My notes on this patch:
* It applies and compiles fine with current qemu on Linux/x86_64 host.
However after building I had to do:
cp pc-bios/optionrom/linuxboot_dma.bin pc-bios/
I wasn't sure if that was supposed to be done automatically.
* It reduces libguestfs appliance boot time by
On Mon, Feb 22, 2016 at 02:56:27PM +0300, Vladimir Sementsov-Ogievskiy wrote:
> Hi, Marc!
>
> Am I right, that this patch is necessary to reproduce your results from
> "QEMU fw_cfg DMA interface"?
>
> >QEMU commit 2be4f242b50a8 and SeaBIOS commit 908a58c1d5ff
> >QEMU startup time: .078
> >BIOS
Hi, Marc!
Am I right, that this patch is necessary to reproduce your results from
"QEMU fw_cfg DMA interface"?
QEMU commit 2be4f242b50a8 and SeaBIOS commit 908a58c1d5ff
QEMU startup time: .078
BIOS startup time: .060
Kernel setup time: .578
Total time: .716
QEMU with this patch series and
On Tue, Feb 02, 2016 at 01:58:14PM +0100, Marc Marí wrote:
> El Tue, 02 Feb 2016 12:06:27 +0100
> Gerd Hoffmann escribió:
> > Hi,
> >
> > > %.img: %.o
> > > - $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -Ttext 0 -e
> > > _start -s -o $@ $<," Building $(TARGET_DIR)$@")
> >
Hi,
> I think the "dma_enabled" property is not exposed to the user.
It is: "-global fw_cfg.dma_enabled=off" works (as in: doesn't throw an
error). Has no effect through as it gets overridden later on.
> The default value of "dma_enabled" in both fw_cfg_io_properties and
>
On 02/03/16 09:44, Gerd Hoffmann wrote:
> Hi,
>
>> I think the "dma_enabled" property is not exposed to the user.
>
> It is: "-global fw_cfg.dma_enabled=off" works (as in: doesn't throw an
> error). Has no effect through as it gets overridden later on.
>
>> The default value of "dma_enabled"
Hi,
> >> I agree that DMA capability should be filtered with machine type.
> >> However, that distinction should not be made using the current
> >> "dma_enabled" properties (i.e., of "fw_cfg_io_properties" and
> >> "fw_cfg_mem_properties". Instead, it should be made in the
> >> board-specific
El Tue, 02 Feb 2016 13:04:55 +0100
Gerd Hoffmann escribió:
> On Di, 2016-02-02 at 12:06 +0100, Gerd Hoffmann wrote:
> > Hi,
> >
> > > %.img: %.o
> > > - $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -Ttext 0 -e
> > > _start -s -o $@ $<," Building $(TARGET_DIR)$@")
> > > +
El Tue, 02 Feb 2016 12:06:27 +0100
Gerd Hoffmann escribió:
> Hi,
>
> > %.img: %.o
> > - $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -Ttext 0 -e
> > _start -s -o $@ $<," Building $(TARGET_DIR)$@")
> > + $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -m elf_i386
> > -Ttext
Hi,
> I don't remember discussing the topic of machine types when touching
> fw_cfg DMA. Which means, it probably slipped amongst the other details.
> But it is now merged and in stable, so it should probably be left as it
> is now.
We have to fix it, it breaks live migration. With fw_cfg_dma
On 02/02/16 15:23, Gerd Hoffmann wrote:
> Hi,
>
>> I don't remember discussing the topic of machine types when touching
>> fw_cfg DMA. Which means, it probably slipped amongst the other details.
>> But it is now merged and in stable, so it should probably be left as it
>> is now.
>
> We have
Hi,
> %.img: %.o
> - $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -Ttext 0 -e _start -s -o $@
> $<," Building $(TARGET_DIR)$@")
> + $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -m elf_i386 -Ttext 0 -e
> _start -s -o $@ $<," Building $(TARGET_DIR)$@")
Hmm, that breaks the windows
On Di, 2016-02-02 at 12:06 +0100, Gerd Hoffmann wrote:
> Hi,
>
> > %.img: %.o
> > - $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -Ttext 0 -e _start -s -o $@
> > $<," Building $(TARGET_DIR)$@")
> > + $(call quiet-command,$(LD) $(LDFLAGS_NOPIE) -m elf_i386 -Ttext 0 -e
> > _start -s -o $@
This optionrom is based on linuxboot.S.
Added changes proposed by Gerd Hoffman, Stefan Hajnoczi and Kevin O'Connor.
All optionroms are now compiled in 32 bits. This also forces to not use any
standard C header because this would need cross-compiling support check and a
big modification on the
On Fri, Jan 29, 2016 at 12:12:24PM +0100, Marc Marí wrote:
> This optionrom is based on linuxboot.S.
>
> Added changes proposed by Gerd Hoffman, Stefan Hajnoczi and Kevin O'Connor.
>
> All optionroms are now compiled in 32 bits. This also forces to not use any
> standard C header because this
15 matches
Mail list logo