Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1544205 > unrolled thread

Uninitialized variable in bfad_im_bsg_els_ct_request

Started byNicolas Iooss <nicolas.iooss_linux@m4x.org>
First post2016-12-18 21:50 +0100
Last post2016-12-18 21:50 +0100
Articles 1 — 1 participant

Back to article view | Back to linux.kernel


Contents

  Uninitialized variable in bfad_im_bsg_els_ct_request Nicolas Iooss <nicolas.iooss_linux@m4x.org> - 2016-12-18 21:50 +0100

#1544205 — Uninitialized variable in bfad_im_bsg_els_ct_request

FromNicolas Iooss <nicolas.iooss_linux@m4x.org>
Date2016-12-18 21:50 +0100
SubjectUninitialized variable in bfad_im_bsg_els_ct_request
Message-ID<sPQz7-6FT-3@gated-at.bofh.it>
Hello,

Currently on Linus master tree and in linux-next [1],
bfad_im_bsg_els_ct_request() code starts with the following code:

int
bfad_im_bsg_els_ct_request(struct bsg_job *job)
{
	/*...*/
	struct fc_bsg_request *bsg_request = bsg_request;
	struct fc_bsg_reply *bsg_reply = job->reply;
	uint32_t command_type = bsg_request->msgcode;

The local variable "bsg_request" is initialized to itself (which would
usually mean it is uninitialized) but it is dereferenced in order to get
its "msgcode" field. As I am quite new to the kernel code and
dereferencing self-initialized local variables looks black magic to me,
could you please describe why this code is valid?

It has recently been introduced by commit 01e0e15c8b3b ("scsi: don't use
fc_bsg_job::request and fc_bsg_job::reply directly").

Thanks,
Nicolas Iooss

[1]
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/tree/drivers/scsi/bfa/bfad_bsg.c?id=06548160dfecd1983ffd9d6795242a5cda095da5#n3356

[toc] | [standalone]


Back to top | Article view | linux.kernel


csiph-web