From: Matthew Wilcox <matthew@wil.cx>
To: Mark Haverkamp <markh@osdl.org>
Cc: linux-scsi@vger.kernel.org,
James Bottomley <James.Bottomley@steeleye.com>,
Andrew Morton <akpm@osdl.org>
Subject: [PATCH 11/10] Fix build when spinlock debugging is enabled
Date: Wed, 29 Mar 2006 14:45:18 -0700 [thread overview]
Message-ID: <20060329214518.GD13590@parisc-linux.org> (raw)
In-Reply-To: <1143585873.1061.44.camel@markh3.pdx.osdl.net>
On Tue, Mar 28, 2006 at 02:44:33PM -0800, Mark Haverkamp wrote:
> I am seeing a compile error here:
> drivers/scsi/sym53c8xx_2/sym_glue.c: In function ???sym_eh_done???:
> drivers/scsi/sym53c8xx_2/sym_glue.c:711: error: size of array ???type name??? is negative
> make[3]: *** [drivers/scsi/sym53c8xx_2/sym_glue.o] Error 1
>
> config attached.
Thanks, Mark.
-----------
When spinlock debugging is turned on, a struct completion grows beyond the
size allowed for the scsi_pointer. So move the struct completion back onto
the stack. The additional memory barriers are to keep us from completing
a random piece of kernel stack if the command happens to complete after
the error handling has finished.
Signed-off-by: Matthew Wilcox <matthew@wil.cx>
diff --git a/drivers/scsi/sym53c8xx_2/sym_glue.c b/drivers/scsi/sym53c8xx_2/sym_glue.c
index 2c4e5f1..9c83b4d 100644
--- a/drivers/scsi/sym53c8xx_2/sym_glue.c
+++ b/drivers/scsi/sym53c8xx_2/sym_glue.c
@@ -140,11 +140,11 @@ static struct scsi_transport_template *s
* Driver private area in the SCSI command structure.
*/
struct sym_ucmd { /* Override the SCSI pointer structure */
- struct completion done;
- void (*old_done)(struct scsi_cmnd *);
- dma_addr_t data_mapping;
- int to_do;
- u_char data_mapped; /* corresponds to data_mapping above */
+ dma_addr_t data_mapping;
+ unsigned char data_mapped;
+ unsigned char to_do; /* For error handling */
+ void (*old_done)(struct scsi_cmnd *); /* For error handling */
+ struct completion *eh_done; /* For error handling */
};
#define SYM_UCMD_PTR(cmd) ((struct sym_ucmd *)(&(cmd)->SCp))
@@ -713,7 +713,7 @@ static void sym_eh_done(struct scsi_cmnd
cmd->scsi_done = ucmd->old_done;
if (ucmd->to_do == SYM_EH_DO_WAIT)
- complete(&ucmd->done);
+ complete(ucmd->eh_done);
}
/*
@@ -728,6 +728,7 @@ static int sym_eh_handler(int op, char *
SYM_QUEHEAD *qp;
int to_do = SYM_EH_DO_IGNORE;
int sts = -1;
+ struct completion eh_done;
dev_warn(&cmd->device->sdev_gendev, "%s operation started.\n", opname);
@@ -742,8 +743,10 @@ static int sym_eh_handler(int op, char *
}
if (to_do == SYM_EH_DO_WAIT) {
- init_completion(&ucmd->done);
+ init_completion(&eh_done);
ucmd->old_done = cmd->scsi_done;
+ ucmd->eh_done = &eh_done;
+ wmb();
cmd->scsi_done = sym_eh_done;
}
@@ -779,8 +782,9 @@ static int sym_eh_handler(int op, char *
spin_unlock_irq(host->host_lock);
if (to_do == SYM_EH_DO_WAIT) {
- if (!wait_for_completion_timeout(&ucmd->done, 5*HZ)) {
+ if (!wait_for_completion_timeout(&eh_done, 5*HZ)) {
ucmd->to_do = SYM_EH_DO_IGNORE;
+ wmb();
sts = -2;
}
}
prev parent reply other threads:[~2006-03-29 21:45 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-03-28 16:03 [PATCH 08/10] Simplify error handling Matthew Wilcox
2006-03-28 22:44 ` Mark Haverkamp
2006-03-29 21:45 ` Matthew Wilcox [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20060329214518.GD13590@parisc-linux.org \
--to=matthew@wil.cx \
--cc=James.Bottomley@steeleye.com \
--cc=akpm@osdl.org \
--cc=linux-scsi@vger.kernel.org \
--cc=markh@osdl.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).