Message ID | 4AEFD544.6040602@gmail.com |
---|---|
State | Not Applicable, archived |
Delegated to: | David Miller |
Headers | show |
Eric Dumazet <eric.dumazet@gmail.com> writes: > Eric Dumazet a écrit : > >> Its curious because in my tests the biggest problems come from >> kernel/sysctl.c (__register_sysctl_paths) consuming 80% of cpu >> in following attempt to create 20.000 devices I bet that is Al's cute glue all the sysctl data structures together patch. It improves readdir and lookup at a small cost at registration time. >> (disable hotplug before trying this, and ipv6 too !) >> modprobe dummy numdummies=20000 >> I believe we should address __register_sysctl_paths() scalability >> problems too. Agreed. >> I dont know what is the 'sentinel' we allocate after each struct ctl_table >> But I suspect we could reduce size requirement of the 'sentinel' to include >> only needed fields for the sentinel (and move them at start of ctl_table) The sentinel is just a NULL terminator. > Here is the patch to reduce ram usage of sysctl : > > [PATCH] sysctl: reduce ram usage by 40 % > > We currently reserve space for a so called sentinel, a full struct ctl_table > for each ctl_table. We can cheat a bit since only needed fields of a sentinel > are ctl_name and procname. Add a new structure (struct ctl_table_sentinel) > that includes a full ctl_table and only required part of a sentinel. Before we address sysctl I would like to get out my patchset that makes sys_sysctl a wrapper around the ascii version of /proc/sys/net. Once that goes in it becomes much easier to do things and perform radical surgery on sysctl. Little things like .ctl_name and .strategy go away. Have you happened to look at the other cost of /proc proper? Hmm. Except for /proc/net/dev_snmp6 it doesn't look like we keep per interface directories in proc so without ivp6 you won't see the proc generic code at all. The practical consequence is if /proc/net/dev_snmp6 is not painful during registration right now we can probably convert all of /proc/sys/net to proc generic after my other changes are in. Eric -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 1e4743e..6a1b1d5 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -1050,8 +1050,10 @@ extern ctl_handler sysctl_ms_jiffies; /* A sysctl table is an array of struct ctl_table: */ struct ctl_table { - int ctl_name; /* Binary ID */ + /* ctl_name and procname must be first fields (check sentinel) */ + int ctl_name; /* Binary ID */ const char *procname; /* Text ID for /proc/sys, or zero */ + void *data; int maxlen; mode_t mode; @@ -1063,6 +1065,15 @@ struct ctl_table void *extra2; }; +/* ctl_table_sentinel : a ctl_table followed by a sentinel + * (null ctl & procname) + */ +struct ctl_table_sentinel { + struct ctl_table table; + int ctl_name; + const char *procname; +}; + struct ctl_table_root { struct list_head root_list; struct ctl_table_set default_set; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 0d949c5..5d29dd8 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2063,7 +2063,8 @@ struct ctl_table_header *__register_sysctl_paths( const struct ctl_path *path, struct ctl_table *table) { struct ctl_table_header *header; - struct ctl_table *new, **prevp; + struct ctl_table_sentinel *new; + struct ctl_table **prevp; unsigned int n, npath; struct ctl_table_set *set; @@ -2080,24 +2081,24 @@ struct ctl_table_header *__register_sysctl_paths( * worry about freeing additional memory in unregister_sysctl_table. */ header = kzalloc(sizeof(struct ctl_table_header) + - (2 * npath * sizeof(struct ctl_table)), GFP_KERNEL); + (npath * sizeof(struct ctl_table_sentinel)), GFP_KERNEL); if (!header) return NULL; - new = (struct ctl_table *) (header + 1); + new = (struct ctl_table_sentinel *) (header + 1); /* Now connect the dots */ prevp = &header->ctl_table; for (n = 0; n < npath; ++n, ++path) { /* Copy the procname */ - new->procname = path->procname; - new->ctl_name = path->ctl_name; - new->mode = 0555; + new->table.procname = path->procname; + new->table.ctl_name = path->ctl_name; + new->table.mode = 0555; - *prevp = new; - prevp = &new->child; + *prevp = &new->table; + prevp = &new->table.child; - new += 2; + new++; } *prevp = table; header->ctl_table_arg = table;