LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
* [PATCH] Define EFLAGS_IF
@ 2007-03-22 3:16 Rusty Russell
2007-03-22 3:24 ` Rusty Russell
2007-04-06 0:20 ` [PATCH] Define EFLAGS_IF H. Peter Anvin
0 siblings, 2 replies; 13+ messages in thread
From: Rusty Russell @ 2007-03-22 3:16 UTC (permalink / raw)
To: Andrew Morton; +Cc: lkml - Kernel Mailing List, virtualization, Andi Kleen
There is now more than one place where we use the fact that bit 9 of
eflags is the interrupt-enabled flag, so define EFLAGS_IF. We make it
512 so it can be used in asm, too.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
--- a/arch/i386/lguest/lguest.c
+++ b/arch/i386/lguest/lguest.c
@@ -107,9 +107,8 @@ static void fastcall irq_disable(void)
static void fastcall irq_enable(void)
{
- /* Linux i386 code expects bit 9 set. */
/* FIXME: Check if interrupt pending... */
- lguest_data.irq_enabled = 512;
+ lguest_data.irq_enabled = EFLAGS_IF;
}
static void fastcall lguest_load_gdt(const struct Xgt_desc_struct *desc)
@@ -394,7 +393,7 @@ static fastcall void lguest_write_idt_en
extern const char start_##name[], end_##name[]; \
asm("start_" #name ": " code "; end_" #name ":")
DEF_LGUEST(cli, "movl $0," LGUEST_IRQ);
-DEF_LGUEST(sti, "movl $512," LGUEST_IRQ);
+DEF_LGUEST(sti, "movl $"__stringify(EFLAGS_IF)"," LGUEST_IRQ);
DEF_LGUEST(popf, "movl %eax," LGUEST_IRQ);
DEF_LGUEST(pushf, "movl " LGUEST_IRQ ",%eax");
DEF_LGUEST(pushf_cli, "movl " LGUEST_IRQ ",%eax; movl $0," LGUEST_IRQ);
===================================================================
--- a/include/asm-i386/irqflags.h
+++ b/include/asm-i386/irqflags.h
@@ -87,6 +87,9 @@ static inline unsigned long __raw_local_
#endif /* __ASSEMBLY__ */
#endif /* CONFIG_PARAVIRT */
+/* Bit 9 of eflags means interrupts are enabled: a raw int for asm. */
+#define EFLAGS_IF 512
+
#ifndef __ASSEMBLY__
#define raw_local_save_flags(flags) \
do { (flags) = __raw_local_save_flags(); } while (0)
@@ -96,7 +99,7 @@ static inline unsigned long __raw_local_
static inline int raw_irqs_disabled_flags(unsigned long flags)
{
- return !(flags & (1 << 9));
+ return !(flags & EFLAGS_IF);
}
static inline int raw_irqs_disabled(void)
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-03-22 3:16 [PATCH] Define EFLAGS_IF Rusty Russell
@ 2007-03-22 3:24 ` Rusty Russell
2007-03-22 3:52 ` [PATCH] Use X86_EFLAGS_IF in irqflags.h, lguest Rusty Russell
2007-04-06 0:20 ` [PATCH] Define EFLAGS_IF H. Peter Anvin
1 sibling, 1 reply; 13+ messages in thread
From: Rusty Russell @ 2007-03-22 3:24 UTC (permalink / raw)
To: Andrew Morton; +Cc: lkml - Kernel Mailing List, virtualization, Andi Kleen
On Thu, 2007-03-22 at 14:16 +1100, Rusty Russell wrote:
> There is now more than one place where we use the fact that bit 9 of
> eflags is the interrupt-enabled flag, so define EFLAGS_IF. We make it
> 512 so it can be used in asm, too.
Belay this: there's a X86_EFLAGS_IF in asm/processor.h which we should
use. Will send patch.
Thanks to Jeremy Fitzhardinge for the clue.
Rusty.
^ permalink raw reply [flat|nested] 13+ messages in thread
* [PATCH] Use X86_EFLAGS_IF in irqflags.h, lguest.
2007-03-22 3:24 ` Rusty Russell
@ 2007-03-22 3:52 ` Rusty Russell
2007-03-22 4:03 ` Keith Owens
0 siblings, 1 reply; 13+ messages in thread
From: Rusty Russell @ 2007-03-22 3:52 UTC (permalink / raw)
To: Andrew Morton; +Cc: lkml - Kernel Mailing List, virtualization, Andi Kleen
On Thu, 2007-03-22 at 14:24 +1100, Rusty Russell wrote:
> Belay this: there's a X86_EFLAGS_IF in asm/processor.h which we should
> use. Will send patch.
How's this. There may be other users, but they're not easy to grep for.
==
Move X86_EFLAGS_IF et al out to a new header: processor-flags.h, so we
can include it from irqflags.h and use it in raw_irqs_disabled_flags().
As a side-effect, we could now use these flags in .S files.
Lguest also modified to use the flags.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
diff -r bd0a803d9948 include/asm-i386/processor-flags.h
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/include/asm-i386/processor-flags.h Thu Mar 22 14:31:41 2007 +1100
@@ -0,0 +1,26 @@
+#ifndef __ASM_I386_PROCESSOR_FLAGS_H
+#define __ASM_I386_PROCESSOR_FLAGS_H
+/* Various flags defined: can be included from assembler. */
+
+/*
+ * EFLAGS bits
+ */
+#define X86_EFLAGS_CF 0x00000001 /* Carry Flag */
+#define X86_EFLAGS_PF 0x00000004 /* Parity Flag */
+#define X86_EFLAGS_AF 0x00000010 /* Auxillary carry Flag */
+#define X86_EFLAGS_ZF 0x00000040 /* Zero Flag */
+#define X86_EFLAGS_SF 0x00000080 /* Sign Flag */
+#define X86_EFLAGS_TF 0x00000100 /* Trap Flag */
+#define X86_EFLAGS_IF 0x00000200 /* Interrupt Flag */
+#define X86_EFLAGS_DF 0x00000400 /* Direction Flag */
+#define X86_EFLAGS_OF 0x00000800 /* Overflow Flag */
+#define X86_EFLAGS_IOPL 0x00003000 /* IOPL mask */
+#define X86_EFLAGS_NT 0x00004000 /* Nested Task */
+#define X86_EFLAGS_RF 0x00010000 /* Resume Flag */
+#define X86_EFLAGS_VM 0x00020000 /* Virtual Mode */
+#define X86_EFLAGS_AC 0x00040000 /* Alignment Check */
+#define X86_EFLAGS_VIF 0x00080000 /* Virtual Interrupt Flag */
+#define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */
+#define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */
+
+#endif /* __ASM_I386_PROCESSOR_FLAGS_H */
diff -r bd0a803d9948 include/asm-i386/irqflags.h
--- a/include/asm-i386/irqflags.h Thu Mar 22 14:13:31 2007 +1100
+++ b/include/asm-i386/irqflags.h Thu Mar 22 14:47:21 2007 +1100
@@ -9,6 +9,7 @@
*/
#ifndef _ASM_IRQFLAGS_H
#define _ASM_IRQFLAGS_H
+#include <asm/processor-flags.h>
#ifndef __ASSEMBLY__
static inline unsigned long native_save_fl(void)
@@ -119,7 +120,7 @@ static inline unsigned long __raw_local_
static inline int raw_irqs_disabled_flags(unsigned long flags)
{
- return !(flags & (1 << 9));
+ return !(flags & X86_EFLAGS_IF);
}
static inline int raw_irqs_disabled(void)
diff -r bd0a803d9948 include/asm-i386/processor.h
--- a/include/asm-i386/processor.h Thu Mar 22 14:13:31 2007 +1100
+++ b/include/asm-i386/processor.h Thu Mar 22 14:31:58 2007 +1100
@@ -21,6 +21,7 @@
#include <asm/percpu.h>
#include <linux/cpumask.h>
#include <linux/init.h>
+#include <asm/processor-flags.h>
/* flag for disabling the tsc */
extern int tsc_disable;
@@ -125,27 +126,6 @@ extern void detect_ht(struct cpuinfo_x86
#else
static inline void detect_ht(struct cpuinfo_x86 *c) {}
#endif
-
-/*
- * EFLAGS bits
- */
-#define X86_EFLAGS_CF 0x00000001 /* Carry Flag */
-#define X86_EFLAGS_PF 0x00000004 /* Parity Flag */
-#define X86_EFLAGS_AF 0x00000010 /* Auxillary carry Flag */
-#define X86_EFLAGS_ZF 0x00000040 /* Zero Flag */
-#define X86_EFLAGS_SF 0x00000080 /* Sign Flag */
-#define X86_EFLAGS_TF 0x00000100 /* Trap Flag */
-#define X86_EFLAGS_IF 0x00000200 /* Interrupt Flag */
-#define X86_EFLAGS_DF 0x00000400 /* Direction Flag */
-#define X86_EFLAGS_OF 0x00000800 /* Overflow Flag */
-#define X86_EFLAGS_IOPL 0x00003000 /* IOPL mask */
-#define X86_EFLAGS_NT 0x00004000 /* Nested Task */
-#define X86_EFLAGS_RF 0x00010000 /* Resume Flag */
-#define X86_EFLAGS_VM 0x00020000 /* Virtual Mode */
-#define X86_EFLAGS_AC 0x00040000 /* Alignment Check */
-#define X86_EFLAGS_VIF 0x00080000 /* Virtual Interrupt Flag */
-#define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */
-#define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */
static inline void native_cpuid(unsigned int *eax, unsigned int *ebx,
unsigned int *ecx, unsigned int *edx)
diff -r bd0a803d9948 arch/i386/lguest/interrupts_and_traps.c
--- a/arch/i386/lguest/interrupts_and_traps.c Thu Mar 22 14:13:31 2007 +1100
+++ b/arch/i386/lguest/interrupts_and_traps.c Thu Mar 22 14:46:15 2007 +1100
@@ -42,7 +42,7 @@ static void reflect_trap(struct lguest *
(it's always 0, since irqs are enabled when guest is running). */
eflags = regs->eflags;
get_user(irq_enable, &lg->lguest_data->irq_enabled);
- eflags |= (irq_enable & 512);
+ eflags |= (irq_enable & X86_EFLAGS_IF);
push_guest_stack(lg, &gstack, eflags);
push_guest_stack(lg, &gstack, regs->cs);
@@ -86,7 +86,7 @@ void maybe_do_interrupt(struct lguest *l
/* If they're halted, we re-enable interrupts. */
if (lg->halted) {
/* Re-enable interrupts. */
- put_user(512, &lg->lguest_data->irq_enabled);
+ put_user(X86_EFLAGS_IF, &lg->lguest_data->irq_enabled);
lg->halted = 0;
} else {
/* Maybe they have interrupts disabled? */
diff -r bd0a803d9948 arch/i386/lguest/lguest.c
--- a/arch/i386/lguest/lguest.c Thu Mar 22 14:13:31 2007 +1100
+++ b/arch/i386/lguest/lguest.c Thu Mar 22 14:26:41 2007 +1100
@@ -107,9 +107,8 @@ static void fastcall irq_disable(void)
static void fastcall irq_enable(void)
{
- /* Linux i386 code expects bit 9 set. */
/* FIXME: Check if interrupt pending... */
- lguest_data.irq_enabled = 512;
+ lguest_data.irq_enabled = X86_EFLAGS_IF;
}
static void fastcall lguest_load_gdt(const struct Xgt_desc_struct *desc)
@@ -394,7 +393,7 @@ static fastcall void lguest_write_idt_en
extern const char start_##name[], end_##name[]; \
asm("start_" #name ": " code "; end_" #name ":")
DEF_LGUEST(cli, "movl $0," LGUEST_IRQ);
-DEF_LGUEST(sti, "movl $512," LGUEST_IRQ);
+DEF_LGUEST(sti, "movl $"__stringify(X86_EFLAGS_IF)"," LGUEST_IRQ);
DEF_LGUEST(popf, "movl %eax," LGUEST_IRQ);
DEF_LGUEST(pushf, "movl " LGUEST_IRQ ",%eax");
DEF_LGUEST(pushf_cli, "movl " LGUEST_IRQ ",%eax; movl $0," LGUEST_IRQ);
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Use X86_EFLAGS_IF in irqflags.h, lguest.
2007-03-22 3:52 ` [PATCH] Use X86_EFLAGS_IF in irqflags.h, lguest Rusty Russell
@ 2007-03-22 4:03 ` Keith Owens
0 siblings, 0 replies; 13+ messages in thread
From: Keith Owens @ 2007-03-22 4:03 UTC (permalink / raw)
To: Rusty Russell
Cc: Andrew Morton, lkml - Kernel Mailing List, virtualization, Andi Kleen
Rusty Russell (on Thu, 22 Mar 2007 14:52:29 +1100) wrote:
>On Thu, 2007-03-22 at 14:24 +1100, Rusty Russell wrote:
>> Belay this: there's a X86_EFLAGS_IF in asm/processor.h which we should
>> use. Will send patch.
>
>How's this. There may be other users, but they're not easy to grep for.
One less set of definitions for KDB to create - Thanks.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-03-22 3:16 [PATCH] Define EFLAGS_IF Rusty Russell
2007-03-22 3:24 ` Rusty Russell
@ 2007-04-06 0:20 ` H. Peter Anvin
2007-04-06 0:22 ` Jeremy Fitzhardinge
1 sibling, 1 reply; 13+ messages in thread
From: H. Peter Anvin @ 2007-04-06 0:20 UTC (permalink / raw)
To: Rusty Russell
Cc: Andrew Morton, lkml - Kernel Mailing List, virtualization, Andi Kleen
Rusty Russell wrote:
> There is now more than one place where we use the fact that bit 9 of
> eflags is the interrupt-enabled flag, so define EFLAGS_IF. We make it
> 512 so it can be used in asm, too.
How about defining all the other EFLAGS in one place?
-hpa
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 0:20 ` [PATCH] Define EFLAGS_IF H. Peter Anvin
@ 2007-04-06 0:22 ` Jeremy Fitzhardinge
2007-04-06 0:29 ` H. Peter Anvin
0 siblings, 1 reply; 13+ messages in thread
From: Jeremy Fitzhardinge @ 2007-04-06 0:22 UTC (permalink / raw)
To: H. Peter Anvin
Cc: Rusty Russell, virtualization, Andrew Morton, Andi Kleen,
lkml - Kernel Mailing List
H. Peter Anvin wrote:
> Rusty Russell wrote:
>
>> There is now more than one place where we use the fact that bit 9 of
>> eflags is the interrupt-enabled flag, so define EFLAGS_IF. We make it
>> 512 so it can be used in asm, too.
>>
>
> How about defining all the other EFLAGS in one place?
>
That patch got dropped, and replaced by one which pulled all the flags
definitions out of <asm/processor.h>
J
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 0:22 ` Jeremy Fitzhardinge
@ 2007-04-06 0:29 ` H. Peter Anvin
2007-04-06 0:48 ` Andi Kleen
0 siblings, 1 reply; 13+ messages in thread
From: H. Peter Anvin @ 2007-04-06 0:29 UTC (permalink / raw)
To: Jeremy Fitzhardinge
Cc: Rusty Russell, virtualization, Andrew Morton, Andi Kleen,
lkml - Kernel Mailing List
Jeremy Fitzhardinge wrote:
>
> That patch got dropped, and replaced by one which pulled all the flags
> definitions out of <asm/processor.h>
>
Saw that a little too late :)
In general, it would be nice if the various CPU constants were all
defined in one place, so I'd rather suggest protecting the appropriate
parts of asm/processor.h with #ifndef __ASSEMBLY__.
-hpa
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 0:29 ` H. Peter Anvin
@ 2007-04-06 0:48 ` Andi Kleen
2007-04-06 1:06 ` H. Peter Anvin
0 siblings, 1 reply; 13+ messages in thread
From: Andi Kleen @ 2007-04-06 0:48 UTC (permalink / raw)
To: H. Peter Anvin
Cc: Jeremy Fitzhardinge, Rusty Russell, virtualization,
Andrew Morton, Andi Kleen, lkml - Kernel Mailing List
On Thu, Apr 05, 2007 at 05:29:52PM -0700, H. Peter Anvin wrote:
> Jeremy Fitzhardinge wrote:
> >
> >That patch got dropped, and replaced by one which pulled all the flags
> >definitions out of <asm/processor.h>
> >
>
> Saw that a little too late :)
>
> In general, it would be nice if the various CPU constants were all
> defined in one place, so I'd rather suggest protecting the appropriate
> parts of asm/processor.h with #ifndef __ASSEMBLY__.
No processor.h is such a hodgepodge of unrelated stuff that any
splitting up is a good thing.
-Andi
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 0:48 ` Andi Kleen
@ 2007-04-06 1:06 ` H. Peter Anvin
2007-04-06 2:24 ` Rusty Russell
2007-04-06 10:30 ` Andi Kleen
0 siblings, 2 replies; 13+ messages in thread
From: H. Peter Anvin @ 2007-04-06 1:06 UTC (permalink / raw)
To: Andi Kleen
Cc: Jeremy Fitzhardinge, Rusty Russell, virtualization,
Andrew Morton, lkml - Kernel Mailing List
Andi Kleen wrote:
>
> No processor.h is such a hodgepodge of unrelated stuff that any
> splitting up is a good thing.
>
Fair enough. However, I'd still like to see the X86_CR* constants
moved, too (and constants added for at least CR0 as well.)
-hpa
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 1:06 ` H. Peter Anvin
@ 2007-04-06 2:24 ` Rusty Russell
2007-04-06 10:30 ` Andi Kleen
1 sibling, 0 replies; 13+ messages in thread
From: Rusty Russell @ 2007-04-06 2:24 UTC (permalink / raw)
To: H. Peter Anvin
Cc: Andi Kleen, Jeremy Fitzhardinge, virtualization, Andrew Morton,
lkml - Kernel Mailing List
On Thu, 2007-04-05 at 18:06 -0700, H. Peter Anvin wrote:
> Andi Kleen wrote:
> >
> > No processor.h is such a hodgepodge of unrelated stuff that any
> > splitting up is a good thing.
> >
>
> Fair enough. However, I'd still like to see the X86_CR* constants
> moved, too (and constants added for at least CR0 as well.)
Agreed. This was on theory of minimum damage, but since it seems to
have received a warm reception, I'd say moving the rest to
processor-flags.h would be a welcome addition.
Cheers,
Rusty.
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 1:06 ` H. Peter Anvin
2007-04-06 2:24 ` Rusty Russell
@ 2007-04-06 10:30 ` Andi Kleen
2007-04-06 15:39 ` H. Peter Anvin
1 sibling, 1 reply; 13+ messages in thread
From: Andi Kleen @ 2007-04-06 10:30 UTC (permalink / raw)
To: H. Peter Anvin
Cc: Andi Kleen, Jeremy Fitzhardinge, Rusty Russell, virtualization,
Andrew Morton, lkml - Kernel Mailing List
On Thu, Apr 05, 2007 at 06:06:16PM -0700, H. Peter Anvin wrote:
> Andi Kleen wrote:
> >
> >No processor.h is such a hodgepodge of unrelated stuff that any
> >splitting up is a good thing.
> >
>
> Fair enough. However, I'd still like to see the X86_CR* constants
> moved, too (and constants added for at least CR0 as well.)
Send patches then.
-Andi
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 10:30 ` Andi Kleen
@ 2007-04-06 15:39 ` H. Peter Anvin
2007-04-09 1:11 ` Rusty Russell
0 siblings, 1 reply; 13+ messages in thread
From: H. Peter Anvin @ 2007-04-06 15:39 UTC (permalink / raw)
To: Andi Kleen
Cc: Jeremy Fitzhardinge, Rusty Russell, virtualization,
Andrew Morton, lkml - Kernel Mailing List
Andi Kleen wrote:
> On Thu, Apr 05, 2007 at 06:06:16PM -0700, H. Peter Anvin wrote:
>> Andi Kleen wrote:
>>> No processor.h is such a hodgepodge of unrelated stuff that any
>>> splitting up is a good thing.
>>>
>> Fair enough. However, I'd still like to see the X86_CR* constants
>> moved, too (and constants added for at least CR0 as well.)
>
> Send patches then.
I will, unless Rusty does, first. No desire to step on each other.
-hpa
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Define EFLAGS_IF
2007-04-06 15:39 ` H. Peter Anvin
@ 2007-04-09 1:11 ` Rusty Russell
0 siblings, 0 replies; 13+ messages in thread
From: Rusty Russell @ 2007-04-09 1:11 UTC (permalink / raw)
To: H. Peter Anvin
Cc: Andi Kleen, Jeremy Fitzhardinge, virtualization, Andrew Morton,
lkml - Kernel Mailing List
On Fri, 2007-04-06 at 08:39 -0700, H. Peter Anvin wrote:
> I will, unless Rusty does, first. No desire to step on each other.
Oh no, please, after you!
Rusty.
^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2007-04-09 1:11 UTC | newest]
Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-03-22 3:16 [PATCH] Define EFLAGS_IF Rusty Russell
2007-03-22 3:24 ` Rusty Russell
2007-03-22 3:52 ` [PATCH] Use X86_EFLAGS_IF in irqflags.h, lguest Rusty Russell
2007-03-22 4:03 ` Keith Owens
2007-04-06 0:20 ` [PATCH] Define EFLAGS_IF H. Peter Anvin
2007-04-06 0:22 ` Jeremy Fitzhardinge
2007-04-06 0:29 ` H. Peter Anvin
2007-04-06 0:48 ` Andi Kleen
2007-04-06 1:06 ` H. Peter Anvin
2007-04-06 2:24 ` Rusty Russell
2007-04-06 10:30 ` Andi Kleen
2007-04-06 15:39 ` H. Peter Anvin
2007-04-09 1:11 ` Rusty Russell
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).