Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Fri, 1 Jul 2022 10:19:06 -0700
From:   Ira Weiny <ira.weiny@intel.com>
To:     "Fabio M. De Francesco" <fmdefrancesco@gmail.com>
CC:     David Sterba <dsterba@suse.com>, Chris Mason <clm@fb.com>,
        Josef Bacik <josef@toxicpanda.com>,
        Nick Terrell <terrelln@fb.com>, <linux-btrfs@vger.kernel.org>,
        <linux-kernel@vger.kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Matthew Wilcox <willy@infradead.org>,
        Kees Cook <keescook@chromium.org>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
        "James E . J . Bottomley" <James.Bottomley@hansenpartnership.com>,
        "Helge Deller" <deller@gmx.de>,
        John David Anglin <dave.anglin@bell.net>,
        <linux-parisc@vger.kernel.org>, Filipe Manana <fdmanana@kernel.org>
Subject: Re: [RESEND PATCH v4 2/2] btrfs: Replace kmap() with
 kmap_local_page() in zstd.c
Message-ID: <Yr8siondUuz08rTZ@iweiny-desk3>
References: <20220616210037.7060-1-fmdefrancesco@gmail.com>
 <20220616210037.7060-3-fmdefrancesco@gmail.com>
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <20220616210037.7060-3-fmdefrancesco@gmail.com>
MIME-Version: 1.0
X-MS-Exchange-AntiSpam-MessageData-ChunkCount: 1
X-MS-Exchange-AntiSpam-MessageData-0: =?us-ascii?Q?k0N/oPJe5X8BkSY7LkxwXakZdZ383MXhNrkBeyNRTSRdCwkR8aho0wrfuF0y?=
 =?us-ascii?Q?WYUGsinAMUZlMYbFV+eMB2xFueTz3HqZnkCPt8CtuaRshgi9Lxk6ZQggPUgN?=
 =?us-ascii?Q?i5mANdGG846Flr85c++F1+ExLbKu+DPv6voLLR0CYIKA7O91+riWGcNZqI4r?=
 =?us-ascii?Q?rYU9WzHNcaSgNF7HgBVOKZ1efwLRWDKpDJteDaRTMymEF3b3iGWnencWeLVV?=
 =?us-ascii?Q?nvLIlfrfpo5oTB9gceB5miEOt7xsa5KywkF6N+cztlhae7iGE+5sYpwfflBk?=
 =?us-ascii?Q?2bnpA7tSDlLcvbCRH6qnunlGF24p6z+BLtGwmX8CIe3M55wH4czoWaVpNtB+?=
 =?us-ascii?Q?hQxgNiqejcVYn1DfZ+YYtIuGDrUyE8Hv+Uq4SDPOHvpmY06e6RDSDhN5AB+X?=
 =?us-ascii?Q?yOTcnlX8VwSUxnxqt4h7CsNyYz1NMvKEvlL6Z8AHVmZH+xOYsKcfxWjJ9Rfl?=
 =?us-ascii?Q?b3cg7EAX9FUZX8fxD4vb+SgY9Mrk8SXI4+Ag+yieptaCPHLw8BRKc9Rdvx0Q?=
 =?us-ascii?Q?aC7iZF5cPscZMpbmZZRPaIuG7I3KnQ8MRz1TxE0ooYfPZk5QxlIEVAyq2Nis?=
 =?us-ascii?Q?x3jt4J4hlbyTuA0xr3/Redk+tT5Vbt+WyeuKEcUiMLXqm9qgQ/Jjq+nBT77i?=
 =?us-ascii?Q?Rs6XC6+5IAGa0I+ryFBPmsbsYTLaUtjKaRiUYu0IEkNujBjltrRr8qEuQofY?=
 =?us-ascii?Q?KUZk/FH7UFc0F7FqfacHtRWV0cZo+sWxiDqSpL5EACVdXun+No5oAXZJ+LQM?=
 =?us-ascii?Q?3I85347TYdkYOpaZQyxMOkFrLklSfF2kHSxeB6hpHQ8Jh72k/KU9R6PnpwPa?=
 =?us-ascii?Q?+X3Ut0N3k7ZlRQRyoLok5S36scYrb3oob0vFOpIw+DXCREVQpK/G2iEUelYi?=
 =?us-ascii?Q?S2trE7T9aGpvgerGzoqSbrJQ1n7GZdM13MQceRlJ4I4c47FF+1xhWwxwuEKK?=
 =?us-ascii?Q?j3+iB5KUm2SGzl47bVT5Q1Os9WGAa1zE8TuEMWxZSPPYMuPM0YqmG7GuraGP?=
 =?us-ascii?Q?zKFNUDeDUAgVBcnMmZG8k8i/lKH1pCrpdo9ahwLVO425rvhqgXHgR804zeo3?=
 =?us-ascii?Q?O0FC6BAB5P3Qy522oEZjonnDWnTRqo+LFiOD2LBdTF9cwhIYJRvFNx01FvjK?=
 =?us-ascii?Q?/c7bnY6t4wdwVt7r0fCG/gJIgB1+8wblH1hfsXhUZfAbsThv79xIygyD2mOY?=
 =?us-ascii?Q?mbbe+62ZfWmr5/o8m+x5lbqGRmccK1gZJIdj0NhJWyMz403TbgjLx/ERSxTv?=
 =?us-ascii?Q?mIyLXQucI5ScIiCsfAbmPZvLbcmSDGIvnITBx6uQpquxWif0GpXL1/W4Ofzb?=
 =?us-ascii?Q?H6QPWmIM4XfgoG+sFtJpG/Vjpjg7Cqcfe/UdG8ZCszaXlFcB0g2sxnHz5CF7?=
 =?us-ascii?Q?FRlvtC3hQDCQw1p1o1d5EccT9vA2H1dWkrvyX652yR9M2eqS/yyakYD7nHom?=
 =?us-ascii?Q?4IcUu2mHb+ef2z1bdZvg4d4MRtjdULSVOobeleGrvzNOeoljsiRm4hPRXF9p?=
 =?us-ascii?Q?hYGK3rqfbgJXZJsiVvW7d3roJDJZQGMvcuKtRqmM7cv81Tfm/0hPnxYgyFjh?=
 =?us-ascii?Q?tqThE5oprx/torWVPSAIRCzdqh/KYH1RtKIYNCecLjyPoEI4cuhx7ifF5GPJ?=
 =?us-ascii?Q?JbaeHm52TIcTcPsAC4oBWf441ij/8jMnwAPc8z6B510K?=
X-MS-Exchange-CrossTenant-Network-Message-Id: a96c92b1-f7b0-4d28-9fbd-08da5b85d1d6
X-MS-Exchange-CrossTenant-AuthSource: DM4PR11MB6311.namprd11.prod.outlook.com
X-MS-Exchange-CrossTenant-AuthAs: Internal
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 01 Jul 2022 17:19:13.7033
 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-CrossTenant-Id: 46c98d88-e344-4ed4-8496-4ed7712e255d
X-MS-Exchange-CrossTenant-MailboxType: HOSTED
X-MS-Exchange-CrossTenant-UserPrincipalName: BbM3W7bpJpsx9FsRICXaD038A+t9cQKHCK5QCwrbPT+ylOjxKzSNVmCdYmPlK+uMCEQSpqNqxDTWo8ZH3qp3YA==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: MN0PR11MB5986
X-OriginatorOrg: intel.com
X-Spam-Status: No, score=-4.8 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH,
        DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,RCVD_IN_DNSWL_MED,
        SPF_HELO_NONE,SPF_NONE,T_SCC_BODY_TEXT_LINE autolearn=ham
        autolearn_force=no version=3.4.6
X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on
        lindbergh.monkeyblade.net
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Thu, Jun 16, 2022 at 11:00:36PM +0200, Fabio M. De Francesco wrote:
> The use of kmap() is being deprecated in favor of kmap_local_page(). With
> kmap_local_page(), the mapping is per thread, CPU local and not globally
> visible.
> 
> Therefore, use kmap_local_page() / kunmap_local() in zstd.c because in
> this file the mappings are per thread and are not visible in other
> contexts; meanwhile refactor zstd_compress_pages() to comply with the
> ordering rules about nested local mapping / unmapping.
> 
> Tested with xfstests on QEMU + KVM 32 bits VM with 4GB of RAM and
> HIGHMEM64G enabled. These changes passed all the tests of the group
> "compress".
> 
> Cc: Filipe Manana <fdmanana@kernel.org>
> Suggested-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
> ---
> 
> v3->v4: Cc Maintainers and lists that had been overlooked when v3 was
>         sent (mostly regarding patch 1/2).
> 
> v2->v3: Remove unnecessary casts to arguments of kunmap_local() now that
>         this API can take pointers to const void.
> 
> v1->v2: No changes.
> 
> Thanks to Ira Weiny for his invaluable help and persevering support.
> Thanks also to Filipe Manana for identifying a fundamental detail I had
> overlooked in RFC:
> https://lore.kernel.org/lkml/20220611093411.GA3779054@falcondesktop/
> 
>  fs/btrfs/zstd.c | 42 +++++++++++++++++++++++-------------------
>  1 file changed, 23 insertions(+), 19 deletions(-)
> 
> diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c
> index 0fe31a6f6e68..5d2ab0bac9d2 100644
> --- a/fs/btrfs/zstd.c
> +++ b/fs/btrfs/zstd.c
> @@ -391,6 +391,8 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  	*out_pages = 0;
>  	*total_out = 0;
>  	*total_in = 0;
> +	workspace->in_buf.src = NULL;
> +	workspace->out_buf.dst = NULL;

I don't think either of these are needed as they are both set straight away
below.

>  
>  	/* Initialize the stream */
>  	stream = zstd_init_cstream(&params, len, workspace->mem,
> @@ -403,7 +405,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  
>  	/* map in the first page of input data */
>  	in_page = find_get_page(mapping, start >> PAGE_SHIFT);
> -	workspace->in_buf.src = kmap(in_page);
> +	workspace->in_buf.src = kmap_local_page(in_page);
>  	workspace->in_buf.pos = 0;
>  	workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE);
>  
> @@ -415,7 +417,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  		goto out;
>  	}
>  	pages[nr_pages++] = out_page;
> -	workspace->out_buf.dst = kmap(out_page);
> +	workspace->out_buf.dst = kmap_local_page(out_page);

Given the conversation in the other thread for zlib_compress_pages(); I think
we should also just use page_address() here.  That simplifies the algorithm
immensely.

I know there was a lot of thought put into how to make this conversion when
this was posted but that is now the cleaner solution.

Ira

>  	workspace->out_buf.pos = 0;
>  	workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
>  
> @@ -450,9 +452,9 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  		if (workspace->out_buf.pos == workspace->out_buf.size) {
>  			tot_out += PAGE_SIZE;
>  			max_out -= PAGE_SIZE;
> -			kunmap(out_page);
> +			kunmap_local(workspace->out_buf.dst);
>  			if (nr_pages == nr_dest_pages) {
> -				out_page = NULL;
> +				workspace->out_buf.dst = NULL;
>  				ret = -E2BIG;
>  				goto out;
>  			}
> @@ -462,7 +464,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  				goto out;
>  			}
>  			pages[nr_pages++] = out_page;
> -			workspace->out_buf.dst = kmap(out_page);
> +			workspace->out_buf.dst = kmap_local_page(out_page);
>  			workspace->out_buf.pos = 0;
>  			workspace->out_buf.size = min_t(size_t, max_out,
>  							PAGE_SIZE);
> @@ -477,15 +479,16 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  		/* Check if we need more input */
>  		if (workspace->in_buf.pos == workspace->in_buf.size) {
>  			tot_in += PAGE_SIZE;
> -			kunmap(in_page);
> +			kunmap_local(workspace->out_buf.dst);
> +			kunmap_local(workspace->in_buf.src);
>  			put_page(in_page);
> -
>  			start += PAGE_SIZE;
>  			len -= PAGE_SIZE;
>  			in_page = find_get_page(mapping, start >> PAGE_SHIFT);
> -			workspace->in_buf.src = kmap(in_page);
> +			workspace->in_buf.src = kmap_local_page(in_page);
>  			workspace->in_buf.pos = 0;
>  			workspace->in_buf.size = min_t(size_t, len, PAGE_SIZE);
> +			workspace->out_buf.dst = kmap_local_page(out_page);
>  		}
>  	}
>  	while (1) {
> @@ -510,9 +513,9 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  
>  		tot_out += PAGE_SIZE;
>  		max_out -= PAGE_SIZE;
> -		kunmap(out_page);
> +		kunmap_local(workspace->out_buf.dst);
>  		if (nr_pages == nr_dest_pages) {
> -			out_page = NULL;
> +			workspace->out_buf.dst = NULL;
>  			ret = -E2BIG;
>  			goto out;
>  		}
> @@ -522,7 +525,7 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  			goto out;
>  		}
>  		pages[nr_pages++] = out_page;
> -		workspace->out_buf.dst = kmap(out_page);
> +		workspace->out_buf.dst = kmap_local_page(out_page);
>  		workspace->out_buf.pos = 0;
>  		workspace->out_buf.size = min_t(size_t, max_out, PAGE_SIZE);
>  	}
> @@ -538,12 +541,12 @@ int zstd_compress_pages(struct list_head *ws, struct address_space *mapping,
>  out:
>  	*out_pages = nr_pages;
>  	/* Cleanup */
> -	if (in_page) {
> -		kunmap(in_page);
> +	if (workspace->out_buf.dst)
> +		kunmap_local(workspace->out_buf.dst);
> +	if (workspace->in_buf.src) {
> +		kunmap_local(workspace->in_buf.src);
>  		put_page(in_page);
>  	}
> -	if (out_page)
> -		kunmap(out_page);
>  	return ret;
>  }
>  
> @@ -567,7 +570,7 @@ int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
>  		goto done;
>  	}
>  
> -	workspace->in_buf.src = kmap(pages_in[page_in_index]);
> +	workspace->in_buf.src = kmap_local_page(pages_in[page_in_index]);
>  	workspace->in_buf.pos = 0;
>  	workspace->in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
>  
> @@ -603,14 +606,15 @@ int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
>  			break;
>  
>  		if (workspace->in_buf.pos == workspace->in_buf.size) {
> -			kunmap(pages_in[page_in_index++]);
> +			kunmap_local(workspace->in_buf.src);
> +			page_in_index++;
>  			if (page_in_index >= total_pages_in) {
>  				workspace->in_buf.src = NULL;
>  				ret = -EIO;
>  				goto done;
>  			}
>  			srclen -= PAGE_SIZE;
> -			workspace->in_buf.src = kmap(pages_in[page_in_index]);
> +			workspace->in_buf.src = kmap_local_page(pages_in[page_in_index]);
>  			workspace->in_buf.pos = 0;
>  			workspace->in_buf.size = min_t(size_t, srclen, PAGE_SIZE);
>  		}
> @@ -619,7 +623,7 @@ int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb)
>  	zero_fill_bio(cb->orig_bio);
>  done:
>  	if (workspace->in_buf.src)
> -		kunmap(pages_in[page_in_index]);
> +		kunmap_local(workspace->in_buf.src);
>  	return ret;
>  }
>  
> -- 
> 2.36.1
>