[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Qemu-devel] [Bug 1321464] Re: qemu/block/qcow2.c:1942: possible perform

From: Max Reitz
Subject: [Qemu-devel] [Bug 1321464] Re: qemu/block/qcow2.c:1942: possible performance problem ?
Date: Tue, 20 May 2014 23:21:05 -0000

I can only speak for qcow2 and qcow, but for those places, I don't think
it is worth fixing. First of all, both are image formats, so the
bottleneck is generally the disk on which the images are stored and not
main memory, so an overeager memset should not cause any problems.

For both, the relevant piece of code is in qcow2/qcow_write_compressed()
which are rarely used anyway (as far as I know) and even if used, they
have additional overhead due to having to compress data first, so
“fixing” the memset() won't make them noticibly faster.

I don't know about the ACPI thing, but to me it seems that it's copying
data to a temporary buffer and then overwriting its beginning with
zeroes. From my very limited ACPI knowledge I'd guess this function is
called at some point during qemu startup, so it doesn't seem worth
optimizing either.

You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.

  qemu/block/qcow2.c:1942: possible performance problem ?

Status in QEMU:

Bug description:
  I just ran static analyser cppcheck over today (20140520) qemu source

  It said many things, including

  [qemu/block/qcow2.c:1942] -> [qemu/block/qcow2.c:1943]: (performance) Buffer 
'pad_buf' is being writ
  ten before its old content has been used.

  Source code is

              memset(pad_buf, 0, s->cluster_size);
              memcpy(pad_buf, buf, nb_sectors * BDRV_SECTOR_SIZE);

  Worth tuning ?

  Similar problem here

  [qemu/block/qcow.c:815] -> [qemu/block/qcow.c:816]: (performance) Buffer 
'pad_buf' is being written 
  before its old content has been used.


  [qemu/hw/i386/acpi-build.c:1265] -> [qemu/hw/i386/acpi-build.c:1267]: 
(performance) Buffer 'dsdt' is
   being written before its old content has been used.

To manage notifications about this bug go to:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]