vendor: bump Pebble to bb2c1501ac23

[jbowens]

bb2c1501 tool/logs: print total flush / ingestion counts and durations
d1cf3412 tool/logs: sort output by (start, node, store)
0a77780a tool/logs: fix total time calculation
0011c057 db: fix data race in SeekPrefixGE
99f35c80 internal/keyspan: gate invariant checks behind build tag
00302bfc internal/keyspan: use pointer receiver for most Span methods
3355a02e internal/rangekey: reduce range-key iteration allocations
4e33626f db: fix bug in interaction between Clone and indexed batches
b14ad704 metamorphic: enable writer parallelism for randomzied metamorphic runs
9bd43143 db: reduce allocations in range key iteration
e32e94d8 db: standardize behavior of indexed batch mutations during iteration
a07efd96 internal/keyspan: lift visibility filtering out of Seek{GE,LE}

Release note: none

[cockroach-teamcity]

This change is 

[nicktrav]

Reviewed 5 of 5 files at r1, all commit messages.
Reviewable status: complete! 1 of 0 LGTMs obtained (waiting on @jbowens)

[jbowens]

TFTR!

bors r=nicktrav

[craig]

Build succeeded:

• GitHub CI (Cockroach)

[yuzefovich]

I tracked down one of the microbenchmark regressions to this commit:

BENCHTIMEOUT=24h PKG=./pkg/bench BENCHES=BenchmarkSQL/MultinodeCockroach/Upsert/count=1000 ./scripts/bench 41c642a60524fd3a8c713a4ed1048ad1e53df2ab^ 41c642a60524fd3a8c713a4ed1048ad1e53df2ab
...
+ ./dev bench ./pkg/bench --timeout=24h --filter=BenchmarkSQL/MultinodeCockroach/Upsert/count=1000 --count=10 --bench-mem -v --stream-output --ignore-cache
name                                         old time/op    new time/op    delta
SQL/MultinodeCockroach/Upsert/count=1000-24    22.9ms ± 6%    24.3ms ± 8%  +6.19%  (p=0.007 n=10+10)

name                                         old alloc/op   new alloc/op   delta
SQL/MultinodeCockroach/Upsert/count=1000-24    11.1MB ±21%    11.2MB ±18%    ~     (p=0.853 n=10+10)

name                                         old allocs/op  new allocs/op  delta
SQL/MultinodeCockroach/Upsert/count=1000-24     56.7k ±15%     55.8k ±15%    ~     (p=0.190 n=10+10)

Is this concerning?

[jbowens]

@yuzefovich Is the regression still present on the current release-22.2 HEAD or master HEAD? There was a regression introduced in this pull request that was very recently fixed.

[yuzefovich]

The regression is still seen against both HEADs. However, there are other causes for the regression, so I think the next timestep is to check whether the fix you have in mind claws back the regression from this commit. The command to use is BENCHTIMEOUT=24h PKG=./pkg/bench BENCHES=BenchmarkSQL/MultinodeCockroach/Upsert/count=1000 ./scripts/bench <sha-before-fix> <sha-with-fix>.