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>] [day] [month] [year] [list]
Message-ID: <f3739191-1c13-481a-af70-f517f2dd75de@lunn.ch>
Date: Wed, 17 Apr 2024 21:07:44 +0200
From: Andrew Lunn <andrew@...n.ch>
To: SIMON BABY <simonkbaby@...il.com>
Cc: netdev@...r.kernel.org
Subject: Re: Support needed for integrating Marvel 88e6390 with microchip
 SAMA7 processor (DSA)

On Wed, Apr 17, 2024 at 10:24:00AM -0700, SIMON BABY wrote:
> Hello Team,
> 
> I am trying to integrate the Marvel 88e6390 switch with the SAMA7 processor for
> the DSA driver to work correctly.  I can get a raw mdio reply from marvel phy
> using the devmem tool in Linux. 
> But I could not get any reply with the macb driver code (drivers/net/ethernet/
> cadence/macb_main.c). 
> 
> 
> Below are some of my logs using devmem:
> 
> root@...a7g5ek-sd:~# devmem2 0xe2800034 w 0x58029800                 (C22 code)
> /dev/mem opened.
> Memory mapped at address 0xb6f86000.
> Read at address  0xE2800034 (0xb6f86034): 0x6396C1E1
> Write at address 0xE2800034 (0xb6f86034): 0x58029800, readback 0x58029800
> root@...a7g5ek-sd:~# devmem2 0xe2800034 w 0x68060000                (C22 data)
> /dev/mem opened.
> Memory mapped at address 0xb6f2a000.
> Read at address  0xE2800034 (0xb6f2a034): 0x58029800
> Write at address 0xE2800034 (0xb6f2a034): 0x68060000, readback 0x68060000
> root@...a7g5ek-sd:~# devmem2 0xe2800034         (read)
> /dev/mem opened.
> Memory mapped at address 0xb6f04000.
> Read at address  0xE2800034 (0xb6f04034): 0x68071E07     (got the correct
> marvel ID)

I don't know what you are seeing here, but the 88E6390 has a product
ID of 0x3900 in register 3.

> Below are the extra debugs executed during my testing(I added these debugs in
> my macb driver code):
> 
> macb: ============================>debug_build: inside macb_mdio_read mii_id=0
> regnum=0x3 bus_id=e2800000.ethernet-ffffffff pdev_id=ffffffff
> macb: ============================>debug_build: inside macb_mdio_read
> macb_write C22 mii_id=0 regnum=0x3
> macb: ============================>debug_build: inside macb_mdio_read status
> after macb_mdio_wait_for_idle = 0
> macb: ============================>debug_build: inside macb_mdio_read status
> after macb_readl = 65535

I would check your reset line. The switch does not respond when held
in reset. Because the MDIO data line has a pull up, if the device does
not respond you read all 1s.

> Attached is the sama7 technical document, marvel document

Marvell documents are under NDA. You should not be publishing them.

	Andrew

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ