linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Sam Bobroff <sbobroff@linux.ibm.com>
To: Alexey Kardashevskiy <aik@ozlabs.ru>
Cc: linuxppc-dev@lists.ozlabs.org
Subject: Re: [PATCH 3/8] powerpc/eeh: Convert PNV_PHB_FLAG_EEH to global flag
Date: Tue, 9 Apr 2019 11:41:57 +1000	[thread overview]
Message-ID: <20190409014156.GA11437@tungsten.ozlabs.ibm.com> (raw)
In-Reply-To: <680b2264-93a8-6c75-88c1-ad0520b8e8a7@ozlabs.ru>

[-- Attachment #1: Type: text/plain, Size: 6774 bytes --]

On Wed, Mar 20, 2019 at 05:02:44PM +1100, Alexey Kardashevskiy wrote:
> 
> 
> On 20/03/2019 13:58, Sam Bobroff wrote:
> > The PHB flag, PNV_PHB_FLAG_EEH, is set (on PowerNV) individually on
> > each PHB once the EEH subsystem is ready. It is the only use of the
> > flags member of the phb struct.
> 
> 
> Then why to keep pnv_phb::flags?

No reason. I'll remove it in the next version.

> > However there is no need to store this separately on each PHB, so
> > convert it to a global flag. For symmetry, the flag is now also set
> > for pSeries; although it is currently unused it may be useful in the
> > future.
> 
> Just using eeh_enabled() instead of (phb->flags & PNV_PHB_FLAG_EEH)
> seems easier and cleaner; also pseries does not use it so there is no
> point defining it there either.

I do want to do that. However, eeh_enabled() seems to be slightly
different from PNV_PHB_FLAG_EEH:

- eeh_enabled() is true as soon as the first device with EEH support is
  detected.
- eeh_phb_enabled() is true after EEH support has been enabled on every
  device that supports it.

So I was concerned that using eeh_enabled() would cause problems in
pnv_pci_config_check_eeh() if EEH was detected *during* the initial PCI
scanning phase when eeh_enabled() was true but EEH had not yet been set
up on the device or PHB where it was detected.

Does that make sense?

Would it be reasonable to keep this patch as it is for now and
investigate cleaning it up in a future patch?

> > 
> > Signed-off-by: Sam Bobroff <sbobroff@linux.ibm.com>
> > ---
> >  arch/powerpc/include/asm/eeh.h               | 11 +++++++++++
> >  arch/powerpc/platforms/powernv/eeh-powernv.c | 14 +++-----------
> >  arch/powerpc/platforms/powernv/pci.c         |  7 +++----
> >  arch/powerpc/platforms/powernv/pci.h         |  2 --
> >  arch/powerpc/platforms/pseries/pci.c         |  4 ++++
> >  5 files changed, 21 insertions(+), 17 deletions(-)
> > 
> > diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h
> > index 3613a56281f2..fe4cf7208890 100644
> > --- a/arch/powerpc/include/asm/eeh.h
> > +++ b/arch/powerpc/include/asm/eeh.h
> > @@ -43,6 +43,7 @@ struct pci_dn;
> >  #define EEH_VALID_PE_ZERO	0x10	/* PE#0 is valid		     */
> >  #define EEH_ENABLE_IO_FOR_LOG	0x20	/* Enable IO for log		     */
> >  #define EEH_EARLY_DUMP_LOG	0x40	/* Dump log immediately		     */
> > +#define EEH_PHB_ENABLED		0x80	/* PHB recovery uses EEH	     */
> >  
> >  /*
> >   * Delay for PE reset, all in ms
> > @@ -245,6 +246,11 @@ static inline bool eeh_enabled(void)
> >  	return eeh_has_flag(EEH_ENABLED) && !eeh_has_flag(EEH_FORCE_DISABLED);
> >  }
> >  
> > +static inline bool eeh_phb_enabled(void)
> > +{
> > +	return eeh_has_flag(EEH_PHB_ENABLED);
> > +}
> > +
> >  static inline void eeh_serialize_lock(unsigned long *flags)
> >  {
> >  	raw_spin_lock_irqsave(&confirm_error_lock, *flags);
> > @@ -332,6 +338,11 @@ static inline bool eeh_enabled(void)
> >          return false;
> >  }
> >  
> > +static inline bool eeh_phb_enabled(void)
> > +{
> > +	return false;
> > +}
> > +
> >  static inline void eeh_probe_devices(void) { }
> >  
> >  static inline void *eeh_dev_init(struct pci_dn *pdn, void *data)
> > diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
> > index 6fc1a463b796..f0a95f663810 100644
> > --- a/arch/powerpc/platforms/powernv/eeh-powernv.c
> > +++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
> > @@ -264,22 +264,14 @@ int pnv_eeh_post_init(void)
> >  		return ret;
> >  	}
> >  
> > -	if (!eeh_enabled())
> > +	if (eeh_enabled())
> > +		eeh_add_flag(EEH_PHB_ENABLED);
> > +	else
> >  		disable_irq(eeh_event_irq);
> >  
> >  	list_for_each_entry(hose, &hose_list, list_node) {
> >  		phb = hose->private_data;
> >  
> > -		/*
> > -		 * If EEH is enabled, we're going to rely on that.
> > -		 * Otherwise, we restore to conventional mechanism
> > -		 * to clear frozen PE during PCI config access.
> > -		 */
> > -		if (eeh_enabled())
> > -			phb->flags |= PNV_PHB_FLAG_EEH;
> > -		else
> > -			phb->flags &= ~PNV_PHB_FLAG_EEH;
> > -
> >  		/* Create debugfs entries */
> >  #ifdef CONFIG_DEBUG_FS
> >  		if (phb->has_dbgfs || !phb->dbgfs)
> > diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
> > index 307181fd8a17..d2b50f3bf6b1 100644
> > --- a/arch/powerpc/platforms/powernv/pci.c
> > +++ b/arch/powerpc/platforms/powernv/pci.c
> > @@ -717,10 +717,9 @@ int pnv_pci_cfg_write(struct pci_dn *pdn,
> >  static bool pnv_pci_cfg_check(struct pci_dn *pdn)
> >  {
> >  	struct eeh_dev *edev = NULL;
> > -	struct pnv_phb *phb = pdn->phb->private_data;
> >  
> >  	/* EEH not enabled ? */
> > -	if (!(phb->flags & PNV_PHB_FLAG_EEH))
> > +	if (!eeh_phb_enabled())
> >  		return true;
> >  
> >  	/* PE reset or device removed ? */
> > @@ -761,7 +760,7 @@ static int pnv_pci_read_config(struct pci_bus *bus,
> >  
> >  	ret = pnv_pci_cfg_read(pdn, where, size, val);
> >  	phb = pdn->phb->private_data;
> > -	if (phb->flags & PNV_PHB_FLAG_EEH && pdn->edev) {
> > +	if (eeh_phb_enabled() && pdn->edev) {
> >  		if (*val == EEH_IO_ERROR_VALUE(size) &&
> >  		    eeh_dev_check_failure(pdn->edev))
> >                          return PCIBIOS_DEVICE_NOT_FOUND;
> > @@ -789,7 +788,7 @@ static int pnv_pci_write_config(struct pci_bus *bus,
> >  
> >  	ret = pnv_pci_cfg_write(pdn, where, size, val);
> >  	phb = pdn->phb->private_data;
> > -	if (!(phb->flags & PNV_PHB_FLAG_EEH))
> > +	if (!eeh_phb_enabled())
> >  		pnv_pci_config_check_eeh(pdn);
> >  
> >  	return ret;
> > diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
> > index 8e36da379252..eb0add61397b 100644
> > --- a/arch/powerpc/platforms/powernv/pci.h
> > +++ b/arch/powerpc/platforms/powernv/pci.h
> > @@ -85,8 +85,6 @@ struct pnv_ioda_pe {
> >  	struct list_head	list;
> >  };
> >  
> > -#define PNV_PHB_FLAG_EEH	(1 << 0)
> > -
> >  struct pnv_phb {
> >  	struct pci_controller	*hose;
> >  	enum pnv_phb_type	type;
> > diff --git a/arch/powerpc/platforms/pseries/pci.c b/arch/powerpc/platforms/pseries/pci.c
> > index 37a77e57893e..7be80882c08d 100644
> > --- a/arch/powerpc/platforms/pseries/pci.c
> > +++ b/arch/powerpc/platforms/pseries/pci.c
> > @@ -244,6 +244,10 @@ void __init pSeries_final_fixup(void)
> >  
> >  	eeh_probe_devices();
> >  	eeh_addr_cache_build();
> > +#ifdef CONFIG_EEH
> > +	if (eeh_enabled())
> > +		eeh_add_flag(EEH_PHB_ENABLED);
> > +#endif
> >  
> >  #ifdef CONFIG_PCI_IOV
> >  	ppc_md.pcibios_sriov_enable = pseries_pcibios_sriov_enable;
> > 
> 
> -- 
> Alexey
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

  reply	other threads:[~2019-04-09  1:43 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-20  2:58 [PATCH 0/8] Sam Bobroff
2019-03-20  2:58 ` [PATCH 1/8] powerpc/64: Adjust order in pcibios_init() Sam Bobroff
2019-03-20  6:03   ` Alexey Kardashevskiy
2019-03-20  2:58 ` [PATCH 2/8] powerpc/eeh: Clear stale EEH_DEV_NO_HANDLER flag Sam Bobroff
2019-03-20  6:02   ` Alexey Kardashevskiy
2019-04-08  6:50     ` Sam Bobroff
2019-03-20  2:58 ` [PATCH 3/8] powerpc/eeh: Convert PNV_PHB_FLAG_EEH to global flag Sam Bobroff
2019-03-20  6:02   ` Alexey Kardashevskiy
2019-04-09  1:41     ` Sam Bobroff [this message]
2019-04-18  9:51   ` Oliver O'Halloran
2019-04-30  5:30     ` Sam Bobroff
2019-03-20  2:58 ` [PATCH 4/8] powerpc/eeh: Improve debug messages around device addition Sam Bobroff
2019-03-20  6:02   ` Alexey Kardashevskiy
2019-03-20  2:58 ` [PATCH 5/8] powerpc/eeh: Add eeh_show_enabled() Sam Bobroff
2019-03-20  6:02   ` Alexey Kardashevskiy
2019-03-20  6:24     ` Oliver
2019-04-09  3:30     ` Sam Bobroff
2019-04-18 10:01   ` Oliver O'Halloran
2019-04-30  5:44     ` Sam Bobroff
2019-03-20  2:58 ` [PATCH 6/8] powerpc/eeh: Initialize EEH address cache earlier Sam Bobroff
2019-03-20  6:02   ` Alexey Kardashevskiy
2019-04-18 10:13   ` Oliver O'Halloran
2019-04-30  5:54     ` Sam Bobroff
2019-03-20  2:58 ` [PATCH 7/8] powerpc/eeh: EEH for pSeries hot plug Sam Bobroff
2019-03-20  6:02   ` Alexey Kardashevskiy
2019-03-20  2:58 ` [PATCH 8/8] powerpc/eeh: Remove eeh_probe_devices() and eeh_addr_cache_build() Sam Bobroff
2019-03-20  6:05   ` Alexey Kardashevskiy
2019-04-09  3:31     ` Sam Bobroff
2019-04-12  0:55 ` [PATCH 0/8] Tyrel Datwyler
2019-04-15  3:41   ` Sam Bobroff
2019-04-19 22:36     ` Tyrel Datwyler

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=20190409014156.GA11437@tungsten.ozlabs.ibm.com \
    --to=sbobroff@linux.ibm.com \
    --cc=aik@ozlabs.ru \
    --cc=linuxppc-dev@lists.ozlabs.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;
as well as URLs for NNTP newsgroup(s).