[PATCH 0/2] mmc_test performance test fixes

public inbox for linux-mmc@vger.kernel.org
 help / color / mirror / Atom feed

* [PATCH 0/2] mmc_test performance test fixes
@ 2010-09-23 11:51 Adrian Hunter
  2010-09-23 11:51 ` [PATCH 1/2] mmc_test: fix memory allocation segment limits Adrian Hunter
  2010-09-23 11:51 ` [PATCH 2/2] mmc_test: fix display of .5 KiB Adrian Hunter
  0 siblings, 2 replies; 4+ messages in thread
From: Adrian Hunter @ 2010-09-23 11:51 UTC (permalink / raw)
  To: Chris Ball; +Cc: linux-mmc Mailing List, Adrian Hunter

Hi

Here are two fixes for mmc_test performance tests.  They apply to the mmc-next
branch of Chris Ball's mmc tree at:
	git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc.git


The following changes since commit 0bff5cae99dfc9f820cdbddcc4f2cf55627ccfac:
  Philip Rakity (1):
        mmc: sdhci: print out controller name for register debug

Adrian Hunter (2):
      mmc_test: fix memory allocation segment limits
      mmc_test: fix display of .5 KiB

 drivers/mmc/card/mmc_test.c |   16 ++++++++--------
 1 files changed, 8 insertions(+), 8 deletions(-)


Regards
Adrian


^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH 1/2] mmc_test: fix memory allocation segment limits
  2010-09-23 11:51 [PATCH 0/2] mmc_test performance test fixes Adrian Hunter
@ 2010-09-23 11:51 ` Adrian Hunter
  2010-09-24 21:15   ` Chris Ball
  2010-09-23 11:51 ` [PATCH 2/2] mmc_test: fix display of .5 KiB Adrian Hunter
  1 sibling, 1 reply; 4+ messages in thread
From: Adrian Hunter @ 2010-09-23 11:51 UTC (permalink / raw)
  To: Chris Ball; +Cc: linux-mmc Mailing List, Adrian Hunter

>From aad98716cd9e8f83831721010d58d2a9f8ea2d2b Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Thu, 23 Sep 2010 12:55:15 +0300
Subject: [PATCH 1/2] mmc_test: fix memory allocation segment limits

Correctly allocate memory to meet the host controller
driver's maximum segment size and count limits.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mmc/card/mmc_test.c |   12 ++++++------
 1 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/mmc/card/mmc_test.c b/drivers/mmc/card/mmc_test.c
index 6580af7..5874040 100644
--- a/drivers/mmc/card/mmc_test.c
+++ b/drivers/mmc/card/mmc_test.c
@@ -318,8 +318,8 @@ static struct mmc_test_mem *mmc_test_alloc_mem(unsigned long min_sz,
 
 	if (max_page_cnt > limit)
 		max_page_cnt = limit;
-	if (max_page_cnt < min_page_cnt)
-		max_page_cnt = min_page_cnt;
+	if (min_page_cnt > max_page_cnt)
+		min_page_cnt = max_page_cnt;
 
 	if (max_seg_page_cnt > max_page_cnt)
 		max_seg_page_cnt = max_page_cnt;
@@ -359,13 +359,13 @@ static struct mmc_test_mem *mmc_test_alloc_mem(unsigned long min_sz,
 		mem->cnt += 1;
 		if (max_page_cnt <= (1UL << order))
 			break;
+		max_page_cnt -= 1UL << order;
+		page_cnt += 1UL << order;
 		if (mem->cnt >= max_segs) {
 			if (page_cnt < min_page_cnt)
 				goto out_free;
 			break;
 		}
-		max_page_cnt -= 1UL << order;
-		page_cnt += 1UL << order;
 	}
 
 	return mem;
@@ -1470,12 +1470,12 @@ static int mmc_test_area_init(struct mmc_test_card *test, int erase, int fill)
 		t->max_tfr = t->max_segs * t->max_seg_sz;
 
 	/*
-	 * Try to allocate enough memory for the whole area.  Less is OK
+	 * Try to allocate enough memory for a max. sized transfer.  Less is OK
 	 * because the same memory can be mapped into the scatterlist more than
 	 * once.  Also, take into account the limits imposed on scatterlist
 	 * segments by the host driver.
 	 */
-	t->mem = mmc_test_alloc_mem(min_sz, t->max_sz, t->max_segs,
+	t->mem = mmc_test_alloc_mem(min_sz, t->max_tfr, t->max_segs,
 				    t->max_seg_sz);
 	if (!t->mem)
 		return -ENOMEM;
-- 
1.7.0.4


^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH 1/2] mmc_test: fix memory allocation segment limits
  2010-09-23 11:51 ` [PATCH 1/2] mmc_test: fix memory allocation segment limits Adrian Hunter
@ 2010-09-24 21:15   ` Chris Ball
  0 siblings, 0 replies; 4+ messages in thread
From: Chris Ball @ 2010-09-24 21:15 UTC (permalink / raw)
  To: Adrian Hunter; +Cc: linux-mmc Mailing List

Hi Adrian,

On Thu, Sep 23, 2010 at 02:51:29PM +0300, Adrian Hunter wrote:
> From: Adrian Hunter <adrian.hunter@nokia.com>
> Date: Thu, 23 Sep 2010 12:55:15 +0300
> Subject: [PATCH 1/2] mmc_test: fix memory allocation segment limits
> 
> Correctly allocate memory to meet the host controller
> driver's maximum segment size and count limits.

Thanks very much, this fixes my problems.  I've pushed it to mmc-next.

Output below, just in case you're curious:

mmc_test mmc0:e624: Card claimed for testing.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 1. Basic write (no data verification)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 2. Basic read (no data verification)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 3. Basic write (with data verification)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 4. Basic read (with data verification)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 5. Multi-block write...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 6. Multi-block read...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 7. Power of two block writes...
mmc0: Result: UNSUPPORTED (by card)
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 8. Power of two block reads...
mmc0: Result: UNSUPPORTED (by card)
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 9. Weird sized block writes...
mmc0: Result: UNSUPPORTED (by card)
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 10. Weird sized block reads...
mmc0: Result: UNSUPPORTED (by card)
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 11. Badly aligned write...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 12. Badly aligned read...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 13. Badly aligned multi-block write...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 14. Badly aligned multi-block read...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 15. Correct xfer_size at write (start failure)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 16. Correct xfer_size at read (start failure)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 17. Correct xfer_size at write (midway failure)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 18. Correct xfer_size at read (midway failure)...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 19. Highmem write...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 20. Highmem read...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 21. Multi-block highmem write...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 22. Multi-block highmem read...
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 23. Best-case read performance...
mmc0: Transfer of 1024 sectors (512 KiB) took 0.033058904 seconds (15859 kB/s, 15487 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 24. Best-case write performance...
mmc0: Transfer of 1024 sectors (512 KiB) took 0.036752587 seconds (14265 kB/s, 13930 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 25. Best-case read performance into scattered pages...
mmc0: Transfer of 8 sectors (4 KiB) took 0.000796178 seconds (5144 kB/s, 5024 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 26. Best-case write performance from scattered pages...
mmc0: Transfer of 8 sectors (4 KiB) took 0.002371540 seconds (1727 kB/s, 1686 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 27. Single read performance by transfer size...
mmc0: Transfer of 1 sectors (0.5 KiB) took 0.000551456 seconds (928 kB/s, 906 KiB/s)
mmc0: Transfer of 2 sectors (1 KiB) took 0.000552924 seconds (1851 kB/s, 1808 KiB/s)
mmc0: Transfer of 4 sectors (2 KiB) took 0.000668650 seconds (3062 kB/s, 2991 KiB/s)
mmc0: Transfer of 8 sectors (4 KiB) took 0.000696796 seconds (5878 kB/s, 5740 KiB/s)
mmc0: Transfer of 16 sectors (8 KiB) took 0.001019670 seconds (8033 kB/s, 7845 KiB/s)
mmc0: Transfer of 32 sectors (16 KiB) took 0.001567431 seconds (10452 kB/s, 10207 KiB/s)
mmc0: Transfer of 64 sectors (32 KiB) took 0.002722948 seconds (12034 kB/s, 11751 KiB/s)
mmc0: Transfer of 128 sectors (64 KiB) took 0.004737135 seconds (13834 kB/s, 13510 KiB/s)
mmc0: Transfer of 256 sectors (128 KiB) took 0.008854402 seconds (14803 kB/s, 14456 KiB/s)
mmc0: Transfer of 512 sectors (256 KiB) took 0.016952058 seconds (15463 kB/s, 15101 KiB/s)
mmc0: Transfer of 1024 sectors (512 KiB) took 0.033126787 seconds (15826 kB/s, 15455 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 28. Single write performance by transfer size...
mmc0: Transfer of 1 sectors (0.5 KiB) took 0.001949055 seconds (262 kB/s, 256 KiB/s)
mmc0: Transfer of 2 sectors (1 KiB) took 0.002415603 seconds (423 kB/s, 413 KiB/s)
mmc0: Transfer of 4 sectors (2 KiB) took 0.002114501 seconds (968 kB/s, 945 KiB/s)
mmc0: Transfer of 8 sectors (4 KiB) took 0.001552827 seconds (2637 kB/s, 2575 KiB/s)
mmc0: Transfer of 16 sectors (8 KiB) took 0.001740555 seconds (4706 kB/s, 4596 KiB/s)
mmc0: Transfer of 32 sectors (16 KiB) took 0.002272057 seconds (7211 kB/s, 7042 KiB/s)
mmc0: Transfer of 64 sectors (32 KiB) took 0.003499233 seconds (9364 kB/s, 9144 KiB/s)
mmc0: Transfer of 128 sectors (64 KiB) took 0.005615027 seconds (11671 kB/s, 11397 KiB/s)
mmc0: Transfer of 256 sectors (128 KiB) took 0.009879663 seconds (13266 kB/s, 12955 KiB/s)
mmc0: Transfer of 512 sectors (256 KiB) took 0.173152083 seconds (1513 kB/s, 1478 KiB/s)
mmc0: Transfer of 1024 sectors (512 KiB) took 0.036976626 seconds (14178 kB/s, 13846 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 29. Single trim performance by transfer size...
mmc0: Result: UNSUPPORTED (by card)
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 30. Consecutive read performance by transfer size...
mmc0: Transfer of 8192 x 1 sectors (8192 x 0.5 KiB) took 3.003465917 seconds (1396 kB/s, 1363 KiB/s)
mmc0: Transfer of 4096 x 2 sectors (4096 x 1 KiB) took 1.941040090 seconds (2160 kB/s, 2110 KiB/s)
mmc0: Transfer of 2048 x 4 sectors (2048 x 2 KiB) took 1.056346906 seconds (3970 kB/s, 3877 KiB/s)
mmc0: Transfer of 1024 x 8 sectors (1024 x 4 KiB) took 0.670899009 seconds (6251 kB/s, 6105 KiB/s)
mmc0: Transfer of 512 x 16 sectors (512 x 8 KiB) took 0.496233273 seconds (8452 kB/s, 8254 KiB/s)
mmc0: Transfer of 256 x 32 sectors (256 x 16 KiB) took 0.383516832 seconds (10936 kB/s, 10680 KiB/s)
mmc0: Transfer of 128 x 64 sectors (128 x 32 KiB) took 0.323005936 seconds (12985 kB/s, 12680 KiB/s)
mmc0: Transfer of 64 x 128 sectors (64 x 64 KiB) took 0.291171741 seconds (14404 kB/s, 14067 KiB/s)
mmc0: Transfer of 32 x 256 sectors (32 x 128 KiB) took 0.275758964 seconds (15210 kB/s, 14853 KiB/s)
mmc0: Transfer of 16 x 512 sectors (16 x 256 KiB) took 0.268021558 seconds (15649 kB/s, 15282 KiB/s)
mmc0: Transfer of 8 x 1024 sectors (8 x 512 KiB) took 0.264064314 seconds (15883 kB/s, 15511 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 31. Consecutive write performance by transfer size...
mmc0: Transfer of 8192 x 1 sectors (8192 x 0.5 KiB) took 11.873861873 seconds (353 kB/s, 344 KiB/s)
mmc0: Transfer of 4096 x 2 sectors (4096 x 1 KiB) took 6.859102388 seconds (611 kB/s, 597 KiB/s)
mmc0: Transfer of 2048 x 4 sectors (2048 x 2 KiB) took 3.290696395 seconds (1274 kB/s, 1244 KiB/s)
mmc0: Transfer of 1024 x 8 sectors (1024 x 4 KiB) took 1.475266866 seconds (2843 kB/s, 2776 KiB/s)
mmc0: Transfer of 512 x 16 sectors (512 x 8 KiB) took 0.822344533 seconds (5100 kB/s, 4980 KiB/s)
mmc0: Transfer of 256 x 32 sectors (256 x 16 KiB) took 0.545847885 seconds (7684 kB/s, 7503 KiB/s)
mmc0: Transfer of 128 x 64 sectors (128 x 32 KiB) took 0.414100853 seconds (10128 kB/s, 9891 KiB/s)
mmc0: Transfer of 64 x 128 sectors (64 x 64 KiB) took 0.340080041 seconds (12333 kB/s, 12044 KiB/s)
mmc0: Transfer of 32 x 256 sectors (32 x 128 KiB) took 0.305309179 seconds (13737 kB/s, 13415 KiB/s)
mmc0: Transfer of 16 x 512 sectors (16 x 256 KiB) took 0.287391903 seconds (14594 kB/s, 14252 KiB/s)
mmc0: Transfer of 8 x 1024 sectors (8 x 512 KiB) took 0.279981325 seconds (14980 kB/s, 14629 KiB/s)
mmc0: Result: OK
mmc0: Tests completed.
mmc0: Starting tests of card mmc0:e624...
mmc0: Test case 32. Consecutive trim performance by transfer size...
mmc0: Result: UNSUPPORTED (by card)
mmc0: Tests completed.

-- 
Chris Ball   <cjb@laptop.org>   <http://printf.net/>
One Laptop Per Child

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [PATCH 2/2] mmc_test: fix display of .5 KiB
  2010-09-23 11:51 [PATCH 0/2] mmc_test performance test fixes Adrian Hunter
  2010-09-23 11:51 ` [PATCH 1/2] mmc_test: fix memory allocation segment limits Adrian Hunter
@ 2010-09-23 11:51 ` Adrian Hunter
  1 sibling, 0 replies; 4+ messages in thread
From: Adrian Hunter @ 2010-09-23 11:51 UTC (permalink / raw)
  To: Chris Ball; +Cc: linux-mmc Mailing List, Adrian Hunter

>From 676448f4d2623103cc5a4d246afe9ce32b7f339d Mon Sep 17 00:00:00 2001
From: Adrian Hunter <adrian.hunter@nokia.com>
Date: Thu, 23 Sep 2010 13:08:21 +0300
Subject: [PATCH 2/2] mmc_test: fix display of .5 KiB

Append .5 to KiB display when there are an odd number of
sectors.

Signed-off-by: Adrian Hunter <adrian.hunter@nokia.com>
---
 drivers/mmc/card/mmc_test.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mmc/card/mmc_test.c b/drivers/mmc/card/mmc_test.c
index 5874040..c38a3a8 100644
--- a/drivers/mmc/card/mmc_test.c
+++ b/drivers/mmc/card/mmc_test.c
@@ -539,7 +539,7 @@ static void mmc_test_print_rate(struct mmc_test_card *test, uint64_t bytes,
 	printk(KERN_INFO "%s: Transfer of %u sectors (%u%s KiB) took %lu.%09lu "
 			 "seconds (%u kB/s, %u KiB/s)\n",
 			 mmc_hostname(test->card->host), sectors, sectors >> 1,
-			 (sectors == 1 ? ".5" : ""), (unsigned long)ts.tv_sec,
+			 (sectors & 1 ? ".5" : ""), (unsigned long)ts.tv_sec,
 			 (unsigned long)ts.tv_nsec, rate / 1000, rate / 1024);
 
 	mmc_test_save_transfer_result(test, 1, sectors, ts, rate);
@@ -563,7 +563,7 @@ static void mmc_test_print_avg_rate(struct mmc_test_card *test, uint64_t bytes,
 	printk(KERN_INFO "%s: Transfer of %u x %u sectors (%u x %u%s KiB) took "
 			 "%lu.%09lu seconds (%u kB/s, %u KiB/s)\n",
 			 mmc_hostname(test->card->host), count, sectors, count,
-			 sectors >> 1, (sectors == 1 ? ".5" : ""),
+			 sectors >> 1, (sectors & 1 ? ".5" : ""),
 			 (unsigned long)ts.tv_sec, (unsigned long)ts.tv_nsec,
 			 rate / 1000, rate / 1024);
 
-- 
1.7.0.4


^ permalink raw reply related	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2010-09-24 21:15 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-09-23 11:51 [PATCH 0/2] mmc_test performance test fixes Adrian Hunter
2010-09-23 11:51 ` [PATCH 1/2] mmc_test: fix memory allocation segment limits Adrian Hunter
2010-09-24 21:15   ` Chris Ball
2010-09-23 11:51 ` [PATCH 2/2] mmc_test: fix display of .5 KiB Adrian Hunter

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox