linux-riscv.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Alex Ghiti <alex@ghiti.fr>
To: Palmer Dabbelt <palmer@dabbelt.com>
Cc: anup@brainfault.org, linux-kernel@vger.kernel.org,
	zong.li@sifive.com, Paul Walmsley <paul.walmsley@sifive.com>,
	linux-riscv@lists.infradead.org, Christoph Hellwig <hch@lst.de>
Subject: Re: [RFC PATCH 2/7] riscv: Allow to dynamically define VA_BITS
Date: Tue, 7 Apr 2020 01:12:31 -0400	[thread overview]
Message-ID: <8922d36f-1b4e-0575-33e7-197e3f0b3817@ghiti.fr> (raw)
In-Reply-To: <mhng-1f8477e3-7cb4-454f-b443-8cf032325385@palmerdabbelt-glaptop1>


On 4/3/20 11:17 AM, Palmer Dabbelt wrote:
> On Sun, 22 Mar 2020 04:00:23 PDT (-0700), alex@ghiti.fr wrote:
>> With 4-level page table folding at runtime, we don't know at compile time
>> the size of the virtual address space so we must set VA_BITS dynamically
>> so that sparsemem reserves the right amount of memory for struct pages.
>>
>> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
>> ---
>>  arch/riscv/Kconfig                 | 10 ----------
>>  arch/riscv/include/asm/pgtable.h   | 10 +++++++++-
>>  arch/riscv/include/asm/sparsemem.h |  2 +-
>>  3 files changed, 10 insertions(+), 12 deletions(-)
>>
>> diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
>> index f5f3d474504d..8e4b1cbcf2c2 100644
>> --- a/arch/riscv/Kconfig
>> +++ b/arch/riscv/Kconfig
>> @@ -99,16 +99,6 @@ config ZONE_DMA32
>>      bool
>>      default y if 64BIT
>>
>> -config VA_BITS
>> -    int
>> -    default 32 if 32BIT
>> -    default 39 if 64BIT
>> -
>> -config PA_BITS
>> -    int
>> -    default 34 if 32BIT
>> -    default 56 if 64BIT
>> -
>>  config PAGE_OFFSET
>>      hex
>>      default 0xC0000000 if 32BIT && MAXPHYSMEM_2GB
>> diff --git a/arch/riscv/include/asm/pgtable.h 
>> b/arch/riscv/include/asm/pgtable.h
>> index 185ffe3723ec..dce401eed1d3 100644
>> --- a/arch/riscv/include/asm/pgtable.h
>> +++ b/arch/riscv/include/asm/pgtable.h
>> @@ -26,6 +26,14 @@
>>  #endif /* CONFIG_64BIT */
>>
>>  #ifdef CONFIG_MMU
>> +#ifdef CONFIG_64BIT
>> +#define VA_BITS        39
>> +#define PA_BITS        56
>> +#else
>> +#define VA_BITS        32
>> +#define PA_BITS        34
> 
> We've moved to 32-bit physical addresses on rv32 in Linux.  The mismatch 
> was
> causing too many issues in generic code.

Ok I missed this one, thanks.

> 
>> +#endif
>> +
>>  /* Number of entries in the page global directory */
>>  #define PTRS_PER_PGD    (PAGE_SIZE / sizeof(pgd_t))
>>  /* Number of entries in the page table */
>> @@ -108,7 +116,7 @@ extern pgd_t swapper_pg_dir[];
>>   * position vmemmap directly below the VMALLOC region.
>>   */
>>  #define VMEMMAP_SHIFT \
>> -    (CONFIG_VA_BITS - PAGE_SHIFT - 1 + STRUCT_PAGE_MAX_SHIFT)
>> +    (VA_BITS - PAGE_SHIFT - 1 + STRUCT_PAGE_MAX_SHIFT)
>>  #define VMEMMAP_SIZE    BIT(VMEMMAP_SHIFT)
>>  #define VMEMMAP_END    (VMALLOC_START - 1)
>>  #define VMEMMAP_START    (VMALLOC_START - VMEMMAP_SIZE)
>> diff --git a/arch/riscv/include/asm/sparsemem.h 
>> b/arch/riscv/include/asm/sparsemem.h
>> index 45a7018a8118..f08d72155bc8 100644
>> --- a/arch/riscv/include/asm/sparsemem.h
>> +++ b/arch/riscv/include/asm/sparsemem.h
>> @@ -4,7 +4,7 @@
>>  #define _ASM_RISCV_SPARSEMEM_H
>>
>>  #ifdef CONFIG_SPARSEMEM
>> -#define MAX_PHYSMEM_BITS    CONFIG_PA_BITS
>> +#define MAX_PHYSMEM_BITS    PA_BITS
>>  #define SECTION_SIZE_BITS    27
>>  #endif /* CONFIG_SPARSEMEM */
> 
> Aside from the 32-bit PA issue:
> 
> Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>

Thanks,

Alex


  reply	other threads:[~2020-04-07  5:12 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-03-22 11:00 [RFC PATCH 0/7] Introduce sv48 support Alexandre Ghiti
2020-03-22 11:00 ` [RFC PATCH 1/7] riscv: Get rid of compile time logic with MAX_EARLY_MAPPING_SIZE Alexandre Ghiti
2020-03-26  6:10   ` Anup Patel
2020-04-03 15:17   ` Palmer Dabbelt
2020-04-07  5:12     ` Alex Ghiti
2020-03-22 11:00 ` [RFC PATCH 2/7] riscv: Allow to dynamically define VA_BITS Alexandre Ghiti
2020-03-26  6:12   ` Anup Patel
2020-04-03 15:17   ` Palmer Dabbelt
2020-04-07  5:12     ` Alex Ghiti [this message]
2020-03-22 11:00 ` [RFC PATCH 3/7] riscv: Simplify MAXPHYSMEM config Alexandre Ghiti
2020-03-26  6:22   ` Anup Patel
2020-03-26  6:34   ` Anup Patel
2020-04-03 15:53   ` Palmer Dabbelt
2020-04-07  5:13     ` Alex Ghiti
2020-03-22 11:00 ` [RFC PATCH 4/7] riscv: Implement sv48 support Alexandre Ghiti
2020-03-26  7:00   ` Anup Patel
2020-03-31 16:31     ` Alex Ghiti
2020-04-03 15:53   ` Palmer Dabbelt
2020-04-07  5:14     ` Alex Ghiti
2020-04-07  5:56       ` Anup Patel
2020-04-08  4:39         ` Alex Ghiti
2020-04-08  5:06           ` Anup Patel
2020-03-22 11:00 ` [RFC PATCH 5/7] riscv: Use pgtable_l4_enabled to output mmu type in cpuinfo Alexandre Ghiti
2020-03-26  7:01   ` Anup Patel
2020-04-03 15:53   ` Palmer Dabbelt
2020-04-07  5:14     ` Alex Ghiti
2020-03-22 11:00 ` [RFC PATCH 6/7] dt-bindings: riscv: Remove "riscv, svXX" property from device-tree Alexandre Ghiti
2020-03-26  7:03   ` Anup Patel
2020-04-03 15:53   ` Palmer Dabbelt
2020-04-07  5:14     ` Alex Ghiti
2020-03-22 11:00 ` [RFC PATCH 7/7] riscv: Explicit comment about user virtual address space size Alexandre Ghiti
2020-03-26  7:05   ` Anup Patel
2020-04-03 15:53   ` Palmer Dabbelt
2020-04-07  5:15     ` Alex Ghiti
2020-03-31 19:53 ` [RFC PATCH 0/7] Introduce sv48 support Palmer Dabbelt

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=8922d36f-1b4e-0575-33e7-197e3f0b3817@ghiti.fr \
    --to=alex@ghiti.fr \
    --cc=anup@brainfault.org \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=palmer@dabbelt.com \
    --cc=paul.walmsley@sifive.com \
    --cc=zong.li@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).