Date: Tue, 15 Oct 2024 06:21:13 +0000 From: Paul Floyd <paulf2718@gmail.com> To: freebsd-current@freebsd.org Subject: Re: Problems with FreeBSD-15.0-CURRENT-amd64-20241003 Message-ID: <48a4c570-9c68-4e90-b468-0c305baccb44@gmail.com> In-Reply-To: <Zwe3oX7HJ-jtsyP_@kib.kiev.ua> References: <7bdb3c71-8a36-444e-8b1d-9c4f789fe638@gmail.com> <ZwYrWwd5XU_TRbEw@kib.kiev.ua> <CALUVJ=AhS1NA_4JNEC-c2hMjBMHhNZh0VzLjdvCVDh5siDrriw@mail.gmail.com> <ZwZD6rP1eQ2ehllX@kib.kiev.ua> <62745a68-9ed7-4f6f-bd6f-0ba3e10629c3@gmail.com> <Zwe3oX7HJ-jtsyP_@kib.kiev.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On 10-10-24 11:16, Konstantin Belousov wrote: > If you can determine not just the number of taken locks, but also the > specific locks that were taken, perhaps you can check that this is > the libc exit lock? For the moment I've turned off the check for FreeBSD 15 and above. At present the ExeContext is not saved for each lock. Each thread has a set of held lock IDs. It might be useful to change that, depending on the memory and time overhead. I'll try to see if there is an efficient way to check that the ExeContext for pthread_mutex_lock is exit/__libc_start1. A+ Paul
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?48a4c570-9c68-4e90-b468-0c305baccb44>