| Message ID | 20230713114915.74671-1-linmiaohe@huawei.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | mm: remove some useless comments of node_stat_item | expand |
On Thu, Jul 13, 2023 at 07:49:15PM +0800, Miaohe Lin wrote: > Some comments of node_stat_item are not that helpful and even confusing, > so remove them. No functional change intended. No, that's very useful and important. Why does it confuse you? > enum node_stat_item { > NR_LRU_BASE, > NR_INACTIVE_ANON = NR_LRU_BASE, /* must match order of LRU_[IN]ACTIVE */ > - NR_ACTIVE_ANON, /* " " " " " */ > - NR_INACTIVE_FILE, /* " " " " " */ > - NR_ACTIVE_FILE, /* " " " " " */ > - NR_UNEVICTABLE, /* " " " " " */ > + NR_ACTIVE_ANON, > + NR_INACTIVE_FILE, > + NR_ACTIVE_FILE, > + NR_UNEVICTABLE, > NR_SLAB_RECLAIMABLE_B, > NR_SLAB_UNRECLAIMABLE_B, > NR_ISOLATED_ANON, /* Temporary isolated pages from anon lru */ > -- > 2.33.0 > >
On 2023/7/13 20:10, Matthew Wilcox wrote: > On Thu, Jul 13, 2023 at 07:49:15PM +0800, Miaohe Lin wrote: >> Some comments of node_stat_item are not that helpful and even confusing, >> so remove them. No functional change intended. > > No, that's very useful and important. Why does it confuse you? Thanks for your quick respond. I just can't figure out what these comments want to tell. Could you help explain these? Thanks.
On Thu, Jul 13, 2023 at 08:18:29PM +0800, Miaohe Lin wrote: > On 2023/7/13 20:10, Matthew Wilcox wrote: > > On Thu, Jul 13, 2023 at 07:49:15PM +0800, Miaohe Lin wrote: > >> Some comments of node_stat_item are not that helpful and even confusing, > >> so remove them. No functional change intended. > > > > No, that's very useful and important. Why does it confuse you? > > Thanks for your quick respond. > > I just can't figure out what these comments want to tell. Could you help explain these? Don't snip the thing you want explained to you! NR_INACTIVE_ANON = NR_LRU_BASE, /* must match order of LRU_[IN]ACTIVE */ - NR_ACTIVE_ANON, /* " " " " " */ - NR_INACTIVE_FILE, /* " " " " " */ - NR_ACTIVE_FILE, /* " " " " " */ - NR_UNEVICTABLE, /* " " " " " */ + NR_ACTIVE_ANON, + NR_INACTIVE_FILE, + NR_ACTIVE_FILE, + NR_UNEVICTABLE, What this is communicating to me is that these five items (NR_INACTIVE_ANON to NR_UNEVICTABLE) must stay in the same order with LRU_INACTIVE and LRU_ACTIVE. By removing the ditto-marks from the subsequent four lines, you've made the comment say that this one line must stay in the same order as LRU_INACTIVE and LRU_ACTIVE ... which makes no sense at all.
On 2023/7/13 20:31, Matthew Wilcox wrote: > On Thu, Jul 13, 2023 at 08:18:29PM +0800, Miaohe Lin wrote: >> On 2023/7/13 20:10, Matthew Wilcox wrote: >>> On Thu, Jul 13, 2023 at 07:49:15PM +0800, Miaohe Lin wrote: >>>> Some comments of node_stat_item are not that helpful and even confusing, >>>> so remove them. No functional change intended. >>> >>> No, that's very useful and important. Why does it confuse you? >> >> Thanks for your quick respond. >> >> I just can't figure out what these comments want to tell. Could you help explain these? > > Don't snip the thing you want explained to you! > > NR_INACTIVE_ANON = NR_LRU_BASE, /* must match order of LRU_[IN]ACTIVE */ > - NR_ACTIVE_ANON, /* " " " " " */ > - NR_INACTIVE_FILE, /* " " " " " */ > - NR_ACTIVE_FILE, /* " " " " " */ > - NR_UNEVICTABLE, /* " " " " " */ > + NR_ACTIVE_ANON, > + NR_INACTIVE_FILE, > + NR_ACTIVE_FILE, > + NR_UNEVICTABLE, > > What this is communicating to me is that these five items > (NR_INACTIVE_ANON to NR_UNEVICTABLE) must stay in the same order with > LRU_INACTIVE and LRU_ACTIVE. By removing the ditto-marks from the > subsequent four lines, you've made the comment say that this one line > must stay in the same order as LRU_INACTIVE and LRU_ACTIVE ... which > makes no sense at all. I see. Many thanks for your kind explanation. :)
On 7/13/23 18:01, Matthew Wilcox wrote: > On Thu, Jul 13, 2023 at 08:18:29PM +0800, Miaohe Lin wrote: >> On 2023/7/13 20:10, Matthew Wilcox wrote: >>> On Thu, Jul 13, 2023 at 07:49:15PM +0800, Miaohe Lin wrote: >>>> Some comments of node_stat_item are not that helpful and even confusing, >>>> so remove them. No functional change intended. >>> >>> No, that's very useful and important. Why does it confuse you? >> >> Thanks for your quick respond. >> >> I just can't figure out what these comments want to tell. Could you help explain these? > > Don't snip the thing you want explained to you! > > NR_INACTIVE_ANON = NR_LRU_BASE, /* must match order of LRU_[IN]ACTIVE */ > - NR_ACTIVE_ANON, /* " " " " " */ > - NR_INACTIVE_FILE, /* " " " " " */ > - NR_ACTIVE_FILE, /* " " " " " */ > - NR_UNEVICTABLE, /* " " " " " */ > + NR_ACTIVE_ANON, > + NR_INACTIVE_FILE, > + NR_ACTIVE_FILE, > + NR_UNEVICTABLE, > > What this is communicating to me is that these five items > (NR_INACTIVE_ANON to NR_UNEVICTABLE) must stay in the same order with > LRU_INACTIVE and LRU_ACTIVE. By removing the ditto-marks from the > subsequent four lines, you've made the comment say that this one line > must stay in the same order as LRU_INACTIVE and LRU_ACTIVE ... which > makes no sense at all. Just wondering - would it be better to repeat these comments in words for each line than use "ditto-marks" ?
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 4106fbc5b4b3..844ed29cc260 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -154,10 +154,10 @@ enum zone_stat_item { enum node_stat_item { NR_LRU_BASE, NR_INACTIVE_ANON = NR_LRU_BASE, /* must match order of LRU_[IN]ACTIVE */ - NR_ACTIVE_ANON, /* " " " " " */ - NR_INACTIVE_FILE, /* " " " " " */ - NR_ACTIVE_FILE, /* " " " " " */ - NR_UNEVICTABLE, /* " " " " " */ + NR_ACTIVE_ANON, + NR_INACTIVE_FILE, + NR_ACTIVE_FILE, + NR_UNEVICTABLE, NR_SLAB_RECLAIMABLE_B, NR_SLAB_UNRECLAIMABLE_B, NR_ISOLATED_ANON, /* Temporary isolated pages from anon lru */
Some comments of node_stat_item are not that helpful and even confusing, so remove them. No functional change intended. Signed-off-by: Miaohe Lin <linmiaohe@huawei.com> --- include/linux/mmzone.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)