From f837283f4221f6e3c870cf63ccc4fc2c6f383fc5 Mon Sep 17 00:00:00 2001 From: Jeffrey Cody Date: Wed, 15 Mar 2017 18:48:38 +0100 Subject: [PATCH 01/12] block/gluster: memory usage: use one glfs instance per volume RH-Author: Jeffrey Cody Message-id: Patchwork-id: 74293 O-Subject: [RHEV-7.3.z qemu-kvm-rhev v2 1/2] block/gluster: memory usage: use one glfs instance per volume Bugzilla: 1413044 RH-Acked-by: Stefan Hajnoczi RH-Acked-by: Markus Armbruster RH-Acked-by: Miroslav Rezanina From: Prasanna Kumar Kalever Currently, for every drive accessed via gfapi we create a new glfs instance (call glfs_new() followed by glfs_init()) which could consume memory in few 100 MB's, from the table below it looks like for each instance ~300 MB VSZ was consumed Before: ------- Disks VSZ RSS 1 1098728 187756 2 1430808 198656 3 1764932 199704 4 2084728 202684 This patch maintains a list of pre-opened glfs objects. On adding a new drive belonging to the same gluster volume, we just reuse the existing glfs object by updating its refcount. With this approch we shrink up the unwanted memory consumption and glfs_new/glfs_init calls for accessing a disk (file) if belongs to same volume. >From below table notice that the memory usage after adding a disk (which will reuse the existing glfs object hence) is in negligible compared to before. After: ------ Disks VSZ RSS 1 1101964 185768 2 1109604 194920 3 1114012 196036 4 1114496 199868 Disks: number of -drive VSZ: virtual memory size of the process in KiB RSS: resident set size, the non-swapped physical memory (in kiloBytes) VSZ and RSS are analyzed using 'ps aux' utility. Signed-off-by: Prasanna Kumar Kalever Reviewed-by: Jeff Cody Message-id: 1477581890-4811-1-git-send-email-prasanna.kalever@redhat.com Signed-off-by: Jeff Cody (cherry picked from commit 6349c15410361d3fe52c9beee309954d606f8ccd) Signed-off-by: Miroslav Rezanina --- block/gluster.c | 94 ++++++++++++++++++++++++++++++++++++++++++++++++--------- 1 file changed, 80 insertions(+), 14 deletions(-) diff --git a/block/gluster.c b/block/gluster.c index a5848aa..0771c40 100644 --- a/block/gluster.c +++ b/block/gluster.c @@ -53,6 +53,19 @@ typedef struct BDRVGlusterReopenState { } BDRVGlusterReopenState; +typedef struct GlfsPreopened { + char *volume; + glfs_t *fs; + int ref; +} GlfsPreopened; + +typedef struct ListElement { + QLIST_ENTRY(ListElement) list; + GlfsPreopened saved; +} ListElement; + +static QLIST_HEAD(glfs_list, ListElement) glfs_list; + static QemuOptsList qemu_gluster_create_opts = { .name = "qemu-gluster-create-opts", .head = QTAILQ_HEAD_INITIALIZER(qemu_gluster_create_opts.head), @@ -181,6 +194,57 @@ static QemuOptsList runtime_tcp_opts = { }, }; +static void glfs_set_preopened(const char *volume, glfs_t *fs) +{ + ListElement *entry = NULL; + + entry = g_new(ListElement, 1); + + entry->saved.volume = g_strdup(volume); + + entry->saved.fs = fs; + entry->saved.ref = 1; + + QLIST_INSERT_HEAD(&glfs_list, entry, list); +} + +static glfs_t *glfs_find_preopened(const char *volume) +{ + ListElement *entry = NULL; + + QLIST_FOREACH(entry, &glfs_list, list) { + if (strcmp(entry->saved.volume, volume) == 0) { + entry->saved.ref++; + return entry->saved.fs; + } + } + + return NULL; +} + +static void glfs_clear_preopened(glfs_t *fs) +{ + ListElement *entry = NULL; + + if (fs == NULL) { + return; + } + + QLIST_FOREACH(entry, &glfs_list, list) { + if (entry->saved.fs == fs) { + if (--entry->saved.ref) { + return; + } + + QLIST_REMOVE(entry, list); + + glfs_fini(entry->saved.fs); + g_free(entry->saved.volume); + g_free(entry); + } + } +} + static int parse_volume_options(BlockdevOptionsGluster *gconf, char *path) { char *p, *q; @@ -318,11 +382,18 @@ static struct glfs *qemu_gluster_glfs_init(BlockdevOptionsGluster *gconf, int old_errno; GlusterServerList *server; + glfs = glfs_find_preopened(gconf->volume); + if (glfs) { + return glfs; + } + glfs = glfs_new(gconf->volume); if (!glfs) { goto out; } + glfs_set_preopened(gconf->volume, glfs); + for (server = gconf->server; server; server = server->next) { if (server->value->type == GLUSTER_TRANSPORT_UNIX) { ret = glfs_set_volfile_server(glfs, @@ -374,7 +445,7 @@ static struct glfs *qemu_gluster_glfs_init(BlockdevOptionsGluster *gconf, out: if (glfs) { old_errno = errno; - glfs_fini(glfs); + glfs_clear_preopened(glfs); errno = old_errno; } return NULL; @@ -716,9 +787,9 @@ out: if (s->fd) { glfs_close(s->fd); } - if (s->glfs) { - glfs_fini(s->glfs); - } + + glfs_clear_preopened(s->glfs); + return ret; } @@ -783,9 +854,8 @@ static void qemu_gluster_reopen_commit(BDRVReopenState *state) if (s->fd) { glfs_close(s->fd); } - if (s->glfs) { - glfs_fini(s->glfs); - } + + glfs_clear_preopened(s->glfs); /* use the newly opened image / connection */ s->fd = reop_s->fd; @@ -810,9 +880,7 @@ static void qemu_gluster_reopen_abort(BDRVReopenState *state) glfs_close(reop_s->fd); } - if (reop_s->glfs) { - glfs_fini(reop_s->glfs); - } + glfs_clear_preopened(reop_s->glfs); g_free(state->opaque); state->opaque = NULL; @@ -932,9 +1000,7 @@ static int qemu_gluster_create(const char *filename, out: g_free(tmp); qapi_free_BlockdevOptionsGluster(gconf); - if (glfs) { - glfs_fini(glfs); - } + glfs_clear_preopened(glfs); return ret; } @@ -1006,7 +1072,7 @@ static void qemu_gluster_close(BlockDriverState *bs) glfs_close(s->fd); s->fd = NULL; } - glfs_fini(s->glfs); + glfs_clear_preopened(s->glfs); } static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs) -- 1.8.3.1