From patchwork Tue Jan 15 13:12:57 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joyce Kong X-Patchwork-Id: 49839 X-Patchwork-Delegate: thomas@monjalon.net Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 8FA225F17; Tue, 15 Jan 2019 14:13:13 +0100 (CET) Received: from foss.arm.com (usa-sjc-mx-foss1.foss.arm.com [217.140.101.70]) by dpdk.org (Postfix) with ESMTP id 2ADCB5B2E; Tue, 15 Jan 2019 14:13:11 +0100 (CET) Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 76FEFEBD; Tue, 15 Jan 2019 05:13:10 -0800 (PST) Received: from net-arm-thunderx2.shanghai.arm.com (net-arm-thunderx2.shanghai.arm.com [10.169.40.106]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id AD6C63F70D; Tue, 15 Jan 2019 05:13:08 -0800 (PST) From: Joyce Kong To: dev@dpdk.org Cc: thomas@monjalon.net, jerinj@marvell.com, hemant.agrawal@nxp.com, bruce.richardson@intel.com, chaozhu@linux.vnet.ibm.com, honnappa.nagarahalli@arm.com, nd@arm.com, Gavin Hu , stable@dpdk.org Date: Tue, 15 Jan 2019 21:12:57 +0800 Message-Id: <1547557979-153169-2-git-send-email-joyce.kong@arm.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1547557979-153169-1-git-send-email-joyce.kong@arm.com> References: <1547557979-153169-1-git-send-email-joyce.kong@arm.com> Subject: [dpdk-dev] [PATCH v2 1/3] rwlock: reimplement with __atomic builtins X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" From: Gavin Hu The __sync builtin based implementation generates full memory barriers ('dmb ish') on Arm platforms. Using C11 atomic builtins to generate one way barriers. Here is the assembly code of __sync_compare_and_swap builtin. __sync_bool_compare_and_swap(dst, exp, src); 0x000000000090f1b0 <+16>: e0 07 40 f9 ldr x0, [sp, #8] 0x000000000090f1b4 <+20>: e1 0f 40 79 ldrh w1, [sp, #6] 0x000000000090f1b8 <+24>: e2 0b 40 79 ldrh w2, [sp, #4] 0x000000000090f1bc <+28>: 21 3c 00 12 and w1, w1, #0xffff 0x000000000090f1c0 <+32>: 03 7c 5f 48 ldxrh w3, [x0] 0x000000000090f1c4 <+36>: 7f 00 01 6b cmp w3, w1 0x000000000090f1c8 <+40>: 61 00 00 54 b.ne 0x90f1d4 // b.any 0x000000000090f1cc <+44>: 02 fc 04 48 stlxrh w4, w2, [x0] 0x000000000090f1d0 <+48>: 84 ff ff 35 cbnz w4, 0x90f1c0 0x000000000090f1d4 <+52>: bf 3b 03 d5 dmb ish 0x000000000090f1d8 <+56>: e0 17 9f 1a cset w0, eq // eq = none Fixes: af75078fece3 ("first public release") Cc: stable@dpdk.org Signed-off-by: Gavin Hu Reviewed-by: Honnappa Nagarahalli Reviewed-by: Ola Liljedahl Reviewed-by: Joyce Kong Tested-by: Joyce Kong Acked-by: Jerin Jacob --- lib/librte_eal/common/include/generic/rte_rwlock.h | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/lib/librte_eal/common/include/generic/rte_rwlock.h b/lib/librte_eal/common/include/generic/rte_rwlock.h index b05d85a..3317a21 100644 --- a/lib/librte_eal/common/include/generic/rte_rwlock.h +++ b/lib/librte_eal/common/include/generic/rte_rwlock.h @@ -64,14 +64,14 @@ rte_rwlock_read_lock(rte_rwlock_t *rwl) int success = 0; while (success == 0) { - x = rwl->cnt; + x = __atomic_load_n(&rwl->cnt, __ATOMIC_RELAXED); /* write lock is held */ if (x < 0) { rte_pause(); continue; } - success = rte_atomic32_cmpset((volatile uint32_t *)&rwl->cnt, - (uint32_t)x, (uint32_t)(x + 1)); + success = __atomic_compare_exchange_n(&rwl->cnt, &x, x+1, 1, + __ATOMIC_ACQUIRE, __ATOMIC_RELAXED); } } @@ -114,7 +114,7 @@ rte_rwlock_read_trylock(rte_rwlock_t *rwl) static inline void rte_rwlock_read_unlock(rte_rwlock_t *rwl) { - rte_atomic32_dec((rte_atomic32_t *)(intptr_t)&rwl->cnt); + __atomic_fetch_sub(&rwl->cnt, 1, __ATOMIC_RELEASE); } /** @@ -156,14 +156,14 @@ rte_rwlock_write_lock(rte_rwlock_t *rwl) int success = 0; while (success == 0) { - x = rwl->cnt; + x = __atomic_load_n(&rwl->cnt, __ATOMIC_RELAXED); /* a lock is held */ if (x != 0) { rte_pause(); continue; } - success = rte_atomic32_cmpset((volatile uint32_t *)&rwl->cnt, - 0, (uint32_t)-1); + success = __atomic_compare_exchange_n(&rwl->cnt, &x, -1, 1, + __ATOMIC_ACQUIRE, __ATOMIC_RELAXED); } } @@ -176,7 +176,7 @@ rte_rwlock_write_lock(rte_rwlock_t *rwl) static inline void rte_rwlock_write_unlock(rte_rwlock_t *rwl) { - rte_atomic32_inc((rte_atomic32_t *)(intptr_t)&rwl->cnt); + __atomic_store_n(&rwl->cnt, 0, __ATOMIC_RELEASE); } /** From patchwork Tue Jan 15 13:12:58 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joyce Kong X-Patchwork-Id: 49840 X-Patchwork-Delegate: thomas@monjalon.net Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id A6E7F5F2F; Tue, 15 Jan 2019 14:13:15 +0100 (CET) Received: from foss.arm.com (usa-sjc-mx-foss1.foss.arm.com [217.140.101.70]) by dpdk.org (Postfix) with ESMTP id BFE435B40; Tue, 15 Jan 2019 14:13:12 +0100 (CET) Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 39046A78; Tue, 15 Jan 2019 05:13:12 -0800 (PST) Received: from net-arm-thunderx2.shanghai.arm.com (net-arm-thunderx2.shanghai.arm.com [10.169.40.106]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id B45ED3F70D; Tue, 15 Jan 2019 05:13:10 -0800 (PST) From: Joyce Kong To: dev@dpdk.org Cc: thomas@monjalon.net, jerinj@marvell.com, hemant.agrawal@nxp.com, bruce.richardson@intel.com, chaozhu@linux.vnet.ibm.com, honnappa.nagarahalli@arm.com, nd@arm.com, stable@dpdk.org Date: Tue, 15 Jan 2019 21:12:58 +0800 Message-Id: <1547557979-153169-3-git-send-email-joyce.kong@arm.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1547557979-153169-1-git-send-email-joyce.kong@arm.com> References: <1547557979-153169-1-git-send-email-joyce.kong@arm.com> Subject: [dpdk-dev] [PATCH v2 2/3] test/rwlock: add perf test case X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Add performance test on all available cores to benchmark the scaling up performance of rw_lock. Fixes: af75078fece3 ("first public release") Cc: stable@dpdk.org Suggested-by: Gavin Hu Signed-off-by: Joyce Kong Reviewed-by: Honnappa Nagarahalli Reviewed-by: Ola Liljedahl Reviewed-by: Gavin Hu Reviewed-by: Ruifeng Wang --- test/test/test_rwlock.c | 73 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) diff --git a/test/test/test_rwlock.c b/test/test/test_rwlock.c index 224f0de..250ea5a 100644 --- a/test/test/test_rwlock.c +++ b/test/test/test_rwlock.c @@ -36,6 +36,7 @@ static rte_rwlock_t sl; static rte_rwlock_t sl_tab[RTE_MAX_LCORE]; +static rte_atomic32_t synchro; enum { LC_TYPE_RDLOCK, @@ -83,6 +84,75 @@ test_rwlock_per_core(__attribute__((unused)) void *arg) return 0; } +static rte_rwlock_t lk = RTE_RWLOCK_INITIALIZER; +static volatile uint64_t rwlock_data; +static uint64_t lock_count[RTE_MAX_LCORE] = {0}; + +#define TIME_MS 100 + +static int +load_loop_fn(__attribute__((unused)) void *arg) +{ + uint64_t time_diff = 0, begin; + uint64_t hz = rte_get_timer_hz(); + uint64_t lcount = 0; + const unsigned int lcore = rte_lcore_id(); + + /* wait synchro for slaves */ + if (lcore != rte_get_master_lcore()) + while (rte_atomic32_read(&synchro) == 0) + ; + + begin = rte_rdtsc_precise(); + while (time_diff < hz * TIME_MS / 1000) { + rte_rwlock_write_lock(&lk); + ++rwlock_data; + rte_rwlock_write_unlock(&lk); + + rte_rwlock_read_lock(&lk); + printf("Core [%u] rwlock_data = %"PRIu64"\n", + lcore, rwlock_data); + rte_rwlock_read_unlock(&lk); + + lcount++; + /* delay to make lock duty cycle slightly realistic */ + rte_pause(); + time_diff = rte_rdtsc_precise() - begin; + } + + lock_count[lcore] = lcount; + return 0; +} + +static int +test_rwlock_perf(void) +{ + unsigned int i; + uint64_t total = 0; + + printf("\nRwlock Perf Test on %u cores...\n", rte_lcore_count()); + + /* clear synchro and start slaves */ + rte_atomic32_set(&synchro, 0); + if (rte_eal_mp_remote_launch(load_loop_fn, NULL, SKIP_MASTER) < 0) + return -1; + + /* start synchro and launch test on master */ + rte_atomic32_set(&synchro, 1); + load_loop_fn(NULL); + + rte_eal_mp_wait_lcore(); + + RTE_LCORE_FOREACH(i) { + printf("Core [%u] count = %"PRIu64"\n", i, lock_count[i]); + total += lock_count[i]; + } + + printf("Total count = %"PRIu64"\n", total); + + return 0; +} + /* * - There is a global rwlock and a table of rwlocks (one per lcore). * @@ -132,6 +202,9 @@ rwlock_test1(void) rte_eal_mp_wait_lcore(); + if (test_rwlock_perf() < 0) + return -1; + return 0; } From patchwork Tue Jan 15 13:12:59 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joyce Kong X-Patchwork-Id: 49841 X-Patchwork-Delegate: thomas@monjalon.net Return-Path: X-Original-To: patchwork@dpdk.org Delivered-To: patchwork@dpdk.org Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 2279F6CC1; Tue, 15 Jan 2019 14:13:17 +0100 (CET) Received: from foss.arm.com (usa-sjc-mx-foss1.foss.arm.com [217.140.101.70]) by dpdk.org (Postfix) with ESMTP id 7E5C05F25; Tue, 15 Jan 2019 14:13:14 +0100 (CET) Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id EB285EBD; Tue, 15 Jan 2019 05:13:13 -0800 (PST) Received: from net-arm-thunderx2.shanghai.arm.com (net-arm-thunderx2.shanghai.arm.com [10.169.40.106]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 75A6E3F70D; Tue, 15 Jan 2019 05:13:12 -0800 (PST) From: Joyce Kong To: dev@dpdk.org Cc: thomas@monjalon.net, jerinj@marvell.com, hemant.agrawal@nxp.com, bruce.richardson@intel.com, chaozhu@linux.vnet.ibm.com, honnappa.nagarahalli@arm.com, nd@arm.com, stable@dpdk.org Date: Tue, 15 Jan 2019 21:12:59 +0800 Message-Id: <1547557979-153169-4-git-send-email-joyce.kong@arm.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1547557979-153169-1-git-send-email-joyce.kong@arm.com> References: <1547557979-153169-1-git-send-email-joyce.kong@arm.com> Subject: [dpdk-dev] [PATCH v2 3/3] test/rwlock: amortize the cost of getting time X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Fixes: af75078fece3 ("first public release") Cc: stable@dpdk.org Signed-off-by: Joyce Kong --- test/test/test_rwlock.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/test/test/test_rwlock.c b/test/test/test_rwlock.c index 250ea5a..d42ac35 100644 --- a/test/test/test_rwlock.c +++ b/test/test/test_rwlock.c @@ -86,9 +86,9 @@ test_rwlock_per_core(__attribute__((unused)) void *arg) static rte_rwlock_t lk = RTE_RWLOCK_INITIALIZER; static volatile uint64_t rwlock_data; -static uint64_t lock_count[RTE_MAX_LCORE] = {0}; +static uint64_t time_count[RTE_MAX_LCORE] = {0}; -#define TIME_MS 100 +#define MAX_LOOP 10000 static int load_loop_fn(__attribute__((unused)) void *arg) @@ -104,7 +104,7 @@ load_loop_fn(__attribute__((unused)) void *arg) ; begin = rte_rdtsc_precise(); - while (time_diff < hz * TIME_MS / 1000) { + while (lcount < MAX_LOOP) { rte_rwlock_write_lock(&lk); ++rwlock_data; rte_rwlock_write_unlock(&lk); @@ -117,10 +117,10 @@ load_loop_fn(__attribute__((unused)) void *arg) lcount++; /* delay to make lock duty cycle slightly realistic */ rte_pause(); - time_diff = rte_rdtsc_precise() - begin; } - lock_count[lcore] = lcount; + time_diff = rte_rdtsc_precise() - begin; + time_count[lcore] = time_diff * 1000000 / hz; return 0; } @@ -144,11 +144,13 @@ test_rwlock_perf(void) rte_eal_mp_wait_lcore(); RTE_LCORE_FOREACH(i) { - printf("Core [%u] count = %"PRIu64"\n", i, lock_count[i]); - total += lock_count[i]; + printf("Core [%u] cost time = %"PRIu64" us\n", + i, time_count[i]); + total += time_count[i]; } - printf("Total count = %"PRIu64"\n", total); + printf("Total cost time = %"PRIu64" us\n", total); + memset(time_count, 0, sizeof(time_count)); return 0; }