| 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
| ||
|
Message-ID: <20250514124131.15d0117f@canb.auug.org.au>
Date: Wed, 14 May 2025 12:41:31 +1000
From: Stephen Rothwell <sfr@...b.auug.org.au>
To: Johannes Berg <johannes@...solutions.net>
Cc: Miriam Rachel Korenblit <miriam.rachel.korenblit@...el.com>, Linux
Kernel Mailing List <linux-kernel@...r.kernel.org>, Linux Next Mailing List
<linux-next@...r.kernel.org>
Subject: Re: linux-next: manual merge of the iwlwifi-next tree with Linus'
tree
Hi all,
On Thu, 8 May 2025 13:24:59 +1000 Stephen Rothwell <sfr@...b.auug.org.au> wrote:
>
> On Wed, 07 May 2025 07:37:46 +0200 Johannes Berg <johannes@...solutions.net> wrote:
> >
> > > So, at your suggestion, I just used the latter versions of these files,
> > > but I then got the following build failure:
> >
> > Ouch, sorry about that. That sounds like it's due to the duplicate
> > commits you had pointed out, which Miri dropped from iwlwifi-next now.
> >
> > On the plus side, I'm going to be the one who has to resolve these
> > conflicts (soon) :)
>
> There are real conficts (see below).
>
> > > So, today I have merged the iwlwifi-next tree from next-20250505 (which
> > > was effectively empty). You may want to try merging the iwlwifi tree
> > > into the iwlwifi-next tree and resolve any conflicts yourself (now that
> > > the iwlwifi tree has been merged into Linus' tree).
> >
> > Not sure what you mean by "iwlwifi tree", I think iwlwifi-next/fixes was
> > empty, so perhaps you meant "wireless tree" instead?
>
> Yeah, that or just commit 175e69e33c66 (which has been merged into
> Linus' tree).
>
> > But yeah, I know what all the code is and what's going on, though
> > evidently less than I thought after the recent updates.
>
> So today, I have applied the following clean up patch, but it does mean
> that the 2 commits mentioned from Linus' tree have been reverted in
> linus-next. This means that linux-next builds today with the
> iwlwifi-next tree merged.
>
> From: Stephen Rothwell <sfr@...b.auug.org.au>
> Date: Thu, 8 May 2025 12:26:09 +1000
> Subject: [PATCH] iwlwifi: remove the left over bits of 2 commits from Linus' tree
>
> Commits
>
> d49437a6afc7 ("wifi: iwlwifi: back off on continuous errors")
> 15220a257319 ("wifi: iwlwifi: don't warn if the NIC is gone in resume")
>
> were partly removed by just taking the MERGE_HAD versions of
>
> drivers/net/wireless/intel/iwlwifi/cfg/sc.c
> drivers/net/wireless/intel/iwlwifi/iwl-config.h
> drivers/net/wireless/intel/iwlwifi/iwl-nvm-parse.c
> drivers/net/wireless/intel/iwlwifi/iwl-trans.c
> drivers/net/wireless/intel/iwlwifi/iwl-trans.h
> drivers/net/wireless/intel/iwlwifi/pcie/drv.c
> drivers/net/wireless/intel/iwlwifi/tests/devinfo.c
>
> This removes the rest.
>
> Signed-off-by: Stephen Rothwell <sfr@...b.auug.org.au>
> ---
> .../net/wireless/intel/iwlwifi/pcie/internal.h | 9 ++++-----
> drivers/net/wireless/intel/iwlwifi/pcie/trans.c | 16 ++++------------
> drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 2 +-
> 3 files changed, 9 insertions(+), 18 deletions(-)
>
> diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
> index b5368b256e20..8adeddeb0e4c 100644
> --- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
> +++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
> @@ -542,10 +542,10 @@ void iwl_trans_pcie_free(struct iwl_trans *trans);
> void iwl_trans_pcie_free_pnvm_dram_regions(struct iwl_dram_regions *dram_regions,
> struct device *dev);
>
> -bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans, bool silent);
> -#define _iwl_trans_pcie_grab_nic_access(trans, silent) \
> +bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans);
> +#define _iwl_trans_pcie_grab_nic_access(trans) \
> __cond_lock(nic_access_nobh, \
> - likely(__iwl_trans_pcie_grab_nic_access(trans, silent)))
> + likely(__iwl_trans_pcie_grab_nic_access(trans)))
>
> void iwl_trans_pcie_check_product_reset_status(struct pci_dev *pdev);
> void iwl_trans_pcie_check_product_reset_mode(struct pci_dev *pdev);
> @@ -1101,8 +1101,7 @@ void iwl_trans_pcie_set_bits_mask(struct iwl_trans *trans, u32 reg,
> int iwl_trans_pcie_read_config32(struct iwl_trans *trans, u32 ofs,
> u32 *val);
> bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans);
> -void __releases(nic_access_nobh)
> -iwl_trans_pcie_release_nic_access(struct iwl_trans *trans);
> +void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans);
>
> /* transport gen 1 exported functions */
> void iwl_trans_pcie_fw_alive(struct iwl_trans *trans);
> diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
> index 799d651ddd8a..4d281c702eec 100644
> --- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
> +++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
> @@ -2333,8 +2333,7 @@ void iwl_trans_pcie_reset(struct iwl_trans *trans, enum iwl_reset_mode mode)
> struct iwl_trans_pcie_removal *removal;
> char _msg = 0, *msg = &_msg;
>
> - if (WARN_ON(mode < IWL_RESET_MODE_REMOVE_ONLY ||
> - mode == IWL_RESET_MODE_BACKOFF))
> + if (WARN_ON(mode < IWL_RESET_MODE_REMOVE_ONLY))
> return;
>
> if (test_bit(STATUS_TRANS_DEAD, &trans->status))
> @@ -2388,7 +2387,7 @@ EXPORT_SYMBOL(iwl_trans_pcie_reset);
> * This version doesn't disable BHs but rather assumes they're
> * already disabled.
> */
> -bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans, bool silent)
> +bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
> {
> int ret;
> struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
> @@ -2440,11 +2439,6 @@ bool __iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans, bool silent)
> if (unlikely(ret < 0)) {
> u32 cntrl = iwl_read32(trans, CSR_GP_CNTRL);
>
> - if (silent) {
> - spin_unlock(&trans_pcie->reg_lock);
> - return false;
> - }
> -
> WARN_ONCE(1,
> "Timeout waiting for hardware access (CSR_GP_CNTRL 0x%08x)\n",
> cntrl);
> @@ -2476,7 +2470,7 @@ bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
> bool ret;
>
> local_bh_disable();
> - ret = __iwl_trans_pcie_grab_nic_access(trans, false);
> + ret = __iwl_trans_pcie_grab_nic_access(trans);
> if (ret) {
> /* keep BHs disabled until iwl_trans_pcie_release_nic_access */
> return ret;
> @@ -2485,8 +2479,7 @@ bool iwl_trans_pcie_grab_nic_access(struct iwl_trans *trans)
> return false;
> }
>
> -void __releases(nic_access_nobh)
> -iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
> +void iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
> {
> struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
>
> @@ -2513,7 +2506,6 @@ iwl_trans_pcie_release_nic_access(struct iwl_trans *trans)
> * scheduled on different CPUs (after we drop reg_lock).
> */
> out:
> - __release(nic_access_nobh);
> spin_unlock_bh(&trans_pcie->reg_lock);
> }
>
> diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
> index f142b8beb396..c876912e9aaf 100644
> --- a/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
> +++ b/drivers/net/wireless/intel/iwlwifi/pcie/tx.c
> @@ -1018,7 +1018,7 @@ static int iwl_pcie_set_cmd_in_flight(struct iwl_trans *trans,
> * returned. This needs to be done only on NICs that have
> * apmg_wake_up_wa set (see above.)
> */
> - if (!_iwl_trans_pcie_grab_nic_access(trans, false))
> + if (!_iwl_trans_pcie_grab_nic_access(trans))
> return -EIO;
>
> /*
> --
> 2.47.2
I keep getting more conflicts in the above files (each time one of them
get modified in the iwlwifi-next tree. Could you please merge commit
ebedf8b7f05b ("wifi: iwlwifi: add support for Killer on MTL")
or the current net tree
(git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git#main)
into the iwlwifi-next tree and fix up the conflicts. I believe that
will clean up all the conflicts - that are really just between bug
fixes for iwlwifi and the new development).
--
Cheers,
Stephen Rothwell
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists