From: Drew Fustini <dfustini@baylibre.com>
To: guoren@kernel.org
Cc: palmer@rivosinc.com, paul.walmsley@sifive.com,
falcon@tinylab.org, bjorn@kernel.org, conor.dooley@microchip.com,
linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-riscv@lists.infradead.org, stable@vger.kernel.org,
Guo Ren <guoren@linux.alibaba.com>
Subject: Re: [PATCH V2 0/2] riscv: stack: Fixup independent softirq/irq stack for CONFIG_FRAME_POINTER=n
Date: Fri, 28 Jul 2023 17:33:22 -0700 [thread overview]
Message-ID: <ZMReUsAVmwcDwEhe@x1> (raw)
In-Reply-To: <20230716001506.3506041-1-guoren@kernel.org>
On Sat, Jul 15, 2023 at 08:15:04PM -0400, guoren@kernel.org wrote:
> From: Guo Ren <guoren@linux.alibaba.com>
>
> The independent softirq/irq stack uses s0 to save & restore sp, but s0
> would be corrupted when CONFIG_FRAME_POINTER=n. So add s0 in the clobber
> list to fix the problem.
>
> <+0>: addi sp,sp,-32
> <+2>: sd s0,16(sp)
> <+4>: sd s1,8(sp)
> <+6>: sd ra,24(sp)
> <+8>: sd s2,0(sp)
> <+10>: mv s0,a0 --> compiler allocate s0 for a0 when CONFIG_FRAME_POINTER=n
> <+12>: jal ra,0xffffffff800bc0ce <irqentry_enter>
> <+16>: ld a5,56(tp) # 0x38
> <+20>: lui a4,0x4
> <+22>: mv s1,a0
> <+24>: xor a5,a5,sp
> <+28>: bgeu a5,a4,0xffffffff800bc092 <do_irq+88>
> <+32>: auipc s2,0x5d
> <+36>: ld s2,1118(s2) # 0xffffffff801194b8 <irq_stack_ptr>
> <+40>: add s2,s2,a4
> <+42>: addi sp,sp,-8
> <+44>: sd ra,0(sp)
> <+46>: addi sp,sp,-8
> <+48>: sd s0,0(sp)
> <+50>: addi s0,sp,16 --> our code clobber the s0
> <+52>: mv sp,s2
> <+54>: mv a0,s0 --> a0 got wrong value for handle_riscv_irq
> <+56>: jal ra,0xffffffff800bbb3a <handle_riscv_irq>
>
> Changelog:
> V2
> - Fixup compile error with CONFIG_FRAME_POINTER=y
> - FIxup stable@vger.kernel.org tag
>
> Guo Ren (2):
> riscv: stack: Fixup independent irq stack for CONFIG_FRAME_POINTER=n
> riscv: stack: Fixup independent softirq stack for
> CONFIG_FRAME_POINTER=n
>
> arch/riscv/kernel/irq.c | 3 +++
> arch/riscv/kernel/traps.c | 3 +++
> 2 files changed, 6 insertions(+)
>
> --
> 2.36.1
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
Tested-by: Drew Fustini <dfustini@baylibre.com>
Xi Ruoyao noticed that the mainline kernel crashed when using a kernel
config with CONFIG_FRAME_POINTER=n. I was able to reproduce this [1].
Emil suggested I trying this patche series. I can confirm that this
resolves the kernel crash on the Sipeed Lichee Pi4a [2].
Thanks,
Drew
[1] https://lore.kernel.org/linux-riscv/ZMNojqwLxcG8FcHN@x1/
[2] https://lore.kernel.org/linux-riscv/ZMQAqUfb0y%2FigQs2@x1/
WARNING: multiple messages have this Message-ID (diff)
From: Drew Fustini <dfustini@baylibre.com>
To: guoren@kernel.org
Cc: palmer@rivosinc.com, paul.walmsley@sifive.com,
falcon@tinylab.org, bjorn@kernel.org, conor.dooley@microchip.com,
linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-riscv@lists.infradead.org, stable@vger.kernel.org,
Guo Ren <guoren@linux.alibaba.com>
Subject: Re: [PATCH V2 0/2] riscv: stack: Fixup independent softirq/irq stack for CONFIG_FRAME_POINTER=n
Date: Fri, 28 Jul 2023 17:33:22 -0700 [thread overview]
Message-ID: <ZMReUsAVmwcDwEhe@x1> (raw)
In-Reply-To: <20230716001506.3506041-1-guoren@kernel.org>
On Sat, Jul 15, 2023 at 08:15:04PM -0400, guoren@kernel.org wrote:
> From: Guo Ren <guoren@linux.alibaba.com>
>
> The independent softirq/irq stack uses s0 to save & restore sp, but s0
> would be corrupted when CONFIG_FRAME_POINTER=n. So add s0 in the clobber
> list to fix the problem.
>
> <+0>: addi sp,sp,-32
> <+2>: sd s0,16(sp)
> <+4>: sd s1,8(sp)
> <+6>: sd ra,24(sp)
> <+8>: sd s2,0(sp)
> <+10>: mv s0,a0 --> compiler allocate s0 for a0 when CONFIG_FRAME_POINTER=n
> <+12>: jal ra,0xffffffff800bc0ce <irqentry_enter>
> <+16>: ld a5,56(tp) # 0x38
> <+20>: lui a4,0x4
> <+22>: mv s1,a0
> <+24>: xor a5,a5,sp
> <+28>: bgeu a5,a4,0xffffffff800bc092 <do_irq+88>
> <+32>: auipc s2,0x5d
> <+36>: ld s2,1118(s2) # 0xffffffff801194b8 <irq_stack_ptr>
> <+40>: add s2,s2,a4
> <+42>: addi sp,sp,-8
> <+44>: sd ra,0(sp)
> <+46>: addi sp,sp,-8
> <+48>: sd s0,0(sp)
> <+50>: addi s0,sp,16 --> our code clobber the s0
> <+52>: mv sp,s2
> <+54>: mv a0,s0 --> a0 got wrong value for handle_riscv_irq
> <+56>: jal ra,0xffffffff800bbb3a <handle_riscv_irq>
>
> Changelog:
> V2
> - Fixup compile error with CONFIG_FRAME_POINTER=y
> - FIxup stable@vger.kernel.org tag
>
> Guo Ren (2):
> riscv: stack: Fixup independent irq stack for CONFIG_FRAME_POINTER=n
> riscv: stack: Fixup independent softirq stack for
> CONFIG_FRAME_POINTER=n
>
> arch/riscv/kernel/irq.c | 3 +++
> arch/riscv/kernel/traps.c | 3 +++
> 2 files changed, 6 insertions(+)
>
> --
> 2.36.1
>
>
> _______________________________________________
> linux-riscv mailing list
> linux-riscv@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-riscv
Tested-by: Drew Fustini <dfustini@baylibre.com>
Xi Ruoyao noticed that the mainline kernel crashed when using a kernel
config with CONFIG_FRAME_POINTER=n. I was able to reproduce this [1].
Emil suggested I trying this patche series. I can confirm that this
resolves the kernel crash on the Sipeed Lichee Pi4a [2].
Thanks,
Drew
[1] https://lore.kernel.org/linux-riscv/ZMNojqwLxcG8FcHN@x1/
[2] https://lore.kernel.org/linux-riscv/ZMQAqUfb0y%2FigQs2@x1/
_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv
next prev parent reply other threads:[~2023-07-29 0:33 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-07-16 0:15 [PATCH V2 0/2] riscv: stack: Fixup independent softirq/irq stack for CONFIG_FRAME_POINTER=n guoren
2023-07-16 0:15 ` guoren
2023-07-16 0:15 ` [PATCH V2 1/2] riscv: stack: Fixup independent irq " guoren
2023-07-16 0:15 ` guoren
2023-07-16 8:14 ` Rémi Denis-Courmont
2023-07-16 8:14 ` Rémi Denis-Courmont
2023-07-16 0:15 ` [PATCH V2 2/2] riscv: stack: Fixup independent softirq " guoren
2023-07-16 0:15 ` guoren
2023-07-29 0:33 ` Drew Fustini [this message]
2023-07-29 0:33 ` [PATCH V2 0/2] riscv: stack: Fixup independent softirq/irq " Drew Fustini
2023-08-17 15:20 ` patchwork-bot+linux-riscv
2023-08-17 15:20 ` patchwork-bot+linux-riscv
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=ZMReUsAVmwcDwEhe@x1 \
--to=dfustini@baylibre.com \
--cc=bjorn@kernel.org \
--cc=conor.dooley@microchip.com \
--cc=falcon@tinylab.org \
--cc=guoren@kernel.org \
--cc=guoren@linux.alibaba.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=palmer@rivosinc.com \
--cc=paul.walmsley@sifive.com \
--cc=stable@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.