jan.kratochvil at redhat dot com
2017-11-02 21:16:50 UTC
https://sourceware.org/bugzilla/show_bug.cgi?id=22389
Bug ID: 22389
Summary: aarch64: watchpoints are not merged regression
Product: gdb
Version: HEAD
Status: NEW
Severity: normal
Priority: P2
Component: tdep
Assignee: unassigned at sourceware dot org
Reporter: jan.kratochvil at redhat dot com
Target Milestone: ---
A tracker only.
As discussed in:
Re: ping#2: [patch] aarch64: PR 19806: watchpoints: false negatives + PR
20207 contiguous ones
https://sourceware.org/ml/gdb-patches/2017-10/msg00878.html
Formerly GDB did merge all watchpoints happening inside an 1/2/4/8 byte aligned
region to 8-byte boundary and then reused the same hardware watchpoint register
as they became the same.
New GDB does not merge them as it can properly prevent false positives (or
formerly false negatives). But then it is now using separate hardware
watchpoint register for each different watchpoint - despite GDB could merge
adjacent watchpoints into a single register (without any false
positives/negatives side effects).
Bug ID: 22389
Summary: aarch64: watchpoints are not merged regression
Product: gdb
Version: HEAD
Status: NEW
Severity: normal
Priority: P2
Component: tdep
Assignee: unassigned at sourceware dot org
Reporter: jan.kratochvil at redhat dot com
Target Milestone: ---
A tracker only.
As discussed in:
Re: ping#2: [patch] aarch64: PR 19806: watchpoints: false negatives + PR
20207 contiguous ones
https://sourceware.org/ml/gdb-patches/2017-10/msg00878.html
Formerly GDB did merge all watchpoints happening inside an 1/2/4/8 byte aligned
region to 8-byte boundary and then reused the same hardware watchpoint register
as they became the same.
New GDB does not merge them as it can properly prevent false positives (or
formerly false negatives). But then it is now using separate hardware
watchpoint register for each different watchpoint - despite GDB could merge
adjacent watchpoints into a single register (without any false
positives/negatives side effects).
--
You are receiving this mail because:
You are on the CC list for the bug.
You are receiving this mail because:
You are on the CC list for the bug.