From: Robin Murphy <robin.murphy@arm.com> To: Tony Battersby <tonyb@cybernetics.com>, linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: iommu@lists.linux-foundation.org, kernel-team@fb.com, Matthew Wilcox <willy@infradead.org>, Keith Busch <kbusch@kernel.org>, Andy Shevchenko <andy.shevchenko@gmail.com>, Tony Lindgren <tony@atomide.com> Subject: Re: [PATCH 04/10] dmapool: improve accuracy of debug statistics Date: Tue, 31 May 2022 20:48:24 +0100 [thread overview] Message-ID: <b97645ed-b524-a505-2993-e04a37b50d35@arm.com> (raw) In-Reply-To: <a922c30f-d6d7-dde2-554f-254441290331@cybernetics.com> On 2022-05-31 19:17, Tony Battersby wrote: > The "total number of blocks in pool" debug statistic currently does not > take the boundary value into account, so it diverges from the "total > number of blocks in use" statistic when a boundary is in effect. Add a > calculation for the number of blocks per allocation that takes the > boundary into account, and use it to replace the inaccurate calculation. > > This depends on the patch "dmapool: fix boundary comparison" for the > calculated blks_per_alloc value to be correct. > > Signed-off-by: Tony Battersby <tonyb@cybernetics.com> > --- > mm/dmapool.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/mm/dmapool.c b/mm/dmapool.c > index 782143144a32..9e30f4425dea 100644 > --- a/mm/dmapool.c > +++ b/mm/dmapool.c > @@ -47,6 +47,7 @@ struct dma_pool { /* the pool */ > struct device *dev; > unsigned int allocation; > unsigned int boundary; > + unsigned int blks_per_alloc; > char name[32]; > struct list_head pools; > }; > @@ -92,8 +93,7 @@ static ssize_t pools_show(struct device *dev, struct device_attribute *attr, cha > /* per-pool info, no real statistics yet */ > temp = scnprintf(next, size, "%-16s %4zu %4zu %4u %2u\n", Nit: if we're tinkering with this, it's probably worth updating the whole function to use sysfs_emit{_at}(). > pool->name, blocks, > - (size_t) pages * > - (pool->allocation / pool->size), > + (size_t) pages * pool->blks_per_alloc, > pool->size, pages); > size -= temp; > next += temp; > @@ -168,6 +168,9 @@ struct dma_pool *dma_pool_create(const char *name, struct device *dev, > retval->size = size; > retval->boundary = boundary; > retval->allocation = allocation; > + retval->blks_per_alloc = > + (allocation / boundary) * (boundary / size) + > + (allocation % boundary) / size; Do we really need to store this? Sure, 4 divisions (which could possibly be fewer given the constraints on boundary) isn't the absolute cheapest calculation, but I still can't imagine anyone would be polling sysfs stats hard enough to even notice. Thanks, Robin. > > INIT_LIST_HEAD(&retval->pools); >
WARNING: multiple messages have this Message-ID (diff)
From: Robin Murphy <robin.murphy@arm.com> To: Tony Battersby <tonyb@cybernetics.com>, linux-mm@kvack.org, linux-kernel@vger.kernel.org Cc: Tony Lindgren <tony@atomide.com>, Andy Shevchenko <andy.shevchenko@gmail.com>, Matthew Wilcox <willy@infradead.org>, iommu@lists.linux-foundation.org, Keith Busch <kbusch@kernel.org>, kernel-team@fb.com Subject: Re: [PATCH 04/10] dmapool: improve accuracy of debug statistics Date: Tue, 31 May 2022 20:48:24 +0100 [thread overview] Message-ID: <b97645ed-b524-a505-2993-e04a37b50d35@arm.com> (raw) In-Reply-To: <a922c30f-d6d7-dde2-554f-254441290331@cybernetics.com> On 2022-05-31 19:17, Tony Battersby wrote: > The "total number of blocks in pool" debug statistic currently does not > take the boundary value into account, so it diverges from the "total > number of blocks in use" statistic when a boundary is in effect. Add a > calculation for the number of blocks per allocation that takes the > boundary into account, and use it to replace the inaccurate calculation. > > This depends on the patch "dmapool: fix boundary comparison" for the > calculated blks_per_alloc value to be correct. > > Signed-off-by: Tony Battersby <tonyb@cybernetics.com> > --- > mm/dmapool.c | 7 +++++-- > 1 file changed, 5 insertions(+), 2 deletions(-) > > diff --git a/mm/dmapool.c b/mm/dmapool.c > index 782143144a32..9e30f4425dea 100644 > --- a/mm/dmapool.c > +++ b/mm/dmapool.c > @@ -47,6 +47,7 @@ struct dma_pool { /* the pool */ > struct device *dev; > unsigned int allocation; > unsigned int boundary; > + unsigned int blks_per_alloc; > char name[32]; > struct list_head pools; > }; > @@ -92,8 +93,7 @@ static ssize_t pools_show(struct device *dev, struct device_attribute *attr, cha > /* per-pool info, no real statistics yet */ > temp = scnprintf(next, size, "%-16s %4zu %4zu %4u %2u\n", Nit: if we're tinkering with this, it's probably worth updating the whole function to use sysfs_emit{_at}(). > pool->name, blocks, > - (size_t) pages * > - (pool->allocation / pool->size), > + (size_t) pages * pool->blks_per_alloc, > pool->size, pages); > size -= temp; > next += temp; > @@ -168,6 +168,9 @@ struct dma_pool *dma_pool_create(const char *name, struct device *dev, > retval->size = size; > retval->boundary = boundary; > retval->allocation = allocation; > + retval->blks_per_alloc = > + (allocation / boundary) * (boundary / size) + > + (allocation % boundary) / size; Do we really need to store this? Sure, 4 divisions (which could possibly be fewer given the constraints on boundary) isn't the absolute cheapest calculation, but I still can't imagine anyone would be polling sysfs stats hard enough to even notice. Thanks, Robin. > > INIT_LIST_HEAD(&retval->pools); > _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
next prev parent reply other threads:[~2022-05-31 19:48 UTC|newest] Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-05-31 18:11 [PATCH 00/10] mpt3sas and dmapool scalability Tony Battersby 2022-05-31 18:11 ` Tony Battersby 2022-05-31 18:12 ` [PATCH 01/10] dmapool: remove checks for dev == NULL Tony Battersby 2022-05-31 18:12 ` Tony Battersby 2022-05-31 18:23 ` Robin Murphy 2022-05-31 18:23 ` Robin Murphy 2022-05-31 18:13 ` [PATCH 02/10] dmapool: cleanup integer types Tony Battersby 2022-05-31 18:13 ` Tony Battersby 2022-05-31 18:14 ` [PATCH 03/10] dmapool: fix boundary comparison Tony Battersby 2022-05-31 18:14 ` Tony Battersby 2022-05-31 18:17 ` [PATCH 04/10] dmapool: improve accuracy of debug statistics Tony Battersby 2022-05-31 18:17 ` Tony Battersby 2022-05-31 19:48 ` Robin Murphy [this message] 2022-05-31 19:48 ` Robin Murphy 2022-05-31 19:52 ` Tony Battersby 2022-05-31 19:52 ` Tony Battersby 2022-05-31 21:55 ` Robin Murphy 2022-05-31 21:55 ` Robin Murphy 2022-05-31 18:18 ` [PATCH 05/10] dmapool: debug: prevent endless loop in case of corruption Tony Battersby 2022-05-31 18:18 ` Tony Battersby 2022-05-31 18:20 ` [PATCH 06/10] dmapool: ignore init_on_free when DMAPOOL_DEBUG enabled Tony Battersby 2022-05-31 18:20 ` Tony Battersby 2022-05-31 18:21 ` [PATCH 07/10] dmapool: speedup DMAPOOL_DEBUG with init_on_alloc Tony Battersby 2022-05-31 18:21 ` Tony Battersby 2022-05-31 18:22 ` [PATCH 08/10] dmapool: cleanup dma_pool_destroy Tony Battersby 2022-05-31 18:22 ` Tony Battersby 2022-05-31 19:33 ` Robin Murphy 2022-05-31 19:33 ` Robin Murphy 2022-05-31 21:40 ` Keith Busch 2022-05-31 21:40 ` Keith Busch 2022-05-31 18:23 ` [PATCH 09/10] dmapool: improve scalability of dma_pool_alloc Tony Battersby 2022-05-31 18:23 ` Tony Battersby 2022-05-31 18:23 ` [PATCH 10/10] dmapool: improve scalability of dma_pool_free Tony Battersby 2022-05-31 18:23 ` Tony Battersby 2022-05-31 21:54 ` Keith Busch 2022-05-31 21:54 ` Keith Busch 2022-05-31 22:10 ` Tony Battersby 2022-05-31 22:10 ` Tony Battersby 2022-06-01 9:44 ` Robin Murphy 2022-06-01 9:44 ` Robin Murphy
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=b97645ed-b524-a505-2993-e04a37b50d35@arm.com \ --to=robin.murphy@arm.com \ --cc=andy.shevchenko@gmail.com \ --cc=iommu@lists.linux-foundation.org \ --cc=kbusch@kernel.org \ --cc=kernel-team@fb.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=tony@atomide.com \ --cc=tonyb@cybernetics.com \ --cc=willy@infradead.org \ /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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.