linux-kernel - Re: [PATCH] nvme: Add a module parameter for users to force simple suspend

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 for Android: free password hash cracker in your pocket

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

Message-ID: <576494fb-c3c0-1072-a461-8a822f079f1d@amd.com>
Date:   Tue, 28 Feb 2023 17:48:28 -0600
From:   Mario Limonciello <mario.limonciello@....com>
To:     Keith Busch <kbusch@...nel.org>
Cc:     Jens Axboe <axboe@...com>, Christoph Hellwig <hch@....de>,
        Sagi Grimberg <sagi@...mberg.me>, elvis.angelaccio@....org,
        linux-nvme@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] nvme: Add a module parameter for users to force simple
 suspend

On 2/28/23 16:34, Keith Busch wrote:
> On Tue, Feb 28, 2023 at 04:11:48PM -0600, Mario Limonciello wrote:
>> +static bool nvme_use_simple_suspend(struct pci_dev *pdev)
>> +{
>> +	if (!simple_suspend)
>> +		return false;
>> +	if (simple_suspend == 1)
>> +		return true;
>> +	return !noacpi && acpi_storage_d3(&pdev->dev);
>> +}
>> +
>>   static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>>   {
>>   	int node, result = -ENOMEM;
>> @@ -3128,11 +3145,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
>>   
>>   	quirks |= check_vendor_combination_bug(pdev);
>>   
>> -	if (!noacpi && acpi_storage_d3(&pdev->dev)) {
>> -		/*
>> -		 * Some systems use a bios work around to ask for D3 on
>> -		 * platforms that support kernel managed suspend.
>> -		 */
>> +	if (nvme_use_simple_suspend(pdev)) {
>>   		dev_info(&pdev->dev,
>>   			 "platform quirk: setting simple suspend\n");
>>   		quirks |= NVME_QUIRK_SIMPLE_SUSPEND;
> 
> Do you want the user setting "never" to override the driver's default quirks?

That hadn't occurred to me, but if offering a 0/1/-1 it certainly makes 
sense.  I'll add something to explicitly clear it if present for a V2.

Another way I've hypothesized that this problem (at least as reported) 
can be approached is to examine if ANY disks in the system have simple 
suspend set to apply a "global" change to all NVME disks.

If that is preferable I'm fine to spin it that way too.