qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Ryan Harper <ryanh@us.ibm.com>
To: qemu-devel@nongnu.org
Cc: Anthony Liguori <aliguori@us.ibm.com>,
	Ryan Harper <ryanh@us.ibm.com>,
	kvm@vger.kernel.org
Subject: [Qemu-devel] [PATCH 3/4] Refactor scsi-disk layer for queue'ing writes
Date: Fri,  3 Oct 2008 17:05:30 -0500	[thread overview]
Message-ID: <1223071531-31817-4-git-send-email-ryanh@us.ibm.com> (raw)
In-Reply-To: <1223071531-31817-1-git-send-email-ryanh@us.ibm.com>

Refactor scsi write path to match the simpler, faster scsi read path.  No need
to call into the write completion function to calculate the request buf length.

Signed-off-by: Ryan Harper <ryanh@us.ibm.com>

diff --git a/hw/scsi-disk.c b/hw/scsi-disk.c
index 16b3215..f2b4814 100644
--- a/hw/scsi-disk.c
+++ b/hw/scsi-disk.c
@@ -208,25 +208,14 @@ static void scsi_write_complete(void * opaque, int ret)
 {
     SCSIRequest *r = (SCSIRequest *)opaque;
     SCSIDeviceState *s = r->dev;
-    uint32_t len;
 
     if (ret) {
         fprintf(stderr, "scsi-disc: IO write error\n");
         exit(1);
     }
-
     r->aiocb = NULL;
-    if (r->sector_count == 0) {
-        scsi_command_complete(r, SENSE_NO_SENSE);
-    } else {
-        len = r->sector_count * 512;
-        if (len > SCSI_DMA_BUF_SIZE) {
-            len = SCSI_DMA_BUF_SIZE;
-        }
-        r->buf_len = len;
-        DPRINTF("Write complete tag=0x%x more=%d\n", r->tag, len);
-        s->completion(s->opaque, SCSI_REASON_DATA, r->tag, len);
-    }
+    DPRINTF("Write complete tag=0x%x more=%d\n", r->tag, r->buf_len);
+    s->completion(s->opaque, SCSI_REASON_DATA, r->tag, r->buf_len);
 }
 
 /* Write data to a scsi device.  Returns nonzero on failure.
@@ -235,7 +224,7 @@ static int scsi_write_data(SCSIDevice *d, uint32_t tag)
 {
     SCSIDeviceState *s = d->state;
     SCSIRequest *r;
-    uint32_t n;
+    uint32_t n, len;
 
     DPRINTF("Write data tag=0x%x\n", tag);
     r = scsi_find_request(s, tag);
@@ -244,8 +233,22 @@ static int scsi_write_data(SCSIDevice *d, uint32_t tag)
         scsi_command_complete(r, SENSE_HARDWARE_ERROR);
         return 1;
     }
+    DPRINTF("Write sector_count=%d\n", r->sector_count);
+    if (r->sector_count == 0) {
+        scsi_command_complete(r, SENSE_NO_SENSE);
+        return 0;
+    }
     if (r->aiocb)
         BADF("Data transfer already in progress\n");
+
+    /* we know the len of the request and with the max size of the scsi buffer,
+     * we can calculate buf_len needed */
+    len = r->sector_count * 512;
+    if (len > SCSI_DMA_BUF_SIZE)
+        len = SCSI_DMA_BUF_SIZE;
+    r->buf_len = len;
+
+    /* number of sectors to submit */
     n = r->buf_len / 512;
     if (n) {
         r->aiocb = bdrv_aio_write(s->bdrv, r->sector, r->dma_buf, n,

  parent reply	other threads:[~2008-10-03 22:05 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-10-03 22:05 [Qemu-devel] [PATCH 0/4] Improve emulated scsi write performance Ryan Harper
2008-10-03 22:05 ` [Qemu-devel] [PATCH 1/4] lsi_queue_command: add dma direction parameter Ryan Harper
2008-10-03 22:05 ` [Qemu-devel] [PATCH 2/4] Refactor lsi_do_command to queue read and write ops Ryan Harper
2008-10-03 22:05 ` Ryan Harper [this message]
2008-10-03 22:05 ` [Qemu-devel] [PATCH 4/4] Reallocate dma buffers in read/write path if needed Ryan Harper
2008-10-03 23:17   ` Paul Brook
2008-10-03 23:35     ` Anthony Liguori
2008-10-04  0:00       ` Paul Brook
2008-10-04 10:00       ` Avi Kivity
     [not found]         ` <20081004135749.pphehrhuw9w4gwsc@imap.linux.ibm.com>
2008-10-04 21:47           ` Ryan Harper
2008-10-04 22:22             ` Anthony Liguori
2008-10-05  5:23               ` Avi Kivity
2008-10-05 23:06               ` Ryan Harper
2008-10-06  7:27                 ` Avi Kivity
2008-10-04 23:00             ` Paul Brook
2008-10-05  5:29             ` Avi Kivity
2008-10-05 23:08 ` [Qemu-devel] [PATCH 0/4] Improve emulated scsi write performance Ryan Harper
2008-10-13 16:15   ` Ryan Harper

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=1223071531-31817-4-git-send-email-ryanh@us.ibm.com \
    --to=ryanh@us.ibm.com \
    --cc=aliguori@us.ibm.com \
    --cc=kvm@vger.kernel.org \
    --cc=qemu-devel@nongnu.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).