core: add support for IOReadIOPSMax and IOWriteIOPSMax

cgroup IO controller supports maximum limits for both bandwidth and IOPS but
systemd resource control currently only supports bandwidth limits.  This patch
adds support for IOReadIOPSMax and IOWriteIOPSMax when unified cgroup hierarchy
is in use.

It isn't difficult to also add BlockIOReadIOPS and BlockIOWriteIOPS for legacy
hierarchies but IO control on legacy hierarchies is half-broken anyway, so
let's leave it alone for now.

6 files changed, 38 insertions, 3 deletions

diff --git a/man/systemd.resource-control.xml b/man/systemd.resource-control.xml
index 313a49a959..8a95e1196b 100644
--- a/man/systemd.resource-control.xml
+++ b/man/systemd.resource-control.xml
@@ -337,6 +337,30 @@
       </varlistentry>
 
       <varlistentry>
+        <term><varname>IOReadIOPSMax=<replaceable>device</replaceable> <replaceable>IOPS</replaceable></varname></term>
+        <term><varname>IOWriteIOPSMax=<replaceable>device</replaceable> <replaceable>IOPS</replaceable></varname></term>
+
+        <listitem>
+          <para>Set the per-device overall block I/O IOs-Per-Second maximum limit for the executed processes, if the
+          unified control group hierarchy is used on the system. This limit is not work-conserving and the executed
+          processes are not allowed to use more even if the device has idle capacity.  Takes a space-separated pair of
+          a file path and an IOPS value to specify the device specific IOPS. The file path may be a path to a block
+          device node, or as any other file in which case the backing block device of the file system of the file is
+          used. If the IOPS is suffixed with K, M, G, or T, the specified IOPS is parsed as KiloIOPS, MegaIOPS,
+          GigaIOPS, or TeraIOPS, respectively, to the base of 1000. (Example:
+          "/dev/disk/by-path/pci-0000:00:1f.2-scsi-0:0:0:0 1K"). This controls the <literal>io.max</literal> control
+          group attributes. Use this option multiple times to set IOPS limits for multiple devices. For details about
+          this control group attribute, see <ulink
+          url="https://www.kernel.org/doc/Documentation/cgroup-v2.txt">cgroup-v2.txt</ulink>.
+          </para>
+
+          <para>Implies <literal>IOAccounting=true</literal>.</para>
+
+          <para>This setting is supported only if the unified control group hierarchy is used.</para>
+        </listitem>
+      </varlistentry>
+
+      <varlistentry>
         <term><varname>BlockIOAccounting=</varname></term>
 
         <listitem>
diff --git a/src/basic/cgroup-util.c b/src/basic/cgroup-util.c
index 8eab100748..7cdc97ee3c 100644
--- a/src/basic/cgroup-util.c
+++ b/src/basic/cgroup-util.c
@@ -2272,11 +2272,15 @@ int cg_weight_parse(const char *s, uint64_t *ret) {
 const uint64_t cgroup_io_limit_defaults[_CGROUP_IO_LIMIT_TYPE_MAX] = {
         [CGROUP_IO_RBPS_MAX]    = CGROUP_LIMIT_MAX,
         [CGROUP_IO_WBPS_MAX]    = CGROUP_LIMIT_MAX,
+        [CGROUP_IO_RIOPS_MAX]   = CGROUP_LIMIT_MAX,
+        [CGROUP_IO_WIOPS_MAX]   = CGROUP_LIMIT_MAX,
 };
 
 static const char* const cgroup_io_limit_type_table[_CGROUP_IO_LIMIT_TYPE_MAX] = {
         [CGROUP_IO_RBPS_MAX]    = "IOReadBandwidthMax",
         [CGROUP_IO_WBPS_MAX]    = "IOWriteBandwidthMax",
+        [CGROUP_IO_RIOPS_MAX]   = "IOReadIOPSMax",
+        [CGROUP_IO_WIOPS_MAX]   = "IOWriteIOPSMax",
 };
 
 DEFINE_STRING_TABLE_LOOKUP(cgroup_io_limit_type, CGroupIOLimitType);
diff --git a/src/basic/cgroup-util.h b/src/basic/cgroup-util.h
index 0d96fb6b76..4bb5291296 100644
--- a/src/basic/cgroup-util.h
+++ b/src/basic/cgroup-util.h
@@ -76,6 +76,8 @@ static inline bool CGROUP_WEIGHT_IS_OK(uint64_t x) {
 typedef enum CGroupIOLimitType {
         CGROUP_IO_RBPS_MAX,
         CGROUP_IO_WBPS_MAX,
+        CGROUP_IO_RIOPS_MAX,
+        CGROUP_IO_WIOPS_MAX,
 
         _CGROUP_IO_LIMIT_TYPE_MAX,
         _CGROUP_IO_LIMIT_TYPE_INVALID = -1
diff --git a/src/core/cgroup.c b/src/core/cgroup.c
index 0b902fa6f7..a54634469f 100644
--- a/src/core/cgroup.c
+++ b/src/core/cgroup.c
@@ -439,7 +439,7 @@ void cgroup_context_apply(CGroupContext *c, CGroupMask mask, const char *path, M
 
                 LIST_FOREACH_SAFE(device_limits, l, next, c->io_device_limits) {
                         char limit_bufs[_CGROUP_IO_LIMIT_TYPE_MAX][DECIMAL_STR_MAX(uint64_t)];
-                        char buf[DECIMAL_STR_MAX(dev_t)*2+2+(5+DECIMAL_STR_MAX(uint64_t)+1)*2];
+                        char buf[DECIMAL_STR_MAX(dev_t)*2+2+(6+DECIMAL_STR_MAX(uint64_t)+1)*4];
                         CGroupIOLimitType type;
                         dev_t dev;
                         unsigned n = 0;
@@ -458,8 +458,9 @@ void cgroup_context_apply(CGroupContext *c, CGroupMask mask, const char *path, M
                                 }
                         }
 
-                        xsprintf(buf, "%u:%u rbps=%s wbps=%s\n", major(dev), minor(dev),
-                                 limit_bufs[CGROUP_IO_RBPS_MAX], limit_bufs[CGROUP_IO_WBPS_MAX]);
+                        xsprintf(buf, "%u:%u rbps=%s wbps=%s riops=%s wiops=%s\n", major(dev), minor(dev),
+                                 limit_bufs[CGROUP_IO_RBPS_MAX], limit_bufs[CGROUP_IO_WBPS_MAX],
+                                 limit_bufs[CGROUP_IO_RIOPS_MAX], limit_bufs[CGROUP_IO_WIOPS_MAX]);
                         r = cg_set_attribute("io", path, "io.max", buf);
                         if (r < 0)
                                 log_full_errno(IN_SET(r, -ENOENT, -EROFS, -EACCES) ? LOG_DEBUG : LOG_WARNING, r,
diff --git a/src/core/dbus-cgroup.c b/src/core/dbus-cgroup.c
index 4372828e30..29c2edaf6b 100644
--- a/src/core/dbus-cgroup.c
+++ b/src/core/dbus-cgroup.c
@@ -213,6 +213,8 @@ const sd_bus_vtable bus_cgroup_vtable[] = {
         SD_BUS_PROPERTY("IODeviceWeight", "a(st)", property_get_io_device_weight, 0, 0),
         SD_BUS_PROPERTY("IOReadBandwidthMax", "a(st)", property_get_io_device_limits, 0, 0),
         SD_BUS_PROPERTY("IOWriteBandwidthMax", "a(st)", property_get_io_device_limits, 0, 0),
+        SD_BUS_PROPERTY("IOReadIOPSMax", "a(st)", property_get_io_device_limits, 0, 0),
+        SD_BUS_PROPERTY("IOWriteIOPSMax", "a(st)", property_get_io_device_limits, 0, 0),
         SD_BUS_PROPERTY("BlockIOAccounting", "b", bus_property_get_bool, offsetof(CGroupContext, blockio_accounting), 0),
         SD_BUS_PROPERTY("BlockIOWeight", "t", NULL, offsetof(CGroupContext, blockio_weight), 0),
         SD_BUS_PROPERTY("StartupBlockIOWeight", "t", NULL, offsetof(CGroupContext, startup_blockio_weight), 0),
diff --git a/src/core/load-fragment-gperf.gperf.m4 b/src/core/load-fragment-gperf.gperf.m4
index ad45611d9d..8193418980 100644
--- a/src/core/load-fragment-gperf.gperf.m4
+++ b/src/core/load-fragment-gperf.gperf.m4
@@ -126,6 +126,8 @@ $1.StartupIOWeight,              config_parse_io_weight,             0,
 $1.IODeviceWeight,               config_parse_io_device_weight,      0,                             offsetof($1, cgroup_context)
 $1.IOReadBandwidthMax,           config_parse_io_limit,              0,                             offsetof($1, cgroup_context)
 $1.IOWriteBandwidthMax,          config_parse_io_limit,              0,                             offsetof($1, cgroup_context)
+$1.IOReadIOPSMax,                config_parse_io_limit,              0,                             offsetof($1, cgroup_context)
+$1.IOWriteIOPSMax,               config_parse_io_limit,              0,                             offsetof($1, cgroup_context)
 $1.BlockIOAccounting,            config_parse_bool,                  0,                             offsetof($1, cgroup_context.blockio_accounting)
 $1.BlockIOWeight,                config_parse_blockio_weight,        0,                             offsetof($1, cgroup_context.blockio_weight)
 $1.StartupBlockIOWeight,         config_parse_blockio_weight,        0,                             offsetof($1, cgroup_context.startup_blockio_weight)