linux-kernel - Re: [PATCH v0] pata_parport: add driver (PARIDE replacement)

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

Date:   Sat, 12 Mar 2022 17:09:54 +0900
From:   Damien Le Moal <damien.lemoal@...nsource.wdc.com>
To:     Ondrej Zary <linux@...y.sk>
Cc:     Christoph Hellwig <hch@....de>, Jens Axboe <axboe@...nel.dk>,
        Tim Waugh <tim@...erelk.net>, linux-block@...r.kernel.org,
        linux-parport@...ts.infradead.org, linux-ide@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v0] pata_parport: add driver (PARIDE replacement)

On 3/12/22 03:55, Ondrej Zary wrote:
> On Friday 11 March 2022 00:59:20 Damien Le Moal wrote:
>> On 3/11/22 06:28, Ondrej Zary wrote:
>>> Add pata_parport (PARIDE replacement) core libata driver.
>>>
>>> The original paride protocol modules are used for now so allow them to
>>> be compiled without old PARIDE core.
>>>
>>> Signed-off-by: Ondrej Zary <linux@...y.sk>
>>> ---
>>>  drivers/Makefile                   |   2 +-
>>>  drivers/ata/Kconfig                |  22 +
>>>  drivers/ata/Makefile               |   2 +
>>>  drivers/ata/parport/Makefile       |   3 +
>>>  drivers/ata/parport/pata_parport.c | 805 +++++++++++++++++++++++++++++
>>>  drivers/ata/parport/pata_parport.h | 108 ++++
>>>  drivers/block/paride/Kconfig       |  32 +-
>>>  drivers/block/paride/paride.h      |   5 +
>>>  8 files changed, 962 insertions(+), 17 deletions(-)
>>>  create mode 100644 drivers/ata/parport/Makefile
>>>  create mode 100644 drivers/ata/parport/pata_parport.c
>>>  create mode 100644 drivers/ata/parport/pata_parport.h
>>>
>>> diff --git a/drivers/Makefile b/drivers/Makefile
>>> index a110338c860c..8ec515f3614e 100644
>>> --- a/drivers/Makefile
>>> +++ b/drivers/Makefile
>>> @@ -98,7 +98,7 @@ obj-$(CONFIG_DIO)		+= dio/
>>>  obj-$(CONFIG_SBUS)		+= sbus/
>>>  obj-$(CONFIG_ZORRO)		+= zorro/
>>>  obj-$(CONFIG_ATA_OVER_ETH)	+= block/aoe/
>>> -obj-$(CONFIG_PARIDE) 		+= block/paride/
>>> +obj-y		 		+= block/paride/
>>>  obj-$(CONFIG_TC)		+= tc/
>>>  obj-$(CONFIG_USB_PHY)		+= usb/
>>>  obj-$(CONFIG_USB)		+= usb/
>>> diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
>>> index e5641e6c52ee..671c27b77a48 100644
>>> --- a/drivers/ata/Kconfig
>>> +++ b/drivers/ata/Kconfig
>>> @@ -1161,6 +1161,28 @@ config PATA_WINBOND_VLB
>>>  	  Support for the Winbond W83759A controller on Vesa Local Bus
>>>  	  systems.
>>>  
>>> +config PATA_PARPORT
>>> +	tristate "Parallel port IDE device support"
>>> +	depends on PARPORT_PC && PARIDE=n
>>
>> This is very confusing. The change above this one switch paride
>> compilation to be unconditional, regardless of CONFIG_PARIDE value, but
>> here, you have the dependency to PARIDE=n. I do not understand... Please
>> clarify.
> 
> pata_parport will use existing paride protocol modules. So the paride/ directory must be processed to compile the protocol modules (if they're enabled) even if paride is not enabled.
> 
> pata_parport and paride are mutually exclusive because the binary protocol modules are incompatible (the struct pi_adapter is different).

So if both CONFIG_PARIDE and CONFIG_PATA_PARPORT are disabled, there
should be no need to compile the core PARIDE code under block/. You
should have something like:

In drivers/Makefile:

-obj-$(CONFIG_PARIDE) 		+= block/paride/
+obj-$(CONFIG_PARIDE_CORE) 	+= block/paride/

And then have CONFIG_PARIDE and CONFIG_PATA_PARPORT select PARIDE_CORE,
with CONFIG_PARIDE and CONFIG_PATA_PARPORT being mutually exclusive
(using "depends on" as you did).

Here, I am assuming that block/paride is the core code used by both
PARIDE and PATA_PARPORT. Not sure what PARPORT_PC does nor what its
dependency on block/paride code is.


-- 
Damien Le Moal
Western Digital Research