* [PATCH] block: badblocks: Remove excess comparisons of p[hi - 1]
@ 2026-09-13 17:25 Sean Young
0 siblings, 0 replies; only message in thread
From: Sean Young @ 2026-09-13 17:25 UTC (permalink / raw)
To: linux-kernel, Jens Axboe; +Cc: Sean Young, linux-block
The binary search in prev_badblocks() is correct, but it is doing
excessive comparisons. Once p[hi - 1] is compared we do not have to
compare it again.
Also remove the if after the loop since the condition is always true;
s is always greater or equal too p[lo]; on function entry, if p[lo]) > s
is true we exit early, and then lo is set to mid if p[mid] < s.
The first issue was found with an LLM, the second on code inspection.
Signed-off-by: Sean Young <sean@mess.org>
---
block/badblocks.c | 9 ++++-----
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/block/badblocks.c b/block/badblocks.c
index ece64e76fe8f..143b488b7fe6 100644
--- a/block/badblocks.c
+++ b/block/badblocks.c
@@ -496,14 +496,14 @@ static int prev_badblocks(struct badblocks *bb, struct badblocks_context *bad,
}
lo = 0;
- hi = bb->count;
+ hi = bb->count - 1;
p = bb->page;
/* The following bisect search might be unnecessary */
if (BB_OFFSET(p[lo]) > s)
return -1;
- if (BB_OFFSET(p[hi - 1]) <= s)
- return hi - 1;
+ if (BB_OFFSET(p[hi]) <= s)
+ return hi;
/* Do bisect search in bad table */
while (hi - lo > 1) {
@@ -521,8 +521,7 @@ static int prev_badblocks(struct badblocks *bb, struct badblocks_context *bad,
hi = mid;
}
- if (BB_OFFSET(p[lo]) <= s)
- ret = lo;
+ ret = lo;
out:
return ret;
}
--
2.55.0
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2026-09-13 17:26 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-09-13 17:25 [PATCH] block: badblocks: Remove excess comparisons of p[hi - 1] Sean Young
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox