linux-kernel - RE: [PATCH v3 1/3] aerdrv: Trace Event for AER

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]

Message-ID: <A961433CDFF2F640A2866803152E61F9516EE7B0@G4W3218.americas.hpqcorp.net>
Date:	Mon, 3 Dec 2012 21:13:49 +0000
From:	"Ortiz, Lance E" <lance.oritz@...com>
To:	Borislav Petkov <bp@...en8.de>
CC:	Mauro Carvalho Chehab <mchehab@...hat.com>,
	"bhelgaas@...gle.com" <bhelgaas@...gle.com>,
	"lance_ortiz@...mail.com" <lance_ortiz@...mail.com>,
	"jiang.liu@...wei.com" <jiang.liu@...wei.com>,
	"tony.luck@...el.com" <tony.luck@...el.com>,
	"rostedt@...dmis.org" <rostedt@...dmis.org>,
	"linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
	"linux-pci@...r.kernel.org" <linux-pci@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Zhang Yanmin <yanmin.zhang@...el.com>
Subject: RE: [PATCH v3 1/3] aerdrv: Trace Event for AER

> 
> Maybe we should ask Yanmin who added those strings in
> 6c2b374d74857e892080ee726184ec1d15e7d4e4; CCed.
> 
> Also, it would make a lot of sense to have those string definitions
> at one place and then reuse them instead of define them again in the
> tracepoint header and they get out of sync and have needless
> duplication
> in the kernel, etc, etc.
> 
> So, instead, you could define some macros which can generate your
> strings above like this:
> 
> #define aer_uncor_err_str(bitnum)	\
> 	{ BIT(bitnum), aer_uncorrectable_error_string(bitnum) }
> 
> and use that macro for the __print_flags flag_array argument.
> 
> Or something to that effect.
> 
> Even better would it be if the error strings in
> <drivers/pci/pcie/aer/aerdrv_errprint.c> could be shared with the
> tracepoint ones. That would require a bit more changes though but
> something like using an array of trace_print_flags instead an array of
> strings could be doable.
> 
> Then, when you need the string, you do uncor_err_array[i].name and so
> on.

This a fine idea.  I'm thinking we might want to implement that in a later patch since it is a bigger change and we still need to see about adding additional strings.  I can send out one more version of the patch with the comment header changes you suggested, and the comments Joe had and maybe we can move forward with that.

Lance