mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jiaxun Yang <jiaxun.yang@flygoat.com>
To: Huacai Chen <chenhc@lemote.com>
Cc: "open list:MIPS" <linux-mips@vger.kernel.org>,
	Rob Herring <robh+dt@kernel.org>,
	Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
	Frank Rowand <frowand.list@gmail.com>,
	Paul Burton <paulburton@kernel.org>,
	Nathan Chancellor <natechancellor@gmail.com>,
	Nick Desaulniers <ndesaulniers@google.com>,
	devicetree@vger.kernel.org, LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 2/5] MIPS: Loongson64: Process ISA Node in DeviceTree
Date: Mon, 20 Jul 2020 19:58:54 +0800	[thread overview]
Message-ID: <bc7f1901-5d1a-8358-f597-fb5d60349d11@flygoat.com> (raw)
In-Reply-To: <CAAhV-H7n3Uy-X6ePgTsC6VfV0HbV22SM-BV9Q0i6BS=oxw__yA@mail.gmail.com>



在 2020/7/20 下午7:44, Huacai Chen 写道:
> Hi, Jiaxun,
>
> On Mon, Jul 20, 2020 at 6:20 PM Jiaxun Yang <jiaxun.yang@flygoat.com> wrote:
>>
>>
>> 在 2020/7/20 下午6:01, Huacai Chen 写道:
>>> Hi, Jiaxun,
>>>
>>> On Mon, Jul 20, 2020 at 3:44 PM Jiaxun Yang <jiaxun.yang@flygoat.com> wrote:
>>>> Previously, we're hardcoding resserved ISA I/O Space in code, now
>>>> we're processing reverved I/O via DeviceTree directly. Using the ranges
>>>> property to determine the size and address of reserved I/O space.
>>> Maybe it is better to reserve a default legacy io range if there is no
>>> "isa" node in the .dts file?
>> As currently all dts is built-in in Kernel, I don't think it's necessary.
>>
>> Also the only ISA driver remaining that can't be probed by dts is i8042.
>> We can convert it to DeviceTree based, and then we'll always be safe.
>>
> If you don't reserve a default legacy io range, then I should define
> an "isa" node for KVM guests even if the VM doesn't have i8042 and
> i8259. Because some pci devices still have legacy i/o ports, such as
> qxl. I have tested your patches and found that KVM guests with qxl
> cannot work without an "isa" node.

Oops, I forgot stdvga. piix4_smbus may face the same situation too.

But I'd still think adding a isa node for QEMU virt target can be better.
My universal target is to kill boot_param and reach full-DT boot, thus I 
don't
want to introduce such regression.


Thanks
- Jiaxun

>
> Huacai
>> Thanks.
>>
>> - Jiaxun
>>
>>> Huacai
>>>> Signed-off-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
>>>> ---
>>>>    arch/mips/loongson64/init.c | 85 ++++++++++++++++++++++++++-----------
>>>>    1 file changed, 60 insertions(+), 25 deletions(-)
>>>>
>>>> diff --git a/arch/mips/loongson64/init.c b/arch/mips/loongson64/init.c
>>>> index 59ddadace83f..028d7b324ec2 100644
>>>> --- a/arch/mips/loongson64/init.c
>>>> +++ b/arch/mips/loongson64/init.c
>>>> @@ -7,6 +7,8 @@
>>>>    #include <linux/irqchip.h>
>>>>    #include <linux/logic_pio.h>
>>>>    #include <linux/memblock.h>
>>>> +#include <linux/of.h>
>>>> +#include <linux/of_address.h>
>>>>    #include <asm/bootinfo.h>
>>>>    #include <asm/traps.h>
>>>>    #include <asm/smp-ops.h>
>>>> @@ -63,41 +65,74 @@ void __init prom_free_prom_memory(void)
>>>>    {
>>>>    }
>>>>
>>>> -static __init void reserve_pio_range(void)
>>>> +static int __init add_legacy_isa_io(struct fwnode_handle *fwnode, phys_addr_t addr,
>>>> +                            resource_size_t    size)
>>>>    {
>>>> +       int ret = 0;
>>>>           struct logic_pio_hwaddr *range;
>>>> +       unsigned long vaddr;
>>>>
>>>>           range = kzalloc(sizeof(*range), GFP_ATOMIC);
>>>>           if (!range)
>>>> -               return;
>>>> +               return -ENOMEM;
>>>>
>>>> -       range->fwnode = &of_root->fwnode;
>>>> -       range->size = MMIO_LOWER_RESERVED;
>>>> -       range->hw_start = LOONGSON_PCIIO_BASE;
>>>> +       range->fwnode = fwnode;
>>>> +       range->size = size;
>>>> +       range->hw_start = addr;
>>>>           range->flags = LOGIC_PIO_CPU_MMIO;
>>>>
>>>> -       if (logic_pio_register_range(range)) {
>>>> -               pr_err("Failed to reserve PIO range for legacy ISA\n");
>>>> -               goto free_range;
>>>> +       ret = logic_pio_register_range(range);
>>>> +       if (ret) {
>>>> +               kfree(range);
>>>> +               return ret;
>>>> +       }
>>>> +
>>>> +       /* Legacy ISA must placed at the start of PCI_IOBASE */
>>>> +       if (range->io_start != 0) {
>>>> +               logic_pio_unregister_range(range);
>>>> +               kfree(range);
>>>> +               return -EINVAL;
>>>>           }
>>>>
>>>> -       if (WARN(range->io_start != 0,
>>>> -                       "Reserved PIO range does not start from 0\n"))
>>>> -               goto unregister;
>>>> -
>>>> -       /*
>>>> -        * i8259 would access I/O space, so mapping must be done here.
>>>> -        * Please remove it when all drivers can be managed by logic_pio.
>>>> -        */
>>>> -       ioremap_page_range(PCI_IOBASE, PCI_IOBASE + MMIO_LOWER_RESERVED,
>>>> -                               LOONGSON_PCIIO_BASE,
>>>> -                               pgprot_device(PAGE_KERNEL));
>>>> -
>>>> -       return;
>>>> -unregister:
>>>> -       logic_pio_unregister_range(range);
>>>> -free_range:
>>>> -       kfree(range);
>>>> +       vaddr = PCI_IOBASE + range->io_start;
>>>> +
>>>> +       ioremap_page_range(vaddr, vaddr + size, addr, pgprot_device(PAGE_KERNEL));
>>>> +
>>>> +       return 0;
>>>> +}
>>>> +
>>>> +static __init void reserve_pio_range(void)
>>>> +{
>>>> +       struct device_node *np;
>>>> +
>>>> +       for_each_node_by_name(np, "isa") {
>>>> +               struct of_pci_range range;
>>>> +               struct of_pci_range_parser parser;
>>>> +
>>>> +               pr_info("ISA Bridge: %pOF\n", np);
>>>> +
>>>> +               if (of_pci_range_parser_init(&parser, np)) {
>>>> +                       pr_info("Failed to parse resources.\n");
>>>> +                       break;
>>>> +               }
>>>> +
>>>> +               for_each_of_pci_range(&parser, &range) {
>>>> +                       switch (range.flags & IORESOURCE_TYPE_BITS) {
>>>> +                       case IORESOURCE_IO:
>>>> +                               pr_info(" IO 0x%016llx..0x%016llx\n",
>>>> +                                       range.cpu_addr,
>>>> +                                       range.cpu_addr + range.size - 1);
>>>> +                               if (add_legacy_isa_io(&np->fwnode, range.cpu_addr, range.size))
>>>> +                                       pr_warn("Failed to reserve legacy IO in Logic PIO\n");
>>>> +                               break;
>>>> +                       case IORESOURCE_MEM:
>>>> +                               pr_info(" MEM 0x%016llx..0x%016llx\n",
>>>> +                                       range.cpu_addr,
>>>> +                                       range.cpu_addr + range.size - 1);
>>>> +                               break;
>>>> +                       }
>>>> +               }
>>>> +       }
>>>>    }
>>>>
>>>>    void __init arch_init_irq(void)
>>>> --
>>>> 2.28.0.rc1
>>>>

  reply	other threads:[~2020-07-20 11:59 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-20  7:42 [PATCH 0/5] " Jiaxun Yang
2020-07-20  7:42 ` [PATCH 1/5] of_address: Add bus type match for pci ranges parser Jiaxun Yang
2020-07-21  0:21   ` Rob Herring
2020-07-20  7:42 ` [PATCH 2/5] MIPS: Loongson64: Process ISA Node in DeviceTree Jiaxun Yang
2020-07-20 10:01   ` Huacai Chen
2020-07-20 10:18     ` Jiaxun Yang
2020-07-20 11:44       ` Huacai Chen
2020-07-20 11:58         ` Jiaxun Yang [this message]
2020-07-21  0:27   ` Rob Herring
2020-07-20  7:42 ` [PATCH 3/5] MIPS: Loongson64: Enlarge IO_SPACE_LIMIT Jiaxun Yang
2020-07-20 10:03   ` Huacai Chen
2020-07-20 10:45   ` Arnd Bergmann
2020-07-20 12:04     ` Jiaxun Yang
2020-07-20 12:20       ` Arnd Bergmann
2020-07-20  7:42 ` [PATCH 4/5] MIPS: Loongson64: DTS: Fix ISA range for RS780E PCH Jiaxun Yang
2020-07-20  7:42 ` [PATCH 5/5] MIPS: Loongson64: Add ISA node for LS7A PCH Jiaxun Yang
2020-07-20 10:04   ` Huacai Chen

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=bc7f1901-5d1a-8358-f597-fb5d60349d11@flygoat.com \
    --to=jiaxun.yang@flygoat.com \
    --cc=chenhc@lemote.com \
    --cc=devicetree@vger.kernel.org \
    --cc=frowand.list@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=natechancellor@gmail.com \
    --cc=ndesaulniers@google.com \
    --cc=paulburton@kernel.org \
    --cc=robh+dt@kernel.org \
    --cc=tsbogend@alpha.franken.de \
    /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

all inboxes | Powered by JetHome®