Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Simon Marchi <simark@simark.ca>
To: John Baldwin <jhb@FreeBSD.org>, gdb-patches@sourceware.org
Subject: Re: [PATCH v2 07/11] Add a helper function to resolve TLS variable addresses for FreeBSD.
Date: Thu, 07 Mar 2019 16:18:00 -0000	[thread overview]
Message-ID: <eddc12ba-1b15-f2ad-1a33-9b413ae28b9a@simark.ca> (raw)
In-Reply-To: <c7e57cc0f4e08580f7f78d5cc5e111b3db7637bf.1549672588.git.jhb@FreeBSD.org>

On 2019-02-08 7:40 p.m., John Baldwin wrote:
> The fbsd_get_thread_local_address function accepts the base address of
> a thread's DTV array and the base address of an object file's link map
> and uses this to compute a TLS variable's address.  FreeBSD
> architectures use an architecture-specific method to determine the
> address of the DTV array pointer and call this helper function to
> perform the rest of the address calculation.

I didn't check against FreeBSD's data structures (I trust you for that 
:)), but in general this LGTM.  Just some formatting nits.

> +/* Per-program-space data for FreeBSD architectures.  */
> +static const struct program_space_data *fbsd_pspace_data_handle;
> +
> +struct fbsd_pspace_data
> +  {
> +    /* Offsets in the runtime linker's 'Obj_Entry' structure.  */
> +    LONGEST off_linkmap;
> +    LONGEST off_tlsindex;
> +    bool rtld_offsets_valid;
> +  };

Unindent the { } and what's between.

> +/* Lookup offsets of fields in the runtime linker's 'Obj_Entry'
> +   structure needed to determine the TLS index of an object file.  */
> +
> +static void
> +fbsd_fetch_rtld_offsets (struct gdbarch *gdbarch, struct fbsd_pspace_data *data)
> +{
> +  TRY
> +    {
> +      /* Fetch offsets from debug symbols in rtld.  */
> +      struct symbol *obj_entry_sym
> +	= lookup_symbol_in_language ("Struct_Obj_Entry", NULL, STRUCT_DOMAIN,
> +				     language_c, NULL).symbol;
> +      if (obj_entry_sym == NULL)
> +	error (_("Unable to find Struct_Obj_Entry symbol"));
> +      data->off_linkmap = lookup_struct_elt (SYMBOL_TYPE(obj_entry_sym),
> +					     "linkmap", 0).offset / 8;
> +      data->off_tlsindex = lookup_struct_elt (SYMBOL_TYPE(obj_entry_sym),
> +					      "tlsindex", 0).offset / 8;
> +      data->rtld_offsets_valid = true;
> +      return;
> +    }
> +  CATCH (e, RETURN_MASK_ERROR)
> +    {
> +      data->off_linkmap = -1;
> +    }
> +  END_CATCH
> +
> +  TRY
> +    {
> +      /* Fetch offsets from global variables in libthr.  Note that
> +	 this does not work for single-threaded processes that are not
> +	 linked against libthr.  */
> +      data->off_linkmap = fbsd_read_integer_by_name(gdbarch,
> +						    "_thread_off_linkmap");
> +      data->off_tlsindex = fbsd_read_integer_by_name(gdbarch,
> +						     "_thread_off_tlsindex");

Missing spaces before parentheses.

> +/* Helper function to fetch the address of a thread-local variable.
> +   DTV_ADDR is the base address of the thread's dtv array.  LM_ADDR is
> +   the address of the link_map structure for the associated object
> +   file.  */
> +
> +extern CORE_ADDR fbsd_get_thread_local_address (struct gdbarch *gdbarch,
> +						CORE_ADDR dtv_addr,
> +						CORE_ADDR lm_addr,
> +						CORE_ADDR offset);

Could you document the offset parameter?  I know it's obvious for 
somebody who already knows how TLS works, but if somebody is still a 
noob, they will appreciate the comment :).

Simon


  reply	other threads:[~2019-03-07 16:18 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-09  0:42 [PATCH v2 00/11] Support for thread-local variables on FreeBSD John Baldwin
2019-02-09  0:42 ` [PATCH v2 06/11] Add a more general version of lookup_struct_elt_type John Baldwin
2019-02-09  1:08   ` John Baldwin
2019-02-11 10:27     ` Philipp Rudo
2019-02-11 17:44       ` John Baldwin
2019-03-07 15:53   ` Simon Marchi
2019-03-08  0:04     ` John Baldwin
2019-03-08  0:32       ` Pedro Alves
2019-03-08 18:39         ` John Baldwin
2019-02-09  0:42 ` [PATCH v2 02/11] Support fs_base and gs_base on FreeBSD/i386 John Baldwin
2019-02-09  0:42 ` [PATCH v2 09/11] Support TLS variables " John Baldwin
2019-02-09  0:42 ` [PATCH v2 03/11] Handle an edge case for minisym TLS variable lookups John Baldwin
2019-02-09  0:42 ` [PATCH v2 11/11] Support TLS variables on FreeBSD/powerpc John Baldwin
2019-03-07 16:26   ` Simon Marchi
2019-02-09  0:42 ` [PATCH v2 07/11] Add a helper function to resolve TLS variable addresses for FreeBSD John Baldwin
2019-03-07 16:18   ` Simon Marchi [this message]
2019-02-09  0:42 ` [PATCH v2 08/11] Support TLS variables on FreeBSD/amd64 John Baldwin
2019-02-09  0:42 ` [PATCH v2 01/11] Support the fs_base and gs_base registers on i386 John Baldwin
2019-02-09  0:50 ` [PATCH v2 10/11] Support TLS variables on FreeBSD/riscv John Baldwin
2019-02-09  0:50 ` [PATCH v2 04/11] Add a new gdbarch method to resolve the address of TLS variables John Baldwin
2019-03-07 16:08   ` Simon Marchi
2019-03-07 23:50     ` John Baldwin
2019-03-08  2:55       ` Simon Marchi
2019-03-08 18:39         ` John Baldwin
2019-02-09  0:50 ` [PATCH v2 05/11] Remove code disabled since at least 1999 from lookup_struct_elt_type John Baldwin
2019-03-07 16:25   ` Simon Marchi
2019-02-22 17:22 ` [PING][PATCH v2 00/11] Support for thread-local variables on FreeBSD John Baldwin
2019-03-12 20:21   ` Simon Marchi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=eddc12ba-1b15-f2ad-1a33-9b413ae28b9a@simark.ca \
    --to=simark@simark.ca \
    --cc=gdb-patches@sourceware.org \
    --cc=jhb@FreeBSD.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox