From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:52452) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aE6vD-0006Ns-Vz for qemu-devel@nongnu.org; Tue, 29 Dec 2015 21:55:33 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aE6vA-0004m3-Pg for qemu-devel@nongnu.org; Tue, 29 Dec 2015 21:55:31 -0500 Received: from [59.151.112.132] (port=35251 helo=heian.cn.fujitsu.com) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aE6vA-0004k1-D8 for qemu-devel@nongnu.org; Tue, 29 Dec 2015 21:55:28 -0500 References: <1451277767-9325-1-git-send-email-caoj.fnst@cn.fujitsu.com> <5682747E.8080301@cn.fujitsu.com> <568277CD.7040004@cn.fujitsu.com> <5682B402.7060808@redhat.com> From: Cao jin Message-ID: <56834830.6050803@cn.fujitsu.com> Date: Wed, 30 Dec 2015 10:57:52 +0800 MIME-Version: 1.0 In-Reply-To: <5682B402.7060808@redhat.com> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v2] bugfix: passing reference instead of value List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini , qemu-devel@nongnu.org Cc: stefano.stabellini@eu.citrix.com, mst@redhat.com Hi Paolo On 12/30/2015 12:25 AM, Paolo Bonzini wrote: >>>> Separated from previous "igd-passthru convert to realize" patch. Since >>>> these two don`t have dependency, can send it solely. >>>> >>>> Not test since it is easy to find out if reading carefully, just >>>> compiled. > > This patch works but the added conversion to LE is conceptually wrong. > The conversion from LE to CPU and vice versa is already done by > host_pci_config_read and pci_default_write_config. You don't have it in > host_pci_config_read because the code is not portable and x86 is > little-endian. > Do you mean, this code section(or this igd passthru device)will only run on x86? If it is, yes, the cpu_to_le32() is not necessary. If it has opportunity to run on BE, I think it is necessary, right? The reason I add the conversion is that I am not sure whether it will run on a BE machine or not, and it will no bad to LE machine like x86. > Generally, functions that accept/return integers take care themselves of > endianness conversions. > > Paolo > > >>>> hw/pci-host/piix.c | 10 ++++++---- >>>> 1 file changed, 6 insertions(+), 4 deletions(-) >>>> >>>> diff --git a/hw/pci-host/piix.c b/hw/pci-host/piix.c >>>> index 715208b..a9cb983 100644 >>>> --- a/hw/pci-host/piix.c >>>> +++ b/hw/pci-host/piix.c >>>> @@ -761,7 +761,7 @@ static const IGDHostInfo igd_host_bridge_infos[] = { >>>> {0xa8, 4}, /* SNB: base of GTT stolen memory */ >>>> }; >>>> >>>> -static int host_pci_config_read(int pos, int len, uint32_t val) >>>> +static int host_pci_config_read(int pos, int len, uint32_t *val) >>>> { >>>> char path[PATH_MAX]; >>>> int config_fd; >>>> @@ -784,12 +784,14 @@ static int host_pci_config_read(int pos, int >>>> len, uint32_t val) >>>> ret = -errno; >>>> goto out; >>>> } >>>> + >>>> do { >>>> - rc = read(config_fd, (uint8_t *)&val, len); >>>> + rc = read(config_fd, (uint8_t *)val, len); >>>> } while (rc < 0 && (errno == EINTR || errno == EAGAIN)); >>>> if (rc != len) { >>>> ret = -errno; >>>> } >>>> + >>>> out: >>>> close(config_fd); >>>> return ret; >>>> @@ -805,11 +807,11 @@ static int igd_pt_i440fx_initfn(struct PCIDevice >>>> *pci_dev) >>>> for (i = 0; i < num; i++) { >>>> pos = igd_host_bridge_infos[i].offset; >>>> len = igd_host_bridge_infos[i].len; >>>> - rc = host_pci_config_read(pos, len, val); >>>> + rc = host_pci_config_read(pos, len, &val); >>>> if (rc) { >>>> return -ENODEV; >>>> } >>>> - pci_default_write_config(pci_dev, pos, val, len); >>>> + pci_default_write_config(pci_dev, pos, cpu_to_le32(val), len); >>>> } >>>> >>>> return 0; >>>> >>> >> > > > . > -- Yours Sincerely, Cao Jin