linux-kernel - Re: [PATCH 07/10] bootmem: add free_bootmem

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20091110064650H.fujita.tomonori@lab.ntt.co.jp>
Date:	Tue, 10 Nov 2009 06:47:05 +0900
From:	FUJITA Tomonori <fujita.tomonori@....ntt.co.jp>
To:	penberg@...helsinki.fi
Cc:	fujita.tomonori@....ntt.co.jp, linux-kernel@...r.kernel.org,
	chrisw@...s-sol.org, dwmw2@...radead.org, joerg.roedel@....com,
	mingo@...e.hu, akpm@...ux-foundation.org, hannes@...xchg.org,
	tj@...nel.org
Subject: Re: [PATCH 07/10] bootmem: add free_bootmem_late

On Mon, 9 Nov 2009 22:21:48 +0200
Pekka Enberg <penberg@...helsinki.fi> wrote:

> On Mon, Nov 9, 2009 at 10:13 PM, Pekka Enberg <penberg@...helsinki.fi> wrote:
> >>        for (cursor = start; cursor < end; cursor += BITS_PER_LONG) {
> >> -               unsigned long idx, vec;
> >> +               unsigned long vec;
> >>
> >> -               idx = cursor - start;
> >> -               vec = ~map[idx / BITS_PER_LONG];
> >> +               if (map) {
> >> +                       unsigned long idx = cursor - start;
> >> +                       vec = ~map[idx / BITS_PER_LONG];
> >> +               } else {
> >> +                       if (end - cursor >= BITS_PER_LONG)
> >> +                               vec = ~0UL;
> >
> > Why do we need the above?
> 
> OK, I figured that out. I'm not sure why you want to play tricks with
> "vec" when you could just add a new helper that calls
> __free_pages_bootmem() for the full contiguous page range.

I'm not sure why Chris did that, but yeah, I think that it's cleaner
to leave free_all_bootmem_core alone.

There is just one user of free_bootmem_late to free some memory so how
about the following simple patch instead 7/10?

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index dd97fb8..b10ec49 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -53,6 +53,7 @@ extern void free_bootmem_node(pg_data_t *pgdat,
 			      unsigned long addr,
 			      unsigned long size);
 extern void free_bootmem(unsigned long addr, unsigned long size);
+extern void free_bootmem_late(unsigned long addr, unsigned long size);
 
 /*
  * Flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE,
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 94ef2e7..b0b393b 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -196,6 +196,30 @@ static unsigned long __init free_bootmem_pages(unsigned long start,
 	return count;
 }
 
+/*
+ * free_bootmem_late - free bootmem pages directly to page allocator
+ * @addr: starting address of the range
+ * @size: size of the range in bytes
+ *
+ * This is only useful when the bootmem allocator has already been torn
+ * down, but we are still initializing the system.  Pages are given directly
+ * to the page allocator, no bootmem metadata is updated because it is gone.
+ */
+void __init free_bootmem_late(unsigned long addr, unsigned long size)
+{
+	unsigned long cursor, end;
+
+	kmemleak_free_part(__va(addr), size);
+
+	cursor = PFN_UP(addr);
+	end = PFN_DOWN(addr + size);
+
+	for (; cursor < end; cursor++) {
+		__free_pages_bootmem(pfn_to_page(cursor), 0);
+		totalram_pages++;
+	}
+}
+
 static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 {
 	struct page *page;


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/