Date:   Wed, 26 Oct 2016 12:31:24 +0200
From:   Paul Menzel <pmenzel@...gen.mpg.de>
To:     Sony Chacko <sony.chacko@...gic.com>,
        Dept-HSGLinuxNICDev@...gic.com, Baoquan He <bhe@...hat.com>
Cc:     netdev@...r.kernel.org, "David S. Miller" <davem@...emloft.net>
Subject: Re: [bnx2] [Regression 4.8] Driver loading fails without firmware

Dear Baoquan, Linux folks,


I am sorry for Thunderbird’s autowrapping messing up the message formatting.


On 10/25/16 16:33, Paul Menzel wrote:

> A server with the Broadcom devices below, fails to load the drivers
> because of missing firmware.
>
>> $ lspci -nn
>> […]
>> 01:00.0 Ethernet controller [0200]: Broadcom Corporation NetXtreme II
>> BCM5709 Gigabit Ethernet [14e4:1639] (rev 20)
>> 01:00.1 Ethernet controller [0200]: Broadcom Corporation NetXtreme II
>> BCM5709 Gigabit Ethernet [14e4:1639] (rev 20)
>> 02:00.0 Ethernet controller [0200]: Broadcom Corporation NetXtreme II
>> BCM5709 Gigabit Ethernet [14e4:1639] (rev 20)
>> 02:00.1 Ethernet controller [0200]: Broadcom Corporation NetXtreme II
>> BCM5709 Gigabit Ethernet [14e4:1639] (rev 20)
>> […]
>
> Here are the error messages from Linux 4.8.4.
>
>> [   10.365839] bnx2: QLogic bnx2 Gigabit Ethernet Driver v2.2.6
>> (January 29, 2014)
>> [   10.366989] bnx2 0000:01:00.0: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.367384] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.367719] bnx2: probe of 0000:01:00.0 failed with error -2
>> [   10.368616] bnx2 0000:01:00.1: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.369015] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.369292] bnx2: probe of 0000:01:00.1 failed with error -2
>> [   10.370194] bnx2 0000:02:00.0: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.374460] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.374743] bnx2: probe of 0000:02:00.0 failed with error -2
>> [   10.375619] bnx2 0000:02:00.1: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.376024] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.376303] bnx2: probe of 0000:02:00.1 failed with error -2
>> [   10.376589] bnx2x: QLogic 5771x/578xx 10/20-Gigabit Ethernet Driver
>> bnx2x 1.712.30-0 (2014/02/10)
>> [   10.509323] bnx2: QLogic bnx2 Gigabit Ethernet Driver v2.2.6
>> (January 29, 2014)
>> [   10.510467] bnx2 0000:01:00.0: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.510855] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.511188] bnx2: probe of 0000:01:00.0 failed with error -2
>> [   10.512087] bnx2 0000:01:00.1: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.512492] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.512773] bnx2: probe of 0000:01:00.1 failed with error -2
>> [   10.513648] bnx2 0000:02:00.0: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.517861] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.518182] bnx2: probe of 0000:02:00.0 failed with error -2
>> [   10.519071] bnx2 0000:02:00.1: Direct firmware load for
>> bnx2/bnx2-mips-09-6.2.1b.fw failed with error -2
>> [   10.519466] bnx2: Can't load firmware file
>> "bnx2/bnx2-mips-09-6.2.1b.fw"
>> [   10.519740] bnx2: probe of 0000:02:00.1 failed with error -2
>> [   10.520039] bnx2x: QLogic 5771x/578xx 10/20-Gigabit Ethernet Driver
>> bnx2x 1.712.30-0 (2014/02/10)
>
> The system boots fine with the *same* initramfs and Linux 4.4.27, and
> Linux 4.7.10.
>
>> $ dmesg | grep bnx # Linux 4.7.10
>> [    9.875426] bnx2: QLogic bnx2 Gigabit Ethernet Driver v2.2.6
>> (January 29, 2014)
>> [    9.876766] bnx2 0000:01:00.0 eth0: Broadcom NetXtreme II BCM5709
>> 1000Base-T (C0) PCI Express found at mem e6000000, IRQ 43, node addr
>> 14:fe:b5:c6:81:01
>> [    9.878065] bnx2 0000:01:00.1 eth1: Broadcom NetXtreme II BCM5709
>> 1000Base-T (C0) PCI Express found at mem e8000000, IRQ 44, node addr
>> 14:fe:b5:c6:81:03
>> [    9.879357] bnx2 0000:02:00.0 eth2: Broadcom NetXtreme II BCM5709
>> 1000Base-T (C0) PCI Express found at mem ea000000, IRQ 45, node addr
>> 14:fe:b5:c6:81:05
>> [    9.880630] bnx2 0000:02:00.1 eth3: Broadcom NetXtreme II BCM5709
>> 1000Base-T (C0) PCI Express found at mem ec000000, IRQ 46, node addr
>> 14:fe:b5:c6:81:07
>> [    9.881129] bnx2x: QLogic 5771x/578xx 10/20-Gigabit Ethernet Driver
>> bnx2x 1.712.30-0 (2014/02/10)
>> [   27.692521] bnx2 0000:01:00.0 net00: renamed from eth0
>> [   27.702703] bnx2 0000:01:00.1 net01: renamed from eth1
>> [   27.719779] bnx2 0000:02:00.0 net02: renamed from eth2
>> [   27.768563] bnx2 0000:02:00.1 net03: renamed from eth3
>> [   28.146396] bnx2 0000:01:00.0 net00: using MSIX
>> [   31.294206] bnx2 0000:01:00.0 net00: NIC Copper Link is Up, 1000
>> Mbps full duplex
>
> There were not many commits between Linux 4.7 and Linux 4.8.
>
> Could the commit below cause this regression?
>
>> $ git log --oneline v4.7...v4.8 -- drivers/net/ethernet/broadcom/bnx2.c
>> 3e1be7a bnx2: Reset device during driver initialization
>> commit 3e1be7ad2d38c6bd6aeef96df9bd0a7822f4e51c
>> Author: Baoquan He <bhe@...hat.com>
>> Date:   Fri Sep 9 22:43:12 2016 +0800
>>
>>     bnx2: Reset device during driver initialization
>>
>>     When system enters into kdump kernel because of kernel panic, it
>> won't
>>     shutdown devices. On-flight DMA will continue transferring data until
>>     device driver initializes. All devices are supposed to reset during
>>     driver initialization. And this property is used to fix the kdump
>>     failure in system with intel iommu. Other systems with hardware iommu
>>     should be similar. Please check commit 091d42e ("iommu/vt-d: Copy
>>     translation tables from old kernel") and those commits around.
>>
>>     But bnx2 driver doesn't reset device during driver initialization.
>> The
>>     device resetting is deferred to net device up stage. This will cause
>>     hardware iommu handling failure on bnx2 device. And its resetting
>> relies
>>     on firmware. So in this patch move the firmware requesting code to
>> earlier
>>     bnx2_init_one(), then next call bnx2_reset_chip to reset device.
>>
>>     Signed-off-by: Baoquan He <bhe@...hat.com>
>>     Signed-off-by: David S. Miller <davem@...emloft.net>
>>
>> diff --git a/drivers/net/ethernet/broadcom/bnx2.c
>> b/drivers/net/ethernet/broadcom/bnx2.c
>> index 8fc3f3c..505ceaf 100644
>> --- a/drivers/net/ethernet/broadcom/bnx2.c
>> +++ b/drivers/net/ethernet/broadcom/bnx2.c
>> @@ -6356,10 +6356,6 @@ bnx2_open(struct net_device *dev)
>>         struct bnx2 *bp = netdev_priv(dev);
>>         int rc;
>>
>> -       rc = bnx2_request_firmware(bp);
>> -       if (rc < 0)
>> -               goto out;
>> -
>>         netif_carrier_off(dev);
>>
>>         bnx2_disable_int(bp);
>> @@ -6428,7 +6424,6 @@ bnx2_open(struct net_device *dev)
>>         bnx2_free_irq(bp);
>>         bnx2_free_mem(bp);
>>         bnx2_del_napi(bp);
>> -       bnx2_release_firmware(bp);
>>         goto out;
>>  }
>>
>> @@ -8575,6 +8570,12 @@ bnx2_init_one(struct pci_dev *pdev, const
>> struct pci_device_id *ent)
>>
>>         pci_set_drvdata(pdev, dev);
>>
>> +       rc = bnx2_request_firmware(bp);
>> +       if (rc < 0)
>> +               goto error;
>> +
>> +
>> +       bnx2_reset_chip(bp, BNX2_DRV_MSG_CODE_RESET);
>>         memcpy(dev->dev_addr, bp->mac_addr, ETH_ALEN);
>>
>>         dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_SG |
>> @@ -8607,6 +8608,7 @@ bnx2_init_one(struct pci_dev *pdev, const struct
>> pci_device_id *ent)
>>         return 0;
>>
>>  error:
>> +       bnx2_release_firmware(bp);
>>         pci_iounmap(pdev, bp->regview);
>>         pci_release_regions(pdev);
>>         pci_disable_device(pdev);

Baoquan, could you please fix this regression. My suggestion is, that 
you add the old code back, but check if the firmware has been loaded. If 
it hasn’t, load it again.

That way, people can update their Linux kernel, and it continues working 
without changing the initramfs, or anything else.


Kind regards,

Paul

Powered by blists - more mailing lists