From: Matthew Wilcox <willy@infradead.org>
To: Josef Bacik <josef@toxicpanda.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Matthew Wilcox <mawilcox@microsoft.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
linux-fsdevel@vger.kernel.org,
Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>,
linux-nilfs@vger.kernel.org
Subject: Re: [PATCH v9 09/61] xarray: Replace exceptional entries
Date: Fri, 16 Mar 2018 12:06:04 -0700 [thread overview]
Message-ID: <20180316190604.GF27498@bombadil.infradead.org> (raw)
In-Reply-To: <20180316185349.c4ebbwuzlhihec5f@destiny>
On Fri, Mar 16, 2018 at 02:53:50PM -0400, Josef Bacik wrote:
> On Tue, Mar 13, 2018 at 06:25:47AM -0700, Matthew Wilcox wrote:
> > @@ -453,18 +449,14 @@ int ida_get_new_above(struct ida *ida, int start, int *id)
> > new += bit;
> > if (new < 0)
> > return -ENOSPC;
> > - if (ebit < BITS_PER_LONG) {
> > - bitmap = (void *)((1UL << ebit) |
> > - RADIX_TREE_EXCEPTIONAL_ENTRY);
> > - radix_tree_iter_replace(root, &iter, slot,
> > - bitmap);
> > - *id = new;
> > - return 0;
> > + if (bit < BITS_PER_XA_VALUE) {
> > + bitmap = xa_mk_value(1UL << bit);
> > + } else {
> > + bitmap = this_cpu_xchg(ida_bitmap, NULL);
> > + if (!bitmap)
> > + return -EAGAIN;
> > + __set_bit(bit, bitmap->bitmap);
> > }
> > - bitmap = this_cpu_xchg(ida_bitmap, NULL);
> > - if (!bitmap)
> > - return -EAGAIN;
> > - __set_bit(bit, bitmap->bitmap);
> > radix_tree_iter_replace(root, &iter, slot, bitmap);
> > }
> >
>
> This threw me off a bit, but we do *id = new below.
Yep. Fortunately, I have a test-suite for the IDA, so I'm relatively
sure this works.
> > @@ -495,9 +487,9 @@ void ida_remove(struct ida *ida, int id)
> > goto err;
> >
> > bitmap = rcu_dereference_raw(*slot);
> > - if (radix_tree_exception(bitmap)) {
> > + if (xa_is_value(bitmap)) {
> > btmp = (unsigned long *)slot;
> > - offset += RADIX_TREE_EXCEPTIONAL_SHIFT;
> > + offset += 1; /* Intimate knowledge of the xa_data encoding */
> > if (offset >= BITS_PER_LONG)
> > goto err;
> > } else {
>
> Ick.
I know. I feel quite ashamed of this code. I do have a rewrite to use
the XArray, but I didn't want to include it as part of *this* merge request.
And that rewrite decodes the value into an unsigned long, sets the bit,
reencodes it as an xa_value and stores it.
> > @@ -393,11 +393,11 @@ void ida_check_conv(void)
> > for (i = 0; i < 1000000; i++) {
> > int err = ida_get_new(&ida, &id);
> > if (err == -EAGAIN) {
> > - assert((i % IDA_BITMAP_BITS) == (BITS_PER_LONG - 2));
> > + assert((i % IDA_BITMAP_BITS) == (BITS_PER_LONG - 1));
> > assert(ida_pre_get(&ida, GFP_KERNEL));
> > err = ida_get_new(&ida, &id);
> > } else {
> > - assert((i % IDA_BITMAP_BITS) != (BITS_PER_LONG - 2));
> > + assert((i % IDA_BITMAP_BITS) != (BITS_PER_LONG - 1));
>
> Can we just use BITS_PER_XA_VALUE here?
Yes! I'll change that.
> Reviewed-by: Josef Bacik <jbacik@fb.com>
Thanks!
next prev parent reply other threads:[~2018-03-16 19:06 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-13 13:25 [PATCH v9 00/61] XArray v9 Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 01/61] mac80211_hwsim: Use DEFINE_IDA Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 02/61] radix tree: Use GFP_ZONEMASK bits of gfp_t for flags Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 03/61] arm64: Turn flush_dcache_mmap_lock into a no-op Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 04/61] unicore32: " Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 05/61] Export __set_page_dirty Matthew Wilcox
2018-03-27 0:28 ` Darrick J. Wong
2018-03-13 13:25 ` [PATCH v9 06/61] fscache: Use appropriate radix tree accessors Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 07/61] xarray: Add the xa_lock to the radix_tree_root Matthew Wilcox
2018-04-12 20:59 ` Ross Zwisler
2018-04-12 21:10 ` Matthew Wilcox
2018-04-12 21:16 ` Ross Zwisler
2018-04-12 21:22 ` Matthew Wilcox
2018-04-12 21:27 ` Ross Zwisler
2018-03-13 13:25 ` [PATCH v9 08/61] page cache: Use xa_lock Matthew Wilcox
2018-03-16 18:06 ` Josef Bacik
2018-03-13 13:25 ` [PATCH v9 09/61] xarray: Replace exceptional entries Matthew Wilcox
2018-03-16 18:53 ` Josef Bacik
2018-03-16 19:06 ` Matthew Wilcox [this message]
2018-03-13 13:25 ` [PATCH v9 10/61] xarray: Change definition of sibling entries Matthew Wilcox
2018-03-16 19:00 ` Josef Bacik
2018-03-13 13:25 ` [PATCH v9 11/61] xarray: Add definition of struct xarray Matthew Wilcox
2018-03-16 19:05 ` Josef Bacik
2018-03-13 13:25 ` [PATCH v9 12/61] xarray: Define struct xa_node Matthew Wilcox
2018-03-16 19:11 ` Josef Bacik
2018-03-13 13:25 ` [PATCH v9 13/61] xarray: Add documentation Matthew Wilcox
2018-03-16 19:12 ` Josef Bacik
2018-03-13 13:25 ` [PATCH v9 14/61] xarray: Add xa_load Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 15/61] xarray: Add xa_get_tag, xa_set_tag and xa_clear_tag Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 16/61] xarray: Add xa_store Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 17/61] xarray: Add xa_cmpxchg and xa_insert Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 18/61] xarray: Add xa_for_each Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 19/61] xarray: Add xa_extract Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 20/61] xarray: Add xa_destroy Matthew Wilcox
2018-03-13 13:25 ` [PATCH v9 21/61] xarray: Add xas_next and xas_prev Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 22/61] xarray: Add xas_create_range Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 23/61] xarray: Add MAINTAINERS entry Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 24/61] page cache: Rearrange address_space Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 25/61] page cache: Convert hole search to XArray Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 26/61] page cache: Add and replace pages using the XArray Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 27/61] page cache: Convert page deletion to XArray Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 28/61] page cache: Convert page cache lookups " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 29/61] page cache: Convert delete_batch " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 30/61] page cache: Remove stray radix comment Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 31/61] page cache: Convert filemap_range_has_page to XArray Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 32/61] mm: Convert page-writeback " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 33/61] mm: Convert workingset " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 34/61] mm: Convert truncate " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 35/61] mm: Convert add_to_swap_cache " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 36/61] mm: Convert delete_from_swap_cache " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 37/61] mm: Convert __do_page_cache_readahead " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 38/61] mm: Convert page migration " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 39/61] mm: Convert huge_memory " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 40/61] mm: Convert collapse_shmem " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 41/61] mm: Convert khugepaged_scan_shmem " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 42/61] pagevec: Use xa_tag_t Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 43/61] shmem: Convert replace to XArray Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 44/61] shmem: Convert shmem_confirm_swap " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 45/61] shmem: Convert find_swap_entry " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 46/61] shmem: Convert shmem_add_to_page_cache " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 47/61] shmem: Convert shmem_alloc_hugepage " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 48/61] shmem: Convert shmem_free_swap " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 49/61] shmem: Convert shmem_partial_swap_usage " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 50/61] memfd: Convert shmem_tag_pins " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 51/61] memfd: Convert shmem_wait_for_pins " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 52/61] shmem: Comment fixups Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 53/61] btrfs: Convert page cache to XArray Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 54/61] fs: Convert buffer " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 55/61] fs: Convert writeback " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 56/61] nilfs2: Convert " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 57/61] f2fs: " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 58/61] lustre: " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 59/61] dax: " Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 60/61] page cache: Finish XArray conversion Matthew Wilcox
2018-03-13 13:26 ` [PATCH v9 61/61] radix tree: Remove unused functions Matthew Wilcox
2018-03-26 22:36 ` [PATCH v9 00/61] XArray v9 Andrew Morton
2018-03-26 23:26 ` Matthew Wilcox
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=20180316190604.GF27498@bombadil.infradead.org \
--to=willy@infradead.org \
--cc=akpm@linux-foundation.org \
--cc=josef@toxicpanda.com \
--cc=konishi.ryusuke@lab.ntt.co.jp \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nilfs@vger.kernel.org \
--cc=mawilcox@microsoft.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).