linux-riscv.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Palmer Dabbelt <palmer@sifive.com>
To: Atish Patra <Atish.Patra@wdc.com>
Cc: Damien Le Moal <Damien.LeMoal@wdc.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	Paul Walmsley <paul.walmsley@sifive.com>,
	linux-riscv@lists.infradead.org, Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH 16/17] riscv: clear the instruction cache and all registers when booting
Date: Mon, 08 Jul 2019 01:26:33 -0700 (PDT)	[thread overview]
Message-ID: <mhng-3f43f4b8-473d-429d-9a09-12d3542e33bc@palmer-si-x1e> (raw)
In-Reply-To: <78919862d11f6d56446f8fffd8a1a8c601ea5c32.camel@wdc.com>

On Mon, 01 Jul 2019 14:26:18 PDT (-0700), Atish Patra wrote:
> On Mon, 2019-06-24 at 07:43 +0200, Christoph Hellwig wrote:
>> When we get booted we want a clear slate without any leaks from
>> previous
>> supervisors or the firmware.  Flush the instruction cache and then
>> clear
>> all registers to known good values.  This is really important for the
>> upcoming nommu support that runs on M-mode, but can't really harm
>> when
>> running in S-mode either.
> 
> That means it should be done for S-mode as well. Right ?
> I see the reset code is enabled only for M-mode only.
> 
>>   Vaguely based on the concepts from opensbi.
>> 
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>> ---
>>  arch/riscv/kernel/head.S | 85
>> ++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 85 insertions(+)
>> 
>> diff --git a/arch/riscv/kernel/head.S b/arch/riscv/kernel/head.S
>> index a4c170e41a34..74feb17737b4 100644
>> --- a/arch/riscv/kernel/head.S
>> +++ b/arch/riscv/kernel/head.S
>> @@ -11,6 +11,7 @@
>>  #include <asm/thread_info.h>
>>  #include <asm/page.h>
>>  #include <asm/csr.h>
>> +#include <asm/hwcap.h>
>>  
>>  __INIT
>>  ENTRY(_start)
>> @@ -19,6 +20,12 @@ ENTRY(_start)
>>  	csrw CSR_XIP, zero
>>  
>>  #ifdef CONFIG_M_MODE
>> +	/* flush the instruction cache */
>> +	fence.i
>> +
>> +	/* Reset all registers except ra, a0, a1 */
>> +	call reset_regs
>> +
>>  	/*
>>  	 * The hartid in a0 is expected later on, and we have no
>> firmware
>>  	 * to hand it to us.
>> @@ -168,6 +175,84 @@ relocate:
>>  	j .Lsecondary_park
>>  END(_start)
>>  
>> +#ifdef CONFIG_M_MODE
>> +ENTRY(reset_regs)
>> +	li	sp, 0
>> +	li	gp, 0
>> +	li	tp, 0
>> +	li	t0, 0
>> +	li	t1, 0
>> +	li	t2, 0
>> +	li	s0, 0
>> +	li	s1, 0
>> +	li	a2, 0
>> +	li	a3, 0
>> +	li	a4, 0
>> +	li	a5, 0
>> +	li	a6, 0
>> +	li	a7, 0
>> +	li	s2, 0
>> +	li	s3, 0
>> +	li	s4, 0
>> +	li	s5, 0
>> +	li	s6, 0
>> +	li	s7, 0
>> +	li	s8, 0
>> +	li	s9, 0
>> +	li	s10, 0
>> +	li	s11, 0
>> +	li	t3, 0
>> +	li	t4, 0
>> +	li	t5, 0
>> +	li	t6, 0
>> +	csrw	sscratch, 0
>> +
>> +#ifdef CONFIG_FPU
>> +	csrr	t0, misa
>> +	andi	t0, t0, (COMPAT_HWCAP_ISA_F | COMPAT_HWCAP_ISA_D)
>> +	bnez	t0, .Lreset_regs_done
>> +
>> +	li	t1, SR_FS
>> +	csrs	sstatus, t1

You need to check that the write stuck and branch around the FP instructions.
Specifically, CONFIG_FPU means there may be an FPU, not there's definately an
FPU.  You should also turn the FPU back off after zeroing the state.

>> +	fmv.s.x	f0, zero
>> +	fmv.s.x	f1, zero
>> +	fmv.s.x	f2, zero
>> +	fmv.s.x	f3, zero
>> +	fmv.s.x	f4, zero
>> +	fmv.s.x	f5, zero
>> +	fmv.s.x	f6, zero
>> +	fmv.s.x	f7, zero
>> +	fmv.s.x	f8, zero
>> +	fmv.s.x	f9, zero
>> +	fmv.s.x	f10, zero
>> +	fmv.s.x	f11, zero
>> +	fmv.s.x	f12, zero
>> +	fmv.s.x	f13, zero
>> +	fmv.s.x	f14, zero
>> +	fmv.s.x	f15, zero
>> +	fmv.s.x	f16, zero
>> +	fmv.s.x	f17, zero
>> +	fmv.s.x	f18, zero
>> +	fmv.s.x	f19, zero
>> +	fmv.s.x	f20, zero
>> +	fmv.s.x	f21, zero
>> +	fmv.s.x	f22, zero
>> +	fmv.s.x	f23, zero
>> +	fmv.s.x	f24, zero
>> +	fmv.s.x	f25, zero
>> +	fmv.s.x	f26, zero
>> +	fmv.s.x	f27, zero
>> +	fmv.s.x	f28, zero
>> +	fmv.s.x	f29, zero
>> +	fmv.s.x	f30, zero
>> +	fmv.s.x	f31, zero
>> +	csrw	fcsr, 0
>> +#endif /* CONFIG_FPU */
>> +.Lreset_regs_done:
>> +	ret
>> +END(reset_regs)
>> +#endif /* CONFIG_M_MODE */
>> +
>>  __PAGE_ALIGNED_BSS
>>  	/* Empty zero page */
>>  	.balign PAGE_SIZE
> 
> -- 
> Regards,
> Atish

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

  reply	other threads:[~2019-07-08  8:26 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-24  5:42 RISC-V nommu support v2 Christoph Hellwig
2019-06-24  5:42 ` [PATCH 01/17] mm: provide a print_vma_addr stub for !CONFIG_MMU Christoph Hellwig
2019-06-24  5:42 ` [PATCH 02/17] mm: stub out all of swapops.h " Christoph Hellwig
2019-06-24  5:42 ` [PATCH 03/17] mm/nommu: fix the MAP_UNINITIALIZED flag Christoph Hellwig
2019-06-24  5:42 ` [PATCH 04/17] irqchip/sifive-plic: set max threshold for ignored handlers Christoph Hellwig
2019-06-24  5:42 ` [PATCH 05/17] riscv: use CSR_SATP instead of the legacy sptbr name in switch_mm Christoph Hellwig
2019-07-01 18:53   ` Atish Patra
2019-06-24  5:43 ` [PATCH 06/17] riscv: refactor the IPI code Christoph Hellwig
2019-06-24  5:43 ` [PATCH 07/17] riscv: abstract out CSR names for supervisor vs machine mode Christoph Hellwig
2019-07-01 18:37   ` Atish Patra
2019-06-24  5:43 ` [PATCH 08/17] riscv: improve the default power off implementation Christoph Hellwig
2019-07-01 21:07   ` Atish Patra
2019-06-24  5:43 ` [PATCH 09/17] riscv: provide a flat entry loader Christoph Hellwig
2019-06-24  5:43 ` [PATCH 10/17] riscv: read the hart ID from mhartid on boot Christoph Hellwig
2019-07-01 21:15   ` Atish Patra
2019-06-24  5:43 ` [PATCH 11/17] riscv: provide native clint access for M-mode Christoph Hellwig
2019-06-24  5:43 ` [PATCH 12/17] riscv: implement remote sfence.i natively " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 13/17] riscv: poison SBI calls " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 14/17] riscv: don't allow selecting SBI-based drivers " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 15/17] riscv: use the correct interrupt levels " Christoph Hellwig
2019-06-24  5:43 ` [PATCH 16/17] riscv: clear the instruction cache and all registers when booting Christoph Hellwig
2019-07-01 21:26   ` Atish Patra
2019-07-08  8:26     ` Palmer Dabbelt [this message]
2019-08-13 15:40       ` Christoph Hellwig
2019-08-13 15:37     ` hch
2019-06-24  5:43 ` [PATCH 17/17] riscv: add nommu support Christoph Hellwig
2019-07-12 14:52   ` Vladimir Murzin
2019-06-24 11:47 ` RISC-V nommu support v2 Vladimir Murzin
2019-06-24 11:54   ` Christoph Hellwig
2019-06-24 13:08     ` Vladimir Murzin
2019-06-24 13:16       ` Christoph Hellwig
2019-06-25  7:31       ` Palmer Dabbelt
2019-06-25 12:37         ` Vladimir Murzin
2019-07-01  6:56 ` Christoph Hellwig
2019-07-01 16:06   ` Paul Walmsley

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=mhng-3f43f4b8-473d-429d-9a09-12d3542e33bc@palmer-si-x1e \
    --to=palmer@sifive.com \
    --cc=Atish.Patra@wdc.com \
    --cc=Damien.LeMoal@wdc.com \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=paul.walmsley@sifive.com \
    /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).