From: Mike Rapoport <rppt@kernel.org>
To: Tiezhu Yang <yangtiezhu@loongson.cn>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
Andrew Morton <akpm@linux-foundation.org>,
Xuefeng Li <lixuefeng@loongson.cn>,
linux-mips@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 3/4] MIPS: Refactor early_parse_memmap() to fix memmap= parameter
Date: Mon, 28 Feb 2022 08:50:34 +0200 [thread overview]
Message-ID: <YhxwuixN+9D8mRF2@kernel.org> (raw)
In-Reply-To: <1646029866-6692-4-git-send-email-yangtiezhu@loongson.cn>
On Mon, Feb 28, 2022 at 02:31:05PM +0800, Tiezhu Yang wrote:
> According to Documentation/admin-guide/kernel-parameters.txt,
> the kernel command-line parameter memmap= means "Force usage
> of a specific region of memory", but when add "memmap=3G@64M"
> to the command-line, kernel boot hangs in sparse_init().
>
> In order to support memmap=limit@base, refactor the function
> early_parse_memmap() and then use memblock_mem_range_remove_map()
> to limit the memory region.
>
> With this patch, when add "memmap=3G@64M" to the command-line,
> the kernel boots successfully, we can see the following messages:
>
> [ 0.000000] Memory limited to 64MB-3136MB
> ...
> [ 0.000000] Early memory node ranges
> [ 0.000000] node 0: [mem 0x0000000004000000-0x000000000effffff]
> [ 0.000000] node 0: [mem 0x0000000090200000-0x00000000ffffffff]
> [ 0.000000] node 0: [mem 0x0000000120000000-0x00000001653fffff]
> ...
> [ 0.000000] Memory: 3070816K/3147776K available (...)
>
> When add "memmap=128M@64M nr_cpus=1 init 3" to the command-line,
> the kernel also boots successfully, we can see the following messages:
>
> [ 0.000000] Memory limited to 64MB-192MB
> ...
> [ 0.000000] Early memory node ranges
> [ 0.000000] node 0: [mem 0x0000000004000000-0x000000000c1fffff]
> ...
> [ 0.000000] Memory: 95312K/133120K available (...)
>
> After login, the output of free command is consistent with the
> above log.
>
> By the way, this commit only supports memmap=limit@base format,
> the other formats such as memmap=limit#base, memmap=limit$base
> and memmap=limit!base can be added if they are necessary in the
> future.
>
> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
> ---
> arch/mips/kernel/setup.c | 42 +++++++++---------------------------------
> 1 file changed, 9 insertions(+), 33 deletions(-)
>
> diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
> index 6b6d718..e3b1f2e 100644
> --- a/arch/mips/kernel/setup.c
> +++ b/arch/mips/kernel/setup.c
> @@ -340,6 +340,7 @@ static void __init bootmem_init(void)
>
> static int usermem __initdata;
> static phys_addr_t memory_limit;
> +static phys_addr_t memory_base;
>
> static int __init early_parse_mem(char *p)
> {
> @@ -355,42 +356,17 @@ early_param("mem", early_parse_mem);
>
> static int __init early_parse_memmap(char *p)
> {
> - char *oldp;
> - u64 start_at, mem_size;
> -
> if (!p)
> - return -EINVAL;
> + return 1;
>
> - if (!strncmp(p, "exactmap", 8)) {
> - pr_err("\"memmap=exactmap\" invalid on MIPS\n");
> - return 0;
> - }
> + memory_limit = memparse(p, &p) & PAGE_MASK;
> + if (*p == '@')
> + memory_base = memparse(p + 1, &p) & PAGE_MASK;
>
> - oldp = p;
> - mem_size = memparse(p, &p);
> - if (p == oldp)
> - return -EINVAL;
> -
> - if (*p == '@') {
> - start_at = memparse(p+1, &p);
> - memblock_add(start_at, mem_size);
> - } else if (*p == '#') {
> - pr_err("\"memmap=nn#ss\" (force ACPI data) invalid on MIPS\n");
> - return -EINVAL;
> - } else if (*p == '$') {
> - start_at = memparse(p+1, &p);
> - memblock_add(start_at, mem_size);
> - memblock_reserve(start_at, mem_size);
> - } else {
> - pr_err("\"memmap\" invalid format!\n");
> - return -EINVAL;
This breaks backward compatibility for systems that use memmap=X#Y and
memmap=X$Y.
For your use case it is enough to implement "memmap=exactmap" that will
drop memory provided by the firmware and only use ranges supplied in
memmap=
> - }
> + pr_notice("Memory limited to %lldMB-%lldMB\n",
> + memory_base >> 20, (memory_base + memory_limit) >> 20);
>
> - if (*p == '\0') {
> - usermem = 1;
> - return 0;
> - } else
> - return -EINVAL;
> + return 0;
> }
> early_param("memmap", early_parse_memmap);
>
> @@ -667,7 +643,7 @@ static void __init arch_mem_init(char **cmdline_p)
> __pa_symbol(&__nosave_end) - __pa_symbol(&__nosave_begin));
>
> /* Limit the memory. */
> - memblock_enforce_memory_limit(memory_limit);
> + memblock_mem_range_remove_map(memory_base, memory_limit);
> memblock_allow_resize();
>
> early_memtest(PFN_PHYS(ARCH_PFN_OFFSET), PFN_PHYS(max_low_pfn));
> --
> 2.1.0
>
--
Sincerely yours,
Mike.
next prev parent reply other threads:[~2022-02-28 6:50 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-28 6:31 [PATCH v2 0/4] MIPS: Modify mem= and memmap= parameter Tiezhu Yang
2022-02-28 6:31 ` [PATCH v2 1/4] MIPS: Refactor early_parse_mem() to fix mem= parameter Tiezhu Yang
2022-02-28 17:34 ` kernel test robot
2022-02-28 6:31 ` [PATCH v2 2/4] memblock: Introduce memblock_mem_range_remove_map() Tiezhu Yang
2022-02-28 6:31 ` [PATCH v2 3/4] MIPS: Refactor early_parse_memmap() to fix memmap= parameter Tiezhu Yang
2022-02-28 6:50 ` Mike Rapoport [this message]
2022-02-28 6:31 ` [PATCH v2 4/4] MIPS: Remove not used variable usermem Tiezhu Yang
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=YhxwuixN+9D8mRF2@kernel.org \
--to=rppt@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mips@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lixuefeng@loongson.cn \
--cc=tsbogend@alpha.franken.de \
--cc=yangtiezhu@loongson.cn \
/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.