Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1462213

[PATCH 3.2 03/94] crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks

From Ben Hutchings <ben@decadent.org.uk>
Newsgroups linux.kernel
Subject [PATCH 3.2 03/94] crypto: s5p-sss - Fix missed interrupts when working with 8 kB blocks
Date 2016-08-14 20:00 +0200
Message-ID <s67Rx-2Hd-73@gated-at.bofh.it> (permalink)
References <s5LnX-4sk-7@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw

3.2.82-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Krzysztof Kozlowski <k.kozlowski@samsung.com>

commit 79152e8d085fd64484afd473ef6830b45518acba upstream.

The tcrypt testing module on Exynos5422-based Odroid XU3/4 board failed on
testing 8 kB size blocks:

	$ sudo modprobe tcrypt sec=1 mode=500
	testing speed of async ecb(aes) (ecb-aes-s5p) encryption
	test 0 (128 bit key, 16 byte blocks): 21971 operations in 1 seconds (351536 bytes)
	test 1 (128 bit key, 64 byte blocks): 21731 operations in 1 seconds (1390784 bytes)
	test 2 (128 bit key, 256 byte blocks): 21932 operations in 1 seconds (5614592 bytes)
	test 3 (128 bit key, 1024 byte blocks): 21685 operations in 1 seconds (22205440 bytes)
	test 4 (128 bit key, 8192 byte blocks):

This was caused by a race issue of missed BRDMA_DONE ("Block cipher
Receiving DMA") interrupt. Device starts processing the data in DMA mode
immediately after setting length of DMA block: receiving (FCBRDMAL) or
transmitting (FCBTDMAL). The driver sets these lengths from interrupt
handler through s5p_set_dma_indata() function (or xxx_setdata()).

However the interrupt handler was first dealing with receive buffer
(dma-unmap old, dma-map new, set receive block length which starts the
operation), then with transmit buffer and finally was clearing pending
interrupts (FCINTPEND). Because of the time window between setting
receive buffer length and clearing pending interrupts, the operation on
receive buffer could end already and driver would miss new interrupt.

User manual for Exynos5422 confirms in example code that setting DMA
block lengths should be the last operation.

The tcrypt hang could be also observed in following blocked-task dmesg:

INFO: task modprobe:258 blocked for more than 120 seconds.
      Not tainted 4.6.0-rc4-next-20160419-00005-g9eac8b7b7753-dirty #42
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
modprobe        D c06b09d8     0   258    256 0x00000000
[<c06b09d8>] (__schedule) from [<c06b0f24>] (schedule+0x40/0xac)
[<c06b0f24>] (schedule) from [<c06b49f8>] (schedule_timeout+0x124/0x178)
[<c06b49f8>] (schedule_timeout) from [<c06b17fc>] (wait_for_common+0xb8/0x144)
[<c06b17fc>] (wait_for_common) from [<bf0013b8>] (test_acipher_speed+0x49c/0x740 [tcrypt])
[<bf0013b8>] (test_acipher_speed [tcrypt]) from [<bf003e8c>] (do_test+0x2240/0x30ec [tcrypt])
[<bf003e8c>] (do_test [tcrypt]) from [<bf008048>] (tcrypt_mod_init+0x48/0xa4 [tcrypt])
[<bf008048>] (tcrypt_mod_init [tcrypt]) from [<c010177c>] (do_one_initcall+0x3c/0x16c)
[<c010177c>] (do_one_initcall) from [<c0191ff0>] (do_init_module+0x5c/0x1ac)
[<c0191ff0>] (do_init_module) from [<c0185610>] (load_module+0x1a30/0x1d08)
[<c0185610>] (load_module) from [<c0185ab0>] (SyS_finit_module+0x8c/0x98)
[<c0185ab0>] (SyS_finit_module) from [<c01078c0>] (ret_fast_syscall+0x0/0x3c)

Fixes: a49e490c7a8a ("crypto: s5p-sss - add S5PV210 advanced crypto engine support")
Signed-off-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
Tested-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
 drivers/crypto/s5p-sss.c | 53 +++++++++++++++++++++++++++++++++++-------------
 1 file changed, 39 insertions(+), 14 deletions(-)

--- a/drivers/crypto/s5p-sss.c
+++ b/drivers/crypto/s5p-sss.c
@@ -258,39 +258,51 @@ static int s5p_set_indata(struct s5p_aes
 	return err;
 }
 
-static void s5p_aes_tx(struct s5p_aes_dev *dev)
+/*
+ * Returns true if new transmitting (output) data is ready and its
+ * address+length have to be written to device (by calling
+ * s5p_set_dma_outdata()). False otherwise.
+ */
+static bool s5p_aes_tx(struct s5p_aes_dev *dev)
 {
 	int err = 0;
+	bool ret = false;
 
 	s5p_unset_outdata(dev);
 
 	if (!sg_is_last(dev->sg_dst)) {
 		err = s5p_set_outdata(dev, sg_next(dev->sg_dst));
-		if (err) {
+		if (err)
 			s5p_aes_complete(dev, err);
-			return;
-		}
-
-		s5p_set_dma_outdata(dev, dev->sg_dst);
+		else
+			ret = true;
 	} else
 		s5p_aes_complete(dev, err);
+
+	return ret;
 }
 
-static void s5p_aes_rx(struct s5p_aes_dev *dev)
+/*
+ * Returns true if new receiving (input) data is ready and its
+ * address+length have to be written to device (by calling
+ * s5p_set_dma_indata()). False otherwise.
+ */
+static bool s5p_aes_rx(struct s5p_aes_dev *dev)
 {
 	int err;
+	bool ret = false;
 
 	s5p_unset_indata(dev);
 
 	if (!sg_is_last(dev->sg_src)) {
 		err = s5p_set_indata(dev, sg_next(dev->sg_src));
-		if (err) {
+		if (err)
 			s5p_aes_complete(dev, err);
-			return;
-		}
-
-		s5p_set_dma_indata(dev, dev->sg_src);
+		else
+			ret = true;
 	}
+
+	return ret;
 }
 
 static irqreturn_t s5p_aes_interrupt(int irq, void *dev_id)
@@ -299,19 +311,32 @@ static irqreturn_t s5p_aes_interrupt(int
 	struct s5p_aes_dev     *dev  = platform_get_drvdata(pdev);
 	uint32_t                status;
 	unsigned long           flags;
+	bool			set_dma_tx = false;
+	bool			set_dma_rx = false;
 
 	spin_lock_irqsave(&dev->lock, flags);
 
 	if (irq == dev->irq_fc) {
 		status = SSS_READ(dev, FCINTSTAT);
 		if (status & SSS_FCINTSTAT_BRDMAINT)
-			s5p_aes_rx(dev);
+			set_dma_rx = s5p_aes_rx(dev);
 		if (status & SSS_FCINTSTAT_BTDMAINT)
-			s5p_aes_tx(dev);
+			set_dma_tx = s5p_aes_tx(dev);
 
 		SSS_WRITE(dev, FCINTPEND, status);
 	}
 
+	/*
+	 * Writing length of DMA block (either receiving or transmitting)
+	 * will start the operation immediately, so this should be done
+	 * at the end (even after clearing pending interrupts to not miss the
+	 * interrupt).
+	 */
+	if (set_dma_tx)
+		s5p_set_dma_outdata(dev, dev->sg_dst);
+	if (set_dma_rx)
+		s5p_set_dma_indata(dev, dev->sg_src);
+
 	spin_unlock_irqrestore(&dev->lock, flags);
 
 	return IRQ_HANDLED;

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 3.2 00/94] 3.2.82-rc1 review Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 13:50 +0200
  [PATCH 3.2 26/94] fs/cifs: correctly to anonymous authentication  for the LANMAN authentication Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 13:50 +0200
  [PATCH 3.2 23/94] x86/PCI: Mark Broadwell-EP Home Agent 1 as  having non-compliant BARs Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 13:50 +0200
  [PATCH 3.2 93/94] tcp: make challenge acks less predictable Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 13:50 +0200
  [PATCH 3.2 62/94] UBIFS: Implement ->migratepage() Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 42/94] x86, build: copy ldlinux.c32 to image.iso Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 54/94] kvm: Fix irq route entries exceeding  KVM_MAX_IRQ_ROUTES Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 13/94] USB: serial: option: add even more ZTE device ids Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 35/94] Input: xpad - prevent spurious input from wired  Xbox 360 controllers Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 43/94] fix d_walk()/non-delayed __d_free() race Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 18/94] powerpc/mm/hash64: Factor out hash preload  psize check Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:00 +0200
  [PATCH 3.2 86/94] proc: prevent stacking filesystems on top Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 34/94] RDMA/cxgb3: device driver frees DMA memory with  different size Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 22/94] net/mlx4_core: Fix access to uninitialized index Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 82/94] xenbus: don't BUG() on user mode induced condition Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 69/94] ALSA: dummy: Fix a use-after-free at closing Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 74/94] NFS: Fix another OPEN_DOWNGRADE bug Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 73/94] ALSA: echoaudio: Fix memory allocation Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 61/94] mm: Export migrate_page_move_mapping and  migrate_page_copy Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  [PATCH 3.2 02/94] ath5k: Change led pin configuration for compaq  c700 laptop Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 14:10 +0200
  Re: [PATCH 3.2 00/94] 3.2.82-rc1 review Guenter Roeck <linux@roeck-us.net> - 2016-08-14 17:10 +0200
  [PATCH 3.2 39/94] ARM: fix PTRACE_SETVFPREGS on SMP systems Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 85/94] fs: limit filesystem stacking depth Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 11/94] tty: vt, return error when con_startup fails Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 91/94] tipc: fix an infoleak in tipc_nl_compat_link_dump Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 15/94] MIPS: Adjust set_pte() SMP fix to handle  R10000_LLSC_WAR Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 60/94] ubi: Make recover_peb power cut aware Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 03/94] crypto: s5p-sss - Fix missed interrupts when  working with 8 kB blocks Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 64/94] xen/pciback: Fix conf_space read/write overlap  check. Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 25/94] fs/cifs: correctly to anonymous authentication  via NTLMSSP Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:00 +0200
  [PATCH 3.2 37/94] Input: pwm-beeper - fix - scheduling while atomic Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 21/94] mmc: mmc: Fix partition switch timeout for some  eMMCs Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 88/94] ALSA: timer: Fix leak in SNDRV_TIMER_IOCTL_PARAMS Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 01/94] regmap: cache: Fix typo in cache_bypass  parameter description Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 63/94] can: fix oops caused by wrong rtnl dellink usage Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 77/94] x86/amd_nb: Fix boot crash on non-AMD systems Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 79/94] bonding: prevent out of bound accesses Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 06/94] crypto: s5p-sss - fix incorrect usage of  scatterlists api Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 59/94] can: at91_can: RX queue could get stuck at high  bus load Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 70/94] fs/nilfs2: fix potential underflow in call to  crc32_le Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 71/94] staging: iio: accel: fix error check Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 84/94] cifs: dynamic allocation of ntlmssp blob Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 52/94] kernel/sysrq, watchdog, sched/core: Reset  watchdog on all CPUs while processing sysrq-w Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 40/94] KVM: x86: fix OOPS after invalid KVM_SET_DEBUGREGS Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 53/94] base: make module_create_drivers_dir race-free Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:10 +0200
  [PATCH 3.2 87/94] USB: usbfs: fix potential infoleak in devio Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:20 +0200
  [PATCH 3.2 48/94] staging:iio: trigger fixes for repeat request  of same trigger and allocation failure Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:20 +0200
  [PATCH 3.2 09/94] ext4: fix hang when processing corrupted  orphaned inode list Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:20 +0200
  [PATCH 3.2 75/94] batman-adv: Fix use-after-free/double-free of  tt_req_node Ben Hutchings <ben@decadent.org.uk> - 2016-08-14 20:20 +0200

csiph-web