All of lore.kernel.org
 help / color / mirror / Atom feed
From: Juergen Gross <jgross@suse.com>
To: minios-devel@lists.xenproject.org, xen-devel@lists.xenproject.org
Cc: samuel.thibault@ens-lyon.org, wl@xen.org,
	Juergen Gross <jgross@suse.com>
Subject: [PATCH v3 03/10] mini-os: don't assume contiguous RAM when initializing in PVH mode
Date: Tue, 21 Dec 2021 09:46:29 +0100	[thread overview]
Message-ID: <20211221084636.6987-4-jgross@suse.com> (raw)
In-Reply-To: <20211221084636.6987-1-jgross@suse.com>

Sizing the available memory should respect memory holes, so look at
the memory map when setting the boundary for the memory allocator.

Signed-off-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
---
V2:
- rename "max" to "start" (Samuel Thibault)
---
 arch/x86/mm.c  |  6 +-----
 e820.c         | 14 ++++++++------
 include/e820.h |  2 +-
 3 files changed, 10 insertions(+), 12 deletions(-)

diff --git a/arch/x86/mm.c b/arch/x86/mm.c
index 8df93da..3bf6170 100644
--- a/arch/x86/mm.c
+++ b/arch/x86/mm.c
@@ -107,7 +107,6 @@ void arch_mm_preinit(void *p)
 {
     long ret;
     domid_t domid = DOMID_SELF;
-    unsigned long max;
 
     pt_base = page_table_base;
     first_free_pfn = PFN_UP(to_phys(&_end));
@@ -117,11 +116,8 @@ void arch_mm_preinit(void *p)
         xprintk("could not get memory size\n");
         do_exit();
     }
-    last_free_pfn = ret;
 
-    max = e820_get_maxpfn();
-    if ( max < last_free_pfn )
-        last_free_pfn = max;
+    last_free_pfn = e820_get_maxpfn(ret);
 }
 #endif
 
diff --git a/e820.c b/e820.c
index 70286cb..8030f43 100644
--- a/e820.c
+++ b/e820.c
@@ -285,10 +285,10 @@ void arch_print_memmap(void)
 }
 #endif
 
-unsigned long e820_get_maxpfn(void)
+unsigned long e820_get_maxpfn(unsigned long pages)
 {
     int i;
-    unsigned long pfn, max = 0;
+    unsigned long pfns, start = 0;
 
     e820_get_memmap();
 
@@ -296,10 +296,12 @@ unsigned long e820_get_maxpfn(void)
     {
         if ( e820_map[i].type != E820_RAM )
             continue;
-        pfn = (e820_map[i].addr + e820_map[i].size) >> PAGE_SHIFT;
-        if ( pfn > max )
-            max = pfn;
+        pfns = e820_map[i].size >> PAGE_SHIFT;
+        start = e820_map[i].addr >> PAGE_SHIFT;
+        if ( pages <= pfns )
+            return start + pages;
+        pages -= pfns;
     }
 
-    return max;
+    return start + pfns;
 }
diff --git a/include/e820.h b/include/e820.h
index af2129f..6a57f05 100644
--- a/include/e820.h
+++ b/include/e820.h
@@ -49,6 +49,6 @@ struct __packed e820entry {
 extern struct e820entry e820_map[];
 extern unsigned e820_entries;
 
-unsigned long e820_get_maxpfn(void);
+unsigned long e820_get_maxpfn(unsigned long pages);
 
 #endif /*__E820_HEADER*/
-- 
2.26.2



  parent reply	other threads:[~2021-12-21  8:47 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-21  8:46 [PATCH v3 00/10] mini-os: add missing PVH features Juergen Gross
2021-12-21  8:46 ` [PATCH v3 01/10] mini-os: split e820 map handling into new source file Juergen Gross
2021-12-21  8:46 ` [PATCH v3 02/10] mini-os: sort and sanitize e820 memory map Juergen Gross
2021-12-21  8:46 ` Juergen Gross [this message]
2021-12-21  8:46 ` [PATCH v3 04/10] mini-os: respect memory map when ballooning up Juergen Gross
2021-12-21  8:46 ` [PATCH v3 05/10] mini-os: don't repeat definition available via header file Juergen Gross
2021-12-21  8:46 ` [PATCH v3 06/10] mini-os: add memory map service functions Juergen Gross
2021-12-21  8:46 ` [PATCH v3 07/10] mini-os: move x86 specific gnttab coding into arch/x86/gnttab.c Juergen Gross
2021-12-21  8:46 ` [PATCH v3 08/10] mini-os: add proper pvh grant table handling Juergen Gross
2021-12-21  8:46 ` [PATCH v3 09/10] mini-os: prepare grantmap entry interface for use by PVH mode Juergen Gross
2021-12-21  8:46 ` [PATCH v3 10/10] mini-os: modify grant mappings to work in " Juergen Gross

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=20211221084636.6987-4-jgross@suse.com \
    --to=jgross@suse.com \
    --cc=minios-devel@lists.xenproject.org \
    --cc=samuel.thibault@ens-lyon.org \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.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.