linux-kernel - Re: SYSFS "errors"

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]

Date:	Mon, 18 Feb 2013 20:46:33 +0200
From:	Felipe Balbi <balbi@...com>
To:	Greg KH <gregkh@...uxfoundation.org>
CC:	Felipe Balbi <balbi@...com>,
	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	Steven Rostedt <rostedt@...dmis.org>,
	Frederic Weisbecker <fweisbec@...il.com>,
	Ingo Molnar <mingo@...hat.com>, <JBottomley@...allels.com>,
	<linux-scsi@...r.kernel.org>, <davem@...emloft.net>,
	<netdev@...r.kernel.org>,
	Doug Thompson <dougthompson@...ssion.com>,
	<linux-edac@...r.kernel.org>, <rjw@...k.pl>,
	<linux-pm@...r.kernel.org>
Subject: Re: SYSFS "errors"

Hi, On Mon, Feb 18, 2013 at 09:49:16AM -0800, Greg KH wrote:
> > Input/output error - /sys/devices/cpu/power/autosuspend_delay_ms
> 
> The issue with this file is, if the power.use_autosuspend flag is not
> set for the device, then it can't be read or written to.  This flag
> changes dynamically with the system state
> (__pm_runtime_use_autosuspend() can change it), so we can't just not
> show the file if the flag is not set properly, sorry.
> 
> So the "error" is correct here, as is the 0644 file value.

hmm... we could create the file at pm_runtime_enable() time and remove
it on pm_runtime_disable() time, no ? Addin Rafael to Cc

> > No such device or address - /sys/devices/pci0000:00/0000:00:1d.7/usb2/2-5/2-5.1/2-5.1.1/2-5.1.1:1.0/host100/target100:0:0/100:0:0:0/block/sdc/trace/pid
> > No such device or address - /sys/devices/pci0000:00/0000:00:1d.7/usb2/2-5/2-5.1/2-5.1.1/2-5.1.1:1.0/host100/target100:0:0/100:0:0:0/block/sdc/trace/end_lba
> > No such device or address - /sys/devices/pci0000:00/0000:00:1d.7/usb2/2-5/2-5.1/2-5.1.1/2-5.1.1:1.0/host100/target100:0:0/100:0:0:0/block/sdc/trace/enable
> > No such device or address - /sys/devices/pci0000:00/0000:00:1d.7/usb2/2-5/2-5.1/2-5.1.1/2-5.1.1:1.0/host100/target100:0:0/100:0:0:0/block/sdc/trace/act_mask
> > No such device or address - /sys/devices/pci0000:00/0000:00:1d.7/usb2/2-5/2-5.1/2-5.1.1/2-5.1.1:1.0/host100/target100:0:0/100:0:0:0/block/sdc/trace/start_lba
> > Invalid argument - /sys/devices/pci0000:00/0000:00:1f.2/ata2/host1/scsi_host/host1/em_buffer
> > Invalid argument - /sys/devices/pci0000:00/0000:00:1f.2/ata2/host1/target1:0:0/1:0:0:0/sw_activity
> > Operation not supported - /sys/devices/pci0000:00/0000:00:1f.2/ata2/host1/target1:0:0/1:0:0:0/unload_heads
> 
> SCSI fun, go poke the SCSI developers about these files, I know nothing
> about them, nor do I really want to :)
> 
> > No such device - /sys/devices/system/edac/mc/mc0/sdram_scrub_rate
> 
> Odd, go ask the edac developers

will do ;-)

> > Input/output error - /sys/devices/system/memory/soft_offline_page
> > Input/output error - /sys/devices/system/memory/hard_offline_page
> 
> These both should be mode 0400, not 0644, they have no 'show' attribute,
> that's a bug, care to make up a patch that I can take?  Or I can, let me
> know.

sure, will send in a bit.

> > Invalid argument - /sys/devices/virtual/net/lo/speed
> > Invalid argument - /sys/devices/virtual/net/lo/duplex
> 
> Ask the network developers, odds are, the loopback device doesn't really
> have a speed :)

right ;-)

> > No such device or address - /sys/devices/virtual/block/loop0/trace/pid
> > No such device or address - /sys/devices/virtual/block/loop0/trace/end_lba
> > No such device or address - /sys/devices/virtual/block/loop0/trace/enable
> > No such device or address - /sys/devices/virtual/block/loop0/trace/act_mask
> > No such device or address - /sys/devices/virtual/block/loop0/trace/start_lba
> 
> Hm, these show up for virtual block devices as well, maybe they aren't
> scsi specific, but rather block specific, go poke the block developers
> then.

yeah, I was looking into these a while ago. They're created by
blk_trace_init_sysfs() which is called by blk_register_queue() but I
wonder if we couldn't defer the creation of those so that:

sysfs_blk_trace_attr_show():

1639         bdev = bdget(part_devt(p));
1640         if (bdev == NULL)
1641                 goto out;
1642 
1643         q = blk_trace_get_queue(bdev);
1644         if (q == NULL)
1645                 goto out_bdput;

never fail (meaning they could be removed), adding some block folks to
the loop.

-- 
balbi

Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)