linux-kernel - [PATCH v3 0/5] EDAC/ghes: Cleanup and reworks

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200519104443.15673-1-rrichter@marvell.com>
Date:   Tue, 19 May 2020 12:44:38 +0200
From:   Robert Richter <rrichter@...vell.com>
To:     Borislav Petkov <bp@...en8.de>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Tony Luck <tony.luck@...el.com>
CC:     James Morse <james.morse@....com>,
        Aristeu Rozanski <aris@...hat.com>,
        Robert Richter <rrichter@...vell.com>,
        Matthias Brugger <mbrugger@...e.com>,
        <linux-edac@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: [PATCH v3 0/5] EDAC/ghes: Cleanup and reworks

This series contains a general cleanup and rework of the edac ghes
driver:

 * Some small code improvements (patches #1, #2).

 * Code in functions ghes_edac_{register,unregister}() is move to new
   functions ghes_mc_{create,destroy}() and ghes_mc_{add,del}() (patch
   #3).

 * Separated 'fake' controller code path (patches #4, #5).

Tested on a Marvell/Cavium ThunderX2 Sabre (dual socket) system.

v3:
 * Rebased onto dc63e28efa19 ("Merge branch 'edac-i10nm' into
   edac-for-next") plus patch "EDAC/ghes: Setup DIMM label from DMI
   and use it in error reports" applied from
   https://lore.kernel.org/patchwork/patch/1243203/
 * Removed v2 patches 01/10 and 02/10 for edac_mc driver from this
   series, both are unrelated.
 * Dropped v2 patch 04/10 "EDAC/ghes: Make SMBIOS handle private data
   to ghes", there is no consent with the maintainer to the code
   introduced to get a private ghes_dimm data structure, nor there was
   any feasible alternative suggested.
 * Taken v2 patch 05/10 "EDAC/ghes: Setup DIMM label from DMI and use
   it in error reports" out of this series and submitted separately
   (see above patchwork link).
 * Dropped v2 patch 06/10, keep rdr_mask variable.
 * Fixed subject of v2 patch 07/10 to 'EDAC/ghes: Cleanup struct
   dimm_fill'.
 * Reworked function interface, there is now
   ghes_mc_{create,destroy}() and ghes_mc_{add,del}().
 * Aligned arguments on the opening brace (ghes_mc_*()).
 * Remove ghes_ prefix from ghes_dimm_* definitions.
 * Use sizeof(struct ghes_pvt) in edac_mc_alloc().
 * Rename struct ghes_mci to struct ghes_pvt.

v2:
 * https://lore.kernel.org/patchwork/cover/1229380/
 * reordered patches to have fixes and struct changes first, code
   refactoring patches last,
 * dropped v1 patches #9 to #11 (multiple conrollers) to handle them
   in a separate series,
 * rewrote patch to remove smbios_handle (based on v1 #9): EDAC/ghes:
   Move smbios_handle from struct dimm_info to ghes private data,
 * added lockdep_assert_held() checkers,
 * renamed struct ghes_dimm_fill to struct dimm_fill,
 * renamed local variable dimms to dimm_list to avoid conflict with
   the global variable,
 * removed dimm list for "fake" controller,
 * fixed return code check to use (rc < 0),
 * added: EDAC/mc: Fix usage of snprintf() and dimm location setup

v1:
 * https://lore.kernel.org/patchwork/cover/1205901/

Robert Richter (5):
  EDAC/ghes: Remove unused members of struct ghes_edac_pvt, rename it to
    ghes_pvt
  EDAC/ghes: Cleanup struct dimm_fill
  EDAC/ghes: Carve out MC device handling into separate functions
  EDAC/ghes: Have a separate code path for creating the fake MC
  EDAC/ghes: Carve out code into ghes_edac_register_{one,fake}()

 drivers/edac/ghes_edac.c | 254 ++++++++++++++++++++++++---------------
 1 file changed, 159 insertions(+), 95 deletions(-)

-- 
2.20.1