From: Juan Quintela <quintela@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: chegu_vinod@hp.com, qemu-devel@nongnu.org,
Alexander Graf <agraf@suse.de>
Subject: Re: [Qemu-devel] [PATCH 36/39] memory: move bitmap synchronization to its own function
Date: Wed, 06 Nov 2013 17:22:12 +0100 [thread overview]
Message-ID: <87ppqdxzyj.fsf@elfo.elfo> (raw)
In-Reply-To: <527A66BA.7030906@redhat.com> (Paolo Bonzini's message of "Wed, 06 Nov 2013 16:56:42 +0100")
Paolo Bonzini <pbonzini@redhat.com> wrote:
> Il 06/11/2013 14:04, Juan Quintela ha scritto:
>> We want to have all the functions that handle directly the dirty
>> bitmap near. We will change it later.
>
> Please move it to exec.c instead.
>
>> Signed-off-by: Juan Quintela <quintela@redhat.com>
>> ---
>> include/exec/memory-physical.h | 31 +++++++++++++++++++++++++++++++
>> kvm-all.c | 27 ++-------------------------
>> 2 files changed, 33 insertions(+), 25 deletions(-)
>>
>> diff --git a/include/exec/memory-physical.h b/include/exec/memory-physical.h
>> index 610c55f..72faf06 100644
>> --- a/include/exec/memory-physical.h
>> +++ b/include/exec/memory-physical.h
>> @@ -72,6 +72,37 @@ static inline void cpu_physical_memory_set_dirty_range(ram_addr_t start,
>> xen_modified_memory(start, length);
>> }
>>
>> +static inline void cpu_physical_memory_set_dirty_lebitmap(unsigned long *bitmap,
>
> Why "le"?
little endian, name is already too long (cpu_physical_memory_ is too
big as a preffix.)
>
> Paolo
>
>> + ram_addr_t start,
>> + ram_addr_t pages)
>> +{
>> + unsigned int i, j;
>> + unsigned long page_number, c;
>> + hwaddr addr;
>> + ram_addr_t ram_addr;
>> + unsigned int len = (pages + HOST_LONG_BITS - 1) / HOST_LONG_BITS;
>> + unsigned long hpratio = getpagesize() / TARGET_PAGE_SIZE;
>> +
>> + /*
>> + * bitmap-traveling is faster than memory-traveling (for addr...)
>> + * especially when most of the memory is not dirty.
>> + */
>> + for (i = 0; i < len; i++) {
>> + if (bitmap[i] != 0) {
>> + c = leul_to_cpu(bitmap[i]);
this is the important part. KVM maintains the bitmap is Little Endian
mode, and PPC? I assme needs this.
>> + do {
>> + j = ffsl(c) - 1;
>> + c &= ~(1ul << j);
>> + page_number = (i * HOST_LONG_BITS + j) * hpratio;
>> + addr = page_number * TARGET_PAGE_SIZE;
>> + ram_addr = start + addr;
>> + cpu_physical_memory_set_dirty_range(ram_addr,
>> + TARGET_PAGE_SIZE
>> * hpratio);
I checked, and it appears that ppc really use the hpratio thing. I
hope that the bitmap optimization also works for ppc, but its
architecture is weird^Winteresting O:-)
Alex?
next prev parent reply other threads:[~2013-11-06 16:22 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-06 13:04 [Qemu-devel] [PATCH v2 00/39] bitmap handling optimization Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 01/39] Move prototypes to memory.h Juan Quintela
2013-11-07 9:36 ` Orit Wasserman
2013-11-06 13:04 ` [Qemu-devel] [PATCH 02/39] memory: cpu_physical_memory_set_dirty_flags() result is never used Juan Quintela
2013-11-07 9:42 ` Orit Wasserman
2013-11-06 13:04 ` [Qemu-devel] [PATCH 03/39] memory: cpu_physical_memory_set_dirty_range() return void Juan Quintela
2013-11-07 9:51 ` Orit Wasserman
2013-11-06 13:04 ` [Qemu-devel] [PATCH 04/39] exec: use accessor function to know if memory is dirty Juan Quintela
2013-11-07 10:00 ` Orit Wasserman
2013-11-06 13:04 ` [Qemu-devel] [PATCH 05/39] memory: create function to set a single dirty bit Juan Quintela
2013-11-07 11:37 ` Orit Wasserman
2013-11-06 13:04 ` [Qemu-devel] [PATCH 06/39] exec: create function to get " Juan Quintela
2013-11-06 23:11 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 07/39] memory: make cpu_physical_memory_is_dirty return bool Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 08/39] exec: simplify notdirty_mem_write() Juan Quintela
2013-11-06 23:12 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 09/39] memory: all users of cpu_physical_memory_get_dirty used only one flag Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 10/39] memory: set single dirty flags when possible Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 11/39] memory: cpu_physical_memory_set_dirty_range() allways dirty all flags Juan Quintela
2013-11-06 23:15 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 12/39] memory: cpu_physical_memory_mask_dirty_range() always clear a single flag Juan Quintela
2013-11-06 23:18 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 13/39] memory: use DIRTY_MEMORY_* instead of *_DIRTY_FLAG Juan Quintela
2013-11-06 23:24 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 14/39] memory: use bit 2 for migration Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 15/39] memory: make sure that client is always inside range Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 16/39] memory: only resize dirty bitmap when memory size increases Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 17/39] memory: cpu_physical_memory_clear_dirty_flag() result is never used Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 18/39] bitmap: Add bitmap_zero_extend operation Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 19/39] memory: split dirty bitmap into three Juan Quintela
2013-11-06 23:39 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 20/39] memory: unfold cpu_physical_memory_clear_dirty_flag() in its only user Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 21/39] memory: unfold cpu_physical_memory_set_dirty() " Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 22/39] memory: unfold cpu_physical_memory_set_dirty_flag() Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 23/39] memory: make cpu_physical_memory_get_dirty() the main function Juan Quintela
2013-11-06 23:44 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 24/39] memory: cpu_physical_memory_get_dirty() is used as returning a bool Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 25/39] memory: s/mask/clear/ cpu_physical_memory_mask_dirty_range Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 26/39] memory: use find_next_bit() to find dirty bits Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 27/39] memory: cpu_physical_memory_set_dirty_range() now uses bitmap operations Juan Quintela
2013-11-06 23:49 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 28/39] memory: cpu_physical_memory_clear_dirty_range() " Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 29/39] memory: s/dirty/clean/ in cpu_physical_memory_is_dirty() Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 30/39] memory: make cpu_physical_memory_reset_dirty() take a length parameter Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 31/39] memory: cpu_physical_memory_set_dirty_tracking() should return void Juan Quintela
2013-11-06 23:55 ` Eric Blake
2013-11-06 13:04 ` [Qemu-devel] [PATCH 32/39] memory: split cpu_physical_memory_* functions to its own include Juan Quintela
2013-11-06 15:54 ` Paolo Bonzini
2013-11-06 13:04 ` [Qemu-devel] [PATCH 33/39] memory: unfold memory_region_test_and_clear() Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 34/39] kvm: use directly cpu_physical_memory_* api for tracking dirty pages Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 35/39] kvm: refactor start address calculation Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 36/39] memory: move bitmap synchronization to its own function Juan Quintela
2013-11-06 15:56 ` Paolo Bonzini
2013-11-06 16:22 ` Juan Quintela [this message]
2013-11-06 16:23 ` Paolo Bonzini
2013-12-18 8:54 ` Alexander Graf
2013-11-06 13:04 ` [Qemu-devel] [PATCH 37/39] memory: syncronize kvm bitmap using bitmaps operations Juan Quintela
2013-11-06 15:58 ` Paolo Bonzini
2013-11-06 13:04 ` [Qemu-devel] [PATCH 38/39] ram: split function that synchronizes a range Juan Quintela
2013-11-06 13:04 ` [Qemu-devel] [PATCH 39/39] migration: synchronize memory bitmap 64bits at a time Juan Quintela
2013-11-06 14:37 ` [Qemu-devel] [PATCH v2 00/39] bitmap handling optimization Gerd Hoffmann
2013-11-06 15:38 ` Juan Quintela
2013-11-07 11:23 ` Gerd Hoffmann
2013-11-06 15:49 ` Paolo Bonzini
2013-11-25 6:39 ` Michael R. Hines
2013-11-25 9:45 ` Paolo Bonzini
2013-11-08 15:18 ` Chegu Vinod
2013-11-25 6:15 ` Michael R. Hines
2013-11-25 6:58 ` Michael R. Hines
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=87ppqdxzyj.fsf@elfo.elfo \
--to=quintela@redhat.com \
--cc=agraf@suse.de \
--cc=chegu_vinod@hp.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.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 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).