Skip to contentSkip to conversation

Loading handle_arch_irq with a PC relative load

PATCH Gilles Chanteperdrix 2 participants 6 messages Jul 13, 2012 – Jul 14, 2012 Arm (32-bit)

lore
Share & tools
lore
#0 lore raw

I do not know if it is really useful, but it seems it would be possible to reduce the number of memory accesses to just one in the irq_handler macro in the case where CONFIG_MULTI_IRQ_HANDLER is enabled, by using a PC relative load, with something like the following patch:

arch/arm/kernel/entry-armv.S +7 -8
index 0d1851c..48ee46a 100644
@@ -37,10 +37,9 @@
  */
 	.macro	irq_handler
 #ifdef CONFIG_MULTI_IRQ_HANDLER
-	ldr	r1, =handle_arch_irq
 	mov	r0, sp
 	adr	lr, BSYM(9997f)
-	ldr	pc, [r1]
+	ldr	pc, handle_arch_irq
 #else
 	arch_irq_handler_default
 #endif
@@ -325,6 +324,12 @@ ENDPROC(__pabt_svc)
 #endif
 .LCfp:
 	.word	fp_enter
+#ifdef CONFIG_MULTI_IRQ_HANDLER
+	.globl	handle_arch_irq
+handle_arch_irq:
+	.space	4
+#endif
+
 
 /*
  * User mode handlers
@@ -1151,9 +1156,3 @@ cr_alignment:
 	.space	4
 cr_no_alignment:
 	.space	4
-
-#ifdef CONFIG_MULTI_IRQ_HANDLER
-	.globl	handle_arch_irq
-handle_arch_irq:
-	.space	4
-#endif
Signature · 1 line
                                                                Gilles.
#1 lore raw
Quoted 5 lines · Gilles Chanteperdrix

I do not know if it is really useful, but it seems it would be possible to reduce the number of memory accesses to just one in the irq_handler macro in the case where CONFIG_MULTI_IRQ_HANDLER is enabled, by using a PC relative load, with something like the following patch:

To be strict with ccode sections, you can't do this. The handle_arch_irq symbol identifies a variable and with your patch you're moving it from the .data section to the .text section. The .text section is meant to be read only, and this is even more true when using a XIP kernel where .text is in ROM, or if we could make the access protection of the kernel ro.

Quoted 49 lines
arch/arm/kernel/entry-armv.S +7 -8
index 0d1851c..48ee46a 100644
@@ -37,10 +37,9 @@
  */
 	.macro	irq_handler
 #ifdef CONFIG_MULTI_IRQ_HANDLER
-	ldr	r1, =handle_arch_irq
 	mov	r0, sp
 	adr	lr, BSYM(9997f)
-	ldr	pc, [r1]
+	ldr	pc, handle_arch_irq
 #else
 	arch_irq_handler_default
 #endif
@@ -325,6 +324,12 @@ ENDPROC(__pabt_svc)
 #endif
 .LCfp:
 	.word	fp_enter
+#ifdef CONFIG_MULTI_IRQ_HANDLER
+	.globl	handle_arch_irq
+handle_arch_irq:
+	.space	4
+#endif
+
 
 /*
  * User mode handlers
@@ -1151,9 +1156,3 @@ cr_alignment:
 	.space	4
 cr_no_alignment:
 	.space	4
-
-#ifdef CONFIG_MULTI_IRQ_HANDLER
-	.globl	handle_arch_irq
-handle_arch_irq:
-	.space	4
-#endif
Signature · 7 lines
-- 
                                                                Gilles.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
#2 lore raw
Quoted 8 lines · Nicolas Pitre
Quoted 5 lines · Gilles Chanteperdrix

I do not know if it is really useful, but it seems it would be possible to reduce the number of memory accesses to just one in the irq_handler macro in the case where CONFIG_MULTI_IRQ_HANDLER is enabled, by using a PC relative load, with something like the following patch:

Quoted 6 lines

To be strict with ccode sections, you can't do this. The handle_arch_irq symbol identifies a variable and with your patch you're moving it from the .data section to the .text section. The .text section is meant to be read only, and this is even more true when using a XIP kernel where .text is in ROM, or if we could make the access protection of the kernel ro.

I understand that but, XIP kernel aside, the handle_arch_irq variable is set only once very early during the boot process, so, almost read-only. Is not Linux using self-modifying code in some cases anyway (booting an SMP kernel on an UP processor for instance).

Signature · 1 line
                                                                Gilles.
#3 lore raw
Quoted 16 lines · Gilles Chanteperdrix
Quoted 8 lines · Nicolas Pitre
Quoted 5 lines · Gilles Chanteperdrix

I do not know if it is really useful, but it seems it would be possible to reduce the number of memory accesses to just one in the irq_handler macro in the case where CONFIG_MULTI_IRQ_HANDLER is enabled, by using a PC relative load, with something like the following patch:

Quoted 6 lines

To be strict with ccode sections, you can't do this. The handle_arch_irq symbol identifies a variable and with your patch you're moving it from the .data section to the .text section. The .text section is meant to be read only, and this is even more true when using a XIP kernel where .text is in ROM, or if we could make the access protection of the kernel ro.

Quoted 4 lines

I understand that but, XIP kernel aside, the handle_arch_irq variable is set only once very early during the boot process, so, almost read-only. Is not Linux using self-modifying code in some cases anyway (booting an SMP kernel on an UP processor for instance).

There are limits to which such tricks should be applied. In the SMP on UP case this is a matter of making the kernel boot at all which is a rather strong reason.

Do you have performance numbers like interrupt latency that show this patch being worth it? Without concrete justifications I don't think we should go down that path.

Nicolas

#4 lore raw
Quoted 23 lines · Nicolas Pitre
Quoted 16 lines · Gilles Chanteperdrix
Quoted 8 lines · Nicolas Pitre
Quoted 5 lines · Gilles Chanteperdrix

I do not know if it is really useful, but it seems it would be possible to reduce the number of memory accesses to just one in the irq_handler macro in the case where CONFIG_MULTI_IRQ_HANDLER is enabled, by using a PC relative load, with something like the following patch:

Quoted 6 lines

To be strict with ccode sections, you can't do this. The handle_arch_irq symbol identifies a variable and with your patch you're moving it from the .data section to the .text section. The .text section is meant to be read only, and this is even more true when using a XIP kernel where .text is in ROM, or if we could make the access protection of the kernel ro.

Quoted 4 lines

I understand that but, XIP kernel aside, the handle_arch_irq variable is set only once very early during the boot process, so, almost read-only. Is not Linux using self-modifying code in some cases anyway (booting an SMP kernel on an UP processor for instance).

Quoted 7 lines

There are limits to which such tricks should be applied. In the SMP on UP case this is a matter of making the kernel boot at all which is a rather strong reason.

Do you have performance numbers like interrupt latency that show this patch being worth it? Without concrete justifications I don't think we should go down that path.

I intend to do some interrupt latency measurements soon. But I suspect CONFIG_MULTI_IRQ_HANDLER will cause more differences due to the fact that the irq handlers are now fat C compiled code instead of carefully optimized assembly code, than because of these two memory accesses.

And in fact, chances are that I will observe nothing at all since the low end platforms I have are AT91 which are not using CONFIG_MULTI_IRQ_HANDLER yet.

Signature · 1 line
                                                                Gilles.
#5 lore raw
In reply to Nicolas Pitre
Quoted 23 lines · Nicolas Pitre
Quoted 16 lines · Gilles Chanteperdrix
Quoted 8 lines · Nicolas Pitre
Quoted 5 lines · Gilles Chanteperdrix

I do not know if it is really useful, but it seems it would be possible to reduce the number of memory accesses to just one in the irq_handler macro in the case where CONFIG_MULTI_IRQ_HANDLER is enabled, by using a PC relative load, with something like the following patch:

Quoted 6 lines

To be strict with ccode sections, you can't do this. The handle_arch_irq symbol identifies a variable and with your patch you're moving it from the .data section to the .text section. The .text section is meant to be read only, and this is even more true when using a XIP kernel where .text is in ROM, or if we could make the access protection of the kernel ro.

Quoted 4 lines

I understand that but, XIP kernel aside, the handle_arch_irq variable is set only once very early during the boot process, so, almost read-only. Is not Linux using self-modifying code in some cases anyway (booting an SMP kernel on an UP processor for instance).

Quoted 7 lines

There are limits to which such tricks should be applied. In the SMP on UP case this is a matter of making the kernel boot at all which is a rather strong reason.

Do you have performance numbers like interrupt latency that show this patch being worth it? Without concrete justifications I don't think we should go down that path.

So, I ran a few tests on at91rm9200, where I expected the differences to be most visible.

First I enabled CONFIG_MULTI_IRQ_HANDLER and wrote the irq decoding handler in plain C. This increases the irq latency of 1.2us (measured with the average irq latency on an idle system).

I rewrote this irq decoding handler in assembly, using the macros in entry-macro.S. This decreases the irq latency of 600ns.

Then I try the trick at the beginning of this thread, and... could not measure any difference, so, you were right.

Anyway, given that on at91rm9200 worst case irq latencies are in the 80us range, all these optimizations are pointless.

Signature · 1 line
                                                                Gilles.