From 9cc63f8bcd560c760d0b12e15bba8f81c86237cf Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 14 Aug 2026 15:43:51 -0700 Subject: Merge branches 'expcb.2026.07.24a', 'misc.2026.07.30a', 'rcu-tasks.2026.07.30a', 'srcu.2026.08.11a' and 'torture.2026.08.14a' into HEAD Changes: Make expedited grace periods expedite normal RCU callbacks Miscellaneous fixes: * Improve diagnostic output with character task states. * Mark accesses to inform KCSAN of concurrency design. * Move from kmalloc() to kmalloc_obj(). * Documentation updates. * Improve handling of RCU deferred quiescent states. * Clean up unused function arguments and structure fields. * Reduce show_rcu_gp_kthreads() stack space. Tasks RCU updates: * Clean up after SRCU re-implementation of Tasks Trace RCU. * Mark accesses to inform KCSAN of concurrency design. * Add ->lazy_timer status to diagnostic output. * Remove an unnecessary memory barrier. * Fix a data race, courtesy of KCSAN. * Documentation updates. * Convert cond_resched_tasks_rcu_qs() from macro to static inline function. SRCU updates: * Add Rust helpers for SRCU. * Avoid losing queued work at cleanup_srcu_struct() time. Torture-test updates: * Preparation work for immediate RCU priority deboosting. * Test RCU readers from real interrupt handlers (as opposed to softirq). * Simplify code through use of cpumask_next_wrap(). * Improve diagnostic output with character task states. * Add rcutorture.nwriters parameter to allow lightweight stall testing, and rcutorture.stall_only to make doing so easier. * Test an RCU Tasks Trace grace period implying an RCU grace period. * Make RCU Tasks Trace torturing track reader batches. * Fix a data race, courtesy of KCSAN. * Plug a shuffle_tmp_mask memory leak on kthread spawn failure. --- arch/arc/lib/strcmp.S | 93 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 93 insertions(+) create mode 100644 arch/arc/lib/strcmp.S (limited to 'arch/arc/lib/strcmp.S') diff --git a/arch/arc/lib/strcmp.S b/arch/arc/lib/strcmp.S new file mode 100644 index 000000000..b20c98fb3 --- /dev/null +++ b/arch/arc/lib/strcmp.S @@ -0,0 +1,93 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com) + */ + +/* This is optimized primarily for the ARC700. + It would be possible to speed up the loops by one cycle / word + respective one cycle / byte by forcing double source 1 alignment, unrolling + by a factor of two, and speculatively loading the second word / byte of + source 1; however, that would increase the overhead for loop setup / finish, + and strcmp might often terminate early. */ + +#include + +ENTRY_CFI(strcmp) + or r2,r0,r1 + bmsk_s r2,r2,1 + brne r2,0,.Lcharloop + mov_s r12,0x01010101 + ror r5,r12 +.Lwordloop: + ld.ab r2,[r0,4] + ld.ab r3,[r1,4] + nop_s + sub r4,r2,r12 + bic r4,r4,r2 + and r4,r4,r5 + brne r4,0,.Lfound0 + breq r2,r3,.Lwordloop +#ifdef __LITTLE_ENDIAN__ + xor r0,r2,r3 ; mask for difference + sub_s r1,r0,1 + bic_s r0,r0,r1 ; mask for least significant difference bit + sub r1,r5,r0 + xor r0,r5,r1 ; mask for least significant difference byte + and_s r2,r2,r0 + and_s r3,r3,r0 +#endif /* LITTLE ENDIAN */ + cmp_s r2,r3 + mov_s r0,1 + j_s.d [blink] + bset.lo r0,r0,31 + + .balign 4 +#ifdef __LITTLE_ENDIAN__ +.Lfound0: + xor r0,r2,r3 ; mask for difference + or r0,r0,r4 ; or in zero indicator + sub_s r1,r0,1 + bic_s r0,r0,r1 ; mask for least significant difference bit + sub r1,r5,r0 + xor r0,r5,r1 ; mask for least significant difference byte + and_s r2,r2,r0 + and_s r3,r3,r0 + sub.f r0,r2,r3 + mov.hi r0,1 + j_s.d [blink] + bset.lo r0,r0,31 +#else /* BIG ENDIAN */ + /* The zero-detection above can mis-detect 0x01 bytes as zeroes + because of carry-propagateion from a lower significant zero byte. + We can compensate for this by checking that bit0 is zero. + This compensation is not necessary in the step where we + get a low estimate for r2, because in any affected bytes + we already have 0x00 or 0x01, which will remain unchanged + when bit 7 is cleared. */ + .balign 4 +.Lfound0: + lsr r0,r4,8 + lsr_s r1,r2 + bic_s r2,r2,r0 ; get low estimate for r2 and get ... + bic_s r0,r0,r1 ; + or_s r3,r3,r0 ; ... high estimate r3 so that r2 > r3 will ... + cmp_s r3,r2 ; ... be independent of trailing garbage + or_s r2,r2,r0 ; likewise for r3 > r2 + bic_s r3,r3,r0 + rlc r0,0 ; r0 := r2 > r3 ? 1 : 0 + cmp_s r2,r3 + j_s.d [blink] + bset.lo r0,r0,31 +#endif /* ENDIAN */ + + .balign 4 +.Lcharloop: + ldb.ab r2,[r0,1] + ldb.ab r3,[r1,1] + nop_s + breq r2,0,.Lcmpend + breq r2,r3,.Lcharloop +.Lcmpend: + j_s.d [blink] + sub r0,r2,r3 +END_CFI(strcmp) -- cgit v1.3.1