linux-kernel - Re: [PATCH RESEND] wifi: mt76: mt7921e: Set memory space enable in PCI

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <cdf612bf-96f6-9b4e-a32c-50007892083c@amd.com>
Date:   Thu, 30 Mar 2023 09:13:37 -0500
From:   "Limonciello, Mario" <mario.limonciello@....com>
To:     Sean Wang <sean.wang@...nel.org>
Cc:     nbd@....name, lorenzo@...nel.org, ryder.lee@...iatek.com,
        shayne.chen@...iatek.com, sean.wang@...iatek.com,
        Matthias Brugger <matthias.bgg@...il.com>,
        Anson Tsao <anson.tsao@....com>, Kalle Valo <kvalo@...nel.org>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        linux-wireless@...r.kernel.org, netdev@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org,
        linux-mediatek@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH RESEND] wifi: mt76: mt7921e: Set memory space enable in
 PCI_COMMAND if unset

On 3/29/2023 18:24, Sean Wang wrote:
> Hi,
> 
> On Wed, Mar 29, 2023 at 1:18 PM Mario Limonciello
> <mario.limonciello@....com> wrote:
>>
>> When the BIOS has been configured for Fast Boot, systems with mt7921e
>> have non-functional wifi.  Turning on Fast boot caused both bus master
>> enable and memory space enable bits in PCI_COMMAND not to get configured.
>>
>> The mt7921 driver already sets bus master enable, but explicitly check
>> and set memory access enable as well to fix this problem.
>>
>> Tested-by: Anson Tsao <anson.tsao@....com>
>> Signed-off-by: Mario Limonciello <mario.limonciello@....com>
>> ---
>> Original patch was submitted ~3 weeks ago with no comments.
>> Link: https://lore.kernel.org/all/20230310170002.200-1-mario.limonciello@amd.com/
>> ---
>>   drivers/net/wireless/mediatek/mt76/mt7921/pci.c | 6 ++++++
>>   1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
>> index cb72ded37256..aa1a427b16c2 100644
>> --- a/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
>> +++ b/drivers/net/wireless/mediatek/mt76/mt7921/pci.c
>> @@ -263,6 +263,7 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
>>          struct mt76_dev *mdev;
>>          u8 features;
>>          int ret;
>> +       u16 cmd;
>>
>>          ret = pcim_enable_device(pdev);
>>          if (ret)
>> @@ -272,6 +273,11 @@ static int mt7921_pci_probe(struct pci_dev *pdev,
>>          if (ret)
>>                  return ret;
>>
>> +       pci_read_config_word(pdev, PCI_COMMAND, &cmd);
>> +       if (!(cmd & PCI_COMMAND_MEMORY)) {
>> +               cmd |= PCI_COMMAND_MEMORY;
>> +               pci_write_config_word(pdev, PCI_COMMAND, cmd);
>> +       }
> 
> If PCI_COMMAND_MEMORY is required in any circumstance, then we don't
> need to add a conditional check and OR it with PCI_COMMAND_MEMORY.

Generally it seemed advantageous to avoid an extra PCI write if it's not 
needed.  For example that's how bus mastering works too (see 
__pci_set_master).


> Also, I will try the patch on another Intel machine to see if it worked.

Thanks.

> 
>       Sean
> 
>>          pci_set_master(pdev);
>>
>>          ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES);
>> --
>> 2.34.1
>>