[{"id":1765222,"web_url":"http://patchwork.ozlabs.org/comment/1765222/","msgid":"<87d171fqzd.fsf@dusky.pond.sub.org>","list_archive_url":null,"date":"2017-09-08T10:02:14","subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","submitter":{"id":2645,"url":"http://patchwork.ozlabs.org/api/people/2645/","name":"Markus Armbruster","email":"armbru@redhat.com"},"content":"Pradeep Jagadeesh writes:\n\n> This patch introduces qmp interfaces for the fsdev\n> devices. This provides two interfaces one\n> for querying info of all the fsdev devices. The second one\n> to set the IO limits for the required fsdev device.\n>\n> Signed-off-by: Pradeep Jagadeesh \n> ---\n> Makefile | 4 ++\n> fsdev/qemu-fsdev-dummy.c | 11 ++++\n> fsdev/qemu-fsdev-throttle.c | 145 ++++++++++++++++++++++++++++++++++++++++++--\n> fsdev/qemu-fsdev-throttle.h | 9 ++-\n> fsdev/qemu-fsdev.c | 30 +++++++++\n> monitor.c | 5 ++\n> qapi-schema.json | 3 +\n> qapi/fsdev.json | 81 +++++++++++++++++++++++++\n> qmp.c | 14 +++++\n> 9 files changed, 294 insertions(+), 8 deletions(-)\n> create mode 100644 qapi/fsdev.json\n>\n> diff --git a/Makefile b/Makefile\n> index 81447b1..ec31ffa 100644\n> --- a/Makefile\n> +++ b/Makefile\n> @@ -414,6 +414,10 @@ qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \\\n> $(SRC_PATH)/qapi/crypto.json $(SRC_PATH)/qapi/rocker.json \\\n> $(SRC_PATH)/qapi/trace.json\n> \n> +ifdef CONFIG_VIRTFS\n> +qapi-modules += $(SRC_PATH)/qapi/fsdev.json\n> +endif\n> +\n\nThe ifdef is a *lie*: qapi.py *will* include this file no matter what.\nLying to make is not a good idea. Drop the ifdef.\n\n> qapi-types.c qapi-types.h :\\\n> $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py)\n> \t$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-types.py \\\n> diff --git a/fsdev/qemu-fsdev-dummy.c b/fsdev/qemu-fsdev-dummy.c\n> index 6dc0fbc..28c82d2 100644\n> --- a/fsdev/qemu-fsdev-dummy.c\n> +++ b/fsdev/qemu-fsdev-dummy.c\n> @@ -14,8 +14,19 @@\n> #include \"qemu-fsdev.h\"\n> #include \"qemu/config-file.h\"\n> #include \"qemu/module.h\"\n> +#include \"qmp-commands.h\"\n> \n> int qemu_fsdev_add(QemuOpts *opts)\n> {\n> return 0;\n> }\n> +\n> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n> +{\n> + return;\n> +}\n> +\n> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n> +{\n> + return NULL;\n> +}\n\nPredates this patch: should this be in stubs/? Might simplify\nfsdev/Makefile.objs.\n\n> diff --git a/fsdev/qemu-fsdev-throttle.c b/fsdev/qemu-fsdev-throttle.c\n> index 0e6fb86..525352e 100644\n> --- a/fsdev/qemu-fsdev-throttle.c\n> +++ b/fsdev/qemu-fsdev-throttle.c\n[Skipping this one...]\n> diff --git a/fsdev/qemu-fsdev-throttle.h b/fsdev/qemu-fsdev-throttle.h\n> index e418643..bfeab40 100644\n> --- a/fsdev/qemu-fsdev-throttle.h\n> +++ b/fsdev/qemu-fsdev-throttle.h\n> @@ -15,8 +15,6 @@\n> #ifndef _FSDEV_THROTTLE_H\n> #define _FSDEV_THROTTLE_H\n> \n> -#include \"block/aio.h\"\n> -#include \"qemu/main-loop.h\"\n> #include \"qemu/coroutine.h\"\n> #include \"qapi/error.h\"\n> #include \"qemu/throttle.h\"\n> @@ -25,6 +23,7 @@ typedef struct FsThrottle {\n> ThrottleState ts;\n> ThrottleTimers tt;\n> ThrottleConfig cfg;\n> + AioContext *ctx;\n> CoQueue throttled_reqs[2];\n> } FsThrottle;\n> \n> @@ -36,4 +35,10 @@ void coroutine_fn fsdev_co_throttle_request(FsThrottle *, bool ,\n> struct iovec *, int);\n> \n> void fsdev_throttle_cleanup(FsThrottle *);\n> +\n> +void fsdev_set_io_throttle(IOThrottle *, FsThrottle *, Error **errp);\n> +\n> +void fsdev_get_io_throttle(FsThrottle *, IOThrottle **iothp,\n> + char *);\n\nInclude the last parameter's name, for consistency with the other\nparameter declarations.\n\n> +\n> #endif /* _FSDEV_THROTTLE_H */\n> diff --git a/fsdev/qemu-fsdev.c b/fsdev/qemu-fsdev.c\n> index 266e442..0f701fc 100644\n> --- a/fsdev/qemu-fsdev.c\n> +++ b/fsdev/qemu-fsdev.c\n> @@ -16,6 +16,7 @@\n> #include \"qemu-common.h\"\n> #include \"qemu/config-file.h\"\n> #include \"qemu/error-report.h\"\n> +#include \"qmp-commands.h\"\n> \n> static QTAILQ_HEAD(FsDriverEntry_head, FsDriverListEntry) fsdriver_entries =\n> QTAILQ_HEAD_INITIALIZER(fsdriver_entries);\n> @@ -98,3 +99,32 @@ FsDriverEntry *get_fsdev_fsentry(char *id)\n> }\n> return NULL;\n> }\n> +\n> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n> +{\n> +\n> + FsDriverEntry *fse;\n> +\n> + fse = get_fsdev_fsentry(arg->has_id ? arg->id : NULL);\n\n!arg->has_id implies !arg->id. Therefore, Just arg->id would do, no\nneed to check arg->has_id first. It's not wrong, though.\n\n> + if (!fse) {\n> + error_setg(errp, \"Not a valid fsdev device\");\n> + return;\n> + }\n> +\n> + fsdev_set_io_throttle(arg, &fse->fst, errp);\n> +}\n> +\n> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n> +{\n> + IOThrottleList *head = NULL, *p_next;\n> + struct FsDriverListEntry *fsle;\n> +\n> + QTAILQ_FOREACH(fsle, &fsdriver_entries, next) {\n> + p_next = g_new0(IOThrottleList, 1);\n> + fsdev_get_io_throttle(&fsle->fse.fst, &p_next->value,\n> + fsle->fse.fsdev_id);\n> + p_next->next = head;\n> + head = p_next;\n> + }\n> + return head;\n> +}\n> diff --git a/monitor.c b/monitor.c\n> index e0f8801..eebda1c 100644\n> --- a/monitor.c\n> +++ b/monitor.c\n> @@ -998,6 +998,11 @@ static void qmp_unregister_commands_hack(void)\n> && !defined(TARGET_S390X)\n> qmp_unregister_command(&qmp_commands, \"query-cpu-definitions\");\n> #endif\n> +#ifndef CONFIG_VIRTFS\n> + qmp_unregister_command(&qmp_commands, \"fsdev-set-io-throttle\");\n> + qmp_unregister_command(&qmp_commands, \"query-fsdev-io-throttle\");\n> +#endif\n> +\n> }\n> \n> void monitor_init_qmp_commands(void)\n> diff --git a/qapi-schema.json b/qapi-schema.json\n> index 802ea53..8cf8140 100644\n> --- a/qapi-schema.json\n> +++ b/qapi-schema.json\n> @@ -81,6 +81,9 @@\n> # QAPI block definitions\n> { 'include': 'qapi/block.json' }\n> \n> +# QAPI fsdev definitions\n> +{ 'include': 'qapi/fsdev.json' }\n> +\n> # QAPI event definitions\n> { 'include': 'qapi/event.json' }\n> \n> diff --git a/qapi/fsdev.json b/qapi/fsdev.json\n> new file mode 100644\n> index 0000000..21d074d\n> --- /dev/null\n> +++ b/qapi/fsdev.json\n> @@ -0,0 +1,81 @@\n> +# -*- Mode: Python -*-\n> +\n> +##\n> +# == QAPI fsdev definitions\n> +##\n> +\n> +##\n> +# @fsdev-set-io-throttle:\n> +#\n> +# Change I/O limits for a 9p/fsdev device.\n> +#\n> +# I/O limits can be enabled by setting throttle value to non-zero number.\n> +#\n> +# I/O limits can be disabled by setting all throttle values to 0.\n> +#\n> +# Returns: Nothing on success\n> +# If @device is not a valid fsdev device, GenericError\n> +#\n> +# Since: 2.10\n\n2.11\n\n> +#\n> +# Example:\n> +#\n> +# -> { \"execute\": \"fsdev-set-io-throttle\",\n> +# \"arguments\": { \"id\": \"id0-1-0\",\n> +# \"bps\": 1000000,\n> +# \"bps_rd\": 0,\n> +# \"bps_wr\": 0,\n> +# \"iops\": 0,\n> +# \"iops_rd\": 0,\n> +# \"iops_wr\": 0,\n> +# \"bps_max\": 8000000,\n> +# \"bps_rd_max\": 0,\n> +# \"bps_wr_max\": 0,\n> +# \"iops_max\": 0,\n> +# \"iops_rd_max\": 0,\n> +# \"iops_wr_max\": 0,\n> +# \"bps_max_length\": 60,\n> +# \"iops_size\": 0 } }\n> +# <- { \"returns\": {} }\n> +##\n> +{ 'command': 'fsdev-set-io-throttle', 'boxed': true,\n> + 'data': 'IOThrottle' }\n> +##\n> +# @query-fsdev-io-throttle:\n> +#\n> +# Returns: a list of @IOThrottle describing I/O throttle\n> +# values of each fsdev device\n> +#\n> +# Since: 2.10\n\n2.11\n\nIOThrottle serves both as parameter of fsdev-set-io-throttle and as\nresult of query-fsdev-io-throttle.\n\nThe former needs many of its members to be optional.\n\nWhich of the optional members are *actually* optional in the result of\nquery-fsdev-io-throttle? Under what conditions can they be absent?\n\nFor what it's worth, the corresponding block layer query doesn't reuse\nIOThrottle. Instead, it repeats the throttling stuff in\nBlockDeviceInfo, with the same members optional. BlockDeviceInfo\ndocumentation neglects to specify under what conditions they can be\nabsent. Berto?\n\n> +#\n> +# Example:\n> +#\n> +# -> { \"Execute\": \"query-fsdev-io-throttle\" }\n> +# <- { \"returns\" : [\n> +# {\n> +# \"id\": \"id0-hd0\",\n> +# \"bps\":1000000,\n> +# \"bps_rd\":0,\n> +# \"bps_wr\":0,\n> +# \"iops\":1000000,\n> +# \"iops_rd\":0,\n> +# \"iops_wr\":0,\n> +# \"bps_max\": 8000000,\n> +# \"bps_rd_max\": 0,\n> +# \"bps_wr_max\": 0,\n> +# \"iops_max\": 0,\n> +# \"iops_rd_max\": 0,\n> +# \"iops_wr_max\": 0,\n> +# \"bps_max_length\": 0,\n> +# \"bps_rd_max_length\": 0,\n> +# \"bps_wr_max_length\": 0,\n> +# \"iops_max_length\": 0,\n> +# \"iops_rd_max_length\": 0,\n> +# \"iops_wr_max_length\": 0,\n> +# \"iops_size\": 0\n> +# }\n> +# ]\n> +# }\n> +#\n> +##\n> +{ 'command': 'query-fsdev-io-throttle', 'returns': [ 'IOThrottle' ] }\n> diff --git a/qmp.c b/qmp.c\n> index b86201e..eed91e5 100644\n> --- a/qmp.c\n> +++ b/qmp.c\n> @@ -130,6 +130,20 @@ void qmp_cpu_add(int64_t id, Error **errp)\n> }\n> }\n> \n> +#if defined(_WIN64) || defined(_WIN32) || defined(__FreeBSD__)\n> +\n> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n> +{\n> + return;\n> +}\n> +\n> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n> +{\n> + return NULL;\n> +}\n> +\n> +#endif\n> +\n> #ifndef CONFIG_VNC\n> /* If VNC support is enabled, the \"true\" query-vnc command is\n> defined in the VNC subsystem */\n\nWhy do we need *two* stubs for these functions, one here and one in\nfsdev/qemu-fsdev-dummy.c?","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","ext-mx05.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx05.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=armbru@redhat.com"],"Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xpYFn2xQTz9s3T\n\tfor ;\n\tFri, 8 Sep 2017 20:19:09 +1000 (AEST)","from localhost ([::1]:44342 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1dqGNP-0003Ys-El\n\tfor incoming@patchwork.ozlabs.org; Fri, 08 Sep 2017 06:19:07 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:45060)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1dqG7I-0005nx-6g\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 06:02:38 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1dqG77-0001Zk-NU\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 06:02:28 -0400","from mx1.redhat.com ([209.132.183.28]:32890)\n\tby eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32)\n\t(Exim 4.71) (envelope-from ) id 1dqG77-0001Z3-FS\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 06:02:17 -0400","from smtp.corp.redhat.com\n\t(int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.13])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby mx1.redhat.com (Postfix) with ESMTPS id 6D20C13A88;\n\tFri, 8 Sep 2017 10:02:16 +0000 (UTC)","from blackfin.pond.sub.org (ovpn-116-75.ams2.redhat.com\n\t[10.36.116.75])\n\tby smtp.corp.redhat.com (Postfix) with ESMTPS id 9E63C614C0;\n\tFri, 8 Sep 2017 10:02:15 +0000 (UTC)","by blackfin.pond.sub.org (Postfix, from userid 1000)\n\tid 275DA1138645; Fri, 8 Sep 2017 12:02:14 +0200 (CEST)"],"DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 6D20C13A88","From":"Markus Armbruster ","To":"Pradeep Jagadeesh ","References":"<1504541267-36954-1-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>","Date":"Fri, 08 Sep 2017 12:02:14 +0200","In-Reply-To":"<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>\n\t(Pradeep Jagadeesh's message of \"Mon, 4 Sep 2017 12:07:46 -0400\")","Message-ID":"<87d171fqzd.fsf@dusky.pond.sub.org>","User-Agent":"Gnus/5.13 (Gnus v5.13) Emacs/25.2 (gnu/linux)","MIME-Version":"1.0","Content-Type":"text/plain","X-Scanned-By":"MIMEDefang 2.79 on 10.5.11.13","X-Greylist":"Sender IP whitelisted, not delayed by milter-greylist-4.5.16\n\t(mx1.redhat.com [10.5.110.29]);\n\tFri, 08 Sep 2017 10:02:16 +0000 (UTC)","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic]\n\t[fuzzy]","X-Received-From":"209.132.183.28","Subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"alberto garcia , qemu-devel@nongnu.org,\n\tgreg kurz ,\n\t\"Dr. David Alan Gilbert\" , \n\tPradeep Jagadeesh ,\n\tjani kokkonen ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"}},{"id":1765284,"web_url":"http://patchwork.ozlabs.org/comment/1765284/","msgid":"<426ebb0f-c603-3d1e-2ba0-963875b14394@huawei.com>","list_archive_url":null,"date":"2017-09-08T12:19:16","subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","submitter":{"id":69883,"url":"http://patchwork.ozlabs.org/api/people/69883/","name":"Pradeep Jagadeesh","email":"pradeep.jagadeesh@huawei.com"},"content":"On 9/8/2017 12:02 PM, Markus Armbruster wrote:\n> Pradeep Jagadeesh writes:\n>\n>> This patch introduces qmp interfaces for the fsdev\n>> devices. This provides two interfaces one\n>> for querying info of all the fsdev devices. The second one\n>> to set the IO limits for the required fsdev device.\n>>\n>> Signed-off-by: Pradeep Jagadeesh \n>> ---\n>> Makefile | 4 ++\n>> fsdev/qemu-fsdev-dummy.c | 11 ++++\n>> fsdev/qemu-fsdev-throttle.c | 145 ++++++++++++++++++++++++++++++++++++++++++--\n>> fsdev/qemu-fsdev-throttle.h | 9 ++-\n>> fsdev/qemu-fsdev.c | 30 +++++++++\n>> monitor.c | 5 ++\n>> qapi-schema.json | 3 +\n>> qapi/fsdev.json | 81 +++++++++++++++++++++++++\n>> qmp.c | 14 +++++\n>> 9 files changed, 294 insertions(+), 8 deletions(-)\n>> create mode 100644 qapi/fsdev.json\n>>\n>> diff --git a/Makefile b/Makefile\n>> index 81447b1..ec31ffa 100644\n>> --- a/Makefile\n>> +++ b/Makefile\n>> @@ -414,6 +414,10 @@ qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \\\n>> $(SRC_PATH)/qapi/crypto.json $(SRC_PATH)/qapi/rocker.json \\\n>> $(SRC_PATH)/qapi/trace.json\n>>\n>> +ifdef CONFIG_VIRTFS\n>> +qapi-modules += $(SRC_PATH)/qapi/fsdev.json\n>> +endif\n>> +\n>\n> The ifdef is a *lie*: qapi.py *will* include this file no matter what.\n> Lying to make is not a good idea. Drop the ifdef.\nYes, thats right. As of these are useless.\nFor the same reason I have these dummy functions in two different places.\n>\n>> qapi-types.c qapi-types.h :\\\n>> $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py)\n>> \t$(call quiet-command,$(PYTHON) $(SRC_PATH)/scripts/qapi-types.py \\\n>> diff --git a/fsdev/qemu-fsdev-dummy.c b/fsdev/qemu-fsdev-dummy.c\n>> index 6dc0fbc..28c82d2 100644\n>> --- a/fsdev/qemu-fsdev-dummy.c\n>> +++ b/fsdev/qemu-fsdev-dummy.c\n>> @@ -14,8 +14,19 @@\n>> #include \"qemu-fsdev.h\"\n>> #include \"qemu/config-file.h\"\n>> #include \"qemu/module.h\"\n>> +#include \"qmp-commands.h\"\n>>\n>> int qemu_fsdev_add(QemuOpts *opts)\n>> {\n>> return 0;\n>> }\n>> +\n>> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n>> +{\n>> + return;\n>> +}\n>> +\n>> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n>> +{\n>> + return NULL;\n>> +}\n>\n> Predates this patch: should this be in stubs/? Might simplify\n> fsdev/Makefile.objs.\nExplained above\n>\n>> diff --git a/fsdev/qemu-fsdev-throttle.c b/fsdev/qemu-fsdev-throttle.c\n>> index 0e6fb86..525352e 100644\n>> --- a/fsdev/qemu-fsdev-throttle.c\n>> +++ b/fsdev/qemu-fsdev-throttle.c\n> [Skipping this one...]\n>> diff --git a/fsdev/qemu-fsdev-throttle.h b/fsdev/qemu-fsdev-throttle.h\n>> index e418643..bfeab40 100644\n>> --- a/fsdev/qemu-fsdev-throttle.h\n>> +++ b/fsdev/qemu-fsdev-throttle.h\n>> @@ -15,8 +15,6 @@\n>> #ifndef _FSDEV_THROTTLE_H\n>> #define _FSDEV_THROTTLE_H\n>>\n>> -#include \"block/aio.h\"\n>> -#include \"qemu/main-loop.h\"\n>> #include \"qemu/coroutine.h\"\n>> #include \"qapi/error.h\"\n>> #include \"qemu/throttle.h\"\n>> @@ -25,6 +23,7 @@ typedef struct FsThrottle {\n>> ThrottleState ts;\n>> ThrottleTimers tt;\n>> ThrottleConfig cfg;\n>> + AioContext *ctx;\n>> CoQueue throttled_reqs[2];\n>> } FsThrottle;\n>>\n>> @@ -36,4 +35,10 @@ void coroutine_fn fsdev_co_throttle_request(FsThrottle *, bool ,\n>> struct iovec *, int);\n>>\n>> void fsdev_throttle_cleanup(FsThrottle *);\n>> +\n>> +void fsdev_set_io_throttle(IOThrottle *, FsThrottle *, Error **errp);\n>> +\n>> +void fsdev_get_io_throttle(FsThrottle *, IOThrottle **iothp,\n>> + char *);\n>\n> Include the last parameter's name, for consistency with the other\n> parameter declarations.\nOK\n>\n>> +\n>> #endif /* _FSDEV_THROTTLE_H */\n>> diff --git a/fsdev/qemu-fsdev.c b/fsdev/qemu-fsdev.c\n>> index 266e442..0f701fc 100644\n>> --- a/fsdev/qemu-fsdev.c\n>> +++ b/fsdev/qemu-fsdev.c\n>> @@ -16,6 +16,7 @@\n>> #include \"qemu-common.h\"\n>> #include \"qemu/config-file.h\"\n>> #include \"qemu/error-report.h\"\n>> +#include \"qmp-commands.h\"\n>>\n>> static QTAILQ_HEAD(FsDriverEntry_head, FsDriverListEntry) fsdriver_entries =\n>> QTAILQ_HEAD_INITIALIZER(fsdriver_entries);\n>> @@ -98,3 +99,32 @@ FsDriverEntry *get_fsdev_fsentry(char *id)\n>> }\n>> return NULL;\n>> }\n>> +\n>> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n>> +{\n>> +\n>> + FsDriverEntry *fse;\n>> +\n>> + fse = get_fsdev_fsentry(arg->has_id ? arg->id : NULL);\n>\n> !arg->has_id implies !arg->id. Therefore, Just arg->id would do, no\n> need to check arg->has_id first. It's not wrong, though.\n>\n>> + if (!fse) {\n>> + error_setg(errp, \"Not a valid fsdev device\");\n>> + return;\n>> + }\n>> +\n>> + fsdev_set_io_throttle(arg, &fse->fst, errp);\n>> +}\n>> +\n>> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n>> +{\n>> + IOThrottleList *head = NULL, *p_next;\n>> + struct FsDriverListEntry *fsle;\n>> +\n>> + QTAILQ_FOREACH(fsle, &fsdriver_entries, next) {\n>> + p_next = g_new0(IOThrottleList, 1);\n>> + fsdev_get_io_throttle(&fsle->fse.fst, &p_next->value,\n>> + fsle->fse.fsdev_id);\n>> + p_next->next = head;\n>> + head = p_next;\n>> + }\n>> + return head;\n>> +}\n>> diff --git a/monitor.c b/monitor.c\n>> index e0f8801..eebda1c 100644\n>> --- a/monitor.c\n>> +++ b/monitor.c\n>> @@ -998,6 +998,11 @@ static void qmp_unregister_commands_hack(void)\n>> && !defined(TARGET_S390X)\n>> qmp_unregister_command(&qmp_commands, \"query-cpu-definitions\");\n>> #endif\n>> +#ifndef CONFIG_VIRTFS\n>> + qmp_unregister_command(&qmp_commands, \"fsdev-set-io-throttle\");\n>> + qmp_unregister_command(&qmp_commands, \"query-fsdev-io-throttle\");\n>> +#endif\n>> +\n>> }\n>>\n>> void monitor_init_qmp_commands(void)\n>> diff --git a/qapi-schema.json b/qapi-schema.json\n>> index 802ea53..8cf8140 100644\n>> --- a/qapi-schema.json\n>> +++ b/qapi-schema.json\n>> @@ -81,6 +81,9 @@\n>> # QAPI block definitions\n>> { 'include': 'qapi/block.json' }\n>>\n>> +# QAPI fsdev definitions\n>> +{ 'include': 'qapi/fsdev.json' }\n>> +\n>> # QAPI event definitions\n>> { 'include': 'qapi/event.json' }\n>>\n>> diff --git a/qapi/fsdev.json b/qapi/fsdev.json\n>> new file mode 100644\n>> index 0000000..21d074d\n>> --- /dev/null\n>> +++ b/qapi/fsdev.json\n>> @@ -0,0 +1,81 @@\n>> +# -*- Mode: Python -*-\n>> +\n>> +##\n>> +# == QAPI fsdev definitions\n>> +##\n>> +\n>> +##\n>> +# @fsdev-set-io-throttle:\n>> +#\n>> +# Change I/O limits for a 9p/fsdev device.\n>> +#\n>> +# I/O limits can be enabled by setting throttle value to non-zero number.\n>> +#\n>> +# I/O limits can be disabled by setting all throttle values to 0.\n>> +#\n>> +# Returns: Nothing on success\n>> +# If @device is not a valid fsdev device, GenericError\n>> +#\n>> +# Since: 2.10\n>\n> 2.11\n>\nok\n>> +#\n>> +# Example:\n>> +#\n>> +# -> { \"execute\": \"fsdev-set-io-throttle\",\n>> +# \"arguments\": { \"id\": \"id0-1-0\",\n>> +# \"bps\": 1000000,\n>> +# \"bps_rd\": 0,\n>> +# \"bps_wr\": 0,\n>> +# \"iops\": 0,\n>> +# \"iops_rd\": 0,\n>> +# \"iops_wr\": 0,\n>> +# \"bps_max\": 8000000,\n>> +# \"bps_rd_max\": 0,\n>> +# \"bps_wr_max\": 0,\n>> +# \"iops_max\": 0,\n>> +# \"iops_rd_max\": 0,\n>> +# \"iops_wr_max\": 0,\n>> +# \"bps_max_length\": 60,\n>> +# \"iops_size\": 0 } }\n>> +# <- { \"returns\": {} }\n>> +##\n>> +{ 'command': 'fsdev-set-io-throttle', 'boxed': true,\n>> + 'data': 'IOThrottle' }\n>> +##\n>> +# @query-fsdev-io-throttle:\n>> +#\n>> +# Returns: a list of @IOThrottle describing I/O throttle\n>> +# values of each fsdev device\n>> +#\n>> +# Since: 2.10\n>\n> 2.11\n>\n> IOThrottle serves both as parameter of fsdev-set-io-throttle and as\n> result of query-fsdev-io-throttle.\n>\n> The former needs many of its members to be optional.\n>\n> Which of the optional members are *actually* optional in the result of\n> query-fsdev-io-throttle? Under what conditions can they be absent?\n>\n> For what it's worth, the corresponding block layer query doesn't reuse\n> IOThrottle. Instead, it repeats the throttling stuff in\n> BlockDeviceInfo, with the same members optional. BlockDeviceInfo\n> documentation neglects to specify under what conditions they can be\n> absent. Berto?\n>\n>> +#\n>> +# Example:\n>> +#\n>> +# -> { \"Execute\": \"query-fsdev-io-throttle\" }\n>> +# <- { \"returns\" : [\n>> +# {\n>> +# \"id\": \"id0-hd0\",\n>> +# \"bps\":1000000,\n>> +# \"bps_rd\":0,\n>> +# \"bps_wr\":0,\n>> +# \"iops\":1000000,\n>> +# \"iops_rd\":0,\n>> +# \"iops_wr\":0,\n>> +# \"bps_max\": 8000000,\n>> +# \"bps_rd_max\": 0,\n>> +# \"bps_wr_max\": 0,\n>> +# \"iops_max\": 0,\n>> +# \"iops_rd_max\": 0,\n>> +# \"iops_wr_max\": 0,\n>> +# \"bps_max_length\": 0,\n>> +# \"bps_rd_max_length\": 0,\n>> +# \"bps_wr_max_length\": 0,\n>> +# \"iops_max_length\": 0,\n>> +# \"iops_rd_max_length\": 0,\n>> +# \"iops_wr_max_length\": 0,\n>> +# \"iops_size\": 0\n>> +# }\n>> +# ]\n>> +# }\n>> +#\n>> +##\n>> +{ 'command': 'query-fsdev-io-throttle', 'returns': [ 'IOThrottle' ] }\n>> diff --git a/qmp.c b/qmp.c\n>> index b86201e..eed91e5 100644\n>> --- a/qmp.c\n>> +++ b/qmp.c\n>> @@ -130,6 +130,20 @@ void qmp_cpu_add(int64_t id, Error **errp)\n>> }\n>> }\n>>\n>> +#if defined(_WIN64) || defined(_WIN32) || defined(__FreeBSD__)\n>> +\n>> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n>> +{\n>> + return;\n>> +}\n>> +\n>> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n>> +{\n>> + return NULL;\n>> +}\n>> +\n>> +#endif\n>> +\n>> #ifndef CONFIG_VNC\n>> /* If VNC support is enabled, the \"true\" query-vnc command is\n>> defined in the VNC subsystem */\n>\n> Why do we need *two* stubs for these functions, one here and one in\n> fsdev/qemu-fsdev-dummy.c?\nAt two different platforms the build fails. So, there are needed.\n\n-Pradeep\n>","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=208.118.235.17; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","Received":["from lists.gnu.org (lists.gnu.org [208.118.235.17])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xpbzv1ZZRz9t6W\n\tfor ;\n\tFri, 8 Sep 2017 22:22:16 +1000 (AEST)","from localhost ([::1]:45208 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1dqIHA-00087V-Ei\n\tfor incoming@patchwork.ozlabs.org; Fri, 08 Sep 2017 08:20:48 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:42207)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1dqIGr-00087A-Jo\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:20:31 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1dqIGn-0001wb-Cl\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:20:29 -0400","from lhrrgout.huawei.com ([194.213.3.17]:23214)\n\tby eggs.gnu.org with esmtps (TLS1.0:RSA_ARCFOUR_SHA1:16) (Exim 4.71)\n\t(envelope-from ) id 1dqIGm-0001o6-Rk\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:20:25 -0400","from 172.18.7.190 (EHLO LHREML712-CAH.china.huawei.com)\n\t([172.18.7.190])\n\tby lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued)\n\twith ESMTP id DOE70712; Fri, 08 Sep 2017 12:19:27 +0000 (GMT)","from [127.0.0.1] (10.210.165.15) by LHREML712-CAH.china.huawei.com\n\t(10.201.108.35) with Microsoft SMTP Server id 14.3.301.0;\n\tFri, 8 Sep 2017 13:19:18 +0100"],"To":"Markus Armbruster , Pradeep Jagadeesh\n\t","References":"<1504541267-36954-1-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<87d171fqzd.fsf@dusky.pond.sub.org>","From":"Pradeep Jagadeesh ","Message-ID":"<426ebb0f-c603-3d1e-2ba0-963875b14394@huawei.com>","Date":"Fri, 8 Sep 2017 14:19:16 +0200","User-Agent":"Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101\n\tThunderbird/45.8.0","MIME-Version":"1.0","In-Reply-To":"<87d171fqzd.fsf@dusky.pond.sub.org>","Content-Type":"text/plain; charset=\"windows-1252\"; format=flowed","Content-Transfer-Encoding":"7bit","X-Originating-IP":"[10.210.165.15]","X-CFilter-Loop":"Reflected","X-Mirapoint-Virus-RAPID-Raw":"score=unknown(0),\n\trefid=str=0001.0A020204.59B28AD0.00A4, ss=1, re=0.000, recu=0.000,\n\treip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0,\n\tso=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32","X-Mirapoint-Loop-Id":"79228641cc23b39fb5dfeb1887dc3318","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.4.x-2.6.x [generic]\n\t[fuzzy]","X-Received-From":"194.213.3.17","Subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"alberto garcia , qemu-devel@nongnu.org,\n\tgreg kurz ,\n\t\"Dr. David Alan Gilbert\" , \n\tjani kokkonen ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"}},{"id":1765292,"web_url":"http://patchwork.ozlabs.org/comment/1765292/","msgid":"<87bmml8j2x.fsf@dusky.pond.sub.org>","list_archive_url":null,"date":"2017-09-08T12:34:46","subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","submitter":{"id":2645,"url":"http://patchwork.ozlabs.org/api/people/2645/","name":"Markus Armbruster","email":"armbru@redhat.com"},"content":"Pradeep Jagadeesh writes:\n\n> On 9/8/2017 12:02 PM, Markus Armbruster wrote:\n>> Pradeep Jagadeesh writes:\n[...]\n>>> diff --git a/qmp.c b/qmp.c\n>>> index b86201e..eed91e5 100644\n>>> --- a/qmp.c\n>>> +++ b/qmp.c\n>>> @@ -130,6 +130,20 @@ void qmp_cpu_add(int64_t id, Error **errp)\n>>> }\n>>> }\n>>>\n>>> +#if defined(_WIN64) || defined(_WIN32) || defined(__FreeBSD__)\n>>> +\n>>> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n>>> +{\n>>> + return;\n>>> +}\n>>> +\n>>> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n>>> +{\n>>> + return NULL;\n>>> +}\n>>> +\n>>> +#endif\n>>> +\n>>> #ifndef CONFIG_VNC\n>>> /* If VNC support is enabled, the \"true\" query-vnc command is\n>>> defined in the VNC subsystem */\n>>\n>> Why do we need *two* stubs for these functions, one here and one in\n>> fsdev/qemu-fsdev-dummy.c?\n> At two different platforms the build fails. So, there are needed.\n\nWe don't add stubs to random places until the linker succeeds for all\nthe configurations we happen to test. We figure out *how* the linker\nfails to correct our idea of where a certain stub belongs until we find\nthe one appropriate home for it.\n\nIf you can't figure that out, that's okay, just tell us how exactly the\nlinker fails for you, and we'll be happy to help.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","ext-mx09.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx09.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=armbru@redhat.com"],"Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xpcHS4Bdbz9t63\n\tfor ;\n\tFri, 8 Sep 2017 22:35:47 +1000 (AEST)","from localhost ([::1]:45251 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1dqIV9-0005KH-Nf\n\tfor incoming@patchwork.ozlabs.org; Fri, 08 Sep 2017 08:35:15 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:46585)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1dqIUo-0005Jw-9o\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:34:55 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1dqIUl-00065U-5c\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:34:54 -0400","from mx1.redhat.com ([209.132.183.28]:55626)\n\tby eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32)\n\t(Exim 4.71) (envelope-from ) id 1dqIUk-00064d-Vz\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:34:51 -0400","from smtp.corp.redhat.com\n\t(int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby mx1.redhat.com (Postfix) with ESMTPS id DDBCD4A6EA;\n\tFri, 8 Sep 2017 12:34:48 +0000 (UTC)","from blackfin.pond.sub.org (ovpn-116-75.ams2.redhat.com\n\t[10.36.116.75])\n\tby smtp.corp.redhat.com (Postfix) with ESMTPS id 7BD6A600CA;\n\tFri, 8 Sep 2017 12:34:48 +0000 (UTC)","by blackfin.pond.sub.org (Postfix, from userid 1000)\n\tid 039C51138645; Fri, 8 Sep 2017 14:34:47 +0200 (CEST)"],"DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com DDBCD4A6EA","From":"Markus Armbruster ","To":"Pradeep Jagadeesh ","References":"<1504541267-36954-1-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<87d171fqzd.fsf@dusky.pond.sub.org>\n\t<426ebb0f-c603-3d1e-2ba0-963875b14394@huawei.com>","Date":"Fri, 08 Sep 2017 14:34:46 +0200","In-Reply-To":"<426ebb0f-c603-3d1e-2ba0-963875b14394@huawei.com> (Pradeep\n\tJagadeesh's message of \"Fri, 8 Sep 2017 14:19:16 +0200\")","Message-ID":"<87bmml8j2x.fsf@dusky.pond.sub.org>","User-Agent":"Gnus/5.13 (Gnus v5.13) Emacs/25.2 (gnu/linux)","MIME-Version":"1.0","Content-Type":"text/plain","X-Scanned-By":"MIMEDefang 2.79 on 10.5.11.11","X-Greylist":"Sender IP whitelisted, not delayed by milter-greylist-4.5.16\n\t(mx1.redhat.com [10.5.110.38]);\n\tFri, 08 Sep 2017 12:34:49 +0000 (UTC)","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic]\n\t[fuzzy]","X-Received-From":"209.132.183.28","Subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"alberto garcia ,\n\t\"Dr. David Alan Gilbert\" ,\n\tPradeep Jagadeesh ,\n\tqemu-devel@nongnu.org, greg kurz ,\n\tjani kokkonen ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"}},{"id":1765309,"web_url":"http://patchwork.ozlabs.org/comment/1765309/","msgid":"<93be19f0-2380-89e2-5c8e-c4dbe5405a9a@huawei.com>","list_archive_url":null,"date":"2017-09-08T12:49:31","subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","submitter":{"id":69883,"url":"http://patchwork.ozlabs.org/api/people/69883/","name":"Pradeep Jagadeesh","email":"pradeep.jagadeesh@huawei.com"},"content":"On 9/8/2017 2:34 PM, Markus Armbruster wrote:\n> Pradeep Jagadeesh writes:\n>\n>> On 9/8/2017 12:02 PM, Markus Armbruster wrote:\n>>> Pradeep Jagadeesh writes:\n> [...]\n>>>> diff --git a/qmp.c b/qmp.c\n>>>> index b86201e..eed91e5 100644\n>>>> --- a/qmp.c\n>>>> +++ b/qmp.c\n>>>> @@ -130,6 +130,20 @@ void qmp_cpu_add(int64_t id, Error **errp)\n>>>> }\n>>>> }\n>>>>\n>>>> +#if defined(_WIN64) || defined(_WIN32) || defined(__FreeBSD__)\n>>>> +\n>>>> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n>>>> +{\n>>>> + return;\n>>>> +}\n>>>> +\n>>>> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n>>>> +{\n>>>> + return NULL;\n>>>> +}\n>>>> +\n>>>> +#endif\n>>>> +\n>>>> #ifndef CONFIG_VNC\n>>>> /* If VNC support is enabled, the \"true\" query-vnc command is\n>>>> defined in the VNC subsystem */\n>>>\n>>> Why do we need *two* stubs for these functions, one here and one in\n>>> fsdev/qemu-fsdev-dummy.c?\n>> At two different platforms the build fails. So, there are needed.\n>\n> We don't add stubs to random places until the linker succeeds for all\n> the configurations we happen to test. We figure out *how* the linker\n> fails to correct our idea of where a certain stub belongs until we find\n> the one appropriate home for it.\n>\nOK, I will figure it out and send it across. But these were discussed \nsince the v1, now I do not remember on top of my mind. I will have to \ndig my old emails and send it across or I should reproduce them again.\n\n> If you can't figure that out, that's okay, just tell us how exactly the\n> linker fails for you, and we'll be happy to help.\nThanks for offering help!\n\n-Pradeep\n>","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xpcdJ0b0Kz9s71\n\tfor ;\n\tFri, 8 Sep 2017 22:51:14 +1000 (AEST)","from localhost ([::1]:45308 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1dqIka-0004Ic-7M\n\tfor incoming@patchwork.ozlabs.org; Fri, 08 Sep 2017 08:51:12 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:51200)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1dqIjz-0004Gg-OI\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:50:36 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1dqIjv-0002sH-3f\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:50:35 -0400","from lhrrgout.huawei.com ([194.213.3.17]:23215)\n\tby eggs.gnu.org with esmtps (TLS1.0:RSA_ARCFOUR_SHA1:16) (Exim 4.71)\n\t(envelope-from ) id 1dqIju-0002qr-Qf\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:50:31 -0400","from 172.18.7.190 (EHLO lhreml707-cah.china.huawei.com)\n\t([172.18.7.190])\n\tby lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued)\n\twith ESMTP id DVA85577; Fri, 08 Sep 2017 12:49:43 +0000 (GMT)","from [127.0.0.1] (10.210.165.15) by lhreml707-cah.china.huawei.com\n\t(10.201.108.48) with Microsoft SMTP Server id 14.3.301.0;\n\tFri, 8 Sep 2017 13:49:32 +0100"],"To":"Markus Armbruster ","References":"<1504541267-36954-1-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<87d171fqzd.fsf@dusky.pond.sub.org>\n\t<426ebb0f-c603-3d1e-2ba0-963875b14394@huawei.com>\n\t<87bmml8j2x.fsf@dusky.pond.sub.org>","From":"Pradeep Jagadeesh ","Message-ID":"<93be19f0-2380-89e2-5c8e-c4dbe5405a9a@huawei.com>","Date":"Fri, 8 Sep 2017 14:49:31 +0200","User-Agent":"Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101\n\tThunderbird/45.8.0","MIME-Version":"1.0","In-Reply-To":"<87bmml8j2x.fsf@dusky.pond.sub.org>","Content-Type":"text/plain; charset=\"windows-1252\"; format=flowed","Content-Transfer-Encoding":"7bit","X-Originating-IP":"[10.210.165.15]","X-CFilter-Loop":"Reflected","X-Mirapoint-Virus-RAPID-Raw":"score=unknown(0),\n\trefid=str=0001.0A0B0207.59B291E9.00A9, ss=1, re=0.000, recu=0.000,\n\treip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0,\n\tso=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32","X-Mirapoint-Loop-Id":"18d631d0913d2f2bd96c3b6aa0930ded","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.4.x-2.6.x [generic]\n\t[fuzzy]","X-Received-From":"194.213.3.17","Subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"alberto garcia ,\n\t\"Dr. David Alan Gilbert\" ,\n\tPradeep Jagadeesh ,\n\tqemu-devel@nongnu.org, greg kurz ,\n\tjani kokkonen ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"}},{"id":1765314,"web_url":"http://patchwork.ozlabs.org/comment/1765314/","msgid":"<8760ct8ib2.fsf@dusky.pond.sub.org>","list_archive_url":null,"date":"2017-09-08T12:51:29","subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","submitter":{"id":2645,"url":"http://patchwork.ozlabs.org/api/people/2645/","name":"Markus Armbruster","email":"armbru@redhat.com"},"content":"Alberto Garcia writes:\n\n> On Fri 08 Sep 2017 12:02:14 PM CEST, Markus Armbruster wrote:\n>\n>>> + fse = get_fsdev_fsentry(arg->has_id ? arg->id : NULL);\n>>\n>> !arg->has_id implies !arg->id.\n>\n> Hey Markus,\n>\n> I have the impression that I've also written code that never uses\n> arg->foo when arg->has_foo is false.\n>\n> Can we then assume that to be NULL/0 in all cases? Also for other data\n> types (int, bool, ...)?\n\nQAPI code always passes zero FOOs along with false has_FOOs. Anything\nthat doesn't is a bug. This is particular important when FOO is a\npointer; we don't want to pass around pointers pointing to random junk\nin the hope that everybody will obediently check their has_FOO before\ndereferencing.\n\nOther code might not always be as well-behaved. In random context,\nchecking has_FOO is probably safer.\n\nI still want to eliminate has_FOO for pointer-valued FOO. So much to\ndo, so little time!","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","ext-mx08.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx08.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=armbru@redhat.com"],"Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xpcgg3TfRz9s71\n\tfor ;\n\tFri, 8 Sep 2017 22:53:19 +1000 (AEST)","from localhost ([::1]:45316 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1dqImb-0006Jl-Io\n\tfor incoming@patchwork.ozlabs.org; Fri, 08 Sep 2017 08:53:17 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:51525)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1dqIky-00053O-V0\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:51:39 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1dqIku-0003ls-3o\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:51:37 -0400","from mx1.redhat.com ([209.132.183.28]:59514)\n\tby eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32)\n\t(Exim 4.71) (envelope-from ) id 1dqIkt-0003ke-UU\n\tfor qemu-devel@nongnu.org; Fri, 08 Sep 2017 08:51:32 -0400","from smtp.corp.redhat.com\n\t(int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby mx1.redhat.com (Postfix) with ESMTPS id E0A8FC057FA6;\n\tFri, 8 Sep 2017 12:51:30 +0000 (UTC)","from blackfin.pond.sub.org (ovpn-116-75.ams2.redhat.com\n\t[10.36.116.75])\n\tby smtp.corp.redhat.com (Postfix) with ESMTPS id AD33B600CA;\n\tFri, 8 Sep 2017 12:51:30 +0000 (UTC)","by blackfin.pond.sub.org (Postfix, from userid 1000)\n\tid 33B331138645; Fri, 8 Sep 2017 14:51:29 +0200 (CEST)"],"DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com E0A8FC057FA6","From":"Markus Armbruster ","To":"Alberto Garcia ","References":"<1504541267-36954-1-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<87d171fqzd.fsf@dusky.pond.sub.org>\n\t","Date":"Fri, 08 Sep 2017 14:51:29 +0200","In-Reply-To":" (Alberto Garcia's\n\tmessage of \"Fri, 08 Sep 2017 13:33:46 +0200\")","Message-ID":"<8760ct8ib2.fsf@dusky.pond.sub.org>","User-Agent":"Gnus/5.13 (Gnus v5.13) Emacs/25.2 (gnu/linux)","MIME-Version":"1.0","Content-Type":"text/plain","X-Scanned-By":"MIMEDefang 2.79 on 10.5.11.11","X-Greylist":"Sender IP whitelisted, not delayed by milter-greylist-4.5.16\n\t(mx1.redhat.com [10.5.110.32]);\n\tFri, 08 Sep 2017 12:51:31 +0000 (UTC)","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic]\n\t[fuzzy]","X-Received-From":"209.132.183.28","Subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"\"Dr. David Alan Gilbert\" ,\n\tPradeep Jagadeesh ,\n\tqemu-devel@nongnu.org, greg kurz ,\n\tPradeep Jagadeesh ,\n\tjani kokkonen ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"}},{"id":1766162,"web_url":"http://patchwork.ozlabs.org/comment/1766162/","msgid":"<51ae9f4f-fa74-9348-c355-ecd0566c0017@huawei.com>","list_archive_url":null,"date":"2017-09-11T09:23:05","subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","submitter":{"id":69883,"url":"http://patchwork.ozlabs.org/api/people/69883/","name":"Pradeep Jagadeesh","email":"pradeep.jagadeesh@huawei.com"},"content":"On 9/8/2017 2:34 PM, Markus Armbruster wrote:\n> Pradeep Jagadeesh writes:\n>\n>> On 9/8/2017 12:02 PM, Markus Armbruster wrote:\n>>> Pradeep Jagadeesh writes:\n> [...]\n>>>> diff --git a/qmp.c b/qmp.c\n>>>> index b86201e..eed91e5 100644\n>>>> --- a/qmp.c\n>>>> +++ b/qmp.c\n>>>> @@ -130,6 +130,20 @@ void qmp_cpu_add(int64_t id, Error **errp)\n>>>> }\n>>>> }\n>>>>\n>>>> +#if defined(_WIN64) || defined(_WIN32) || defined(__FreeBSD__)\n>>>> +\n>>>> +void qmp_fsdev_set_io_throttle(IOThrottle *arg, Error **errp)\n>>>> +{\n>>>> + return;\n>>>> +}\n>>>> +\n>>>> +IOThrottleList *qmp_query_fsdev_io_throttle(Error **errp)\n>>>> +{\n>>>> + return NULL;\n>>>> +}\n>>>> +\n>>>> +#endif\n>>>> +\n>>>> #ifndef CONFIG_VNC\n>>>> /* If VNC support is enabled, the \"true\" query-vnc command is\n>>>> defined in the VNC subsystem */\n>>>\n>>> Why do we need *two* stubs for these functions, one here and one in\n>>> fsdev/qemu-fsdev-dummy.c?\n>> At two different platforms the build fails. So, there are needed.\n>\n> We don't add stubs to random places until the linker succeeds for all\n> the configurations we happen to test. We figure out *how* the linker\n> fails to correct our idea of where a certain stub belongs until we find\n> the one appropriate home for it.\n>\n> If you can't figure that out, that's okay, just tell us how exactly the\n> linker fails for you, and we'll be happy to help.\nIf I do not have those stubs in qmp.c.The linking fails when I compile \nlike \"make docker-test-mingw@fedora\"\n\nThe stubs in qemu-fsdev-dummy.c required because when I configure on x86 \nas --disable-virtfs, then the compilation fails.\n\nSo, I need the stubs at both the places. Please let me know if I can \navoid some how.\n\nRegards,\nPradeep","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":"ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xrMvf0YRBz9s75\n\tfor ;\n\tMon, 11 Sep 2017 19:24:44 +1000 (AEST)","from localhost ([::1]:56303 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1drKxO-0002Xo-JD\n\tfor incoming@patchwork.ozlabs.org; Mon, 11 Sep 2017 05:24:42 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:43623)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1drKwy-0002XP-F2\n\tfor qemu-devel@nongnu.org; Mon, 11 Sep 2017 05:24:17 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1drKwu-0007Sn-GB\n\tfor qemu-devel@nongnu.org; Mon, 11 Sep 2017 05:24:16 -0400","from lhrrgout.huawei.com ([194.213.3.17]:23369)\n\tby eggs.gnu.org with esmtps (TLS1.0:RSA_ARCFOUR_SHA1:16) (Exim 4.71)\n\t(envelope-from ) id 1drKwu-0007QJ-8Z\n\tfor qemu-devel@nongnu.org; Mon, 11 Sep 2017 05:24:12 -0400","from 172.18.7.190 (EHLO LHREML714-CAH.china.huawei.com)\n\t([172.18.7.190])\n\tby lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued)\n\twith ESMTP id DVE83097; Mon, 11 Sep 2017 09:23:22 +0000 (GMT)","from [127.0.0.1] (10.210.166.117) by LHREML714-CAH.china.huawei.com\n\t(10.201.108.37) with Microsoft SMTP Server id 14.3.301.0;\n\tMon, 11 Sep 2017 10:23:11 +0100"],"To":"Markus Armbruster ","References":"<1504541267-36954-1-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<1504541267-36954-6-git-send-email-pradeep.jagadeesh@huawei.com>\n\t<87d171fqzd.fsf@dusky.pond.sub.org>\n\t<426ebb0f-c603-3d1e-2ba0-963875b14394@huawei.com>\n\t<87bmml8j2x.fsf@dusky.pond.sub.org>","From":"Pradeep Jagadeesh ","Message-ID":"<51ae9f4f-fa74-9348-c355-ecd0566c0017@huawei.com>","Date":"Mon, 11 Sep 2017 11:23:05 +0200","User-Agent":"Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101\n\tThunderbird/45.8.0","MIME-Version":"1.0","In-Reply-To":"<87bmml8j2x.fsf@dusky.pond.sub.org>","Content-Type":"text/plain; charset=\"windows-1252\"; format=flowed","Content-Transfer-Encoding":"7bit","X-Originating-IP":"[10.210.166.117]","X-CFilter-Loop":"Reflected","X-Mirapoint-Virus-RAPID-Raw":"score=unknown(0),\n\trefid=str=0001.0A020204.59B6560A.00FE, ss=1, re=0.000, recu=0.000,\n\treip=0.000, cl=1, cld=1, fgs=0, ip=0.0.0.0,\n\tso=2013-06-18 04:22:30, dmn=2013-03-21 17:37:32","X-Mirapoint-Loop-Id":"18d631d0913d2f2bd96c3b6aa0930ded","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.4.x-2.6.x [generic]\n\t[fuzzy]","X-Received-From":"194.213.3.17","Subject":"Re: [Qemu-devel] [PATCH v10 5/6] fsdev: QMP interface for throttling","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"alberto garcia ,\n\t\"Dr. David Alan Gilbert\" ,\n\tPradeep Jagadeesh ,\n\tqemu-devel@nongnu.org, greg kurz ,\n\tjani kokkonen ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"}}]