPE_irq_gate (14)
> +#define DESC_TYPE_trap_gate (15)
> +
> struct desc_struct {
> u32 a, b;
> };
> diff --git a/xen/include/asm-x86/ldt.h b/xen/include/asm-x86/ldt.h
> index b6f7beb..4ba6d32 100644
> --- a/xen/include/asm-x86/ldt.h
> +++ b/xen/include/asm-x86/ldt.h
> @@ -18,7 +18,7 @@ static inline void load_LDT(struct vcpu *v)
> desc = (!is_pv_32on64_vcpu(v)
> ? this_cpu(gdt_table) : this_cpu(compat_gdt_table))
> + LDT_ENTRY - FIRST_RESERVED_GDT_ENTRY;
> - _set_tssldt_desc(desc, LDT_VIRT_START(v), ents*8-1, 2);
> + _set_tssldt_desc(desc, LDT_VIRT_START(v), ents*8-1, DESC_TYPE_ldt);
> __asm__ __volatile__ ( "lldt %%ax" : : "a" (LDT_ENTRY << 3) );
> }
> }
next prev parent reply other threads:[~2014-05-15 9:56 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-15 9:48 [PATCH RFC 0/9] x86: Improvements to trap handling Andrew Cooper
2014-05-15 9:48 ` [PATCH RFC 1/9] x86/traps: Names for system descriptor types Andrew Cooper
2014-05-15 9:56 ` Andrew Cooper [this message]
2014-05-15 10:08 ` Jan Beulich
2014-05-15 10:26 ` Andrew Cooper
2014-05-15 12:10 ` Jan Beulich
2014-05-15 9:48 ` [PATCH RFC 2/9] x86/traps: Make panic and reboot paths safe during early boot Andrew Cooper
2014-05-15 10:19 ` Jan Beulich
2014-05-15 10:53 ` Andrew Cooper
2014-05-15 12:12 ` Jan Beulich
2014-05-15 15:46 ` Andrew Cooper
2014-05-15 15:59 ` Jan Beulich
2014-05-15 9:48 ` [PATCH RFC 3/9] x86/traps: Make the main trap handlers safe for use early during Xen boot Andrew Cooper
2014-05-15 10:20 ` Jan Beulich
2014-05-15 9:48 ` [PATCH RFC 4/9] x86/misc: Early cleanup Andrew Cooper
2014-05-15 10:32 ` Jan Beulich
2014-05-15 10:38 ` Andrew Cooper
2014-05-15 9:48 ` [PATCH RFC 5/9] x86/traps: Functional prep work Andrew Cooper
2014-05-15 10:36 ` Jan Beulich
2014-05-15 10:45 ` Andrew Cooper
2014-05-15 12:15 ` Jan Beulich
2014-05-15 12:42 ` Andrew Cooper
2014-05-15 9:48 ` [PATCH RFC 6/9] x86/boot: Install trap handlers much earlier on boot Andrew Cooper
2014-05-15 10:53 ` Jan Beulich
2014-05-15 11:05 ` Andrew Cooper
2014-05-15 12:21 ` Jan Beulich
2014-05-15 9:48 ` [PATCH RFC 7/9] x86/boot: Drop pre-C IDT patching Andrew Cooper
2014-05-15 9:48 ` [PATCH RFC 8/9] x86/irqs: Move interrupt-stub generation out of C Andrew Cooper
2014-05-15 13:06 ` Jan Beulich
2014-05-15 9:48 ` [PATCH RFC 9/9] x86/misc: Post cleanup Andrew Cooper
2014-05-15 13:14 ` Jan Beulich
2014-05-15 13:17 ` Andrew Cooper
2014-05-16 8:49 ` [PATCH RFC 0/9] x86: Improvements to trap handling Wu, Feng
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=53748F44.2050801@citrix.com \
--to=andrew.cooper3@citrix.com \
--cc=JBeulich@suse.com \
--cc=keir@xen.org \
--cc=xen-devel@lists.xen.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).