ath9k: add preliminary support for the AR913x SoCs
v3 cover ath9k: add preliminary support for the AR913x SoCs
This patch set introduces preliminary support for the Atheros' AR9130/AR9132 SoCs built-in WMAC device. Tested with compat-wireless-2009-01-12 and 2.6.27.10 kernel on the following devices:
* Atheros AP83-050 reference board (AR9132 SoC, AR9106 RF)
* TRENDnet TEW-632BRP router (AR9130 SoC, AR9102 RF)
* Mikrotik RB433 board (AR7161 SoC) + AR5416 based miniPCI card
Changes since v2:
* parentheses around RTC register's names has been removed
* rebased against master-2009-01-12 of wireless-testing
Changes since v1:
* enabling support of the AR9100 has been moved to the end of the series
* bus specific cleanup routine has been rewritten
* PCI code converted to use 'softc.irq' as well
* setup of 'hw->flags' and 'hw->wiphy->interface_modes' has been removed from ath_ahb_probe()
* unregistering of the ath_ahb_diver has been fixed
Changes since RFC:
* bus specific DMA routines removed
* bus specific register access routines removed
* irq value is stored in softc
* rebased against master-2009-01-06 of wireless-testing
11 patches will follow as replies:
Gabor Juhos (11):
ath9k: convert to struct device
ath9k: convert to use bus-agnostic DMA routines
ath9k: introduce bus specific cache size routine
ath9k: introduce bus specific cleanup routine
ath9k: move PCI code into separate file
ath9k: introduce platform driver for AHB bus support
ath9k: get EEPROM contents from platform data on AHB bus
ath9k: remove (u16) casts from rtc register access
ath9k: fix ar5416Addac_9100 values
ath9k: fix null pointer dereference in ani monitor code
ath9k: enable support for AR9100
drivers/net/wireless/ath9k/Makefile | 2 drivers/net/wireless/ath9k/ahb.c | 187 +++++++++++++++++++ drivers/net/wireless/ath9k/ani.c | 6 - drivers/net/wireless/ath9k/beacon.c | 28 +-- drivers/net/wireless/ath9k/core.h | 47 +++++ drivers/net/wireless/ath9k/eeprom.c | 51 ----- drivers/net/wireless/ath9k/hw.c | 17 +- drivers/net/wireless/ath9k/initvals.h | 2 drivers/net/wireless/ath9k/main.c | 326 +++++---------------------------- drivers/net/wireless/ath9k/pci.c | 305 +++++++++++++++++++++++++++++++ drivers/net/wireless/ath9k/recv.c | 24 +- drivers/net/wireless/ath9k/reg.h | 4 drivers/net/wireless/ath9k/xmit.c | 14 + include/linux/ath9k_platform.h | 28 +++ 14 files changed, 665 insertions(+), 376 deletions(-) create mode 100644 drivers/net/wireless/ath9k/ahb.c create mode 100644 drivers/net/wireless/ath9k/pci.c create mode 100644 include/linux/ath9k_platform.h
PATCH v3 1/11 ath9k: convert to struct device
Convert 'struct pci_dev' to 'struct device' to make it usable on the AHB bus as well.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/beacon.c | 17 ++++++++++------- drivers/net/wireless/ath9k/core.h | 2 +- drivers/net/wireless/ath9k/main.c | 21 +++++++++++---------- drivers/net/wireless/ath9k/recv.c | 15 +++++++++------ drivers/net/wireless/ath9k/xmit.c | 7 ++++--- 5 files changed, 35 insertions(+), 27 deletions(-)
drivers/net/wireless/ath9k/beacon.c +10 -7
@@ -164,7 +164,7 @@ static struct ath_buf *ath_beacon_generate(struct ath_softc *sc, int if_id) bf = avp->av_bcbuf; skb = (struct sk_buff *)bf->bf_mpdu;if (skb) { -pci_unmap_single (sc->pdev, bf->bf_dmacontext, +pci_unmap_single (to_pci_dev (sc->dev), bf->bf_dmacontext, skb->len,PCI_DMA_TODEVICE );dev_kfree_skb_any (skb);
@@ -188,10 +188,11 @@ static struct ath_buf *ath_beacon_generate(struct ath_softc *sc, int if_id) } bf->bf_buf_addr = bf->bf_dmacontext = -pci_map_single (sc->pdev, skb->data, +pci_map_single (to_pci_dev (sc->dev), skb->data, skb->len,PCI_DMA_TODEVICE ); -if (unlikely (pci_dma_mapping_error (sc->pdev, bf->bf_buf_addr))) { +if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), + bf->bf_buf_addr))) {dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ; DPRINTF(sc, ATH_DBG_CONFIG,
@@ -343,7 +344,7 @@ int ath_beacon_alloc(struct ath_softc *sc, int if_id) bf = avp->av_bcbuf;if (bf->bf_mpdu !=NULL ) { skb = (struct sk_buff *)bf->bf_mpdu; -pci_unmap_single (sc->pdev, bf->bf_dmacontext, +pci_unmap_single (to_pci_dev (sc->dev), bf->bf_dmacontext, skb->len,PCI_DMA_TODEVICE );dev_kfree_skb_any (skb);
@@ -402,10 +403,11 @@ int ath_beacon_alloc(struct ath_softc *sc, int if_id) bf->bf_mpdu = skb; bf->bf_buf_addr = bf->bf_dmacontext = -pci_map_single (sc->pdev, skb->data, +pci_map_single (to_pci_dev (sc->dev), skb->data, skb->len,PCI_DMA_TODEVICE ); -if (unlikely (pci_dma_mapping_error (sc->pdev, bf->bf_buf_addr))) { +if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), + bf->bf_buf_addr))) {dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ; DPRINTF(sc, ATH_DBG_CONFIG,
@@ -429,7 +431,8 @@ void ath_beacon_return(struct ath_softc *sc, struct ath_vap *avp) bf = avp->av_bcbuf;if (bf->bf_mpdu !=NULL ) {struct sk_buff *skb = (struct sk_buff *)bf->bf_mpdu; -pci_unmap_single (sc->pdev, bf->bf_dmacontext, +pci_unmap_single (to_pci_dev (sc->dev), + bf->bf_dmacontext, skb->len,PCI_DMA_TODEVICE );dev_kfree_skb_any (skb);
drivers/net/wireless/ath9k/core.h +1 -1
@@ -695,7 +695,7 @@ enum PROT_MODE {struct ath_softc {struct ieee80211_hw *hw; -struct pci_dev *pdev; +struct device *dev;struct tasklet_struct intr_tq;struct tasklet_struct bcon_tasklet;struct ath_hal *sc_ah;
drivers/net/wireless/ath9k/main.c +11 -10
@@ -46,7 +46,8 @@ static void bus_read_cachesize(struct ath_softc *sc, int *csz) {u8 u8tmp; -pci_read_config_byte (sc->pdev,PCI_CACHE_LINE_SIZE , (u8 *)&u8tmp); +pci_read_config_byte (to_pci_dev (sc->dev),PCI_CACHE_LINE_SIZE , + (u8 *)&u8tmp); *csz = (int )u8tmp;/*
@@ -1267,11 +1268,11 @@ static int ath_start_rfkill_poll(struct ath_softc *sc)/* Deinitialize the device */ ath_detach (sc); -if (sc->pdev->irq) -free_irq (sc->pdev->irq, sc); -pci_iounmap (sc->pdev, sc->mem); -pci_release_region (sc->pdev,0 ); -pci_disable_device (sc->pdev); +if (to_pci_dev (sc->dev)->irq) +free_irq (to_pci_dev (sc->dev)->irq, sc); +pci_iounmap (to_pci_dev (sc->dev), sc->mem); +pci_release_region (to_pci_dev (sc->dev),0 ); +pci_disable_device (to_pci_dev (sc->dev));ieee80211_free_hw (sc->hw);return -EIO ; }else {
@@ -1714,7 +1715,7 @@ int ath_descdma_setup(struct ath_softc *sc, struct ath_descdma *dd, }/* allocate descriptors */ - dd->dd_desc =pci_alloc_consistent (sc->pdev, + dd->dd_desc =pci_alloc_consistent (to_pci_dev (sc->dev), dd->dd_desc_len, &dd->dd_desc_paddr);if (dd->dd_desc ==NULL ) {
@@ -1762,7 +1763,7 @@ int ath_descdma_setup(struct ath_softc *sc, struct ath_descdma *dd, }return 0 ; fail2: -pci_free_consistent (sc->pdev, +pci_free_consistent (to_pci_dev (sc->dev), dd->dd_desc_len, dd->dd_desc, dd->dd_desc_paddr); fail:memset (dd,0 ,sizeof (*dd));
@@ -1776,7 +1777,7 @@ void ath_descdma_cleanup(struct ath_softc *sc,struct ath_descdma *dd, struct list_head *head) { - pci_free_consistent(sc->pdev, + pci_free_consistent(to_pci_dev(sc->dev), dd->dd_desc_len, dd->dd_desc, dd->dd_desc_paddr);INIT_LIST_HEAD (head);
@@ -2620,7 +2621,7 @@ static int ath_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) sc = hw->priv; sc->hw = hw; - sc->pdev = pdev; + sc->dev = &pdev->dev; sc->mem = mem;if (ath_attach (id->device, sc) !=0 ) {
drivers/net/wireless/ath9k/recv.c +9 -6
@@ -291,10 +291,11 @@ int ath_rx_init(struct ath_softc *sc, int nbufs) } bf->bf_mpdu = skb; - bf->bf_buf_addr =pci_map_single (sc->pdev, skb->data, + bf->bf_buf_addr =pci_map_single (to_pci_dev (sc->dev), + skb->data, sc->rx.bufsize,PCI_DMA_FROMDEVICE ); -if (unlikely (pci_dma_mapping_error (sc->pdev, +if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), bf->bf_buf_addr))) {dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ;
@@ -524,7 +525,8 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush)* 1. accessing the frame * 2. requeueing the same buffer to h/w */ -pci_dma_sync_single_for_cpu (sc->pdev, bf->bf_buf_addr, +pci_dma_sync_single_for_cpu (to_pci_dev (sc->dev), + bf->bf_buf_addr, sc->rx.bufsize,PCI_DMA_FROMDEVICE );
@@ -557,7 +559,7 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush) goto requeue;/* Unmap the frame */ -pci_unmap_single (sc->pdev, bf->bf_buf_addr, +pci_unmap_single (to_pci_dev (sc->dev), bf->bf_buf_addr, sc->rx.bufsize,PCI_DMA_FROMDEVICE );
@@ -605,10 +607,11 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush)/* We will now give hardware our shiny new allocated skb */ bf->bf_mpdu = requeue_skb; - bf->bf_buf_addr =pci_map_single (sc->pdev, requeue_skb->data, + bf->bf_buf_addr =pci_map_single (to_pci_dev (sc->dev), + requeue_skb->data, sc->rx.bufsize,PCI_DMA_FROMDEVICE ); -if (unlikely (pci_dma_mapping_error (sc->pdev, +if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), bf->bf_buf_addr))) {dev_kfree_skb_any (requeue_skb); bf->bf_mpdu =NULL ;
drivers/net/wireless/ath9k/xmit.c +4 -3
@@ -340,7 +340,7 @@ static void ath_tx_complete_buf(struct ath_softc *sc, }/* Unmap this frame */ -pci_unmap_single (sc->pdev, +pci_unmap_single (to_pci_dev (sc->dev), bf->bf_dmacontext, skb->len,PCI_DMA_TODEVICE );
@@ -1716,9 +1716,10 @@ static int ath_tx_setup_buffer(struct ath_softc *sc, struct ath_buf *bf,/* DMA setup */ bf->bf_mpdu = skb; - bf->bf_dmacontext =pci_map_single (sc->pdev, skb->data, + bf->bf_dmacontext =pci_map_single (to_pci_dev (sc->dev), skb->data, skb->len,PCI_DMA_TODEVICE ); -if (unlikely (pci_dma_mapping_error (sc->pdev, bf->bf_dmacontext))) { +if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), + bf->bf_dmacontext))) { bf->bf_mpdu =NULL ;DPRINTF (sc,ATH_DBG_CONFIG ,"pci_dma_mapping_error() on TX\n" );
Signature · 1 line
1.5.3.2
PATCH v3 2/11 ath9k: convert to use bus-agnostic DMA routines
In reply to Gabor JuhosConvert to use bus-agnostic DMA routines to make it usable on AHB bus as well.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/beacon.c | 31 ++++++++++++++----------------- drivers/net/wireless/ath9k/main.c | 13 ++++++------- drivers/net/wireless/ath9k/recv.c | 27 ++++++++++++--------------- drivers/net/wireless/ath9k/xmit.c | 15 ++++++--------- 4 files changed, 38 insertions(+), 48 deletions(-)
drivers/net/wireless/ath9k/beacon.c +14 -17
@@ -164,9 +164,9 @@ static struct ath_buf *ath_beacon_generate(struct ath_softc *sc, int if_id) bf = avp->av_bcbuf; skb = (struct sk_buff *)bf->bf_mpdu;if (skb) { -pci_unmap_single (to_pci_dev (sc->dev), bf->bf_dmacontext, +dma_unmap_single (sc->dev, bf->bf_dmacontext, skb->len, -PCI_DMA_TODEVICE ); +DMA_TO_DEVICE );dev_kfree_skb_any (skb); }
@@ -188,15 +188,14 @@ static struct ath_buf *ath_beacon_generate(struct ath_softc *sc, int if_id) } bf->bf_buf_addr = bf->bf_dmacontext = -pci_map_single (to_pci_dev (sc->dev), skb->data, +dma_map_single (sc->dev, skb->data, skb->len, -PCI_DMA_TODEVICE ); -if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), - bf->bf_buf_addr))) { +DMA_TO_DEVICE ); +if (unlikely (dma_mapping_error (sc->dev, bf->bf_buf_addr))) {dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ;DPRINTF (sc,ATH_DBG_CONFIG , -"pci_dma_mapping_error() on beaconing\n" ); +"dma_mapping_error() on beaconing\n" );return NULL ; }
@@ -344,9 +343,9 @@ int ath_beacon_alloc(struct ath_softc *sc, int if_id) bf = avp->av_bcbuf;if (bf->bf_mpdu !=NULL ) { skb = (struct sk_buff *)bf->bf_mpdu; -pci_unmap_single (to_pci_dev (sc->dev), bf->bf_dmacontext, +dma_unmap_single (sc->dev, bf->bf_dmacontext, skb->len, -PCI_DMA_TODEVICE ); +DMA_TO_DEVICE );dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ; }
@@ -403,15 +402,14 @@ int ath_beacon_alloc(struct ath_softc *sc, int if_id) bf->bf_mpdu = skb; bf->bf_buf_addr = bf->bf_dmacontext = -pci_map_single (to_pci_dev (sc->dev), skb->data, +dma_map_single (sc->dev, skb->data, skb->len, -PCI_DMA_TODEVICE ); -if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), - bf->bf_buf_addr))) { +DMA_TO_DEVICE ); +if (unlikely (dma_mapping_error (sc->dev, bf->bf_buf_addr))) {dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ;DPRINTF (sc,ATH_DBG_CONFIG , -"pci_dma_mapping_error() on beacon alloc\n" ); +"dma_mapping_error() on beacon alloc\n" );return -ENOMEM ; }
@@ -431,10 +429,9 @@ void ath_beacon_return(struct ath_softc *sc, struct ath_vap *avp) bf = avp->av_bcbuf;if (bf->bf_mpdu !=NULL ) {struct sk_buff *skb = (struct sk_buff *)bf->bf_mpdu; -pci_unmap_single (to_pci_dev (sc->dev), - bf->bf_dmacontext, +dma_unmap_single (sc->dev, bf->bf_dmacontext, skb->len, -PCI_DMA_TODEVICE ); +DMA_TO_DEVICE );dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ; }
drivers/net/wireless/ath9k/main.c +6 -7
@@ -1715,9 +1715,8 @@ int ath_descdma_setup(struct ath_softc *sc, struct ath_descdma *dd, }/* allocate descriptors */ - dd->dd_desc =pci_alloc_consistent (to_pci_dev (sc->dev), - dd->dd_desc_len, - &dd->dd_desc_paddr); + dd->dd_desc =dma_alloc_coherent (sc->dev, dd->dd_desc_len, + &dd->dd_desc_paddr,GFP_ATOMIC );if (dd->dd_desc ==NULL ) { error = -ENOMEM ; goto fail;
@@ -1763,8 +1762,8 @@ int ath_descdma_setup(struct ath_softc *sc, struct ath_descdma *dd, }return 0 ; fail2: -pci_free_consistent (to_pci_dev (sc->dev), - dd->dd_desc_len, dd->dd_desc, dd->dd_desc_paddr); +dma_free_coherent (sc->dev, dd->dd_desc_len, dd->dd_desc, + dd->dd_desc_paddr); fail:memset (dd,0 ,sizeof (*dd));return error;
@@ -1777,8 +1776,8 @@ void ath_descdma_cleanup(struct ath_softc *sc,struct ath_descdma *dd, struct list_head *head) { - pci_free_consistent(to_pci_dev(sc->dev), - dd->dd_desc_len, dd->dd_desc, dd->dd_desc_paddr); + dma_free_coherent(sc->dev, dd->dd_desc_len, dd->dd_desc, + dd->dd_desc_paddr);INIT_LIST_HEAD (head);kfree (dd->dd_bufptr);
drivers/net/wireless/ath9k/recv.c +12 -15
@@ -291,16 +291,15 @@ int ath_rx_init(struct ath_softc *sc, int nbufs) } bf->bf_mpdu = skb; - bf->bf_buf_addr =pci_map_single (to_pci_dev (sc->dev), - skb->data, + bf->bf_buf_addr =dma_map_single (sc->dev, skb->data, sc->rx.bufsize, -PCI_DMA_FROMDEVICE ); -if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), +DMA_FROM_DEVICE ); +if (unlikely (dma_mapping_error (sc->dev, bf->bf_buf_addr))) {dev_kfree_skb_any (skb); bf->bf_mpdu =NULL ;DPRINTF (sc,ATH_DBG_CONFIG , -"pci_dma_mapping_error() on RX init\n" ); +"dma_mapping_error() on RX init\n" ); error = -ENOMEM ;break ; }
@@ -525,10 +524,9 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush)* 1. accessing the frame * 2. requeueing the same buffer to h/w */ -pci_dma_sync_single_for_cpu (to_pci_dev (sc->dev), - bf->bf_buf_addr, +dma_sync_single_for_cpu (sc->dev, bf->bf_buf_addr, sc->rx.bufsize, -PCI_DMA_FROMDEVICE ); +DMA_FROM_DEVICE );/* * If we're asked to flush receive queue, directly
@@ -559,9 +557,9 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush) goto requeue;/* Unmap the frame */ -pci_unmap_single (to_pci_dev (sc->dev), bf->bf_buf_addr, +dma_unmap_single (sc->dev, bf->bf_buf_addr, sc->rx.bufsize, -PCI_DMA_FROMDEVICE ); +DMA_FROM_DEVICE );skb_put (skb, ds->ds_rxstat.rs_datalen); skb->protocol =cpu_to_be16 (ETH_P_CONTROL );
@@ -607,16 +605,15 @@ int ath_rx_tasklet(struct ath_softc *sc, int flush)/* We will now give hardware our shiny new allocated skb */ bf->bf_mpdu = requeue_skb; - bf->bf_buf_addr =pci_map_single (to_pci_dev (sc->dev), - requeue_skb->data, + bf->bf_buf_addr =dma_map_single (sc->dev, requeue_skb->data, sc->rx.bufsize, -PCI_DMA_FROMDEVICE ); -if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), +DMA_FROM_DEVICE ); +if (unlikely (dma_mapping_error (sc->dev, bf->bf_buf_addr))) {dev_kfree_skb_any (requeue_skb); bf->bf_mpdu =NULL ;DPRINTF (sc,ATH_DBG_CONFIG , -"pci_dma_mapping_error() on RX\n" ); +"dma_mapping_error() on RX\n" );break ; } bf->bf_dmacontext = bf->bf_buf_addr;
drivers/net/wireless/ath9k/xmit.c +6 -9
@@ -340,10 +340,8 @@ static void ath_tx_complete_buf(struct ath_softc *sc, }/* Unmap this frame */ -pci_unmap_single (to_pci_dev (sc->dev), - bf->bf_dmacontext, - skb->len, -PCI_DMA_TODEVICE ); +dma_unmap_single (sc->dev, bf->bf_dmacontext, skb->len,DMA_TO_DEVICE ); +/* complete this frame */ ath_tx_complete (sc, skb, &tx_status);
@@ -1716,13 +1714,12 @@ static int ath_tx_setup_buffer(struct ath_softc *sc, struct ath_buf *bf,/* DMA setup */ bf->bf_mpdu = skb; - bf->bf_dmacontext =pci_map_single (to_pci_dev (sc->dev), skb->data, - skb->len,PCI_DMA_TODEVICE ); -if (unlikely (pci_dma_mapping_error (to_pci_dev (sc->dev), - bf->bf_dmacontext))) { + bf->bf_dmacontext =dma_map_single (sc->dev, skb->data, + skb->len,DMA_TO_DEVICE ); +if (unlikely (dma_mapping_error (sc->dev, bf->bf_dmacontext))) { bf->bf_mpdu =NULL ;DPRINTF (sc,ATH_DBG_CONFIG , -"pci_dma_mapping_error() on TX\n" ); +"dma_mapping_error() on TX\n" );return -ENOMEM ; }
Signature · 1 line
1.5.3.2
PATCH v3 3/11 ath9k: introduce bus specific cache size routine
In reply to Gabor JuhosThe PCI specific bus_read_cachesize routine won't work on the AHB bus, we have to replace it with a suitable one later.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/core.h | 10 ++++++++++ drivers/net/wireless/ath9k/main.c | 9 +++++++-- 2 files changed, 17 insertions(+), 2 deletions(-)
drivers/net/wireless/ath9k/core.h +10 -0
@@ -693,6 +693,10 @@ enum PROT_MODE {#define SC_OP_RFKILL_SW_BLOCKED BIT(12)#define SC_OP_RFKILL_HW_BLOCKED BIT(13) +struct ath_bus_ops { +void (*read_cachesize)(struct ath_softc *sc,int *csz); +}; +struct ath_softc {struct ieee80211_hw *hw;struct device *dev;
@@ -743,6 +747,7 @@ struct ath_softc {#ifdef CONFIG_ATH9K_DEBUG struct ath9k_debug sc_debug;#endif +struct ath_bus_ops *bus_ops; };int ath_reset (struct ath_softc *sc,bool retry_tx);
@@ -750,4 +755,9 @@ int ath_get_hal_qnum(u16 queue, struct ath_softc *sc);int ath_get_mac80211_qnum (u32 queue,struct ath_softc *sc);int ath_cabq_update (struct ath_softc *); +static inline void ath_read_cachesize (struct ath_softc *sc,int *csz) +{ + sc->bus_ops->read_cachesize (sc, csz); +} +#endif /* CORE_H */
drivers/net/wireless/ath9k/main.c +7 -2
@@ -42,7 +42,7 @@ static void ath_detach(struct ath_softc *sc);/* return bus cachesize in 4B word units */ -static void bus_read_cachesize (struct ath_softc *sc,int *csz) +static void ath_pci_read_cachesize (struct ath_softc *sc,int *csz) {u8 u8tmp;
@@ -1338,7 +1338,7 @@ static int ath_init(u16 devid, struct ath_softc *sc)* Cache line size is used to size and align various * structures used to communicate with the hardware. */ -bus_read_cachesize (sc, &csz); +ath_read_cachesize (sc, &csz);/* XXX assert csz is non-zero */ sc->sc_cachelsz = csz <<2 ;/* convert to bytes */
@@ -2534,6 +2534,10 @@ ath_rf_name(u16 rf_version)return "????" ; } +static struct ath_bus_ops ath_pci_bus_ops = { + .read_cachesize = ath_pci_read_cachesize, +}; +static int ath_pci_probe (struct pci_dev *pdev,const struct pci_device_id *id) {void __iomem *mem;
@@ -2622,6 +2626,7 @@ static int ath_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) sc->hw = hw; sc->dev = &pdev->dev; sc->mem = mem; + sc->bus_ops = &ath_pci_bus_ops;if (ath_attach (id->device, sc) !=0 ) { ret = -ENODEV ;
Signature · 1 line
1.5.3.2
PATCH v3 4/11 ath9k: introduce bus specific cleanup routine
In reply to Gabor JuhosWe have left only some PCI specific cleanup code. We have to convert them as well.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/core.h | 7 +++++++ drivers/net/wireless/ath9k/main.c | 37 +++++++++++++++++++++++-------------- 2 files changed, 30 insertions(+), 14 deletions(-)
drivers/net/wireless/ath9k/core.h +7 -0
@@ -695,6 +695,7 @@ enum PROT_MODE {struct ath_bus_ops {void (*read_cachesize)(struct ath_softc *sc,int *csz); +void (*cleanup)(struct ath_softc *sc); };struct ath_softc {
@@ -704,6 +705,7 @@ struct ath_softc {struct tasklet_struct bcon_tasklet;struct ath_hal *sc_ah;void __iomem *mem; +int irq;spinlock_t sc_resetlock;struct mutex mutex;
@@ -760,4 +762,9 @@ static inline void ath_read_cachesize(struct ath_softc *sc, int *csz) sc->bus_ops->read_cachesize(sc, csz); } +static inline void ath_bus_cleanup (struct ath_softc *sc) +{ + sc->bus_ops->cleanup (sc); +} +#endif /* CORE_H */
drivers/net/wireless/ath9k/main.c +23 -14
@@ -39,6 +39,7 @@ static struct pci_device_id ath_pci_id_table[] __devinitdata = { };static void ath_detach (struct ath_softc *sc); +static void ath_cleanup (struct ath_softc *sc);/* return bus cachesize in 4B word units */
@@ -1267,13 +1268,7 @@ static int ath_start_rfkill_poll(struct ath_softc *sc)rfkill_free (sc->rf_kill.rfkill);/* Deinitialize the device */ -ath_detach (sc); -if (to_pci_dev (sc->dev)->irq) -free_irq (to_pci_dev (sc->dev)->irq, sc); -pci_iounmap (to_pci_dev (sc->dev), sc->mem); -pci_release_region (to_pci_dev (sc->dev),0 ); -pci_disable_device (to_pci_dev (sc->dev)); -ieee80211_free_hw (sc->hw); +ath_cleanup (sc);return -EIO ; }else { sc->sc_flags |=SC_OP_RFKILL_REGISTERED ;
@@ -1284,6 +1279,14 @@ static int ath_start_rfkill_poll(struct ath_softc *sc) }#endif /* CONFIG_RFKILL */ +static void ath_cleanup (struct ath_softc *sc) +{ +ath_detach (sc); +free_irq (sc->irq, sc); +ath_bus_cleanup (sc); +ieee80211_free_hw (sc->hw); +} +static void ath_detach (struct ath_softc *sc) {struct ieee80211_hw *hw = sc->hw;
@@ -2534,8 +2537,18 @@ ath_rf_name(u16 rf_version)return "????" ; } +static void ath_pci_cleanup (struct ath_softc *sc) +{ +struct pci_dev *pdev =to_pci_dev (sc->dev); + +pci_iounmap (pdev, sc->mem); +pci_release_region (pdev,0 ); +pci_disable_device (pdev); +} +static struct ath_bus_ops ath_pci_bus_ops = { .read_cachesize = ath_pci_read_cachesize, + .cleanup = ath_pci_cleanup, };static int ath_pci_probe (struct pci_dev *pdev,const struct pci_device_id *id)
@@ -2642,6 +2655,8 @@ static int ath_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto bad4; } + sc->irq = pdev->irq; + ah = sc->sc_ah;printk (KERN_INFO "%s: Atheros AR%s MAC/BB Rev:%x "
@@ -2672,13 +2687,7 @@ static void ath_pci_remove(struct pci_dev *pdev)struct ieee80211_hw *hw =pci_get_drvdata (pdev);struct ath_softc *sc = hw->priv; -ath_detach (sc); -if (pdev->irq) -free_irq (pdev->irq, sc); -pci_iounmap (pdev, sc->mem); -pci_release_region (pdev,0 ); -pci_disable_device (pdev); -ieee80211_free_hw (hw); +ath_cleanup (sc); }#ifdef CONFIG_PM
Signature · 1 line
1.5.3.2
PATCH v3 5/11 ath9k: move PCI code into separate file
In reply to Gabor JuhosNow that we have converted all bus specific routines to replaceable, we can move the PCI specific codes into a separate file.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/Makefile | 1 + drivers/net/wireless/ath9k/core.h | 19 ++- drivers/net/wireless/ath9k/main.c | 300 +++-------------------------------- drivers/net/wireless/ath9k/pci.c | 287 +++++++++++++++++++++++++++++++++ 4 files changed, 328 insertions(+), 279 deletions(-) create mode 100644 drivers/net/wireless/ath9k/pci.c
drivers/net/wireless/ath9k/Makefile +1 -0
@@ -11,6 +11,7 @@ ath9k-y += hw.o \ xmit.o \ rc.o +ath9k-$(CONFIG_PCI) += pci.o ath9k-$(CONFIG_ATH9K_DEBUG) += debug.o obj-$(CONFIG_ATH9K) += ath9k.o
drivers/net/wireless/ath9k/core.h +18 -1
@@ -18,7 +18,7 @@#define CORE_H #include <linux/etherdevice.h> -#include <linux/pci.h> +#include <linux/device.h> #include <net/mac80211.h> #include <linux/leds.h> #include <linux/rfkill.h>
@@ -767,4 +767,21 @@ static inline void ath_bus_cleanup(struct ath_softc *sc) sc->bus_ops->cleanup(sc); } +extern struct ieee80211_ops ath9k_ops; + +irqreturn_t ath_isr (int irq,void *dev); +void ath_cleanup (struct ath_softc *sc); +int ath_attach (u16 devid,struct ath_softc *sc); +void ath_detach (struct ath_softc *sc); +const char *ath_mac_bb_name (u32 mac_bb_version); +const char *ath_rf_name (u16 rf_version); + +#ifdef CONFIG_PCI +int ath_pci_init (void ); +void ath_pci_exit (void ); +#else +static inline int ath_pci_init (void ) {return 0 ; }; +static inline void ath_pci_exit (void ) {}; +#endif +#endif /* CORE_H */
drivers/net/wireless/ath9k/main.c +22 -278
370 lines not on this page · raw message
drivers/net/wireless/ath9k/pci.c new file +287 -0
287 lines not on this page · raw message
Signature · 1 line
1.5.3.2
PATCH v3 6/11 ath9k: introduce platform driver for AHB bus support
In reply to Gabor JuhosThis patch adds the platform_driver itself, and modifies the main driver to register it.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/Makefile | 1 + drivers/net/wireless/ath9k/ahb.c | 160 +++++++++++++++++++++++++++++++++++ drivers/net/wireless/ath9k/core.h | 8 ++ drivers/net/wireless/ath9k/main.c | 10 ++ 4 files changed, 179 insertions(+), 0 deletions(-) create mode 100644 drivers/net/wireless/ath9k/ahb.c
drivers/net/wireless/ath9k/Makefile +1 -0
@@ -12,6 +12,7 @@ ath9k-y += hw.o \ rc.o ath9k-$(CONFIG_PCI) += pci.o +ath9k-$(CONFIG_ATHEROS_AR71XX) += ahb.o ath9k-$(CONFIG_ATH9K_DEBUG) += debug.o obj-$(CONFIG_ATH9K) += ath9k.o
drivers/net/wireless/ath9k/ahb.c new file +160 -0
160 lines not on this page · raw message
drivers/net/wireless/ath9k/core.h +8 -0
@@ -784,4 +784,12 @@ static inline int ath_pci_init(void) { return 0; };static inline void ath_pci_exit (void ) {};#endif +#ifdef CONFIG_ATHEROS_AR71XX +int ath_ahb_init (void ); +void ath_ahb_exit (void ); +#else +static inline int ath_ahb_init (void ) {return 0 ; }; +static inline void ath_ahb_exit (void ) {}; +#endif +#endif /* CORE_H */
drivers/net/wireless/ath9k/main.c +10 -0
@@ -2527,8 +2527,17 @@ static int __init ath9k_init(void) goto err_rate_unregister; } + error =ath_ahb_init (); +if (error <0 ) { + error = -ENODEV ; + goto err_pci_exit; + } +return 0 ; + err_pci_exit: +ath_pci_exit (); + err_rate_unregister:ath_rate_control_unregister (); err_out:
@@ -2538,6 +2547,7 @@ module_init(ath9k_init); static void __exit ath9k_exit(void) { +ath_ahb_exit ();ath_pci_exit ();ath_rate_control_unregister ();printk (KERN_INFO "%s: Driver unloaded\n" , dev_info);
Signature · 1 line
1.5.3.2
PATCH v3 7/11 ath9k: get EEPROM contents from platform data on AHB bus
In reply to Gabor JuhosOn the AR913x SOCs we have to provide EEPROM contents via platform_data, because accessing the flash via MMIO is not safe. Additionally different boards may store the radio calibration data at different locations.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/ahb.c | 27 ++++++++++++++++++ drivers/net/wireless/ath9k/core.h | 1 + drivers/net/wireless/ath9k/eeprom.c | 51 ++-------------------------------- drivers/net/wireless/ath9k/pci.c | 18 ++++++++++++ include/linux/ath9k_platform.h | 28 +++++++++++++++++++ 5 files changed, 77 insertions(+), 48 deletions(-) create mode 100644 include/linux/ath9k_platform.h
drivers/net/wireless/ath9k/ahb.c +27 -0
@@ -18,6 +18,7 @@#include <linux/nl80211.h> #include <linux/platform_device.h> +#include <linux/ath9k_platform.h> #include "core.h" #include "reg.h" #include "hw.h"
@@ -33,9 +34,29 @@ static void ath_ahb_cleanup(struct ath_softc *sc)iounmap (sc->mem); } +static bool ath_ahb_eeprom_read (struct ath_hal *ah,u32 off,u16 *data) +{ +struct ath_softc *sc = ah->ah_sc; +struct platform_device *pdev =to_platform_device (sc->dev); +struct ath9k_platform_data *pdata; + + pdata = (struct ath9k_platform_data *) pdev->dev.platform_data; +if (off >= (ARRAY_SIZE (pdata->eeprom_data))) { +DPRINTF (ah->ah_sc,ATH_DBG_FATAL , +"%s: flash read failed, offset %08x is out of range\n" , + __func__, off); +return false; + } + + *data = pdata->eeprom_data[off]; +return true; +} +static struct ath_bus_ops ath_ahb_bus_ops = { .read_cachesize = ath_ahb_read_cachesize, .cleanup = ath_ahb_cleanup, + + .eeprom_read = ath_ahb_eeprom_read, };static int ath_ahb_probe (struct platform_device *pdev)
@@ -48,6 +69,12 @@ static int ath_ahb_probe(struct platform_device *pdev)int ret =0 ;struct ath_hal *ah; +if (!pdev->dev.platform_data) { +dev_err (&pdev->dev,"no platform data specified\n" ); + ret = -EINVAL ; + goto err_out; + } + res =platform_get_resource (pdev,IORESOURCE_MEM ,0 );if (res ==NULL ) {dev_err (&pdev->dev,"no memory resource found\n" );
drivers/net/wireless/ath9k/core.h +1 -0
@@ -696,6 +696,7 @@ enum PROT_MODE {struct ath_bus_ops {void (*read_cachesize)(struct ath_softc *sc,int *csz);void (*cleanup)(struct ath_softc *sc); +bool (*eeprom_read)(struct ath_hal *ah,u32 off,u16 *data); };struct ath_softc {
drivers/net/wireless/ath9k/eeprom.c +3 -48
65 lines not on this page · raw message
drivers/net/wireless/ath9k/pci.c +18 -0
@@ -58,9 +58,27 @@ static void ath_pci_cleanup(struct ath_softc *sc)pci_disable_device (pdev); } +static bool ath_pci_eeprom_read (struct ath_hal *ah,u32 off,u16 *data) +{ + (void )REG_READ (ah,AR5416_EEPROM_OFFSET + (off <<AR5416_EEPROM_S )); + +if (!ath9k_hw_wait (ah, +AR_EEPROM_STATUS_DATA , +AR_EEPROM_STATUS_DATA_BUSY | +AR_EEPROM_STATUS_DATA_PROT_ACCESS ,0 )) { +return false; + } + + *data =MS (REG_READ (ah,AR_EEPROM_STATUS_DATA ), +AR_EEPROM_STATUS_DATA_VAL ); + +return true; +} +static struct ath_bus_ops ath_pci_bus_ops = { .read_cachesize = ath_pci_read_cachesize, .cleanup = ath_pci_cleanup, + .eeprom_read = ath_pci_eeprom_read, };static int ath_pci_probe (struct pci_dev *pdev,const struct pci_device_id *id)
include/linux/ath9k_platform.h new file +28 -0
@@ -0,0 +1,28 @@ +/* +* Copyright (c) 2008 Atheros Communications Inc. +* Copyright (c) 2009 Gabor Juhos <[email protected]> +* Copyright (c) 2009 Imre Kaloz <[email protected]> +* +* Permission to use, copy, modify, and/or distribute this software for any +* purpose with or without fee is hereby granted, provided that the above +* copyright notice and this permission notice appear in all copies. +* +* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES +* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF +* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR +* ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES +* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN +* ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF +* OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. +*/ + +#ifndef _LINUX_ATH9K_PLATFORM_H +#define _LINUX_ATH9K_PLATFORM_H + +#define ATH9K_PLAT_EEP_MAX_WORDS 2048 + +struct ath9k_platform_data { +u16 eeprom_data[ATH9K_PLAT_EEP_MAX_WORDS ]; +}; + +#endif /* _LINUX_ATH9K_PLATFORM_H */
Signature · 1 line
1.5.3.2
PATCH v3 8/11 ath9k: remove (u16) casts from rtc register access
In reply to Gabor JuhosThe RTC register offsets don't fit into 'u16' on the AR913x, so we have to remove the existing casts.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/hw.c | 14 +++++++------- drivers/net/wireless/ath9k/reg.h | 4 ++-- 2 files changed, 9 insertions(+), 9 deletions(-)
drivers/net/wireless/ath9k/hw.c +7 -7
@@ -1007,7 +1007,7 @@ static void ath9k_hw_init_pll(struct ath_hal *ah, pll |=SM (0xb ,AR_RTC_PLL_DIV ); } } -REG_WRITE (ah, (u16) (AR_RTC_PLL_CONTROL ), pll); +REG_WRITE (ah,AR_RTC_PLL_CONTROL , pll);udelay (RTC_PLL_SETTLE_DELAY );
@@ -1546,11 +1546,11 @@ static bool ath9k_hw_set_reset(struct ath_hal *ah, int type) rst_flags |=AR_RTC_RC_MAC_COLD ; } -REG_WRITE (ah, (u16) (AR_RTC_RC ), rst_flags); +REG_WRITE (ah,AR_RTC_RC , rst_flags);udelay (50 ); -REG_WRITE (ah, (u16) (AR_RTC_RC ),0 ); -if (!ath9k_hw_wait (ah, (u16) (AR_RTC_RC ),AR_RTC_RC_M ,0 )) { +REG_WRITE (ah,AR_RTC_RC ,0 ); +if (!ath9k_hw_wait (ah,AR_RTC_RC ,AR_RTC_RC_M ,0 )) {DPRINTF (ah->ah_sc,ATH_DBG_RESET ,"RTC stuck in MAC reset\n" );return false;
@@ -1572,8 +1572,8 @@ static bool ath9k_hw_set_reset_power_on(struct ath_hal *ah)REG_WRITE (ah,AR_RTC_FORCE_WAKE ,AR_RTC_FORCE_WAKE_EN |AR_RTC_FORCE_WAKE_ON_INT ); -REG_WRITE (ah, (u16) (AR_RTC_RESET ),0 ); -REG_WRITE (ah, (u16) (AR_RTC_RESET ),1 ); +REG_WRITE (ah,AR_RTC_RESET ,0 ); +REG_WRITE (ah,AR_RTC_RESET ,1 ); if (!ath9k_hw_wait(ah, AR_RTC_STATUS,
@@ -2615,7 +2615,7 @@ static void ath9k_set_power_sleep(struct ath_hal *ah, int setChip)if (!AR_SREV_9100 (ah))REG_WRITE (ah,AR_RC ,AR_RC_AHB |AR_RC_HOSTIF ); -REG_CLR_BIT (ah, (u16) (AR_RTC_RESET ), +REG_CLR_BIT (ah, (AR_RTC_RESET ),AR_RTC_RESET_EN ); } }
drivers/net/wireless/ath9k/reg.h +2 -2
@@ -953,7 +953,7 @@ enum {#define AR_RTC_BASE 0x00020000#define AR_RTC_RC \ - (AR_SREV_9100(ah)) ? (AR_RTC_BASE + 0x0000) : 0x7000 + ((AR_SREV_9100(ah)) ? (AR_RTC_BASE + 0x0000) : 0x7000)#define AR_RTC_RC_M 0x00000003#define AR_RTC_RC_MAC_WARM 0x00000001#define AR_RTC_RC_MAC_COLD 0x00000002
@@ -961,7 +961,7 @@ enum {#define AR_RTC_RC_WARM_RESET 0x00000008#define AR_RTC_PLL_CONTROL \ - (AR_SREV_9100(ah)) ? (AR_RTC_BASE + 0x0014) : 0x7014 + ((AR_SREV_9100(ah)) ? (AR_RTC_BASE + 0x0014) : 0x7014)#define AR_RTC_PLL_DIV 0x0000001f#define AR_RTC_PLL_DIV_S 0
Signature · 1 line
1.5.3.2
PATCH v3 9/11 ath9k: fix ar5416Addac_9100 values
In reply to Gabor JuhosWriting the register at offset 0x98c4 causes a deadlock on the AR913x SoCs. Although i don't have detailed knowledge about these registers, but if i change the register offset according to the 'ar5416Addac' table, it works. Additionally there is no reference to the 0x98c4 elsewhere.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/initvals.h | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)
drivers/net/wireless/ath9k/initvals.h +1 -1
@@ -658,7 +658,7 @@ static const u32 ar5416Addac_9100[][2] = { {0x0000989c, 0x00000000 }, {0x0000989c, 0x00000000 }, {0x0000989c, 0x00000000 }, - {0x000098c4, 0x00000000 }, + {0x000098cc, 0x00000000 }, }; static const u32 ar5416Modes[][6] = {
Signature · 1 line
1.5.3.2
PATCH v3 10/11 ath9k: fix null pointer dereference in ani monitor code
In reply to Gabor JuhosIn 'ath9k_ani_reset' the 'ahp->ah_curani' will be initialized only if 'DO_ANI(ah)' true. In 'ath9k_hw_ani_monitor' we are using 'ahp->ah_curani' unconditionally, and it will cause a NULL pointer dereference on AR9100.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/ani.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-)
drivers/net/wireless/ath9k/ani.c +3 -3
@@ -551,6 +551,9 @@ void ath9k_hw_ani_monitor(struct ath_hal *ah,struct ar5416AniState *aniState;int32_t listenTime; +if (!DO_ANI (ah)) +return ; + aniState = ahp->ah_curani; ahp->ah_stats.ast_nodestats = *stats;
@@ -610,9 +613,6 @@ void ath9k_hw_ani_monitor(struct ath_hal *ah, aniState->cckPhyErrCount = cckPhyErrCnt; } -if (!DO_ANI (ah)) -return ; -if (aniState->listenTime >5 * ahp->ah_aniPeriod) { if (aniState->ofdmPhyErrCount <= aniState->listenTime * aniState->ofdmTrigLow / 1000 &&
Signature · 1 line
1.5.3.2
PATCH v3 11/11 ath9k: enable support for AR9100
In reply to Gabor JuhosBecause we have support for the AR9100 devices now, we can enable them.
Changes-licensed-under: ISC
--- drivers/net/wireless/ath9k/hw.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)
drivers/net/wireless/ath9k/hw.c +3 -0
@@ -373,6 +373,8 @@ static const char *ath9k_hw_devname(u16 devid)return "Atheros 5418" ;case AR9160_DEVID_PCI :return "Atheros 9160" ; +case AR5416_AR9100_DEVID : +return "Atheros 9100" ;case AR9280_DEVID_PCI :case AR9280_DEVID_PCIE :return "Atheros 9280" ;
@@ -1175,6 +1177,7 @@ struct ath_hal *ath9k_hw_attach(u16 devid, struct ath_softc *sc,switch (devid) {case AR5416_DEVID_PCI :case AR5416_DEVID_PCIE : +case AR5416_AR9100_DEVID :case AR9160_DEVID_PCI :case AR9280_DEVID_PCI :case AR9280_DEVID_PCIE :
Signature · 1 line
1.5.3.2
[ath9k-devel] 2.6.29-rc3 reiser problem?
Hi,
Finally got back to testing the wireless kernel, distracted for a bit:)
make/make modules/install all work.
when I run mkinitrd I get the output below.
Not sure why the modules are not found?
Found a message to say that these messages may occur if the module is now compiled in.
In a message from Linus re rc3 http://lkml.org/lkml/2009/1/28/277 there is a note
Alexey Dobriyan (37):
fs/Kconfig: move reiserfs out
Now I do not understand exactly what that means, but my system with the rc3 kernel will not boot.
My whole system runs fine on reiserfs with 2.6.27.7-9-default . i.e. the 11N has been up for a month with no problems.
Are there changes I have to make in the config for reiserfs?
Running SUSE 11.1
Thanks,
Brian.
mkinitrd -k vmlinuz-2.6.29-rc3-wl-W -i initrd-2.6.29-rc3-wl-W
Kernel image: /boot/vmlinuz-2.6.29-rc3-wl-W Initrd image: /boot/initrd-2.6.29-rc3-wl-W Root device: /dev/sda7 (mounted on / as reiserfs) Resume device: /dev/disk/by-id/ata-IC25N060ATMR04-0_MRG30HKCJUH7WH-part5
(/dev/sda5)
FATAL: Module ata_piix not found.
WARNING: no dependencies for kernel module 'ata_piix' found.
FATAL: Module ata_generic not found.
WARNING: no dependencies for kernel module 'ata_generic' found.
FATAL: Module reiserfs not found.
WARNING: no dependencies for kernel module 'reiserfs' found.
FATAL: Module edd not found.
WARNING: no dependencies for kernel module 'edd' found.
FATAL: Module sd_mod not found.
WARNING: no dependencies for kernel module 'sd_mod' found.
FATAL: Module usbcore not found.
WARNING: no dependencies for kernel module 'usbcore' found.
FATAL: Module ohci_hcd not found.
WARNING: no dependencies for kernel module 'ohci_hcd' found.
FATAL: Module uhci_hcd not found.
WARNING: no dependencies for kernel module 'uhci-hcd' found.
FATAL: Module ehci_hcd not found.
WARNING: no dependencies for kernel module 'ehci_hcd' found. Kernel Modules: processor thermal ide-core piix ide-pci-generic fan hid usbhid
Features: block usb resume.userspace resume.kernel Bootsplash: openSUSE (1600x1200), openSUSE (800x600)
31210 blocks
Signature · 1 line
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
[ath9k-devel] 2.6.29-rc3 reiser problem?
Quoted 22 lines · Brian
Hi,
Finally got back to testing the wireless kernel, distracted for a bit:)make/make modules/install all work.
when I run mkinitrd I get the output below.
Not sure why the modules are not found?
Found a message to say that these messages may occur if the module is now compiled in.In a message from Linus re rc3 http://lkml.org/lkml/2009/1/28/277 there is a note
Alexey Dobriyan (37):
fs/Kconfig: move reiserfs outNow I do not understand exactly what that means, but my system with the rc3 kernel will not boot.
My whole system runs fine on reiserfs with 2.6.27.7-9-default . i.e. the 11N has been up for a month with no problems.Are there changes I have to make in the config for reiserfs?
Quoted 8 lines
Running SUSE 11.1
Thanks,
Brian.mkinitrd -k vmlinuz-2.6.29-rc3-wl-W -i initrd-2.6.29-rc3-wl-W
What's the big -W at the end for? Are you sure that's not a typo?
Luis
[ath9k-devel] 2.6.29-rc3 reiser problem?
Luis,
Quoted 8 lines
Quoted 6 lines
Are there changes I have to make in the config for reiserfs?
Running SUSE 11.1
mkinitrd -k vmlinuz-2.6.29-rc3-wl-W -i initrd-2.6.29-rc3-wl-W
What's the big -W at the end for? Are you sure that's not a typo?
No that is my addition to make the kernel unique.
I have been using it for months, it works.
This parm
(-W) Local version - append to kernel release
under general setup in menuconfig.
Brian
Signature · 1 line
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
[ath9k-devel] 2.6.29-rc3 reiser problem?
Quoted 9 lines · Brian
Luis,
Quoted 7 lines
Quoted 6 lines
Are there changes I have to make in the config for reiserfs?
Running SUSE 11.1
mkinitrd -k vmlinuz-2.6.29-rc3-wl-W -i initrd-2.6.29-rc3-wl-W
Quoted 8 lines
Quoted 1 line
What's the big -W at the end for? Are you sure that's not a typo?
No that is my addition to make the kernel unique.
I have been using it for months, it works.This parm
(-W) Local version - append to kernel release
under general setup in menuconfig.
Oh OK cool, do you have your modules all in place before calling mkinitrd under
/lib/modules/2.6.29-rc3-wl-W/kernel ?
Luis
[ath9k-devel] 2.6.29-rc3 reiser problem?
Luis,
Quoted 12 lines
Quoted 6 lines
No that is my addition to make the kernel unique.
I have been using it for months, it works.This parm
(-W) Local version - append to kernel release
under general setup in menuconfig.Oh OK cool, do you have your modules all in place before calling mkinitrd under
/lib/modules/2.6.29-rc3-wl-W/kernel?
Well spotted. The short answer is NO. I would estimate that over 50% od the files are missing?
In the default kernel under the FS directory I have about 45 dirs including reiser.
Under the new FS I have 8 and reiserfs is not one of them, so the boot was never goiong to work.
So I see two options.
1) You know of a possible explanation as to why the make/make install did not work.
I would be happy to understand why:)
2) I should do the compile again paying more attention to the logs etc.
I must admit this is the first time I have ever had a problem(must have compiled a 100 times).
In fact it was so reliable I left it running while I had tea:)
To be clear. What I did was to go into my original wireless testing directory and did a git pull from there.
I assumed this would bring the wireless testing directory up todate. I then did the make clean, make menuconfig, etc.
Thanks,
Brian
Signature · 1 line
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
[ath9k-devel] 2.6.29-rc3 reiser problem?
In reply to BrianQuoted 10 lines · Brian
Hi,
Finally got back to testing the wireless kernel, distracted for a bit:)make/make modules/install all work.
when I run mkinitrd I get the output below.
Not sure why the modules are not found?
Found a message to say that these messages may occur if the module is now compiled in.
Quoted 7 lines
In a message from Linus re rc3 http://lkml.org/lkml/2009/1/28/277 there is a note
Alexey Dobriyan (37):
fs/Kconfig: move reiserfs outNow I do not understand exactly what that means, but my system with the rc3 kernel will not boot.
It means that reiserfs parts of fs/Kconfig were moved to fs/reiserfs/Kconfig.
Quoted 4 lines
My whole system runs fine on reiserfs with 2.6.27.7-9-default . i.e. the 11N has been up for a month with no problems.
Are there changes I have to make in the config for reiserfs?
No configuration changes should be needed.
Quoted 17 lines
mkinitrd -k vmlinuz-2.6.29-rc3-wl-W -i initrd-2.6.29-rc3-wl-W
Kernel image: /boot/vmlinuz-2.6.29-rc3-wl-W Initrd image: /boot/initrd-2.6.29-rc3-wl-W Root device: /dev/sda7 (mounted on / as reiserfs) Resume device: /dev/disk/by-id/ata-IC25N060ATMR04-0_MRG30HKCJUH7WH-part5(
/dev/sda5)
FATAL: Moduleata_piixnot found.
WARNING: no dependencies for kernel module 'ata_piix' found.
FATAL: Moduleata_genericnot found.
WARNING: no dependencies for kernel module 'ata_generic' found.
FATAL: Module reiserfs not found.
WARNING: no dependencies for kernel module 'reiserfs' found.
FATAL: Module edd not found.
WARNING: no dependencies for kernel module 'edd' found.
FATAL: Modulesd_modnot found.
WARNING: no dependencies for kernel module 'sd_mod' found.
Quoted 8 lines
FATAL: Module usbcore not found.
WARNING: no dependencies for kernel module 'usbcore' found.
FATAL: Moduleohci_hcdnot found.
WARNING: no dependencies for kernel module 'ohci_hcd' found.
FATAL: Moduleuhci_hcdnot found.
WARNING: no dependencies for kernel module 'uhci-hcd' found.
FATAL: Moduleehci_hcdnot found.
WARNING: no dependencies for kernel module 'ehci_hcd' found.
Make sure that all the above code is compiled into the kernel. Alternatively, make them all modules.
As for the problem, you need to capture the boot error. You'll get better help in a group more relevant to that message. Maybe try a list for OpenSUSE.
Signature · 2 lines
Regards, Pavel Roskin
[ath9k-devel] 2.6.29-rc3 reiser problem?
Pavel,
Quoted 6 lines
Quoted 6 lines
In a message from Linus re rc3 http://lkml.org/lkml/2009/1/28/277 there is
a note
Alexey Dobriyan (37):
fs/Kconfig: move reiserfs out
Quoted 6 lines
Quoted 2 lines
Now I do not understand exactly what that means, but my system with the rc3 kernel will not boot.
It means that reiserfs parts of fs/Kconfig were moved to
fs/reiserfs/Kconfig.
Thanks for that, my understanding of how things hang together is going up all the time.
Quoted 6 lines
Quoted 4 lines
My whole system runs fine on reiserfs with 2.6.27.7-9-default . i.e. the 11N has been up for a month with no problems.
Are there changes I have to make in the config for reiserfs?
No configuration changes should be needed.
Great. Given the errors I thought I might have missed something.
Quoted 3 lines
Make sure that all the above code is compiled into the kernel. Alternatively, make them all modules.
As Luis pointed out they were not.
So I did a Make clean again then a make bzImage.
This time I watched the whole thing and the reiserfs and everything else appeared to compile ok.
However, when I did a make modules, reiserfs and other modules were not done.
So naturally the modules_install did not install them either.
Is there some configuration option I need to change?
Also I just did a git pull from the directory of my 11.0 system I am now on 11.1.
Do I need to scrap what I have and download the whole wireless-testing again?
Quoted 5 lines
As for the problem, you need to capture the boot error. You'll get better help in a group more relevant to that message. Maybe try a list for OpenSUSE.
Obviously given the above I was not even getting to first base.
Thanks,
Brian
Signature · 1 line
Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
[ath9k-devel] 2.6.29-rc3 reiser problem?
Quoted 11 lines · Brian
Quoted 2 lines
Make sure that all the above code is compiled into the kernel. Alternatively, make them all modules.
As Luis pointed out they were not.
So I did a Make clean again then a make bzImage.
This time I watched the whole thing and the reiserfs and everything else appeared to compile ok.
However, when I did a make modules, reiserfs and other modules were not done.
So naturally themodules_installdid not install them either.Is there some configuration option I need to change?
Remove all lines containing "REISER" from .config and run "make" again. No need to run "make bzImage" and "make modules" separately, "make" implies both. You will be asked whether to enable reiserfs. Select "m" for it to be a module. Likewise, compile other required options as modules to be sure they are enabled.
Quoted 4 lines
Also I just did a git pull from the directory of my 11.0 system I am now on 11.1.
Do I need to scrap what I have and download the whole wireless-testing again?
No. It's a purely configuration problem.
Quoted 5 lines
Quoted 4 lines
As for the problem, you need to capture the boot error. You'll get better help in a group more relevant to that message. Maybe try a list for OpenSUSE.
Obviously given the above I was not even getting to first base.
That's because you chose not to boot that kernel. Unless you had a fatal installation error, but you chose not to show output of "make install", so I don't know.
Sorry for offtopic. It's the last time I reply to this thread.
Signature · 2 lines
Regards, Pavel Roskin
v3 cover ath9k: add preliminary support for the AR913x SoCs
In reply to Gabor JuhosQuoted 6 lines · Gabor Juhos
This patch set introduces preliminary support for the Atheros' AR9130/AR9132 SoCs built-in WMAC device. Tested with compat-wireless-2009-01-12 and 2.6.27.10 kernel on the following devices:
* Atheros AP83-050 reference board (AR9132 SoC, AR9106 RF)
* TRENDnet TEW-632BRP router (AR9130 SoC, AR9102 RF)
* Mikrotik RB433 board (AR7161 SoC) + AR5416 based miniPCI card
I confirm that my AR5416 miniPCI is still working with this patch on top of yesterdays wireless-testing (master-2009-01-12).
Unfortunately, patch named "ath9k: move PCI code into separate file" doesn't apply to the today's wireless-testing and needs to be rebased.
Current checkpatch.pl gets confused by several patches because an "#endif" in the unchanged part somehow causes the preprocessor stack to underflow:
Can't use an undefined value as an ARRAY reference at scripts/checkpatch.pl line 453.
This appears to be a bug in checkpatch.pl and should not be considered a problem in the patches.
Signature · 2 lines
Regards, Pavel Roskin
v3 cover ath9k: add preliminary support for the AR913x SoCs
Pavel Roskin ?rta:
Quoted 8 lines · Gabor Juhos
Quoted 6 lines
This patch set introduces preliminary support for the Atheros' AR9130/AR9132 SoCs built-in WMAC device. Tested with compat-wireless-2009-01-12 and 2.6.27.10 kernel on the following devices:
* Atheros AP83-050 reference board (AR9132 SoC, AR9106 RF)
* TRENDnet TEW-632BRP router (AR9130 SoC, AR9102 RF)
* Mikrotik RB433 board (AR7161 SoC) + AR5416 based miniPCI card
Quoted 5 lines
I confirm that my AR5416 miniPCI is still working with this patch on top of yesterdays wireless-testing (master-2009-01-12).
Unfortunately, patch named "ath9k: move PCI code into separate file" doesn't apply to the today's wireless-testing and needs to be rebased.
Yes. I will post a new set.
Quoted 11 lines
Current checkpatch.pl gets confused by several patches because an "#endif" in the unchanged part somehow causes the preprocessor stack to underflow:
Can't use an undefined value as an ARRAY reference at
scripts/checkpatch.plline 453.This appears to be a bug in checkpatch.pl and should not be considered a problem in the patches.
I have noticed it, but i'm not a Perl expert, so I have checked the patches with checkpatch.pl from 2.6.28.
v3 cover ath9k: add preliminary support for the AR913x SoCs
Quoted 1 line · Gabor Juhos
Yes. I will post a new set.
Thanks!
Quoted 2 lines
I have noticed it, but i'm not a Perl expert, so I have checked the patches with checkpatch.pl from 2.6.28.
I've got a reply from checkpatch.pl maintainer. The fixed file is at
http://www.kernel.org/pub/linux/kernel/people/apw/checkpatch/checkpatch.pl-testing
It has no problems with the new patchset.
Signature · 2 lines
Regards, Pavel Roskin