File 0039-xen_disk-Avoid-repeated-memory-allo.patch of Package qemu

Overview Repositories Revisions Requests Users Attributes Meta

File 0039-xen_disk-Avoid-repeated-memory-allo.patch of Package qemu (Revision 19fafad1fed18df2caa2f163b3941b7a)

Currently displaying revision 19fafad1fed18df2caa2f163b3941b7a , Show latest

From f88c2f1b54da7a92420724189a6df37e6b3faed3 Mon Sep 17 00:00:00 2001
From: Tim Smith <tim.smith@citrix.com>
Date: Fri, 2 Nov 2018 10:01:09 +0000
Subject: [PATCH] xen_disk: Avoid repeated memory allocation

xen_disk currently allocates memory to hold the data for each ioreq
as that ioreq is used, and frees it afterwards. Because it requires
page-aligned blocks, this interacts poorly with non-page-aligned
allocations and balloons the heap.

Instead, allocate the maximum possible requirement, which is
BLKIF_MAX_SEGMENTS_PER_REQUEST pages (currently 11 pages) when
the ioreq is created, and keep that allocation until it is destroyed.
Since the ioreqs themselves are re-used via a free list, this
should actually improve memory usage.

Signed-off-by: Tim Smith <tim.smith@citrix.com>
[BSC#1100408]
Signed-off-by: Bruce Rogers <brogers@suse.com>
---
 hw/block/xen_disk.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

diff --git a/hw/block/xen_disk.c b/hw/block/xen_disk.c
index d9a55faf27..e9501ee34c 100644
--- a/hw/block/xen_disk.c
+++ b/hw/block/xen_disk.c
@@ -110,7 +110,6 @@ static void ioreq_reset(struct ioreq *ioreq)
     memset(&ioreq->req, 0, sizeof(ioreq->req));
     ioreq->status = 0;
     ioreq->start = 0;
-    ioreq->buf = NULL;
     ioreq->size = 0;
     ioreq->presync = 0;
 
@@ -135,6 +134,7 @@ static struct ioreq *ioreq_start(struct XenBlkDev *blkdev)
         /* allocate new struct */
         ioreq = g_malloc0(sizeof(*ioreq));
         ioreq->blkdev = blkdev;
+        ioreq->buf = qemu_memalign(XC_PAGE_SIZE, BLKIF_MAX_SEGMENTS_PER_REQUEST * XC_PAGE_SIZE);
         blkdev->requests_total++;
         qemu_iovec_init(&ioreq->v, 1);
     } else {
@@ -317,14 +317,12 @@ static void qemu_aio_complete(void *opaque, int ret)
         if (ret == 0) {
             ioreq_grant_copy(ioreq);
         }
-        qemu_vfree(ioreq->buf);
         break;
     case BLKIF_OP_WRITE:
     case BLKIF_OP_FLUSH_DISKCACHE:
         if (!ioreq->req.nr_segments) {
             break;
         }
-        qemu_vfree(ioreq->buf);
         break;
     default:
         break;
@@ -392,12 +390,10 @@ static int ioreq_runio_qemu_aio(struct ioreq *ioreq)
 {
     struct XenBlkDev *blkdev = ioreq->blkdev;
 
-    ioreq->buf = qemu_memalign(XC_PAGE_SIZE, ioreq->size);
     if (ioreq->req.nr_segments &&
         (ioreq->req.operation == BLKIF_OP_WRITE ||
          ioreq->req.operation == BLKIF_OP_FLUSH_DISKCACHE) &&
         ioreq_grant_copy(ioreq)) {
-        qemu_vfree(ioreq->buf);
         goto err;
     }
 
@@ -1007,6 +1003,7 @@ static int blk_free(struct XenDevice *xendev)
         ioreq = QLIST_FIRST(&blkdev->freelist);
         QLIST_REMOVE(ioreq, list);
         qemu_iovec_destroy(&ioreq->v);
+        qemu_vfree(ioreq->buf);
         g_free(ioreq);
     }

Places

File 0039-xen_disk-Avoid-repeated-memory-allo.patch of Package qemu (Revision 19fafad1fed18df2caa2f163b3941b7a)

Places