From: Thomas Huth <thuth@redhat.com>
To: "Paolo Bonzini" <pbonzini@redhat.com>,
"Philippe Mathieu-Daudé" <philmd@redhat.com>,
qemu-devel@nongnu.org, "Eduardo Habkost" <ehabkost@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
"Gerd Hoffmann" <kraxel@redhat.com>
Cc: libvir-list@redhat.com
Subject: Re: [PATCH v2 4/4] hw/display: Remove "rombar" hack from vga-pci and vmware_vga
Date: Mon, 9 Dec 2019 14:49:38 +0100 [thread overview]
Message-ID: <d00b48c6-1e07-0fb8-9c68-cecd4c628baf@redhat.com> (raw)
In-Reply-To: <bedf0734-7d00-9e82-56b6-509dff45a946@redhat.com>
On 09/12/2019 14.30, Paolo Bonzini wrote:
> On 09/12/19 14:12, Philippe Mathieu-Daudé wrote:
>> On 12/9/19 1:52 PM, Thomas Huth wrote:
>>> Now that the old pc-0.x machine types have been removed, we do not need
>>> the old "rombar" hacks anymore.
>>>
>>> Signed-off-by: Thomas Huth <thuth@redhat.com>
>>> ---
>>> hw/display/vga-pci.c | 5 -----
>>> hw/display/vga.c | 4 +---
>>> hw/display/vmware_vga.c | 5 -----
>>> 3 files changed, 1 insertion(+), 13 deletions(-)
>>>
>>> diff --git a/hw/display/vga-pci.c b/hw/display/vga-pci.c
>>> index a27b88122d..cfe095713e 100644
>>> --- a/hw/display/vga-pci.c
>>> +++ b/hw/display/vga-pci.c
>>> @@ -264,11 +264,6 @@ static void pci_std_vga_realize(PCIDevice *dev,
>>> Error **errp)
>>> pci_register_bar(&d->dev, 2,
>>> PCI_BASE_ADDRESS_SPACE_MEMORY, &d->mmio);
>>> }
>>> -
>>> - if (!dev->rom_bar) {
>>> - /* compatibility with pc-0.13 and older */
>>> - vga_init_vbe(s, OBJECT(dev), pci_address_space(dev));
>>> - }
>>> }
>>> static void pci_std_vga_init(Object *obj)
>>> diff --git a/hw/display/vga.c b/hw/display/vga.c
>>> index 82ebe53610..636586a476 100644
>>> --- a/hw/display/vga.c
>>> +++ b/hw/display/vga.c
>>> @@ -2304,9 +2304,7 @@ void vga_init(VGACommonState *s, Object *obj,
>>> MemoryRegion *address_space,
>>> void vga_init_vbe(VGACommonState *s, Object *obj, MemoryRegion
>>> *system_memory)
>>> {
>>> - /* With pc-0.12 and below we map both the PCI BAR and the fixed
>>> VBE region,
>>> - * so use an alias to avoid double-mapping the same region.
>>> - */
>>> + /* Use an alias to avoid double-mapping the same region */
>>
>> If I understand the comment correctly, we can now revert commit
>> 8294a64d7f9 and and remove the alias, isn't it?
>
> Yes, even inline vga_init_vbe and remove vbe_mapped.
Sounds like a good idea!
> ----------------- 8< ----------------
> From: Paolo Bonzini <pbonzini@redhat.com>
> Subject: [PATCH] vga: cleanup mapping of VRAM for non-PCI VGA
>
> vga_init_vbe is now used only from ISA VGA cards. Since the alias is
> not needed anymore, remove it (effectively reverting commit 8294a64d7f,
> "vga: fix vram double-mapping with -vga std and -M pc-0.12", 2012-05-29)
> and the new unused vbe_mapped field of VGACommonState. The function now
> consists of a single memory_region_add_subregion call, so we can inline
> it and avoid incorrect usage from PCI cards.
>
> Suggested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
>
> diff --git a/hw/display/vga-isa-mm.c b/hw/display/vga-isa-mm.c
> index e9c43e5530..06f93c4ef5 100644
> --- a/hw/display/vga-isa-mm.c
> +++ b/hw/display/vga-isa-mm.c
> @@ -106,6 +106,9 @@ int isa_vga_mm_init(hwaddr vram_base,
>
> s->vga.con = graphic_console_init(NULL, 0, s->vga.hw_ops, s);
>
> - vga_init_vbe(&s->vga, NULL, address_space);
> + memory_region_add_subregion(system_memory,
Use address_space instead of system_memory ?
> + VBE_DISPI_LFB_PHYSICAL_ADDRESS,
> + &s->vram);
> +
> return 0;
> }
> diff --git a/hw/display/vga-isa.c b/hw/display/vga-isa.c
> index 873e5e9706..5b0b567835 100644
> --- a/hw/display/vga-isa.c
> +++ b/hw/display/vga-isa.c
> @@ -76,7 +76,9 @@ static void vga_isa_realizefn(DeviceState *dev, Error **errp)
> memory_region_set_coalescing(vga_io_memory);
> s->con = graphic_console_init(DEVICE(dev), 0, s->hw_ops, s);
>
> - vga_init_vbe(s, OBJECT(dev), isa_address_space(isadev));
> + memory_region_add_subregion(system_memory,
isa_address_space(isadev) instead of system_memory ?
> + VBE_DISPI_LFB_PHYSICAL_ADDRESS,
> + &s->vram);
> /* ROM BIOS */
> rom_add_vga(VGABIOS_FILENAME);
> }
> diff --git a/hw/display/vga.c b/hw/display/vga.c
> index 636586a476..061fd9ab8f 100644
> --- a/hw/display/vga.c
> +++ b/hw/display/vga.c
> @@ -2301,15 +2301,3 @@ void vga_init(VGACommonState *s, Object *obj, MemoryRegion *address_space,
> portio_list_add(&s->vbe_port_list, address_space_io, 0x1ce);
> }
> }
> -
> -void vga_init_vbe(VGACommonState *s, Object *obj, MemoryRegion *system_memory)
> -{
> - /* Use an alias to avoid double-mapping the same region */
> - memory_region_init_alias(&s->vram_vbe, obj, "vram.vbe",
> - &s->vram, 0, memory_region_size(&s->vram));
> - /* XXX: use optimized standard vga accesses */
> - memory_region_add_subregion(system_memory,
> - VBE_DISPI_LFB_PHYSICAL_ADDRESS,
> - &s->vram_vbe);
> - s->vbe_mapped = 1;
> -}
> diff --git a/hw/display/vga_int.h b/hw/display/vga_int.h
> index 55c418eab5..3081be445d 100644
> --- a/hw/display/vga_int.h
> +++ b/hw/display/vga_int.h
> @@ -106,7 +106,6 @@ typedef struct VGACommonState {
> uint32_t vbe_start_addr;
> uint32_t vbe_line_offset;
> uint32_t vbe_bank_mask;
> - int vbe_mapped;
I think you can now also remove "vram_vbe" from this file.
Thomas
next prev parent reply other threads:[~2019-12-09 13:50 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-09 12:52 [PATCH v2 0/4] Remove deprecated pc-0.x machine types and related hacks Thomas Huth
2019-12-09 12:52 ` [PATCH v2 1/4] hw/i386: Remove the deprecated machines 0.12 up to 0.15 Thomas Huth
2019-12-09 12:52 ` [PATCH v2 2/4] hw/audio: Remove the "use_broken_id" hack from the AC97 device Thomas Huth
2019-12-09 12:52 ` [PATCH v2 3/4] hw/pci: Remove the "command_serr_enable" property Thomas Huth
2019-12-09 12:52 ` [PATCH v2 4/4] hw/display: Remove "rombar" hack from vga-pci and vmware_vga Thomas Huth
2019-12-09 13:12 ` Philippe Mathieu-Daudé
2019-12-09 13:30 ` Paolo Bonzini
2019-12-09 13:49 ` Thomas Huth [this message]
2019-12-09 13:30 ` Gerd Hoffmann
2019-12-09 13:06 ` [PATCH v2 0/4] Remove deprecated pc-0.x machine types and related hacks Paolo Bonzini
2019-12-09 18:18 ` [libvirt] " no-reply
2019-12-09 18:23 ` no-reply
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=d00b48c6-1e07-0fb8-9c68-cecd4c628baf@redhat.com \
--to=thuth@redhat.com \
--cc=ehabkost@redhat.com \
--cc=kraxel@redhat.com \
--cc=libvir-list@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@redhat.com \
--cc=philmd@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).