From: Mike Christie <mchristi@redhat.com>
To: "Nicholas A. Bellinger" <nab@linux-iscsi.org>,
"Bryant G. Ly" <bryantly@linux.vnet.ibm.com>
Cc: seroyer@linux.vnet.ibm.com, linux-scsi@vger.kernel.org,
target-devel@vger.kernel.org, "Bryant G. Ly" <bgly@us.ibm.com>
Subject: Re: [PATCH v4 0/5] tcmu: Add Type of reconfig into netlink
Date: Mon, 12 Jun 2017 01:43:46 -0500 [thread overview]
Message-ID: <593E3822.5030408@redhat.com> (raw)
In-Reply-To: <593DAFE8.80705@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 2124 bytes --]
On 06/11/2017 04:02 PM, Mike Christie wrote:
> On 06/09/2017 01:11 AM, Nicholas A. Bellinger wrote:
>> Hi Bryant & Co,
>>
>> On Tue, 2017-06-06 at 09:28 -0500, Bryant G. Ly wrote:
>>> From: "Bryant G. Ly" <bgly@us.ibm.com>
>>>
>>> This patch consists of adding a netlink to allow for reconfiguration
>>> of a device in tcmu.
>>>
>>> It also changes and adds some attributes that are reconfigurable:
>>> write_cache, device size, and device path.
>>>
>>> V2 - Fixes kfree in tcmu: Make dev_config configurable
>>> V3 - Fixes spelling error
>>> V4 - change strcpy to strlcpy for tcmu_dev_path_store and move
>>> tcmu_reconfig_type into target_core_user.h
>>>
>>>
>>> Bryant G. Ly (5):
>>> tcmu: Support emulate_write_cache
>>> tcmu: Add netlink for device reconfiguration
>>> tcmu: Make dev_size configurable via userspace
>>> tcmu: Make dev_config configurable
>>> tcmu: Add Type of reconfig into netlink
>>>
>>> drivers/target/target_core_user.c | 152 ++++++++++++++++++++++++++++++++--
>>> include/uapi/linux/target_core_user.h | 9 ++
>>> 2 files changed, 155 insertions(+), 6 deletions(-)
>>>
>>
>> AFAICT, it looks like all of the review comments have been addressed in
>> -v4.
>>
>> Applied to target-pending/for-next, with MNC's (pseudo) Reviewed-by's
>> added for #3-#5.
>>
>> Please let me know if anything else needs to be changed.
>>
>
> The patches look ok. Thanks. Could you just merge the attached patch
> into "[PATCH v4 5/5] tcmu: Add Type of reconfig into netlink" or into
> the patchset after it? It just makes some of the names a little less
> generic and only returns the reconfig attr for reconfig commands.
>
Actually Nick, do not merge the last patch. I have a lot more fixes/changes.
Bryant, could you test and adapt your userspace patches for the attached
patch build over Nicks for-next branch.
Basically, the patch just has use pass the value being reconfigured with
the netlink event. Along the way, it fixes a couple bugs.
Nick, when we have tested the patch, then I can submit as a formal
patchset, or you can fold into the existing patches or whatever you prefer.
[-- Attachment #2: 0001-tcmu-reconfigure-netlink-attr-changes.patch --]
[-- Type: text/x-patch, Size: 7087 bytes --]
>From dad38c89d92a83b19b285431268818a74fe48fab Mon Sep 17 00:00:00 2001
From: Mike Christie <mchristi@redhat.com>
Date: Mon, 12 Jun 2017 01:34:28 -0500
Subject: [PATCH 1/1] tcmu: reconfigure netlink attr changes v2
1. TCMU_ATTR_TYPE is too generic. Drop it and use per reconfig
type attrs.
2. Only return the reconfig type when it is a TCMU_CMD_RECONFIG_DEVICE command.
3. CONFIG_* type is not needed. We can pass the value along with an
ATTR to userspace, so it does not need to read sysfs/configfs.
4. Fix leak in tcmu_dev_path_store and rename to dev_config to
reflect it is more than just a path that can be changed.
6. Don't update kernel struct value if netlink sending fails.
Signed-off-by: Mike Christie <mchristi@redhat.com>
---
drivers/target/target_core_user.c | 73 +++++++++++++++++++++--------------
include/uapi/linux/target_core_user.h | 12 ++----
2 files changed, 48 insertions(+), 37 deletions(-)
diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c
index afc1fd6..1712c42 100644
--- a/drivers/target/target_core_user.c
+++ b/drivers/target/target_core_user.c
@@ -1177,7 +1177,8 @@ static int tcmu_release(struct uio_info *info, struct inode *inode)
}
static int tcmu_netlink_event(enum tcmu_genl_cmd cmd, const char *name,
- int minor, int type)
+ int minor, int reconfig_attr,
+ const void *reconfig_data)
{
struct sk_buff *skb;
void *msg_header;
@@ -1199,9 +1200,27 @@ static int tcmu_netlink_event(enum tcmu_genl_cmd cmd, const char *name,
if (ret < 0)
goto free_skb;
- ret = nla_put_u32(skb, TCMU_ATTR_TYPE, type);
- if (ret < 0)
- goto free_skb;
+ if (cmd == TCMU_CMD_RECONFIG_DEVICE) {
+ switch (reconfig_attr) {
+ case TCMU_ATTR_DEV_CFG:
+ ret = nla_put_string(skb, reconfig_attr, reconfig_data);
+ break;
+ case TCMU_ATTR_DEV_SIZE:
+ ret = nla_put_u64_64bit(skb, reconfig_attr,
+ *((u64 *)reconfig_data),
+ TCMU_ATTR_PAD);
+ break;
+ case TCMU_ATTR_WRITECACHE:
+ ret = nla_put_u8(skb, reconfig_attr,
+ *((u8 *)reconfig_data));
+ break;
+ default:
+ BUG();
+ }
+
+ if (ret < 0)
+ goto free_skb;
+ }
genlmsg_end(skb, msg_header);
@@ -1306,7 +1325,7 @@ static int tcmu_configure_device(struct se_device *dev)
kref_get(&udev->kref);
ret = tcmu_netlink_event(TCMU_CMD_ADDED_DEVICE, udev->uio_info.name,
- udev->uio_info.uio_dev->minor, NO_RECONFIG);
+ udev->uio_info.uio_dev->minor, 0, NULL);
if (ret)
goto err_netlink;
@@ -1388,7 +1407,7 @@ static void tcmu_free_device(struct se_device *dev)
if (tcmu_dev_configured(udev)) {
tcmu_netlink_event(TCMU_CMD_REMOVED_DEVICE, udev->uio_info.name,
- udev->uio_info.uio_dev->minor, NO_RECONFIG);
+ udev->uio_info.uio_dev->minor, 0, NULL);
uio_unregister_device(&udev->uio_info);
}
@@ -1553,7 +1572,7 @@ static ssize_t tcmu_cmd_time_out_store(struct config_item *item, const char *pag
}
CONFIGFS_ATTR(tcmu_, cmd_time_out);
-static ssize_t tcmu_dev_path_show(struct config_item *item, char *page)
+static ssize_t tcmu_dev_config_show(struct config_item *item, char *page)
{
struct se_dev_attrib *da = container_of(to_config_group(item),
struct se_dev_attrib, da_group);
@@ -1562,37 +1581,34 @@ static ssize_t tcmu_dev_path_show(struct config_item *item, char *page)
return snprintf(page, PAGE_SIZE, "%s\n", udev->dev_config);
}
-static ssize_t tcmu_dev_path_store(struct config_item *item, const char *page,
- size_t count)
+static ssize_t tcmu_dev_config_store(struct config_item *item, const char *page,
+ size_t count)
{
struct se_dev_attrib *da = container_of(to_config_group(item),
struct se_dev_attrib, da_group);
struct tcmu_dev *udev = TCMU_DEV(da->da_dev);
- char *copy = NULL;
- int ret;
+ int ret, len;
- copy = kstrdup(page, GFP_KERNEL);
- if (!copy) {
- kfree(copy);
+ len = strlen(page);
+ if (!len || len > TCMU_CONFIG_LEN - 1)
return -EINVAL;
- }
- strlcpy(udev->dev_config, copy, TCMU_CONFIG_LEN);
/* Check if device has been configured before */
if (tcmu_dev_configured(udev)) {
ret = tcmu_netlink_event(TCMU_CMD_RECONFIG_DEVICE,
udev->uio_info.name,
udev->uio_info.uio_dev->minor,
- CONFIG_PATH);
+ TCMU_ATTR_DEV_CFG, page);
if (ret) {
pr_err("Unable to reconfigure device\n");
return ret;
}
}
+ strlcpy(udev->dev_config, page, TCMU_CONFIG_LEN);
return count;
}
-CONFIGFS_ATTR(tcmu_, dev_path);
+CONFIGFS_ATTR(tcmu_, dev_config);
static ssize_t tcmu_dev_size_show(struct config_item *item, char *page)
{
@@ -1609,26 +1625,25 @@ static ssize_t tcmu_dev_size_store(struct config_item *item, const char *page,
struct se_dev_attrib *da = container_of(to_config_group(item),
struct se_dev_attrib, da_group);
struct tcmu_dev *udev = TCMU_DEV(da->da_dev);
- unsigned long val;
+ u64 val;
int ret;
- ret = kstrtoul(page, 0, &val);
+ ret = kstrtou64(page, 0, &val);
if (ret < 0)
return ret;
- udev->dev_size = val;
/* Check if device has been configured before */
if (tcmu_dev_configured(udev)) {
ret = tcmu_netlink_event(TCMU_CMD_RECONFIG_DEVICE,
udev->uio_info.name,
udev->uio_info.uio_dev->minor,
- CONFIG_SIZE);
+ TCMU_ATTR_DEV_SIZE, &val);
if (ret) {
pr_err("Unable to reconfigure device\n");
return ret;
}
}
-
+ udev->dev_size = val;
return count;
}
CONFIGFS_ATTR(tcmu_, dev_size);
@@ -1648,33 +1663,33 @@ static ssize_t tcmu_emulate_write_cache_store(struct config_item *item,
struct se_dev_attrib *da = container_of(to_config_group(item),
struct se_dev_attrib, da_group);
struct tcmu_dev *udev = TCMU_DEV(da->da_dev);
- int val;
+ u8 val;
int ret;
- ret = kstrtouint(page, 0, &val);
+ ret = kstrtou8(page, 0, &val);
if (ret < 0)
return ret;
- da->emulate_write_cache = val;
-
/* Check if device has been configured before */
if (tcmu_dev_configured(udev)) {
ret = tcmu_netlink_event(TCMU_CMD_RECONFIG_DEVICE,
udev->uio_info.name,
udev->uio_info.uio_dev->minor,
- CONFIG_WRITECACHE);
+ TCMU_ATTR_WRITECACHE, &val);
if (ret) {
pr_err("Unable to reconfigure device\n");
return ret;
}
}
+
+ da->emulate_write_cache = val;
return count;
}
CONFIGFS_ATTR(tcmu_, emulate_write_cache);
struct configfs_attribute *tcmu_attrib_attrs[] = {
&tcmu_attr_cmd_time_out,
- &tcmu_attr_dev_path,
+ &tcmu_attr_dev_config,
&tcmu_attr_dev_size,
&tcmu_attr_emulate_write_cache,
NULL,
diff --git a/include/uapi/linux/target_core_user.h b/include/uapi/linux/target_core_user.h
index 5b00e35..4bfc9a1 100644
--- a/include/uapi/linux/target_core_user.h
+++ b/include/uapi/linux/target_core_user.h
@@ -139,16 +139,12 @@ enum tcmu_genl_attr {
TCMU_ATTR_UNSPEC,
TCMU_ATTR_DEVICE,
TCMU_ATTR_MINOR,
- TCMU_ATTR_TYPE,
+ TCMU_ATTR_PAD,
+ TCMU_ATTR_DEV_CFG,
+ TCMU_ATTR_DEV_SIZE,
+ TCMU_ATTR_WRITECACHE,
__TCMU_ATTR_MAX,
};
#define TCMU_ATTR_MAX (__TCMU_ATTR_MAX - 1)
-enum tcmu_reconfig_types {
- NO_RECONFIG,
- CONFIG_PATH,
- CONFIG_SIZE,
- CONFIG_WRITECACHE,
-};
-
#endif
--
1.8.3.1
next prev parent reply other threads:[~2017-06-12 6:43 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-06 14:28 [PATCH v4 0/5] tcmu: Add Type of reconfig into netlink Bryant G. Ly
2017-06-06 14:28 ` [PATCH v4 1/5] tcmu: Support emulate_write_cache Bryant G. Ly
2017-06-06 14:28 ` [PATCH v4 2/5] tcmu: Add netlink for device reconfiguration Bryant G. Ly
2017-06-06 14:28 ` [PATCH v4 3/5] tcmu: Make dev_size configurable via userspace Bryant G. Ly
2017-06-06 14:28 ` [PATCH v4 4/5] tcmu: Make dev_config configurable Bryant G. Ly
2017-06-06 14:28 ` [PATCH v4 5/5] tcmu: Add Type of reconfig into netlink Bryant G. Ly
2017-06-09 6:11 ` [PATCH v4 0/5] " Nicholas A. Bellinger
2017-06-11 21:02 ` Mike Christie
2017-06-12 6:43 ` Mike Christie [this message]
2017-06-30 7:31 ` Nicholas A. Bellinger
2017-06-30 16:58 ` Mike Christie
2017-06-30 17:51 ` Mike Christie
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=593E3822.5030408@redhat.com \
--to=mchristi@redhat.com \
--cc=bgly@us.ibm.com \
--cc=bryantly@linux.vnet.ibm.com \
--cc=linux-scsi@vger.kernel.org \
--cc=nab@linux-iscsi.org \
--cc=seroyer@linux.vnet.ibm.com \
--cc=target-devel@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.