[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH RFC PKS/PMEM 48/58] drivers/md: Utilize new kmap_thread()
- To: ira.weiny@xxxxxxxxx
- From: Coly Li <colyli@xxxxxxx>
- Date: Sat, 10 Oct 2020 10:20:34 +0800
- Autocrypt: addr=colyli@xxxxxxx; keydata= mQINBFYX6S8BEAC9VSamb2aiMTQREFXK4K/W7nGnAinca7MRuFUD4JqWMJ9FakNRd/E0v30F qvZ2YWpidPjaIxHwu3u9tmLKqS+2vnP0k7PRHXBYbtZEMpy3kCzseNfdrNqwJ54A430BHf2S GMVRVENiScsnh4SnaYjFVvB8SrlhTsgVEXEBBma5Ktgq9YSoy5miatWmZvHLFTQgFMabCz/P j5/xzykrF6yHo0rHZtwzQzF8rriOplAFCECp/t05+OeHHxjSqSI0P/G79Ll+AJYLRRm9til/ K6yz/1hX5xMToIkYrshDJDrUc8DjEpISQQPhG19PzaUf3vFpmnSVYprcWfJWsa2wZyyjRFkf J51S82WfclafNC6N7eRXedpRpG6udUAYOA1YdtlyQRZa84EJvMzW96iSL1Gf+ZGtRuM3k49H 1wiWOjlANiJYSIWyzJjxAd/7Xtiy/s3PRKL9u9y25ftMLFa1IljiDG+mdY7LyAGfvdtIkanr iBpX4gWXd7lNQFLDJMfShfu+CTMCdRzCAQ9hIHPmBeZDJxKq721CyBiGAhRxDN+TYiaG/UWT 7IB7LL4zJrIe/xQ8HhRO+2NvT89o0LxEFKBGg39yjTMIrjbl2ZxY488+56UV4FclubrG+t16 r2KrandM7P5RjR+cuHhkKseim50Qsw0B+Eu33Hjry7YCihmGswARAQABtBhDb2x5IExpIDxj b2x5bGlAc3VzZS5kZT6JAlYEEwEIAEACGyMHCwkIBwMCAQYVCAIJCgsEFgIDAQIeAQIXgBYh BOo+RS/0+Uhgjej60Mc5B5Nrffj8BQJcR84dBQkY++fuAAoJEMc5B5Nrffj8ixcP/3KAKg1X EcoW4u/0z+Ton5rCyb/NpAww8MuRjNW82UBUac7yCi1y3OW7NtLjuBLw5SaVG5AArb7IF3U0 qTOobqfl5XHsT0o5wFHZaKUrnHb6y7V3SplsJWfkP3JmOooJsQB3z3K96ZTkFelsNb0ZaBRu gV+LA4MomhQ+D3BCDR1it1OX/tpvm2uaDF6s/8uFtcDEM9eQeqATN/QAJ49nvU/I8zDSY9rc 0x9mP0x+gH4RccbnoPu/rUG6Fm1ZpLrbb6NpaYBBJ/V1BC4lIOjnd24bsoQrQmnJn9dSr60X 1MY60XDszIyzRw7vbJcUn6ZzPNFDxFFT9diIb+wBp+DD8ZlD/hnVpl4f921ZbvfOSsXAJrKB 1hGY17FPwelp1sPcK2mDT+pfHEMV+OQdZzD2OCKtza/5IYismJJm3oVUYMogb5vDNAw9X2aP XgwUuG+FDEFPamFMUwIfzYHcePfqf0mMsaeSgtA/xTxzx/0MLjUJHl46Bc0uKDhv7QUyGz0j Ywgr2mHTvG+NWQ/mDeHNGkcnsnp3IY7koDHnN2xMFXzY4bn9m8ctqKo2roqjCzoxD/njoAhf KBzdybLHATqJG/yiZSbCxDA1n/J4FzPyZ0rNHUAJ/QndmmVspE9syFpFCKigvvyrzm016+k+ FJ59Q6RG4MSy/+J565Xj+DNY3/dCuQINBFYX6S8BEADZP+2cl4DRFaSaBms08W8/smc5T2CO YhAoygZn71rB7Djml2ZdvrLRjR8Qbn0Q/2L2gGUVc63pJnbrjlXSx2LfAFE0SlfYIJ11aFdF 9w7RvqWByQjDJor3Z0fWvPExplNgMvxpD0U0QrVT5dIGTx9hadejCl/ug09Lr6MPQn+a4+qs aRWwgCSHaIuDkH3zI1MJXiqXXFKUzJ/Fyx6R72rqiMPHH2nfwmMu6wOXAXb7+sXjZz5Po9GJ g2OcEc+rpUtKUJGyeQsnCDxUcqJXZDBi/GnhPCcraQuqiQ7EGWuJfjk51vaI/rW4bZkA9yEP B9rBYngbz7cQymUsfxuTT8OSlhxjP3l4ZIZFKIhDaQeZMj8pumBfEVUyiF6KVSfgfNQ/5PpM R4/pmGbRqrAAElhrRPbKQnCkGWDr8zG+AjN1KF6rHaFgAIO7TtZ+F28jq4reLkur0N5tQFww wFwxzROdeLHuZjL7eEtcnNnzSkXHczLkV4kQ3+vr/7Gm65mQfnVpg6JpwpVrbDYQeOFlxZ8+ GERY5Dag4KgKa/4cSZX2x/5+KkQx9wHwackw5gDCvAdZ+Q81nm6tRxEYBBiVDQZYqO73stgT ZyrkxykUbQIy8PI+g7XMDCMnPiDncQqgf96KR3cvw4wN8QrgA6xRo8xOc2C3X7jTMQUytCz9 0MyV1QARAQABiQI8BBgBCAAmAhsMFiEE6j5FL/T5SGCN6PrQxzkHk2t9+PwFAlxHziAFCRj7 5/EACgkQxzkHk2t9+PxgfA//cH5R1DvpJPwraTAl24SUcG9EWe+NXyqveApe05nk15zEuxxd e4zFEjo+xYZilSveLqYHrm/amvQhsQ6JLU+8N60DZHVcXbw1Eb8CEjM5oXdbcJpXh1/1BEwl 4phsQMkxOTns51bGDhTQkv4lsZKvNByB9NiiMkT43EOx14rjkhHw3rnqoI7ogu8OO7XWfKcL CbchjJ8t3c2XK1MUe056yPpNAT2XPNF2EEBPG2Y2F4vLgEbPv1EtpGUS1+JvmK3APxjXUl5z 6xrxCQDWM5AAtGfM/IswVjbZYSJYyH4BQKrShzMb0rWUjkpXvvjsjt8rEXpZEYJgX9jvCoxt oqjCKiVLpwje9WkEe9O9VxljmPvxAhVqJjX62S+TGp93iD+mvpCoHo3+CcvyRcilz+Ko8lfO hS9tYT0HDUiDLvpUyH1AR2xW9RGDevGfwGTpF0K6cLouqyZNdhlmNciX48tFUGjakRFsxRmX K0Jx4CEZubakJe+894sX6pvNFiI7qUUdB882i5GR3v9ijVPhaMr8oGuJ3kvwBIA8lvRBGVGn 9xvzkQ8Prpbqh30I4NMp8MjFdkwCN6znBKPHdjNTwE5PRZH0S9J0o67IEIvHfH0eAWAsgpTz +jwc7VKH7vkvgscUhq/v1/PEWCAqh9UHy7R/jiUxwzw/288OpgO+i+2l11Y=
- Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>, Thomas Gleixner <tglx@xxxxxxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, Borislav Petkov <bp@xxxxxxxxx>, Andy Lutomirski <luto@xxxxxxxxxx>, Peter Zijlstra <peterz@xxxxxxxxxxxxx>, Kent Overstreet <kent.overstreet@xxxxxxxxx>, x86@xxxxxxxxxx, Dave Hansen <dave.hansen@xxxxxxxxxxxxxxx>, Dan Williams <dan.j.williams@xxxxxxxxx>, Fenghua Yu <fenghua.yu@xxxxxxxxx>, linux-doc@xxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, linux-nvdimm@xxxxxxxxxxxx, linux-fsdevel@xxxxxxxxxxxxxxx, linux-mm@xxxxxxxxx, linux-kselftest@xxxxxxxxxxxxxxx, linuxppc-dev@xxxxxxxxxxxxxxxx, kvm@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxxxxxx, bpf@xxxxxxxxxxxxxxx, kexec@xxxxxxxxxxxxxxxxxxx, linux-bcache@xxxxxxxxxxxxxxx, linux-mtd@xxxxxxxxxxxxxxxxxxx, devel@xxxxxxxxxxxxxxxxxxxx, linux-efi@xxxxxxxxxxxxxxx, linux-mmc@xxxxxxxxxxxxxxx, linux-scsi@xxxxxxxxxxxxxxx, target-devel@xxxxxxxxxxxxxxx, linux-nfs@xxxxxxxxxxxxxxx, ceph-devel@xxxxxxxxxxxxxxx, linux-ext4@xxxxxxxxxxxxxxx, linux-aio@xxxxxxxxx, io-uring@xxxxxxxxxxxxxxx, linux-erofs@xxxxxxxxxxxxxxxx, linux-um@xxxxxxxxxxxxxxxxxxx, linux-ntfs-dev@xxxxxxxxxxxxxxxxxxxxx, reiserfs-devel@xxxxxxxxxxxxxxx, linux-f2fs-devel@xxxxxxxxxxxxxxxxxxxxx, linux-nilfs@xxxxxxxxxxxxxxx, cluster-devel@xxxxxxxxxx, ecryptfs@xxxxxxxxxxxxxxx, linux-cifs@xxxxxxxxxxxxxxx, linux-btrfs@xxxxxxxxxxxxxxx, linux-afs@xxxxxxxxxxxxxxxxxxx, linux-rdma@xxxxxxxxxxxxxxx, amd-gfx@xxxxxxxxxxxxxxxxxxxxx, dri-devel@xxxxxxxxxxxxxxxxxxxxx, intel-gfx@xxxxxxxxxxxxxxxxxxxxx, drbd-dev@xxxxxxxxxxxxxxxx, linux-block@xxxxxxxxxxxxxxx, xen-devel@xxxxxxxxxxxxxxxxxxxx, linux-cachefs@xxxxxxxxxx, samba-technical@xxxxxxxxxxxxxxx, intel-wired-lan@xxxxxxxxxxxxxxxx
- Delivery-date: Sat, 10 Oct 2020 02:21:23 +0000
- List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
On 2020/10/10 03:50, ira.weiny@xxxxxxxxx wrote:
> From: Ira Weiny <ira.weiny@xxxxxxxxx>
>
> These kmap() calls are localized to a single thread. To avoid the over
> head of global PKRS updates use the new kmap_thread() call.
>
Hi Ira,
There were a number of options considered.
1) Attempt to change all the thread local kmap() calls to kmap_atomic()
2) Introduce a flags parameter to kmap() to indicate if the mapping
should be global or not
3) Change ~20-30 call sites to 'kmap_global()' to indicate that they
require a global mapping of the pages
4) Change ~209 call sites to 'kmap_thread()' to indicate that the
mapping is to be used within that thread of execution only
I copied the above information from patch 00/58 to this message. The
idea behind kmap_thread() is fine to me, but as you said the new api is
very easy to be missed in new code (even for me). I would like to be
supportive to option 2) introduce a flag to kmap(), then we won't forget
the new thread-localized kmap method, and people won't ask why a
_thread() function is called but no kthread created.
Thanks.
Coly Li
> Cc: Coly Li <colyli@xxxxxxx> (maintainer:BCACHE (BLOCK LAYER CACHE))
> Cc: Kent Overstreet <kent.overstreet@xxxxxxxxx> (maintainer:BCACHE (BLOCK
> LAYER CACHE))
> Signed-off-by: Ira Weiny <ira.weiny@xxxxxxxxx>
> ---
> drivers/md/bcache/request.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
> index c7cadaafa947..a4571f6d09dd 100644
> --- a/drivers/md/bcache/request.c
> +++ b/drivers/md/bcache/request.c
> @@ -44,10 +44,10 @@ static void bio_csum(struct bio *bio, struct bkey *k)
> uint64_t csum = 0;
>
> bio_for_each_segment(bv, bio, iter) {
> - void *d = kmap(bv.bv_page) + bv.bv_offset;
> + void *d = kmap_thread(bv.bv_page) + bv.bv_offset;
>
> csum = bch_crc64_update(csum, d, bv.bv_len);
> - kunmap(bv.bv_page);
> + kunmap_thread(bv.bv_page);
> }
>
> k->ptr[KEY_PTRS(k)] = csum & (~0ULL >> 1);
>
|