From: shijie.huang@arm.com (Huang Shijie)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 6/9] arm64: entry: adjust el1_sync so that a function can be called
Date: Mon, 30 May 2016 10:52:27 +0800 [thread overview]
Message-ID: <1464576750-25160-7-git-send-email-shijie.huang@arm.com> (raw)
In-Reply-To: <1464576750-25160-1-git-send-email-shijie.huang@arm.com>
To implement the hardirq flag tracing properly on arm64,
the trace_hardirqs_on() should be called before interrupts are turned on.
But the trace_hardirqs_on() may clobber the registers, such as x0 ~ 18.
Just as the el0_sync does, this patch uses the callee-saved registers:
1.) uses x25 to save the esr_el1,
2.) uses the x26 to save the far_el1.
And copy x25/x26 back to x1/x0 in the proper places, such as
el1_da/el1_dbg/el1_sp_pc.
Signed-off-by: Huang Shijie <shijie.huang@arm.com>
---
arch/arm64/kernel/entry.S | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
index 98d8dd1..63bf7ad 100644
--- a/arch/arm64/kernel/entry.S
+++ b/arch/arm64/kernel/entry.S
@@ -355,8 +355,8 @@ ENDPROC(el1_error_invalid)
.align 6
el1_sync:
kernel_entry 1
- mrs x1, esr_el1 // read the syndrome register
- lsr x24, x1, #ESR_ELx_EC_SHIFT // exception class
+ mrs x25, esr_el1 // read the syndrome register
+ lsr x24, x25, #ESR_ELx_EC_SHIFT // exception class
cmp x24, #ESR_ELx_EC_DABT_CUR // data abort in EL1
b.eq el1_da
cmp x24, #ESR_ELx_EC_SYS64 // configurable trap
@@ -374,12 +374,14 @@ el1_da:
/*
* Data abort handling
*/
- mrs x0, far_el1
+ mrs x26, far_el1
enable_dbg
// re-enable interrupts if they were enabled in the aborted context
tbnz x23, #7, 1f // PSR_I_BIT
enable_irq
1:
+ mov x0, x26
+ mov x1, x25
mov x2, sp // struct pt_regs
bl do_mem_abort
@@ -392,6 +394,7 @@ el1_sp_pc:
*/
mrs x0, far_el1
enable_dbg
+ mov x1, x25
mov x2, sp
b do_sp_pc_abort
el1_undef:
@@ -409,6 +412,7 @@ el1_dbg:
cinc x24, x24, eq // set bit '0'
tbz x24, #0, el1_inv // EL1 only
mrs x0, far_el1
+ mov x1, x25
mov x2, sp // struct pt_regs
bl do_debug_exception
kernel_exit 1
--
2.5.5
next prev parent reply other threads:[~2016-05-30 2:52 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-30 2:52 [PATCH 0/9] arm64: add more accurate code for tracing the hardirq flags changes Huang Shijie
2016-05-30 2:52 ` [PATCH 1/9] arm64: entry: use the callee-saved registers to save some registers Huang Shijie
2016-05-30 2:52 ` [PATCH 2/9] arm64: entry: add a new macro to restore the registers for syscall Huang Shijie
2016-05-30 2:52 ` [PATCH 3/9] arm64: entry: record the hardirq changes on the entry/exit code Huang Shijie
2016-05-30 2:52 ` [PATCH 4/9] arm64: entry: remove the duplicated code for tracing the IRQ flags Huang Shijie
2016-05-30 2:52 ` [PATCH 5/9] arm64: entry: only restore the syscall registers once Huang Shijie
2016-05-30 2:52 ` Huang Shijie [this message]
2016-05-30 2:52 ` [PATCH 7/9] arm64: entry: save the x0 back into the stack before disabling the interrupt Huang Shijie
2016-05-30 2:52 ` [PATCH 8/9] arm64: add the hardirq flags trace code for irq macros Huang Shijie
2016-05-30 2:52 ` [PATCH 9/9] arm64: entry: remove the duplicated hardirq flags trace code Huang Shijie
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=1464576750-25160-7-git-send-email-shijie.huang@arm.com \
--to=shijie.huang@arm.com \
--cc=linux-arm-kernel@lists.infradead.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).