[Qemu-devel] [PATCH] scsi-disk: Move active request asserts

[Qemu-devel] [PATCH] scsi-disk: Move active request asserts

[Stefan Hajnoczi]

SCSI read/write requests should not be re-issued before the current
fragment of I/O completes.  There are asserts in scsi-disk.c that guard
this constraint but they trigger on SPARC Linux 2.4.  It turns out that
the asserts are too early in the code path and don't allow for read
requests to terminate.

Only the read assert needs to be moved but move the write assert too for
consistency.

Reported-by: Nigel Horne <njh@bandsman.co.uk>
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
---
 hw/scsi-disk.c |   12 ++++++------
 1 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/hw/scsi-disk.c b/hw/scsi-disk.c
index dc71957..7d85899 100644
--- a/hw/scsi-disk.c
+++ b/hw/scsi-disk.c
@@ -170,6 +170,9 @@ static void scsi_read_request(SCSIDiskReq *r)
         return;
     }
 
+    /* No data transfer may already be in progress */
+    assert(r->req.aiocb == NULL);
+
     n = r->sector_count;
     if (n > SCSI_DMA_BUF_SIZE / 512)
         n = SCSI_DMA_BUF_SIZE / 512;
@@ -197,9 +200,6 @@ static void scsi_read_data(SCSIDevice *d, uint32_t tag)
         return;
     }
 
-    /* No data transfer may already be in progress */
-    assert(r->req.aiocb == NULL);
-
     scsi_read_request(r);
 }
 
@@ -269,6 +269,9 @@ static void scsi_write_request(SCSIDiskReq *r)
     SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, r->req.dev);
     uint32_t n;
 
+    /* No data transfer may already be in progress */
+    assert(r->req.aiocb == NULL);
+
     n = r->iov.iov_len / 512;
     if (n) {
         qemu_iovec_init_external(&r->qiov, &r->iov, 1);
@@ -298,9 +301,6 @@ static int scsi_write_data(SCSIDevice *d, uint32_t tag)
         return 1;
     }
 
-    /* No data transfer may already be in progress */
-    assert(r->req.aiocb == NULL);
-
     scsi_write_request(r);
 
     return 0;
-- 
1.7.2.3

[Qemu-devel] Re: [PATCH] scsi-disk: Move active request asserts

[Kevin Wolf]

Am 12.11.2010 10:57, schrieb Stefan Hajnoczi:
> SCSI read/write requests should not be re-issued before the current
> fragment of I/O completes.  There are asserts in scsi-disk.c that guard
> this constraint but they trigger on SPARC Linux 2.4.  It turns out that
> the asserts are too early in the code path and don't allow for read
> requests to terminate.
> 
> Only the read assert needs to be moved but move the write assert too for
> consistency.
> 
> Reported-by: Nigel Horne <njh@bandsman.co.uk>
> Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>

Thanks, applied to the block branch.

Kevin